who guidelines on the quality, safety, and efficacy of biotherapeutic products prepared by...
TRANSCRIPT
1 WHO/BS/2013.2213 2
ENGLISH ONLY 3
4
5
WHO Guidelines on the Quality, Safety, and Efficacy of 6
Biotherapeutic Products Prepared by Recombinant DNA 7
Technology 8
9
10
Proposed guidelines 11
12
NOTE: 13
14
This document has been prepared for the purpose of inviting comments and suggestions 15
on the proposals contained therein, which will then be considered by the Expert 16
Committee on Biological Standardization (ECBS). Publication of this draft is to provide 17
information about the proposed WHO Guidelines on the Quality, Safety, and Efficacy of 18
Biotherapeutic Products Prepared by Recombinant DNA Technology to a broad audience 19
and to improve transparency of the consultation process. 20
21
The text in its present form does not necessarily represent an agreed formulation of 22
the Expert Committee. Written comments proposing modifications to this text MUST 23 be received by 20 September 2013 in the Comment Form available separately and 24
should be addressed to the World Health Organization, 1211 Geneva 27, Switzerland, 25
attention: Department of Essential Medicines and Health Products (EMP). Comments may 26
also be submitted electronically to the Responsible Officer: Dr Hye-Na Kang at email: 27
29
The outcome of the deliberations of the Expert Committee will be published in the WHO 30
Technical Report Series. The final agreed formulation of the document will be edited to be 31
in conformity with the "WHO style guide" (WHO/IMD/PUB/04.1). 32
33
34
© World Health Organization 2013 35
All rights reserved. Publications of the World Health Organization can be obtained from WHO Press, World 36 Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 37 791 4857; e-mail: [email protected]). Requests for permission to reproduce or translate WHO 38 publications – whether for sale or for non-commercial distribution – should be addressed to WHO Press, at 39 the above address (fax: +41 22 791 4806; e-mail: [email protected]). 40
WHO/BS/2013.2213 Page 2
The designations employed and the presentation of the material in this publication do not imply the 1 expression of any opinion whatsoever on the part of the World Health Organization concerning the legal 2 status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers 3 or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full 4 agreement. 5 6 The mention of specific companies or of certain manufacturers’ products does not imply that they are 7 endorsed or recommended by the World Health Organization in preference to others of a similar nature that 8 are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by 9 initial capital letters. 10 11 All reasonable precautions have been taken by the World Health Organization to verify the information 12 contained in this publication. However, the published material is being distributed without warranty of any 13 kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with 14 the reader. In no event shall the World Health Organization be liable for damages arising from its use. 15
16 The named authors [or editors as appropriate] alone are responsible for the views expressed in this 17 publication. 18 19
20
21
22
23
24
Recommendations and guidelines published by WHO are intended to be scientific and
advisory in nature. Each of the following sections constitutes guidance for national
regulatory authorities (NRAs) and for manufacturers of biological products. If a NRA so
desires, these Guidelines may be adopted as definitive national requirements, or
modifications may be justified and made by the NRA. It is recommended that
modifications to these Guidelines made only on condition that modifications ensure that
the product is at least as safe and efficacious as that prepared in accordance with the
recommendations set out below.
WHO/BS/2013.2213 Page 3
Table of contents 1
2
Introduction ··································································································· 6 3
Background ···································································································· 6 4
Scope ············································································································ 9 5
Glossary ······································································································ 10 6
7
Part A. Manufacturing and quality control ·························································· 16 8
A.1 Definitions ····························································································· 16 9
A.1.1 International name and proper name ·························································· 16 10
A.1.2 Descriptive definition ··········································································· 16 11
A.1.3 International standards and reference materials ············································· 16 12
A.2 General manufacturing guidelines ·································································· 16 13
A.3 Control of starting/source materials ································································ 17 14
A.3.1 Expression vector and host cell ································································ 17 15
A.3.2 Cell bank system ················································································· 18 16
A.3.3 Cell culture medium/other materials ·························································· 21 17
A.4 Control manufacturing process ······································································ 22 18
A.4.1 Cell culture ························································································ 22 19
A.4.2 Purification ······················································································· 24 20
A.5 Control of drug substance and drug product ····················································· 26 21
A.5.1 Characterization ·················································································· 26 22
A.5.2 Routine control ··················································································· 27 23
A.6 Filling and container ················································································· 28 24
A.7 Records, retained samples, labelling, distribution and transport ······························ 29 25
A.8 Stability, storage and expiry date ·································································· 29 26
A.8.1 Stability studies ·················································································· 29 27
A.8.2 Drug product requirements ····································································· 31 28
A.9 Manufacturing process changes ···································································· 32 29
30
Part B. Nonclinical evaluation ··········································································· 34 31
B.1 Introduction···························································································· 34 32
B.1.1 Objectives of the nonclinical evaluation ······················································ 35 33
WHO/BS/2013.2213 Page 4
B.1.2 Product development and characterization ··················································· 35 1
B.1.3 Good laboratory practice ········································································ 36 2
B.2 Pharmacodynamics ··················································································· 36 3
B.2.1 Primary and secondary pharmacodynamics/biological activities ························· 36 4
B.2.2 Safety pharmacology ············································································ 37 5
B.3 Pharmacokinetics/Toxicokinetics ·································································· 37 6
B.3.1 General principles ················································································ 37 7
B.3.2 Assay ······························································································· 38 8
B.3.3 Distribution ······················································································· 38 9
B.3.4 Metabolism ······················································································· 39 10
B.4 Toxicity studies ······················································································· 39 11
B.4.1 General principles ················································································ 39 12
B.4.2 Single dose toxicity studies ····································································· 42 13
B.4.3 Repeat dose toxicity studies ···································································· 42 14
B.4.4 Genotoxicity studies ············································································· 43 15
B.4.5 Carcinogenicity studies ········································································· 44 16
B.4.6 Reproductive performance and developmental toxicity studies ··························· 46 17
B.4.7 Local tolerance studies ·········································································· 50 18
B.4.8 Other toxicity studies ············································································ 50 19
20
Part C. Clinical evaluation ··············································································· 54 21
C.1 Good clinical pracitice ··············································································· 54 22
C.2 Clinical pharmacology (Phase I) ··································································· 54 23
C.2.1 Initial safety and tolerability studeis ·························································· 54 24
C.2.2 Pharmacogenomics ·············································································· 56 25
C.2.3 Pharmacokinetics ················································································ 56 26
C.2.4 Pharmacodynamics ·············································································· 62 27
C.2.5 Pharmacokinetics/Pharmacodynamics relationship ········································· 62 28
C.2.6 Modifications of PK and PD profiles of therapeutic proteins ····························· 63 29
C.3 Efficacy ································································································· 63 30
C.3.1 Phase II ···························································································· 63 31
C.3.2 Confirmatory phase III ·········································································· 65 32
C.3.3 Biomarkers ······················································································· 67 33
WHO/BS/2013.2213 Page 5
C.3.4 Manufacturing and formulation changes ····················································· 67 1
C.3.5 Special populations ·············································································· 68 2
C.3.6 Post-marketing: Phase VI ······································································· 69 3
C.4 Statistical considerations ············································································ 69 4
C.4.1 General considerations ·········································································· 69 5
C.4.2 Special considerations for rDNA-derived biotherapeutics ································· 70 6
C.5 Safety ··································································································· 72 7
C.5.1 Special populations ·············································································· 74 8
C.6 Immunogenicity ······················································································ 75 9
C.7 Pharmacovigilance and risk managament planning ············································ 77 10
C.8 Additional guidance ·················································································· 79 11
12
Authors ········································································································ 80 13
Acknowledgements ·························································································· 84 14
References ···································································································· 85 15
16
Appendix 1. Manufacturing process validation ························································· 90 17
Appendix 2. Characterization of rDNA-derived biotherapeutics ····································· 93 18
Appendix 3. Routine control of rDNA-derived biotherapeutics ····································· 102 19
Appendix 4. Product specific guidance in nonclinical evaluation (examples) ····················· 105 20
Appendix 5. Animal species/model selection ··························································· 107 21
Appendix 6. Explanatory notes ··········································································· 111 22
Appendix 7. List of abbreviations ········································································ 114 23
24
25
26
WHO/BS/2013.2213 Page 6
Introduction 1
These guidelines are intended to provide national regulatory authorities (NRAs) and 2
manufacturers with guidance on the quality, safety and efficacy of biotherapeutic 3
products prepared by recombinant deoxyribonucleic acid (DNA) technology (rDNA-4
derived biotherapeutics) and intended for use in humans. They are based on experience 5
gained over the past 25 years or so in this technically demanding field and replace 6
“Guidelines for assuring the quality of pharmaceutical and biological products prepared 7
by recombinant DNA technology” (1). 8
9
Part A sets out updated guidelines for the manufacture and quality control of rDNA-10
derived biotherapeutics, including consideration of the effects of manufacturing changes 11
and of devices used in delivery on the product and its stability. Part B is new and 12
provides guidelines on nonclinical evaluation: Part C, also new, provides guidance on 13
clinical evaluation. The nature and extent of characterization and testing (Part A) 14
required for a product undergoing nonclinical and clinical studies will vary depending on 15
the nature of the product and its stage of development . Part A may also apply to vaccines 16
prepared by rDNA technology. However, neither Part B nor C applies to the vaccines. 17
Detail guidance on nonclinical and clinical evaluation of vaccines (2, 3) as well as other 18
product specific WHO recommendations and guidelines related to vaccines are available 19
elsewhere (http://www.who.int/biologicals/vaccines/en/). 20
21
Background 22
Developments in molecular genetics and nucleic acid chemistry have enabled genes 23
encoding natural biologically active proteins to be identified, modified and transferred 24
from one organism to another so as to obtain highly efficient synthesis of their products. 25
This has led to the production of new rDNA-derived biological medicines using a range 26
of different expression systems such as bacteria, yeast, transformed cell lines of 27
mammalian origin, insect and plant cells, as well as transgenic animals. rDNA technology 28
is also used to produce non-native biologically active proteins such as chimeric, 29
WHO/BS/2013.2213 Page 7
humanized or fully human monoclonal antibodies, or antibody-related proteins or other 1
engineered biological medicines such as fusion proteins. 2
3
There has also been great progress in the ability to purify biologically active 4
macromolecules. In addition, analytical technologies have improved tremendously since 5
the early days of biotechnology, allowing the detailed characterization of many biological 6
macromolecules including its protein, lipid and oligosaccharide components. 7
8
Together these technologies have enabled the production of large quantities of medicinal 9
products that are difficult to prepare from natural sources or were previously unavailable. 10
Nevertheless, it is still not possible to fully predict biological properties and clinical 11
performance of these macromolecules from physicochemical characteristics alone. In 12
addition, the production processes are biological systems which are known to be 13
inherently variable, a feature which has important consequences for the safety and 14
efficacy of the resulting product. A pre-requisite, therefore, for introducing such 15
biologicals into the clinic is to ensure consistency of quality from lot to lot and for this 16
purpose robust manufacturing processes are developed based on process understanding 17
and characterization, including appropriate in-process controls. Process understanding 18
and consistency is critical since slight changes can occasionally lead to major adverse 19
effects, such as immunogenicity, with serious safety implications associated with 20
immunogenicity. 21
22
As with many other new technologies, a new set of safety issues for consideration both 23
by industry and NRAs has been generated by these particular biotechnologies. Potential 24
safety concerns arose from the novel processes used in manufacture, from product and 25
process related impurities and from the complex structural and biological properties of 26
the products themselves. Factors that have received particular attention include the 27
possible presence of contaminating oncogenic host cell DNA in products derived from 28
transformed mammalian cells (4), and the presence of adventitious viruses (4). Since the 29
nature and production of these products are highly sophisticated, they require similar 30
sophisticated laboratory techniques to ensure their proper standardization and control. 31
WHO/BS/2013.2213 Page 8
Although comprehensive characterization of the drug product is expected, considerable 1
emphasis must also be given to process validation and in-process control. Adequate 2
control measures relating to the starting materials and manufacturing process are, 3
therefore, as important as analysis of the drug product. Thus data on the host cell quality, 4
purity, freedom from adventitious agents, adequate in-process testing during production, 5
and effectiveness of test methods are required for licensing. 6
7
At a very early stage in the development of rDNA-derived medicines, the European 8
Medicines Agency and the US Food and Drug Administration produced guidelines and 9
points to consider, respectively, for the development and evaluation of these new 10
products (5, 6). Such guidelines, based as they were on long experience with traditional 11
biologicals, set the scene for regulatory expectations both for clinical trials and for 12
licensing. At the global level, the WHO produced a series of guidance documents on the 13
quality, safety and efficacy of rDNA-derived products, including specific guidance for 14
products such as interferons and monoclonal antibodies (1, 7-9). These regulatory 15
concepts have been instrumental in establishing the quality, safety and efficacy of rDNA-16
derived biotherapeutics which now play a major role in today’s medical practice. 17
18
As patents and data protection measures on biotechnology products have expired, or 19
neared expiration, considerable attention has turned to producing copies of the innovator 20
products with the view to making more affordable products which may improve global 21
access to these medicines. Since by definition it is not possible to produce identical 22
biologicals, the normal method of licensing generic medicines, which relies primarily on 23
bioequivalence data, is not appropriate for licensing such products and the term similar 24
biological product, or biosimilar product, came into existence (10, 11). The concept of 25
similar biological medicinal products was introduced first by the European Medicines 26
Agency (10) and subsequently by other national regulatory authorities (although the 27
actual term used has varied slightly from agency to agency). WHO guidelines on the 28
evaluation of similar biotherapeutic products were produced in 2010 (11), and provided a 29
set of globally acceptable principles regarding the regulatory evaluation of biosimilars, 30
although it was recognized that they will not by themselves resolve all issues. During 31
WHO/BS/2013.2213 Page 9
international consultations on the development of the biosimilar guidelines and also their 1
implementation, it became clear that there was a need to update WHO guidance on the 2
quality, safety and efficacy of rDNA-derived medicines and biotechnology products in 3
general (12). In 2010, the International Conference of Drug Regulatory Authorities noted 4
that WHO should supplement its guidance on the evaluation of similar biotherapeutic 5
products by providing up-to-date guidelines for the evaluation of biotherapeutic products 6
in general. 7
8
The present guidelines have been developed through international consultation and are 9
intended as a replacement of those in Annex 3, TRS No 814, 1991. They are considered 10
to be a replacement and not a revision of those guidelines because they contain new 11
sections on nonclinical and clinical evaluation of rDNA-derived biotherapeutics which 12
were lacking in the original document. In addition, a section on issues related to 13
manufacturing changes both during development and once the product is on the market 14
has also been introduced since considerable improvements to the production process and 15
to the product itself can take place during the later stages of development and post 16
licensure, especially in the immediate post licensing years. These changes can 17
unintentionally impact the clinical performance of the product and need to be handled 18
carefully from a regulatory perspective. 19
20
Scope 21
These guidelines apply, in principle, to all biologically active protein products used in the 22
treatment of human diseases and which are prepared by recombinant DNA technology. 23
They also apply to protein products used in diagnosis (e.g. for monoclonal antibody 24
products including in vivo diagnosis and ex vivo treatment, but excluding in vitro 25
diagnosis) and those intentionally modified by for example pegylation or modification or 26
rDNA sequences. They set out regulatory expectations both for clinical trials and for 27
licensing, as well as for changes in products already on the market. However, the level of 28
data submitted for a product for clinical trials will have to take into account the nature of 29
the product and its stage of development. 30
31
WHO/BS/2013.2213 Page 10
Special considerations for biosimilar products are available in the WHO guidelines on 1
evaluation of similar biotherapeutic products, adopted by the WHO Expert Committee 2
on Biological Standardization in 2009 (11). 3
4
Although the principles outlined in this document (e.g. in Part A) apply to vaccines made 5
by rDNA technology, there are more detailed guidelines/recommendations on vaccine 6
evaluation in terms of quality, safety, and efficacy (2, 3). For example, vaccines such as 7
yeast derived hepatitis B vaccine or malaria vaccine produced by rDNA technology (13, 8
14) are available in the WHO Technical Report Series 9
(http://www.who.int/biologicals/vaccines/en/). 10
11
The present guidelines are not intended to apply to genetically modified live organisms 12
designed to be used directly in humans, such as recombinant viral vectors (15) or live 13
attenuated vaccines, nor to gene transfer products. A WHO guideline on DNA vaccines 14
for therapeutic as well as for prophylactic use, adopted by the WHO Expert Committee 15
on Biological Standardization in 2005, are available (16). Products produced in 16
transgenic animals are also excluded. 17
18
Glossary (alphabetical order) 19
The definitions given below apply to the terms used in this document. They may have 20
different meaning in other contexts. 21
22
Acceptance criteria 23
Numerical limits, ranges, or other suitable measures for acceptance of the results of 24
analytical procedures which the drug substance or drug product or materials at other 25
stages of their manufacture should meet. 26
27
Biomarkers 28
A biomarker is defined as a laboratory measurement that reflects the activity of a disease 29
process, correlates (either directly or inversely) with disease progression, and may also be 30
WHO/BS/2013.2213 Page 11
an indicator of a therapeutic response. A genomic biomarker is a measurable DNA and/or 1
RNA marker that measures the expression, function or regulation of a gene. 2
3
Biotherapeutic 4
A biological medicinal product with the indication of treating human diseases. 5
6
Comparability exercise 7
The activities, including study design, conduct of studies, and evaluation of data, that are 8
designed to investigate whether the products are comparable. 9
10
Critical quality attribute 11
A physical, chemical, biological or microbiological property or characteristic that is 12
selected for its ability to help indicate the consistent quality of the product within an 13
appropriate limit, range, or distribution to ensure the desired product quality. 14
15
Drug product 16
A pharmaceutical product type in a defined container closure system that contains a drug 17
substance, generally in association with excipients. 18
19
Drug substance 20
The active pharmaceutical ingredient and associated molecules that may be subsequently 21
formulated, with excipients, to produce the drug product. It may be composed of the 22
desired product, products-related substances, and product- and process-related impurities. 23
It may also contain other component such as buffers. 24
25
Good clinical practice (GCP) 26
An international ethical and scientific quality standard for designing, conducting, 27
recording and reporting trials that involve the participation of human subjects. 28
Compliance with this standard provides public assurance that the rights, safety and well-29
being of trial subjects are protected, consistent with the principles that have their origin in 30
the Declaration of Helsinki, and that the clinical trial data are credible. 31
WHO/BS/2013.2213 Page 12
1
Good laboratory practice (GLP) 2
A quality system concerned with the organizational process and conditions under which 3
nonclinical health and environmental safety studies are planned, performed, monitored, 4
recorded, archived and reported. 5
6
Good manufacturing practice (GMP) 7
That part of the pharmaceutical quality assurance process which ensures that products are 8
consistently produced and to meet to the quality standards appropriate to their intended 9
use and as required by the marketing authorization. In these guidelines, GMP refers to the 10
current GMP guidelines published by WHO. 11
12
Immunogenicity 13
The ability of a substance to trigger an immune response or reaction (e.g. development of 14
specific antibodies, T cell response, allergic or anaphylactic reaction). 15
16
Impurity 17
Any component present in the drug substance or drug product that is not the desired 18
product, a product-related substance, or excipient including buffer components. It may be 19
either process- or product-related. 20
21
in-silico modeling 22
A computer-simulated model. 23
24
in-process control 25
Checks performed during production in order to monitor and, if necessary, to adjust the 26
process to ensure that the intermediate or product conforms to its specifications. The 27
control of the environment or equipment may also be regarded as a part of in-process 28
control. 29
30
Master cell bank (MCB) 31
WHO/BS/2013.2213 Page 13
A quantity of well-characterized cells of animal or other origin, derived from a cell seed 1
at a specific population doubling level (PDL) or passage level, dispensed into multiple 2
containers, cryopreserved, and stored frozen under defined conditions, such as the vapour 3
or liquid phase of liquid nitrogen in aliquots of uniform composition. The master cell 4
bank is prepared from a single homogeneously mixed pool of cells. 5
6
Non-human primates (NHPs) 7
Primates used as models for the study of the effects of drugs in humans, prior to clinical 8
studies. 9
10
P450 (CYP) enzymes 11
Indicates the family of metabolising enzymes which is the most common group. 12
13
Pharmacodynamics (PD) 14
The study of the biochemical and physiological effects of drugs on the body and the 15
mechanisms of drug action and the relationship between drug concentration and effect. 16
One dominant example is drug-receptor interactions. PD is often summarized as the study 17
of what a drug does to the body, as opposed to pharmacokinetics which is the study of 18
what the body does to a drug. 19
20
Pharmacogenomics 21
The study of the pharmacologic correlation between drug response and variations in 22
genetic elements has become of increasing importance for drug development. Such 23
variations can have effects on the risk of developing adverse drug reactions as well as on 24
the response to treatment; variations in drug pharmacokinetics and metabolic pathways 25
can cause higher drug concentrations in some patients resulting in increased drug toxicity, 26
and/or lower drug concentrations in some patients resulting in decreased drug effects. 27
28
Pharmacokinetics (PK) 29
WHO/BS/2013.2213 Page 14
Pharmacokinetics is the study and characterization of the time course of drug absorption, 1
distribution, metabolism, and elimination (ADME). Pharmacokinetics is a quantitative 2
analysis of how living systems handle foreign compounds. 3
4
Pharmacovigilance (PhV) 5
The activities that are carried out after a medicinal product is marketed to observe and 6
manage in a continuous manner the safety and the efficacy of the products. 7
8
QT/QTc 9
QT interval is a measure of the time between the start of the Q wave and the end of the T 10
wave in the heart's electrical cycle on the electrocardiogram. It measures the conduction 11
speed between the atria and the ventricles. There is a genetic predisposition to the 12
prolongation of the QT interval which can be triggered by several factors, including 13
various medicinal products by themselves or due to their metabolic interaction. It is 14
critical to understand whether a particular drug or a biological trigger the prolongation, as 15
any prolongation of the QT interval outside of the normal limits determined for 16
electrocardiograms indicates potential for arrhythmia (disturbed heart rhythm) which is a 17
serious adverse event during drug therapy. In extreme cases, this can lead to sudden death. 18
Since the QT interval is affected by the heart rate, “corrected” QT-QTc should also be 19
used. 20
21
rDNA-derived biotherapeutics 22
Biotherapeutics prepared by recombinant DNA technology. All biologically active 23
protein products used in the treatment of human diseases and which are prepared by 24
rDNA technology. These include recombinant protein biotherapeutics, recombinant blood 25
products, recombinant monoclonal antibodies and recombinant enzymes. 26
27
Recombinant DNA technology 28
Technology joining together (recombine) DNA segments from two or more different 29
DNA molecules that are inserted into a host organism to produce new genetic 30
combinations. Also referred to as gene manipulation or genetic engineering, as the 31
WHO/BS/2013.2213 Page 15
original gene is artificially altered and changed. These new gene(s) when inserted into the 1
expression system form the basis for the production of rDNA-derived protein(s). 2
3
Risk management plan (RMP) 4
The activities that will, in a continuous manner ensure that patients continue to be safe 5
and benefit from a medicinal ingredient. These plans include PhV plans amongst many 6
other elements. 7
8
Source material/starting material 9
Any substance of a defined quality used in the production of a biological medicinal 10
product, but excluding packaging materials. 11
12
Specification 13
A list of tests, references to analytical procedures, and appropriate acceptance criteria 14
which are numerical limits, ranges, or other criteria for the tests described. Specifications 15
are critical quality standards that are proposed and justified by the manufacturer and 16
approved by regulatory authorities. 17
18
Working cell bank (WCB) 19
A quantity of well-characterized cells of animal or other origin, derived from the master 20
cell bank at a specific PDL or passage level, dispensed into multiple containers, 21
cryopreserved, and stored frozen under defined conditions, such as in the vapour or liquid 22
phase of liquid nitrogen in aliquots of uniform composition. The working cell bank is 23
prepared from a single homogeneously mixed pool of cells. One or more of the WCB 24
containers is used for each production culture. 25
26
WHO/BS/2013.2213 Page 16
Part A. Manufacturing and quality control 1
2
A.1 Definitions 3
A.1.1 International name and proper name 4
Where an International Non-Proprietary Name (INN) for a rDNA-derived biotherapeutic 5
is available, it should be used (17). The proper name should be the equivalent of the INN 6
in the language of the country of origin. 7
8
A.1.2 Descriptive definition 9
The description of a rDNA-derived biotherapeutic should indicate the biological system 10
in which it is produced (e.g. bacterial, fungal or mammalian cells) as well as the 11
presentation of the drug product. 12
13
A.1.3 International standards and reference materials 14
International standards and reference preparations have been established for a wide range 15
of biologicals prepared by rDNA technology. They are used to calibrate assays either 16
directly or for calibration of secondary standards or manufacturers working standards. A 17
list of such materials is available on WHO website 18
(http://www.who.int/bloodproducts/catalogue/AlphFeb2013.pdf). Each 19
standard/reference preparation is held by one of the WHO custodian laboratories, e.g. the 20
National Institute for Biological Standards and Control, Potters Bar, United Kingdom. 21
22
A.2 General manufacturing guidelines 23
The present Guidelines cover the following three main areas: 24
1) Control of starting/source materials, including data both on the host cell and on 25
the source, nature and sequence of the gene used in production. 26
2) Control of the manufacturing process. 27
3) Control of the drug substance and the drug product. 28
29
WHO/BS/2013.2213 Page 17
In this respect, rDNA-derived products are considered to be like biologicals produced by 1
traditional methods, such as bacterial and viral vaccines, where the quality, safety and 2
efficacy of the product relies heavily on adequate control of the starting/source materials 3
and on the manufacturing process, in addition to control tests on the drug substance and 4
drug products themselves. These guidelines therefore place considerable emphasis on the 5
characterization and testing of host cell lines and other materials used during 6
manufacturing and in validating the ability of the purification processes to remove or 7
inactivate unwanted materials, especially possible viral contaminants and process related 8
impurities such as proteins and DNA. They also cover in-process controls in 9
manufacturing and comprehensive characterization of the drug substance and the drug 10
product. 11
12
Information should therefore be provided to adequately describe the starting/source 13
materials, manufacturing process and in-process controls. The description of the 14
manufacturing process should be provided in the form of a flow diagram and sequential 15
procedural narrative and the in-process controls for each step or stage of the process 16
should be indicated in this description. Also, an explanation should be provided of how 17
batches of the drug substance and drug product are defined (e.g. splitting and pooling of 18
harvests or intermediates). Details of batch size or scale and batch numbering should also 19
be included. 20
21
The general recommendations for manufacturing establishments contained in the WHO 22
Good manufacturing practices: main principles for pharmaceuticals preparations (18) 23
and the Good manufacturing practices for biological products (19) as well as those in the 24
WHO Recommendations for the evaluation of animal cell cultures as substrates for the 25
manufacture of biological medicinal products and for the characterization of cell banks 26
(4) should apply to establishments manufacturing rDNA-derived biotherapeutics. 27
28
A.3 Control of starting/source materials 29
A 3.1 Expression vector and host cell 30
WHO/BS/2013.2213 Page 18
A description of the host cell, its source and history, and of the expression vector used in 1
production, including source and history, should be given. This should include details of 2
the origin and identity of the gene being cloned as well as the construction, genetic 3
elements contained and structure of the expression vector. An explanation of the source 4
and function of the component parts of the vector, such as the origins of replication, 5
promoters, or antibiotic markers, should be provided as well as a restriction-enzyme map 6
indicating at least those sites used in construction. 7
8
Methods used to amplify the expression constructs, transform expression constructs into 9
host cells, and rationale used to select the cell clone for production should be fully 10
described. The vector within the cell, whether integrated or extrachromosomal, and copy 11
number, should be analyzed. A host cell containing an expression vector should be cloned 12
and used to establish a master cell bank (MCB) and the correct identity of the vector 13
construct in the cell bank should be established. The genetic stability of the host-vector 14
combination should be documented (see below). 15
16
The nucleotide sequence of the cloned gene insert and of the flanking control regions of 17
the expression vector should be indicated and all relevant expressed sequences clearly 18
delineated. 19
20
Measures to promote and control the expression of the cloned gene in the host cell during 21
production should be described in detail. 22
23
A.3.2 Cell bank system 24
Typically, rDNA-derived biotherapeutics are produced using a cell bank system which 25
involves a manufacturer’s working cell bank (WCB) derived from a MCB. It is 26
acknowledged that a WCB may not always be established in early phases of 27
development. 28
29
The type of banking system used, the size of the cell bank(s), it’s life expectancy, the 30
container (vials, ampoules, or other appropriate vessels) and closure system used, the 31
WHO/BS/2013.2213 Page 19
methods used for preparation of the cell bank(s) including the cryoprotectants and media 1
used, and the conditions employed for cryopreservation or long term storage conditions 2
should all be documented and described in detail. 3
4
Evidence for banked cell stability under defined storage conditions should be provided. 5
Such evidence can be generated during production of material from the banked cells and 6
supported by a programme for monitoring stability indicating attributes over time (e.g. 7
data on cell viability upon thawing, stability of the host-vector expression system in the 8
cell bank). Available data should be clearly documented and the proposed stability 9
monitoring programme described in the marketing application. Evidence for the stability 10
of the host-vector expression in the cell bank under storage as well as under recovery 11
conditions should be provided. 12
13
For animal cells and animal derived cell banks, reference should be made to the WHO 14
Recommendations for the evaluation of animal cell cultures as substrates for the 15
manufacture of biological medicinal products and for the characterization of cell banks 16
(4). 17
18
A.3.2.1 Control of cell banks 19
The characterization and testing of banked cell substrates is a critical component of the 20
control of rDNA-derived biotherapeutics. Cell banks should be tested to confirm the 21
identity, purity, and suitability of the cell substrate for the intended manufacturing use. 22
The MCB should be characterized for relevant phenotypic and genotypic markers which 23
should include the expression of the recombinant protein and/or presence of the 24
expression construct. The testing program chosen for a given cell substrate will vary 25
according to the nature and biological properties of the cells (e.g. growth requirements) 26
and its cultivation history (including use of human-derived or animal-derived biological 27
reagents). The extent of characterization of a cell substrate may influence the type or 28
level of routine testing needed at later stages of manufacturing. Molecular methods 29
should be used to analyse the expression construct for copy number, for insertions or 30
deletions, and for the number of integration sites. Requirements for bacterial systems 31
WHO/BS/2013.2213 Page 20
expressing the protein from a plasmid or mammalian epigenetic expression should be 1
distinguished from mammalian cell systems. The nucleic acid sequence should be shown 2
to be identical to that determined for the expression construct and should correspond to 3
that expected for the protein sequence. 4
5
Animal cell substrates are subject to contamination and have the capacity to propagate 6
extraneous, adventitious organisms, such as mycoplasma and viruses. In addition, animal 7
cells contain endogenous agents such as retroviruses that may raise safety concerns. 8
Testing of cell substrates for both endogenous (e.g. retroviruses) and adventitious agents 9
is critical. A strategy for testing cell banks for adventitious agents should be developed. 10
This strategy should also involve an assessment of specific viruses and the families of 11
viruses that may potentially contaminated the cell substrate. Such testing is described in 12
detail in the WHO Recommendations for the evaluation of animal cell cultures as 13
substrates for the manufacture of biological medicinal products and for the 14
characterization of cell banks (4) and the International Conference on Harmonization 15
(ICH) guidelines Q5A viral safety evaluation of biotechnology products derived from cell 16
lines of human or animal origin (20). 17
18
In general, cell substrates contaminated with microbial agents are not suitable for 19
production. However, there are exceptions. For example, some murine cell lines that are 20
widely used for the production of rDNA-derived biotherapeutics express endogenous 21
retroviral particles. In such circumstances, risk mitigating strategies should be 22
implemented. These include removal of such agents and/or their inactivation by physical, 23
enzymatic and/or chemical treatment during processing of the rDNA-derived 24
biotherapeutics. 25
26
In addition, tests of purity and limited tests of identity should be performed once on each 27
WCB. For the WCB, a specification including test methods and acceptance criteria 28
should be established. A protocol for establishing future WCB should be provided. Each 29
new WCB should comply with the established WCB specification. 30
31
WHO/BS/2013.2213 Page 21
A.3.2.2 Cell substrate genetic stability 1
The limit of in vitro cell age for production should be defined by the time of registration, 2
and based on data derived from production cells expanded under pilot plant scale or 3
commercial scale conditions to the proposed limit of in vitro cell age for production use 4
or beyond. Generally, the production cells are obtained by expansion of cells from the 5
WCB. 6
7
Specific traits of cells, which may include, for example, morphological characteristics, 8
growth characteristics, biochemical markers, immunological markers, productivity of the 9
desired product, or other relevant genotypic or phenotypic markers may be useful for the 10
assessment of cell substrate stability during culture phase. The nucleotide sequence of the 11
insert encoding the rDNA-derived biotherapeutic should be determined at least once after 12
a full-scale culture for each MCB. 13
14
In some cases, multiple harvests from long fermentations could lead to a drift in some 15
quality attributes such as glycosylation, with the appearance of "new" variants with 16
possible impact on quality, safety and efficacy of the product. The management of such 17
drift should be appropriately addressed in process evaluation/validation studies. The 18
molecular integrity of the gene being expressed and the phenotypic and genotypic 19
characteristics of the host cell after long-term cultivation should be established and 20
defined by the time of registration. 21
22
A.3.3 Cell culture medium/other materials 23
Materials used in the manufacture of the drug substance (e.g. solvents, reagents, 24
enzymes) should be listed identifying where each material is used in the process. 25
Information on the source, quality and control of these materials should be provided. 26
Information demonstrating that materials (including biologically-sourced materials, e.g. 27
media components, monoclonal antibodies, enzymes) meet standards appropriate for their 28
intended use (including the clearance or control of adventitious agents) and for the cases 29
when suppliers of materials change should be provided, as appropriate. 30
31
WHO/BS/2013.2213 Page 22
Media and other components should comply with current WHO Guidelines on 1
transmissible spongiform encephalopathies in relation to biological and pharmaceutical 2
products (21). The latest version of the WHO Guidelines on tissue infectivity distribution 3
in transmissible spongiform encephalopathies (22) should also be consulted. These tables 4
are periodically updated as new data becomes available (e.g. 23). 5
6
A.4 Control of the manufacturing process 7
Adequate design of a process and knowledge of its capability are part of the strategy used 8
to develop a manufacturing process which is controlled and reproducible, yielding a drug 9
substance and drug product that consistently meet specifications. In this respect, limits 10
are justified based on information gained from the entire process from early development 11
through commercial scale production. 12
In-process controls are performed at critical decision making steps and at other steps 13
where data serve to ensure the appropriate performance of the manufacturing process, and 14
to demonstrate adequate quality during the production of both the drug substance and the 15
drug product. Those process parameters that are found to impact the quality attributes of 16
the drug substance or drug product should be controlled by suitable acceptance limits. 17
Where appropriate, in-process controls may alleviate the need for routine testing of some 18
quality attribute(s) at the level of the drug substance and/or drug product. 19
20
A.4.1 Cell culture 21
A 4.1.1 Production at finite passage 22
Procedures and materials used both for cell growth and for the induction of the product 23
should be described in detail. For each production run, data on the extent and nature of 24
any microbial contamination of culture vessels should be provided. Acceptable limits for 25
potential contamination should be set and the sensitivity of the methods used to detect it 26
indicated. Microbial and fungal contamination should be monitored according to Part A 27
Section 5.2 of General requirements for the sterility of biological substances (24) or by 28
methods approved by the NRA. 29
30
WHO/BS/2013.2213 Page 23
Data on the consistency of culture conditions and culture growth and on the maintenance 1
of product yield should be presented. Criteria for the rejection of culture lots should be 2
established. The maximum number of cell doublings or passage levels to be permitted 3
during production should be specified taking into account the limit of in-vitro cell age. 4
For a process demonstrating consistent growth characteristics over the proposed cell age 5
range for production, it may also be acceptable to define the cell age limit based on the 6
maximum permitted days in culture from thaw to end of production. 7
8
Host-cell / vector characteristics at the end of production cycles should be monitored to 9
establish consistency, for which purpose information on plasmid copy number or degree 10
of retention of the expression vector within the host cell may be of value, as may 11
restriction enzyme mapping of the vector containing the gene insert. If the vector is 12
present in multiple copies integrated into the host cell genome, confirming the rDNA 13
sequence directly may be difficult. In such cases, alternative approaches to confirming 14
the sequence of insert encoding the rDNA-derived biotherapeutics should be considered 15
and defined by the time of registration (e.g. restriction fragment length polymorphism 16
(RFLP), fluorescence in situ hybridization (FISH), polymerase chain reaction-single-17
strand conformation polymorphism (PCR-SSCP), Southern Blot). For example, 18
confirmation of protein sequence by peptide mapping is an appropriate alternative to 19
rDNA sequencing. 20
21
A.4.1.2 Continuous culture production 22
As recommended above, all procedures and materials used for cell culture and induction 23
of the product should be described in detail. In addition, particular consideration should 24
be given to the procedures used in production control. Monitoring is necessary 25
throughout the life of the culture, although the frequency and type of monitoring required 26
depend on the nature of both the production system and product. 27
28
Evidence should be produced to show that variations in yield or other culture parameters 29
do not exceed specified limits. The acceptance of harvests for further processing should 30
be clearly linked to the monitoring schedule in use, and a clear definition of “batch” of 31
WHO/BS/2013.2213 Page 24
product for further processing should be established. Criteria for the rejection of harvests 1
or termination of the culture should also be established. Tests for microbial contamination 2
should be performed as appropriate to the harvesting strategy. 3
4
The maximum period of continuous culture should be specified, based on information on 5
the stability of the system and consistency of the product during and after this period. In 6
long-term continuous culture, the cell line and product should be fully re-evaluated at 7
intervals determined by information on the stability of the host-vector system and the 8
characteristics of the product. 9
10
A.4.2 Purification 11
The methods used for harvesting, extraction and purification of the product and related 12
in-process controls, including their acceptance criteria, should be described in detail. 13
Special attention should be given to the elimination of viruses, nucleic acid, host cell 14
proteins and impurities considered to pose an immunogenicity risk. 15
16
The ability of the purification procedure to remove unwanted product-related or process-17
related impurities (e.g. host-cell derived proteins, nucleic acid, carbohydrates, viruses and 18
other impurities, including media derived compounds and undesirable chemicals 19
introduced by the purification process itself) should be investigated thoroughly, as should 20
the reproducibility of the process. Particular attention should be given to demonstrating 21
the removal and/or inactivation of possible contaminating viruses and residual DNA from 22
products manufactured using continuous cell lines. 23
24
A.4.2.1 Residual cellular DNA from continuous cell lines (rcDNA) 25
The ability of the manufacturing process to reduce the amount of rcDNA to an acceptable 26
level, to reduce the size of the rcDNA or to chemically inactivate the biological activity 27
of this DNA should be demonstrated. 28
29
Acceptable limits on the amount of rcDNA as well as points to be considered concerning 30
the size of rcDNA in a rDNA- derived biotherapeutic are discussed in WHO 31
WHO/BS/2013.2213 Page 25
recommendations for the evaluation of animal cell substrates for the manufacture of 1
biological medicinal products and for the characterization of cell banks (4). These should 2
be set taking into consideration the characteristics of the cell substrate, the intended use 3
and route of administration of the rDNA-derived biotherapeutics and, most importantly, 4
the effect of the manufacturing process on the size, quantity and biological activity of the 5
residual host cell DNA fragments. In general it has been possible to reduce rcDNA levels 6
in rDNA-derived biotherapeutics to <10 ng per dose. 7
8
A.4.2.2 Virus clearance 9
For cell substrates of human or animal origin, virus clearance or inactivation processes, 10
individually and overall, should be shown to be able to adequately remove/inactivate any 11
contaminating viruses and to ensure viral safety in the drug substance. 12
13
Where appropriate, validation studies (see Appendix 1) should be undertaken using small 14
scale studies with carefully selected model viruses to evaluate the virus 15
clearance/inactivation capability of selected process steps and overall, aiming at a 16
significant safety margins. The results will indicate the extent to which these 17
contaminants can theoretically be inactivated and removed during purification. 18
19
The overall manufacturing process, including the testing and selection of the cells and 20
source materials, as well as the validation of the ability of the purification process to 21
adequately remove possible contaminants, should ensure the absence of infectious agents 22
in the drug product. Nevertheless, to complement such approaches, testing of the product 23
itself at appropriate steps in the production process for the absence of contaminating 24
infectious viruses is also recommended. A sample of the unprocessed bulk following 25
fermentation constitutes one of the most suitable levels at which the possibility of 26
detecting adventitious virus contamination can be determined with a high probability of 27
detection. A programme of ongoing assessment of adventitious viruses in production 28
batches should be undertaken. The scope, extent and frequency of virus testing on the 29
unprocessed bulk should take into account the nature of the cell lines used, the results and 30
extent of virus testing performed during the qualification of the MCB and WCB, the 31
WHO/BS/2013.2213 Page 26
cultivation method, the source materials used and the results of virus clearance studies. In 1
vitro screening tests using one or more cell lines are generally used to test unprocessed 2
bulk. If appropriate, a PCR test or other suitable methods may be used. 3
4
If contamination by adventitious viruses is detected in the unprocessed bulk, the 5
manufacturing process should be carefully checked to determine the cause of the 6
contamination and to decide on the appropriate action to take. 7
8
Further considerations of the detection, elimination and inactivation of viruses in animal 9
cell substrates used in the production of rDNA-derived biotherapeutics, as well as the 10
problem of rcDNA, can be found in the WHO Recommendations for the evaluation of 11
animal cell cultures as substrates for the manufacture of biological medicinal products 12
and for the characterization of cell banks (4) as well as in the ICH guidelines Q5A viral 13
safety evaluation of biotechnology products derived from cell lines of human or animal 14
origin (20). 15
16
A.5 Control of drug substance and drug product 17
A.5.1 Characterization 18
Rigorous characterization of the rDNA-derived biotherapeutics by chemical, 19
physicochemical and biological methods is essential. Characterization is typically 20
performed in the development phase to determine the physicochemical properties, 21
biological activity, immunochemical properties, purity and impurities of the product, and 22
following significant process changes and/or for periodic monitoring to confirm the 23
quality of the product. Characterization allows relevant specifications to be established. 24
25
Particular attention should be given to using a wide range of analytical techniques 26
exploiting different physiochemical properties of the molecule (size, charge, isoelectric 27
point, amino acid composition, hydrophobicity). Post-translational modifications, such as 28
glycosylation should be identified and adequately characterized. It may also be necessary 29
to include suitable tests to establish that the product has the desired conformation, state of 30
aggregation and/or degradation, as well as higher order structure. The rationale for 31
WHO/BS/2013.2213 Page 27
selection of the methods used for characterization should be provided and their suitability 1
should be justified bearing in mind that the characterization of the product is intended to 2
identify attributes that may be important to the overall safety and efficacy of the product. 3
Details of the expected characterization of a rDNA-derived biotherapeutic and techniques 4
suitable for such purposes are set out in Appendix 2. The specific technical approach 5
employed will vary from product to product and alternative approaches, other than those 6
included in the appendix, will be appropriate in many cases. New analytical technologies 7
and modifications to existing technologies are continuously being developed and should 8
be utilized when appropriate. 9
10
Where relevant and possible, characteristics of the properties of the product should be 11
compared with its natural counterpart. For example, post-translational modifications, 12
such as glycosylation are likely to differ from those found in the natural counterpart and 13
may influence the biological, pharmacological and immunological properties of the 14
rDNA-derived biotherapeutics. 15
16
A.5.2 Routine control 17
Not all the characterization and testing described above in A 5.1 and in Appendix 2, 18
needs to be carried out on each batch of drug substance and drug product prior release on 19
the market. Some tests may need to be performed only initially and/or periodically to 20
establish or verify the validity or acceptability of a product and its manufacturing process. 21
Others may be required on a routine basis. A comprehensive analysis of the initial 22
production batches is expected in order to establish consistency with regard to identity, 23
purity and potency. A more limited series of tests is appropriate for routine control as 24
outlined below and in more detail in Appendix 3. Tests for use in routine control should 25
be chosen to confirm quality. The rationale and justification for including and/or 26
excluding testing for specific quality attributes should be provided. 27
28
An acceptable number of consecutive batches should be characterized to determine 29
consistency of analytical parameters. Any differences between one batch and another 30
should be noted. Data obtained from such studies as well as knowledge gained from 31
WHO/BS/2013.2213 Page 28
clinical and nonclinical development and during stability studies should be used as the 1
basis for establishing product specifications. 2
3
The selection of tests to be included in the routine control programme will be product 4
specific and should take into account the quality attributes (e.g. potential influence on 5
safety, efficacy or stability), the process performance (e.g. clearance capability, content), 6
the controls in place through the manufacturing process (e.g. multiple testing points), and 7
the material used in relevant nonclinical and clinical studies. These tests should include 8
criteria such as potency, the nature and quantity of product-related substances, product-9
related impurities, process-related impurities, and absence of contaminants. 10
11
A.6 Filling and container 12
The general requirements concerning filling and containers given in the WHO Guidelines 13
on good manufacturing practices for biological products (19) should apply. 14
15
A description of the container closure systems for the drug substance and the drug 16
product should be provided including a specification for their component materials. 17
Evidence exists to show that formulated proteins can interact chemically with the 18
formulation excipients and/or the container closure system, and can, for example, lead to 19
the formation of potentially immunogenic complexes. The suitability of the container 20
closure system should be evaluated and described for its intended use. This should cover 21
evaluation of the compatibility of the materials of construction with the formulated 22
product, including adsorption to the container, leaching and other chemical or physical 23
interaction between the product and the contacting materials. The integrity of the closure 24
and its ability to protect the formulation from contamination and maintain sterility needs 25
to be ensured. 26
27
When a delivery device is presented as part of the drug product (e.g. prefilled syringe, 28
single use autoinjector), it is important to demonstrate the functionality of such a 29
combination, such as the reproducibility and accuracy of the dispensed dose under testing 30
conditions which should simulate the use of the drug product as closely as possible. For 31
WHO/BS/2013.2213 Page 29
multi-use containers such as vials or cartridges for a pen injector, proper in-use stability 1
studies should be performed to evaluate the impact of the in-use period of the vial or the 2
assembled device on the formulation and the functionality of the pen injector. Dose 3
accuracy should be demonstrated for the first and last dose delivered. In addition, the 4
effect of multiple injections/withdrawals on the closure should be evaluated. 5
6
A.7 Records, retained samples, labelling, distribution and transport 7
The requirements given in the WHO Guidelines on good manufacturing practices for 8
biological products (19) should apply. 9
10
The conditions of shipping should be such as to ensure that the products are maintained at 11
the appropriate environment. 12
13
A.8 Stability, storage and expiry date 14
A.8.1 Stability studies 15
For proteins, maintenance of biological activity is generally dependent on maintaining 16
molecular conformation. Such products can be particularly sensitive to environmental 17
factors such as temperature changes, oxidation, and light exposure. In order to ensure 18
maintenance of biological activity and to avoid degradation, appropriate conditions for 19
their storage are usually necessary. 20
21
A detailed protocol for the assessment of the stability of both drug substance and drug 22
product in support of the proposed storage conditions and expiration dating periods 23
should be developed. This should include all necessary information which demonstrates 24
the stability of the rDNA-derived biotherapeutics throughout the proposed shelf life 25
including, for example, well-defined specifications and test intervals. 26
27
Each product should retain its specification within established limits for stability-28
indicating attributes, including potency throughout its proposed shelf-life. Specifications 29
should be derived from all available information using appropriate statistical methods. 30
There is no single stability-indicating assay or parameter that profiles the stability 31
WHO/BS/2013.2213 Page 30
characteristics of a rDNA-derived biotherapeutic. Consequently, the manufacturer should 1
develop a stability-indicating programme that provides assurance that changes in the 2
quality and potency of the product will be detected. 3
4
Primary data to support a requested storage period for either drug substance or drug 5
product should be based on long-term, real-time, and real-condition stability studies, 6
covering up to or beyond the claimed shelf-life. In cases where the stability of the product 7
is influenced by storage of intermediates (e.g. significant degradation trend observed 8
during storage of an intermediate), cumulative stability study should be considered. This 9
study should include all intermediates stored at the longest storage time claimed, or 10
selection of the most storage sensitive intermediates, as appropriate. Considering the time 11
necessary to generate the data, such cumulative study could be presented and justified in 12
the proposed stability programme at the time of licensing. 13
Also, stability studies should include an evaluation of the impact of the container closure 14
system on the formulated rDNA- derived biotherapeutics throughout the shelf life. In 15
order to ensure that the formulated product is in contact with all material of the container 16
closure system, stability studies should include samples maintained in the inverted or 17
horizontal position (i.e. in contract with the closure), as well as in the upright position, to 18
determine the effects of the closure on product quality. Data should be supplied for all 19
different container closure combinations that will be marketed. 20
21
Stability information should be provided on at least 3 batches for which manufacture and 22
storage are representative of the commercial process. 23
24
When shelf-lives of 1 year or less are proposed, real-time stability studies should be 25
conducted monthly for the first 3 months and at 3 month intervals thereafter. For products 26
with proposed shelf-lives of greater than 1 year, the studies should be conducted every 3 27
months during the first year of storage, every 6 months during the second year, and 28
annually thereafter. A minimum of 6 months data at the time of submission should be 29
submitted in cases where storage periods greater than 6 months are requested, unless 30
WHO/BS/2013.2213 Page 31
otherwise justified. For storage periods of less than 6 months, the minimum amount of 1
stability data in the initial submission should be determined on a case-by-case basis. 2
3
It is recommended that stability studies under accelerated and stress conditions, including 4
the impact of the container closure system (see A.6), should also be conducted on the 5
drug product. Studies under accelerated conditions may provide useful supportive data 6
for establishing the expiry date, provide product stability information for future product 7
development (e.g. preliminary assessment of proposed manufacturing changes such as 8
changes in formulation or scale-up), assist in validation of analytical methods for the 9
stability program, or generate information which may help elucidate the degradation 10
profile of the rDNA-derived biotherapeutics. Studies under stress conditions may also be 11
useful in determining whether accidental exposures to conditions other than those 12
proposed (e.g. during transportation) are deleterious to the product and for evaluating 13
which specific test parameters may be the best indicators of product stability. 14
15
Further guidance on both general and specific aspects of stability testing of a rDNA -16
derived biotherapeutic can be obtained by consulting the WHO guidelines on the stability 17
testing of active pharmaceutical ingredients and finished pharmaceutical products (25), 18
as well as the WHO Guidelines for stability evaluation of vaccines (26). 19
20
A.8.2 Drug product requirements 21
Stability information should be provided on at least 3 batches of drug product 22
representative of that which will be used in commercial manufacture, and presented in the 23
final container. Where possible, the drug product batches included in stability testing 24
should be derived from different batches of drug substance. 25
26
Where one product is distributed in multiple presentations, the samples to be entered into 27
the stability program may be selected on the basis of a matrix system and/or by 28
bracketing. Where the same strength and exact container/closure system is used for 3 or 29
more fill contents, the manufacturer may elect to place only the smallest and largest 30
container size into the stability program, i.e. bracketing. The design of a protocol that 31
WHO/BS/2013.2213 Page 32
incorporates bracketing assumes that the stability of the intermediate condition samples is 1
represented by those at the extremes. In certain cases, data may be needed to demonstrate 2
that all samples are properly represented by data collected for the extremes. 3
Matrixing, i.e. the statistical design of a stability study in which account is taken of 4
factors such as the tests, process characteristics, presentation characteristics and different 5
testing time points, should only be applied when appropriate documentation is provided 6
that confirms that the stability of the samples tested represents the stability of all samples. 7
The differences in the samples for the same drug product should be identified as, for 8
example, covering different batches, different strengths, different sizes of the same 9
closure and possibly, in some cases, different container/closure systems. Matrixing 10
should not be applied to samples with differences that may affect stability, such as 11
different strengths and different containers/closures, where it cannot be confirmed that 12
the products respond similarly under storage conditions. 13
14
For preparations intended for use after reconstitution, dilution or mixing, in-use stability 15
data should be obtained. The stability should be demonstrated up to and beyond the 16
storage conditions and the maximum storage period claimed. 17
18
In addition to the standard data necessary for a conventional single-use vial, it should be 19
shown that the closure used with a multiple-dose vial is capable of withstanding the 20
conditions of repeated insertions and withdrawals so that the product retains its identity, 21
strength, potency, purity, and quality for the maximum period specified in the 22
instructions-for-use on containers, packages, and/or package inserts. 23
24
A.9 Manufacturing process changes 25
Changes to the manufacturing processes of a rDNA-derived biotherapeutic often occur 26
both during development and after approval. The reasons for such changes include 27
improving the manufacturing process, increasing scale, site change, improving product 28
stability, or complying with changes in regulatory requirements. When substantial 29
changes are made to the manufacturing process, a comparability exercise to evaluate the 30
impact of the change(s) on the quality, safety and efficacy of the rDNA-derived 31
WHO/BS/2013.2213 Page 33
biotherapeutics should be undertaken. The demonstration of comparability does not 1
necessarily mean that the quality attributes of the pre-change and post-change product are 2
identical, but that they are highly similar and that the existing knowledge is sufficiently 3
predictive to ensure that any differences in quality attributes have no adverse impact upon 4
safety or efficacy of the rDNA-derived biotherapeutics. The reason for each significant 5
change should be explained, together with an assessment of its potential to impact on 6
quality, safety and efficacy. 7
8
The extent of a comparability exercise depends on the potential impact of the process 9
change(s) on the quality, safety and efficacy of the product. It can range from analytical 10
testing alone (e.g. where process changes lead to no changes in any quality attribute) to a 11
comprehensive exercise requiring nonclinical and clinical bridging studies (e.g. the 12
establishment of a new host cell line with altered properties resulting in more pronounced 13
changes in quality attributes). If assurance of comparability can be shown through 14
analytical studies alone, nonclinical or clinical studies with the post-change product may 15
not be necessary. However, where the relationship between specific quality attributes and 16
safety and efficacy has not been established, and differences between quality attributes of 17
the pre- and post-change product are observed, it might be appropriate to include a 18
combination of quality, nonclinical, and/or clinical studies in the comparability exercise. 19
20
Further considerations of manufacturing changes can be found in guidelines provided by 21
the ICH (27), the European Medicines Agency (EMA) (28), the United States Food and 22
Drug Administration (US FDA) (29) and other major NRAs. 23
24
WHO/BS/2013.2213 Page 34
Part B. Nonclinical evaluation 1
2
B.1 Introduction 3
The general aim of nonclinical evaluation is to determine whether new medicinal 4
products possess the desired pharmacodynamic (PD) activity and have the potential to 5
cause unexpected and undesirable effects. However, classical PD, safety or toxicological 6
testing, as recommended for chemical drugs, may be of only limited relevance for rDNA-7
derived biotherapeutics due to their unique and diverse structural and biological 8
properties including species specificity, immunogenicity, and unpredicted pleiotropic 9
activities. These properties pose particular problems in relation to nonclinical testing in 10
animals, and their pharmacological and safety evaluation will have to take a large number 11
of factors into account. Thus, a flexible approach is necessary for the nonclinical 12
evaluation of rDNA-derived biotherapeutics. For example, certain proteins, e.g. 13
interferons, are highly species-specific, so that the human protein is pharmacologically 14
much more active in humans than in any animal species. Furthermore, human proteins 15
frequently produce immunological responses in animal species which may ultimately 16
modify their biological effects and may result in toxicity, e.g. due to immune complex 17
formation. Such toxicity has little bearing on the safety of the product in the intended 18
human host. 19
20
Although some safety testing will be required for most products, the range of tests that 21
need to be carried out should be decided on a case-by-case basis (e.g. Appendix 4) in 22
consultation with the NRA/NCL. A wide range of pharmacological, biochemical, 23
immunological, toxicological and histopathological investigative techniques should be 24
used, where appropriate, in the assessment of a product’s effect, over an appropriate 25
range of doses and, in accordance with the desired clinical indication(s), during both 26
acute and chronic exposure. However, the points made above concerning species-27
specificity and antibody formation should always be taken into consideration. 28
29
Additional information on specific safety issues, as for example carcinogenic potential, 30
reproductive toxicity, or safety pharmacology, is provided in respective ICH safety 31
WHO/BS/2013.2213 Page 35
guidelines (30-32). Recommendations concerning timing and interplay of nonclinical and 1
clinical studies in drug development are given in the ICH Guidance on nonclinical safety 2
studies for the conduct for human clinical trials and marketing authorization for 3
pharmaceuticals (33) and in ICH guideline Preclinical safety evaluation of 4
biotechnology-derived pharmaceuticals (34). 5
6
B.1.1 Objectives of the nonclinical evaluation 7
The objectives of the nonclinical studies are to define pharmacological and toxicological 8
effects throughout clinical development, not only prior to initiation of human studies. 9
The primary goals are to: 1) identify an initial safe dose and subsequent dose escalation 10
schemes in humans; 2) identify potential target organs for toxicity and for the study of 11
whether such toxicity is reversible; and 3) identify safety parameters for clinical 12
monitoring. 13
14
Nonclinical evaluation should consider: 1) selection of the pharmacologically or 15
toxicologically relevant animal species; 2) age of animals; 3) physiological state of 16
animals (e.g. whether healthy/diseased animals are used, whether treatment naïve animals 17
are used); 4) weight of animals; 5) the manner of delivery, including dose, route of 18
administration, and treatment regimen; and 6) stability of the test material under the 19
conditions of use. 20
Both in vitro and in vivo studies can contribute to this characterization. 21
rDNA-derived biotherapeutics that belong structurally and pharmacologically to a (the 22
same) product class for which there is wide experience in clinical practice may need less 23
extensive toxicity testing. 24
25
B.1.2 Product development and characterization 26
In general, the product that is used in the definitive pharmacology and toxicology studies 27
should be the same as the product proposed for the initial clinical studies. However, it is 28
appreciated that during the course of development programs, changes normally occur in 29
the manufacturing process in order to improve product quality and yields. The potential 30
WHO/BS/2013.2213 Page 36
impact of such changes for extrapolation of the animal findings to humans should be 1
considered, including the impact of post-translational modifications. 2
The comparability of the test material should be demonstrated when a new or modified 3
manufacturing process or other significant changes in the product or formulation are 4
made in an ongoing development program. Comparability can be evaluated on the basis 5
of biochemical and biological characterization (i.e. identity, purity, stability, and potency). 6
In some cases, additional studies may be needed (i.e. PK, PD and/or safety). The 7
scientific rationale for the approach taken should be provided. 8
9
B.1.3 Good laboratory practice 10
Pivotal (toxicity) studies should be performed in compliance with good laboratory 11
practice (GLP). However, it is recognized that some studies employing specialized test 12
systems which are often needed for rDNA-derived biotherapeutics may not comply fully 13
with GLP. Areas of non-compliance should be identified and their significance evaluated 14
relative to the overall nonclinical assessment. In some cases, lack of full GLP compliance 15
does not necessarily mean that the data from these studies cannot be used to support 16
clinical trials and marketing authorization. However, justification which is supported with 17
data, such as method validation should be provided for the data quality assurance. 18
19
B.2 Pharmacodynamics 20
B.2.1 Primary and secondary pharmacodynamics/Biological activity 21
Biological activity may be evaluated using in vitro assays to determine which effects of 22
the product may be related to clinical activity. The use of cell lines and/or primary cell 23
cultures can be useful to examine the direct effects on cellular phenotype and 24
proliferation. Due to the species specificity of many rDNA-derived biotherapeutics, it is 25
important to select relevant animal species for testing (see Appendix 5). Non-human 26
primates (NHPs) are often the only pharmacologically or toxicologically relevant species; 27
however, other species should also be evaluated for relevant biological activity. In vitro 28
cell lines derived from mammalian cells can be used to predict specific aspects of in vivo 29
activity and to assess quantitatively the relative sensitivity of various species to the 30
biotherapeutics, including human. Such studies may be designed to determine, for 31
WHO/BS/2013.2213 Page 37
example, receptor occupancy, receptor affinity, and/or pharmacological effects, and to 1
assist in the selection of an appropriate animal species for further in vivo pharmacology 2
and toxicology studies. The combined results from in vitro and in vivo studies assist in 3
the extrapolation of the findings to humans. In vivo studies to assess pharmacological 4
activity, including defining mechanism(s) of action, are often used to support the 5
rationale for the proposed use of the product in clinical studies. When feasible, PD 6
endpoints can be incorporated into general toxicity studies (e.g. hemoglobin blood 7
concentration in repeated dose toxicity studies with erythropoetins). 8
9
B.2.2 Safety pharmacology 10
Based on the target or mechanism of action of the product, it is important to investigate 11
the potential for undesirable pharmacological activity in appropriate animal models. The 12
aim of the safety pharmacology studies is to reveal any functional effects on the major 13
physiological systems (e.g. cardiovascular, respiratory, and central nervous systems). 14
These functional indices may be investigated in separate studies or incorporated in the 15
design of toxicity studies and/or clinical studies. Investigations may include the use of 16
isolated organs or other test systems not involving intact animals. All of these studies 17
may allow for a mechanistically-based explanation of specific organ effects/toxicities, 18
which should be considered carefully with respect to their applicability for human use 19
and indication(s). 20
21
B.3 Pharmacokinetics/Toxicokinetics 22
B.3.1 General principles 23
It is difficult to establish uniform guidelines for PK studies for rDNA-derived 24
biotherapeutics. Single and multiple dose PK, toxicokinetics (TK), and tissue distribution 25
studies in relevant species are useful; however, routine studies that attempt to assess mass 26
balance are not useful. Differences in PK among animal species may have a significant 27
impact on the predictiveness of animal studies or on the assessment of dose response 28
relationships in toxicity studies. Scientific justification should be provided for the 29
selection of the animal species used for PK/TK evaluation, taking into account that the 30
PK profile in the chosen animal species should, ideally, reflect the PK profile in humans. 31
WHO/BS/2013.2213 Page 38
Alterations in the PK profile due to immune-mediated clearance mechanisms may affect 1
the kinetic profiles and the interpretation of the toxicity data (see also B.4.8.1). For some 2
products there may also be inherent significant delays in the expression of PD effects 3
relative to the PK profile (e.g. cytokines) or there may be prolonged expression of PD 4
effects relative to plasma levels. 5
6
PK studies should, whenever possible, utilize preparations that are representative of that 7
intended for toxicity testing and clinical use, and employ a route of administration that is 8
relevant to the anticipated clinical studies. Patterns of absorption may be influenced by 9
formulation, active substance concentration, application site, and/or application volume. 10
Whenever possible, systemic exposure should be monitored during the toxicity studies. 11
When feasible, PK/TK evaluations can be incorporated into general toxicity studies. 12
Some information on absorption, disposition and clearance in relevant animal models 13
should be available prior to clinical studies in order to predict margins of safety based 14
upon exposure and dose. Understanding the behaviour of the biotherapeutic in the 15
biologic matrix, (e.g. plasma, serum, cerebral spinal fluid) and the possible influence of 16
binding proteins is important for understanding the PD effect. 17
18
B.3.2 Assays 19
The use of one or more assay methods should be addressed on a case-by-case basis and 20
the scientific rationale should be provided. One validated method is usually considered 21
sufficient. For example, quantitation of trichloracetic acid (TCA)-precipitable 22
radioactivity following administration of a radiolabeled protein may provide adequate 23
information, but a specific assay for the analyte is preferred. Ideally, the assay methods 24
should be the same for animal and human studies. The possible influence of plasma 25
binding proteins and/or antibodies in plasma/serum on the assay performance should be 26
determined. 27
28
B.3.3 Distribution 29
Unlike small chemical drugs that readily diffuse, rDNA-derived biotherapeutics, due to 30
their molecular weight, usually do not readily diffuse, but are, following intravenous 31
WHO/BS/2013.2213 Page 39
application, initially confined to the vascular system. However, with time they may 1
distribute to the extravascular space by various factors, including bulk flow and active 2
transport. 3
4
As a supplement to standard tissue distribution studies, complimentary information about 5
tissue distribution of molecular targets for rDNA-derived biotherapeutics may be 6
obtained from tissue cross-reactivity (TCR) studies (see B.4.8.3). 7
8
Tissue concentrations of radioactivity and/or autoradiography data using radiolabeled 9
proteins may be difficult to interpret due to rapid in vivo protein metabolism or unstable 10
radiolabeled linkage. Care should be taken in interpreting studies using radioactive 11
tracers incorporated into specific amino acids because of recycling of amino acids into 12
non-drug related proteins/peptides. 13
14
B.3.4 Metabolism 15
The expected consequence of metabolism of rDNA-derived biotherapeutics is the 16
degradation to small peptides and individual amino acids. Therefore, the metabolic 17
pathways are generally understood. Classical biotransformation studies, as performed for 18
pharmaceuticals, are not needed. 19
20
B.4 Toxicity studies 21
B.4.1 General principles 22
Number/Gender of animals 23
The number of animals used per dose has a direct bearing on the ability to detect toxicity. 24
A small sample size may lead to failure to observe toxic events due to observed 25
frequency alone regardless of severity. The limitations that are imposed by sample size, 26
as often is the case for NHP studies, may be in part compensated by increasing the 27
frequency and duration of monitoring. Both genders should generally be used or 28
justification given for specific omissions. As an example, the minimum sample size for a 29
pivotal GLP toxicity study in NHPs is considered to be three animals per sex, and, if a 30
WHO/BS/2013.2213 Page 40
recovery group is included in the study, an additional minimum of two animals per sex 1
would be included. 2
3
It is desirable to apply the “3R principles” (i.e. reduction, replacement, refinement) to 4
minimize the use of animals for ethical reasons and consideration should be given to the 5
use of appropriate in vitro alternative methods for safety evaluation to reduce the use of 6
animals (35). 7
8
Administration/Dose selection and application of PK/PD principles 9
The route and frequency of administration should be as close as possible to that proposed 10
for clinical use. Consideration should be given to pharmacokinetics and bioavailability of 11
the product in the species being used, and the volume which can be safely and humanely 12
administered to the test animals. For example, the frequency of administration in 13
laboratory animals may be increased compared to the proposed schedule for the human 14
clinical studies in order to compensate for faster clearance rates or low solubility of the 15
active ingredient. In these cases, the level of exposure of the test animal relative to the 16
clinical exposure should be defined. Consideration should also be given to the effects of 17
application volume, active substance concentration, formulation, and site of 18
administration. The use of routes of administration other than those used clinically may 19
be acceptable if the route must be modified due to limited bioavailability, limitations due 20
to the route of administration, or to size/physiology of the used animal species. 21
If feasible, dosage levels should be selected to provide information on a dose-response 22
relationship, including a toxic dose and a no observed adverse effect level (NOAEL). 23
However, for oncology drugs where significant toxicity is anticipated, studies are often 24
designed to identify a “maximum tolerated dose (MTD)” rather than a NOAEL. 25
The toxicity of most rDNA-derived biotherapeutics is related to their targeted mechanism 26
of action; therefore, relatively high doses can elicit adverse effects which are apparent as 27
exaggerated pharmacology. 28
For some classes of products which show little to no toxicity it may not be possible to 29
define a specific maximum dose. In these cases, a scientific justification of the rationale 30
for the dose selection and projected multiples of human exposure should be provided. To 31
WHO/BS/2013.2213 Page 41
justify high dose selection, consideration should be given to the expected 1
pharmacological/physiological effects, and the intended clinical use. Where a product has 2
a lower affinity for, or potency in, the cells of the selected species than in human cells, 3
testing of higher doses may be important. The multiples of the human dose that are 4
needed to determine adequate safety margins may vary with each class of rDNA-derived 5
biotherapeutics and its clinical indication(s). 6
7
A rationale should be provided for dose selection taking into account the characteristics 8
of the dose-response relationship. PK-PD approaches (e.g. simple exposure-response 9
relationships or more complex modeling and simulation approaches) can assist in high 10
dose selection by identifying (i) a dose which provides the maximum intended 11
pharmacological effect in the selected animal species; and (ii) a dose which provides an 12
approximately 10-fold exposure multiple over the maximum exposure to be achieved in 13
the clinic. The higher of these two doses should be chosen for the high dose group in 14
nonclinical toxicity studies unless there is a justification for using a lower dose (e.g. 15
maximum feasible dose). 16
Where in vivo/ex vivo PD endpoints are not available, the high dose selection can be 17
based on PK data and available in vitro binding and/or pharmacology data. Corrections 18
for differences in target binding and in vitro pharmacological activity between the 19
nonclinical species and humans should be taken into account to adjust the exposure 20
margin over the highest anticipated clinical exposure. For example, a large relative 21
difference in binding affinity and/or in vitro potency might suggest that testing higher 22
doses in the nonclinical studies is appropriate. In the event that toxicity cannot be 23
demonstrated at the doses selected using this approach, then additional toxicity studies at 24
higher multiples of human dosing are unlikely to provide additional useful information. 25
26
Use of one or two species 27
Concerning the use of one or two species for toxicity studies, see Appendix 5. 28
29
Study duration 30
WHO/BS/2013.2213 Page 42
For chronic use products, repeat dose toxicity studies of 6 months duration in rodents or 1
non-rodents are usually considered sufficient, providing the high dose is selected in 2
accordance with the principles above. Studies of longer duration have not generally 3
provided useful information that changed the clinical course of development (see also 4
B.4.3). 5
For chronic use of rDNA-derived biotherapeutics developed for patients with advanced 6
cancer, see Appendix 4. 7
8
Evaluation of immunogenicity 9
Many rDNA-derived biotherapeutics intended for human use are immunogenic in 10
animals. Therefore, an immunogenicity assessment should be performed when 11
conducting repeated dose toxicity studies in order to aid in the interpretation of these 12
studies (for details, see B.4.8.1). 13
14
B.4.2 Single dose toxicity studies 15
In general, single dose toxicity studies should only be pursued in cases where significant 16
toxicity is anticipated and the information is needed to select doses for repeated dose 17
studies (33, 34). 18
Single dose studies may generate useful data to describe the relationship of dose to 19
systemic and/or local toxicity. These data can be used to select doses for repeated dose 20
toxicity studies. Information on dose-response relationships may be gathered through the 21
conduct of a single dose toxicity study, as a component of pharmacology or animal model 22
efficacy studies. The incorporation of safety pharmacology parameters in the design of 23
these studies should be considered. 24
25
B.4.3 Repeated dose toxicity studies 26
For consideration of the selection of animal species for repeated dose studies, see B.4.1. 27
The route and dosing regimen (e.g. daily versus intermittent dosing) should reflect the 28
intended clinical use or exposure. When feasible, these studies should include TK 29
measurements, but interpretation should consider the formation of possible antidrug 30
antibodies (see B.4.8.1). 31
WHO/BS/2013.2213 Page 43
1
Study duration 2
The duration of repeated dose studies should be based on the intended duration of clinical 3
exposure and disease indication. Duration of animal dosing has generally been 1-3 4
months for most rDNA-derived biotherapeutics. For rDNA-derived biotherapeutics 5
intended for short-term use (e.g. < 7 days) and for acute life-threatening diseases, 6
repeated dose studies up to two weeks duration have been considered adequate to support 7
clinical studies as well as marketing authorization. For those rDNA-derived 8
biotherapeutics intended for chronic indications, studies of 6 months duration have 9
generally been appropriate although in some cases shorter or longer durations have 10
supported marketing authorizations. For rDNA-derived biotherapeutics intended for 11
chronic use, the duration of long term toxicity studies should be scientifically justified. 12
13
Recovery period 14
Recovery from pharmacological and toxicological effects with potential adverse clinical 15
impact should be understood when they occur at clinically relevant exposures. This 16
information can be obtained by an understanding that the particular effect observed is 17
generally reversible/nonreversible or by including a non-dosing period in at least one 18
study, at least at one dose level, to be justified by the sponsor. The purpose of the non-19
dosing period is to examine reversibility of these effects, not to assess delayed toxicity. 20
The demonstration of complete recovery is not considered essential. The addition of a 21
recovery period, for the sole purpose of assessing the potential for immunogenicity, is not 22
required. 23
24
B.4.4 Genotoxicity studies 25
The range and type of genotoxicity studies routinely conducted for pharmaceuticals are 26
not applicable to rDNA-derived biotherapeutics and therefore are not needed. Moreover, 27
the administration of large quantities of peptides/proteins may yield uninterpretable 28
results. It is not expected that these substances would interact directly with DNA or other 29
chromosomal material. 30
WHO/BS/2013.2213 Page 44
With some rDNA-derived biotherapeutics there is a potential concern about accumulation 1
of spontaneously mutated cells (e.g. via facilitating a selective advantage of proliferation) 2
leading to carcinogenicity. The standard battery of genotoxicity tests is not designed to 3
detect these conditions. Alternative in vitro or in vivo models to address such concerns 4
may have to be developed and evaluated (see B.4.5). 5
Studies in available and relevant systems, including newly developed systems, should be 6
performed in those cases where there is cause for concern about the product (e.g. because 7
of the presence of an organic linker molecule in a conjugated protein product). 8
The use of standard genotoxicity studies for assessing the genotoxic potential of process 9
contaminants is usually not considered appropriate. If performed for this purpose, 10
however, the rationale should be provided. 11
12
B.4.5 Carcinogenicity studies 13
General principles 14
Carcinogenicity in the strict sense is increased probability of development of new tumors. 15
However, activation of proliferation and progression of existing tumor cells/tumors 16
should also be considered. 17
18
The need for a product-specific assessment of the carcinogenic potential for rDNA-19
derived biotherapeutics should be determined with regard to the intended clinical 20
population and treatment duration (e.g. 31). When an assessment is warranted, the 21
sponsor should design a strategy to address the potential hazard. This strategy could be 22
based on a review of relevant data from a variety of sources. The data sources can include 23
published data (e.g. information from transgenic, knock-out or animal disease models, 24
human genetic diseases), information on class effects, detailed information on target 25
biology and mechanism of action, in vitro data, data from chronic toxicity studies and 26
clinical data. In some cases, the available information can be sufficient to address 27
carcinogenic potential and inform clinical risk without additional nonclinical studies. 28
The mechanism of action of some rDNA-derived biotherapeutics might raise concern 29
regarding potential for carcinogenicity (e.g. immunosuppressives and growth factors). If 30
the review of all available data (see above) supports the concern regarding carcinogenic 31
WHO/BS/2013.2213 Page 45
potential, rodent bioassays are not warranted. In this case potential hazard can be best 1
addressed by product labeling and risk management practices. However, if the potential 2
for carcinogenicity remains still unclear after a review of all available data, the sponsor 3
can propose additional studies that could mitigate the mechanism-based concern (e.g. 34). 4
5
For products where there is insufficient knowledge about specific product characteristics 6
and mode of action in relation to carcinogenic potential, a more extensive assessment 7
might be appropriate (e.g. understanding of target biology related to potential 8
carcinogenic concern, inclusion of additional endpoints in toxicity studies). 9
If the review of all data from this more extensive assessment does not suggest a 10
carcinogenic potential, no additional nonclinical testing is recommended. Alternatively, if 11
the review of all data available suggests a concern about carcinogenic potential, then the 12
sponsor can propose additional nonclinical studies that could mitigate the concern (see 13
above), or the label should reflect the concern. 14
15
Use of homologous proteins 16
A homologous protein is defined as a mouse, rat, or primate etc. protein that recognizes 17
the appropriate mouse, rat or primate etc. target(s) with similar potency to the clinical 18
candidate that recognizes the corresponding human target(s) (36). 19
Rodent bioassays (or short-term carcinogenicity studies) with homologous products are 20
generally of limited value to assess carcinogenic potential of the clinical candidate. Since 21
the production process, range of impurities/contaminants, pharmacokinetics, and exact 22
pharmacological mechanism(s) may differ between the homologous form and the product 23
intended for clinical use, studies with homologous proteins are generally not useful for 24
quantitative risk assessment (see also Appendix 5). 25
26
Risk communication 27
The product-specific assessment of carcinogenic potential is used to communicate risk 28
and provide input to the risk management plan along with labeling proposals, clinical 29
monitoring, post-marketing surveillance, or a combination of these approaches. 30
31
WHO/BS/2013.2213 Page 46
B.4.6 Reproductive performance and developmental toxicity studies 1
B.4.6.1 General principles 2
The need for reproductive/developmental toxicity studies is dependent upon the product, 3
clinical indication and intended patient population. The specific study design and dosing 4
schedule may be modified based on issues related to species specificity, immunogenicity, 5
biological activity and/or a long elimination half-life. For example, concerns regarding 6
potential developmental immunotoxicity, which may apply particularly to certain 7
monoclonal antibodies with prolonged immunological effects, could be addressed in a 8
study design modified to assess immune function of the neonate. 9
10
(i) Products with expected/probable adverse effects on fertility/pregnancy outcome 11
When the available data (e.g. mechanism of action, phenotypic data from genetically 12
modified animals, class effects) clearly suggest that there will be an adverse effect on 13
fertility or pregnancy outcome, these data can provide adequate information to 14
communicate risk to reproduction and, under appropriate circumstances, additional 15
nonclinical studies might not be warranted. 16
There may be extensive public information available regarding potential reproductive 17
and/or developmental effects of a particular class of compounds (e.g. interferons) where 18
the only relevant species is the non-human primate. In such cases, mechanistic studies 19
indicating that similar effects are likely to be caused by a new but related molecule, may 20
obviate the need for formal reproductive/developmental toxicity studies. In each case, the 21
scientific basis for assessing the potential for possible effects on reproduction/ 22
development should be provided. 23
24
(ii) Products with unclear potential for adverse effects on fertility/pregnancy outcome 25
The specific study design and dosing schedule can be modified based on an 26
understanding of species specificity, the nature of the product and mechanism of action, 27
immunogenicity and/or PK behavior and embryo-fetal exposure. 28
29
Species selection 30
WHO/BS/2013.2213 Page 47
An assessment of reproductive toxicity of the clinical candidate should usually be 1
conducted only in pharmacologically relevant species. When the clinical candidate is 2
pharmacologically active in rodents and rabbits, both species should be used for embryo-3
fetal development (EFD) studies, unless embryo-fetal lethality or teratogenicity has been 4
identified in one species. 5
Developmental toxicity studies should only be conducted in NHPs when they are the only 6
relevant species. When the clinical candidate is pharmacologically active only in NHPs, 7
there is still a preference to test the clinical candidate. However, an alternative model can 8
be used in place of NHPs if appropriate scientific justification is provided. 9
10
Alternative evaluation in the absence of a relevant species 11
When no relevant animal species exist(s) for testing the clinical candidate, the use of 12
transgenic mice expressing the human target or homologous protein in a species 13
expressing an ortholog of the human target can be considered, assuming that sufficient 14
background knowledge exists for the model (e.g. historical background data). 15
16
(iii) Products for which adverse effects on fertility/pregnancy outcome are not expected 17
For products that are directed at a foreign target such as bacteria and viruses, in general 18
no reproductive toxicity studies would be expected. 19
20
B.4.6.2 Fertility 21
For products where mice and rats are pharmacologically relevant species, an assessment 22
of fertility can be made in one of these rodent species (30). Study designs can be adapted 23
for other species provided they are pharmacologically relevant and should be amended as 24
appropriate, for example to address the nature of the product and the potential for 25
immunogenicity. 26
It is recognized that mating studies are not practical for NHPs. However, when the NHP 27
is the only relevant species, the potential for effects on male and female fertility can be 28
assessed by evaluation of the reproductive tract (organ weights and histopathological 29
evaluation) in repeat dose toxicity studies of at least 3 months duration using sexually 30
mature NHPs. If there is a specific cause for concern based on pharmacological activity 31
WHO/BS/2013.2213 Page 48
or previous findings, specialized assessments such as menstrual cyclicity, sperm count, 1
sperm morphology/motility, and male or female reproductive hormone levels can be 2
evaluated in a repeat dose toxicity study. 3
If there is a specific concern from the pharmacological activity about potential effects on 4
conception/implantation and the NHP is the only relevant species, the concern should be 5
addressed experimentally. A homologous product or transgenic model could be the only 6
practical means to assess potential effects on conception or implantation when those are 7
of specific concern. However, it is not recommended to produce a homologous product or 8
transgenic model solely to conduct mating studies in rodents. In absence of nonclinical 9
information, the risk to patients should be mitigated through clinical trial management 10
procedures, informed consent and appropriate product labeling. 11
12
B.4.6.3 Embryo-fetal development (EFD) and pre/post-natal development (PPND) 13
Selection of study design 14
Potential differences in placental transfer of rDNA-derived biotherapeutics should be 15
considered in the design and interpretation of developmental toxicity studies (see Note 1 16
of Appendix 6). 17
For products pharmacologically active only in NHPs, several study designs can be 18
considered based on intended clinical use and expected pharmacology. Separate EFD 19
and/or PPND studies, or other study designs (justified by the sponsor) can be appropriate, 20
particularly when there is some concern that the mechanism of action might lead to an 21
adverse effect on EFD or pregnancy loss. However, one well-designed study in NHPs 22
which includes dosing from day 20 of gestation to birth (enhanced PPND, ePPND) can be 23
considered, rather than separate EFD and/or PPND studies. 24
25
Enhanced pre/post-natal development (ePPND) studies 26
For the single ePPND study design described above, no Caesarian section group is 27
warranted, but assessment of pregnancy outcome at natural delivery should be performed. 28
This study should also evaluate offspring viability, external malformations, skeletal 29
effects (e.g. by X-ray) and, ultimately, visceral morphology at necropsy. Ultrasound is 30
useful to track maintenance of pregnancy but is not appropriate for detecting 31
WHO/BS/2013.2213 Page 49
malformations. These latter data are derived from post-partum observations. Because of 1
confounding effects on maternal care of offspring, dosing of the mother post-partum is 2
generally not recommended. Other endpoints in the offspring can also be evaluated if 3
relevant for the pharmacological activity. The duration of the post-natal phase will be 4
dependent on which additional endpoints are considered relevant based on mechanism of 5
action (see Note 2 of Appendix 6). 6
Developmental toxicity studies in NHPs can only provide hazard identification. The 7
number of animals per group should be sufficient to allow meaningful interpretation of 8
the data (see Note3 of Appendix 6). 9
Study design should be justified if species other than the cynomolgus monkey are used. 10
The developmental toxicity studies in NHPs, as outlined above, are just hazard 11
identification studies; therefore it might be possible to conduct these studies using a 12
control group and one dose group, provided there is a scientific justification for the dose 13
level selected (see Note 4 of Appendix 6). 14
15
B.4.6.4 Timing of studies 16
If women of child-bearing potential are included in clinical trials prior to acquiring 17
information on effects on EFD, suitable clinical risk management is appropriate, such as 18
the use of highly effective methods of contraception (33). For rDNA-derived 19
biotherapeutics pharmacologically active only in NHPs, where there are sufficient 20
precautions to prevent pregnancy, an EFD or ePPND study can be conducted during 21
Phase III, and the report submitted at the time of marketing application. When a sponsor 22
cannot take sufficient precaution to prevent pregnancy in clinical trials, either a complete 23
report of an EFD study or an interim report of an ePPND study should be submitted 24
before initiation of Phase III (see Note 5 of Appendix 6). Where the product is 25
pharmacologically active only in NHPs and its mechanism of action raises serious 26
concern for embryo-fetal development, the label should reflect the concern without 27
warranting a developmental toxicity study in NHPs and therefore administration to 28
women of child-bearing potential should be avoided. 29
If the rodent or rabbits is a relevant species, timing of reproductive toxicity/fertility 30
studies should follow the recommendations given (33). 31
WHO/BS/2013.2213 Page 50
For oncology products, see Appendix 4. 1
2
B.4.7 Local tolerance studies 3
Local tolerance should be evaluated. Ideally, the formulation intended for marketing 4
should be tested; however, in certain justified cases, the testing of representative 5
formulations may be acceptable. If feasible, the potential adverse effects of the product 6
can be evaluated in single or repeated dose toxicity studies, thus obviating the need for 7
separate local tolerance studies. 8
9
B.4.8 Other toxicity studies 10
B.4.8.1 Antibody Formation 11
Immunogenicity assessments in animals should only be conducted to assist in the 12
interpretation of the study results and to improve the design of subsequent studies. Such 13
analyses in animal studies are usually not relevant in terms of predicting potential 14
immunogenicity of human or humanized proteins in humans. Since antibody formation to 15
human proteins in animal studies is usually not predictive of the clinical situation, 16
concerns regarding antibody formation to the endogenous hormones, e.g. in case of 17
erythropoietin or somatropin, will have to be addressed on a clinical safety level. 18
19
Measurement of anti-drug antibodies (ADA) in nonclinical studies should be evaluated 20
when there is: 1) evidence of altered PD activity; 2) unexpected changes in exposure in 21
the absence of a PD marker; or 3) evidence of immune-mediated reactions (immune 22
complex disease, vasculitis, anaphylaxis, etc.). 23
Since it is difficult to predict prior to study completion whether such analysis will be 24
necessary, it is often useful to obtain appropriate samples during the course of the study, 25
which can subsequently be analyzed when warranted to aid in interpretation of the study 26
results. 27
When ADAs are detected, their impact on the interpretation of the study results should be 28
assessed. Antibody responses should be characterized (e.g. titer, number of responding 29
animals, neutralizing or non-neutralizing activity), and their appearance should be 30
correlated with any pharmacological and/or toxicological changes. Specifically, the 31
WHO/BS/2013.2213 Page 51
effects of antibody formation on PK/PD parameters, incidence and/or severity of adverse 1
effects, complement activation, or the emergence of new toxic effects should be 2
considered when interpreting the data. Attention should also be paid to the evaluation of 3
possible pathological changes related to immune complex formation and deposition. 4
Characterization of neutralizing potential is warranted when ADAs are detected and there 5
is no PD marker to demonstrate sustained activity in the in vivo toxicology studies. 6
Neutralizing antibody activity can be assessed indirectly with an ex vivo bioactivity assay 7
or an appropriate combination of assay formats for PK-PD, or directly in a specific 8
neutralizing antibody assay. 9
The detection of antibodies should not be the sole criterion for the early termination of a 10
nonclinical safety study or modification in the duration of the study design, unless the 11
immune response neutralizes the pharmacological and/or toxicological effects of the 12
rDNA-derived biotherapeutics in a large proportion of the animals. In most cases, the 13
immune response to rDNA-derived biotherapeutics is variable, similarly to that observed 14
in humans. If the interpretation of the data from the safety study is not compromised by 15
these issues, then no special significance should be ascribed to the antibody response. 16
17
Anaphylaxis tests 18
The occurrence of severe anaphylactic responses to rDNA-derived biotherapeutics is 19
uncommon in humans. In this regard, the results of guinea pig anaphylaxis tests, which 20
are generally positive for protein products, are usually not predictive for reactions in 21
humans and are usually not conducted. 22
23
B.4.8.2 Immunotoxicity studies 24
One aspect of immunotoxicological evaluation includes assessment of potential 25
immunogenicity (see B.4.1 and B.4.8.1). Many rDNA-derived biotherapeutics are 26
intended to stimulate or suppress the immune system and, therefore, may affect humoral 27
as well as cell-mediated immunity. Inflammatory reactions at the injection site may be 28
indicative of a stimulatory response. It is important, however, to recognize that simple 29
injection trauma and/or specific toxic effects caused by the formulation vehicle may 30
result in toxic changes at the injection site. The expression of surface antigens on target 31
WHO/BS/2013.2213 Page 52
cells may be altered, with implications for autoimmune potential. Immunotoxicological 1
testing strategies may require screening studies followed by mechanistic studies to clarify 2
such issues. Routine tiered testing approaches or standard testing batteries, however, are 3
not recommended for rDNA-derived biotherapeutics. 4
5
The following modes of action might require special attention (37): 6
• A mode of action that involves a target which is connected to multiple signaling 7
pathways (target with pleiotropic effects), e.g. leading to various physiological effects, 8
or targets that are ubiquitously expressed, as often seen in the immune system. 9
• A biological cascade or cytokine release including those leading to an amplification 10
of an effect that might not be sufficiently controlled by a physiologic feedback 11
mechanism (e.g., in the immune system or blood coagulation system). The so-called 12
“cytokine release syndrome” (CRS) is characterized by the uncontrolled release of 13
cytokines (such as interleukin-6, tumor necrosis factor or interferon gamma). CD3 or 14
CD28 (super-) agonists might serve as an example. In severe cases, a “cytokine 15
storm” (hypercytokinemia) with potentially fatal consequences might be induced (38). 16
Currently available tests for prediction of the potential of a rDNA-derived biotherapeutic 17
with imunmodulatory properties to induce a CRS could for example include, on a case-18
by-case basis, whole blood assays, peripheral blood mononuclear cell (PBMC)-based 19
assays and biomimetic cell models (39). 20
21
B.4.8.3 Tissue cross-reactivity studies 22
Tissue cross-reactivity (TCR) studies are in vitro tissue-binding assays employing 23
immunohistochemical (IHC) techniques conducted to characterize binding of monoclonal 24
antibodies and related antibody-like products to antigenic determinants in tissues. Other 25
technologies can be employed in place of IHC techniques to demonstrate distribution to 26
target/binding site. 27
A TCR study with a panel of human tissues is a recommended component of the safety 28
assessment package supporting initial clinical dosing of these products (e.g. 40, 41). 29
However, in some cases the clinical candidate is not a good IHC reagent and a TCR study 30
might not be technically feasible. 31
WHO/BS/2013.2213 Page 53
TCR studies can provide useful information to supplement knowledge of target 1
distribution and can provide information on potential unexpected binding. Tissue binding 2
per se does not indicate biological activity in vivo. In addition, binding to areas not 3
typically accessible to the active substance in vivo (i.e. cytoplasm) is generally not 4
therapeutically relevant. Findings should be evaluated and interpreted in the context of 5
the overall pharmacology and safety assessment data package. 6
When there is unexpected binding (i.e. cross reactivity) to human tissues, a TCR 7
evaluation of selected tissues for the animal species chosen for the nonclinical toxicity 8
studies can provide supplemental information regarding potential correlations or lack 9
thereof, with preclinical toxicity. TCR using a full panel of animal tissues is not 10
recommended. 11
When a bi-specific antibody product will be evaluated in a TCR study using a panel of 12
human tissues, there is no need to study the individual binding components. 13
Evaluating the tissue binding of homologous products does not provide additional value 14
when TCR studies have been conducted with the clinical candidate in a human tissue 15
panel, and is not recommended. 16
TCR studies are not expected to detect subtle changes in critical quality attributes. 17
Therefore TCR studies are not recommended for assessing comparability of the test 18
article as a result of process changes over the course of a development program. 19
20
B.4.8.4 Impurities 21
Safety concerns may arise from the presence of impurities or contaminants. There are 22
potential risks associated with host cell contaminants whether derived from bacteria, 23
yeast, insect, plants, or mammalian cells. The presence of cellular host contaminants can 24
result in allergic reactions and other immunopathological effects. The adverse effects 25
associated with nucleic acid contaminants are theoretical but include potential integration 26
into the host genome (4). For products derived from insect, plant and mammalian cells, or 27
transgenic plants and animals there may be an additional risk of viral infections. 28
However, it is preferable to rely on manufacturing and quality control processes to deal 29
with these issues (section Part A) rather than to establish a preclinical testing program for 30
their qualification. 31
32
WHO/BS/2013.2213 Page 54
Part C. Clinical evaluation 1
2
C.1 Good Clinical Practice (GCP) 3
All clinical trials should be conducted under the principles described in the WHO 4
guidelines for good clinical practice for trials on pharmaceutical product (42). 5
6
C.2 Clinical Pharmacology (Phase I) 7
C.2.1 Initial safety and tolerability studies 8
Initial safety and tolerability studies start the first-in-human studies of drugs after the 9
completion of essential nonclinical studies (30-34, 43). The safety of clinical study 10
participants is the paramount consideration in proceeding to first-in-human studies. 11
Decisions on strategies for the development of a new medicine and the experimental 12
approaches used to assemble information relevant to the safety of first-in-human studies 13
must be science-based and ethically acceptable. Such studies should be closely monitored 14
and conducted with small numbers of healthy volunteers. However, products with 15
significant potential toxicity and those used for rare and/or life-threatening diseases are 16
normally studied in patients only. Study protocols should define stopping rules for 17
individual subjects, cohorts and the trial itself. Initial safety and tolerability studies are 18
designed to detect common adverse reactions, the tolerated dose range and the potential 19
drug effect. The ultimate goal of the studies is to obtain adequate safety and 20
pharmacokinetic data to permit the design of sufficiently valid phase II studies. 21
22
Initial safety and tolerability studies should preferably be randomised, placebo-controlled 23
studies but may also be single-arm studies with no comparator; they may range from 24
single dose studies to studies involving multiple doses and lasting for an extended period 25
of time. Drug doses usually start at low levels, and study participants are monitored very 26
carefully as the dose is escalated. In some settings and depending on the study protocol, 27
individual participants receive only one dose (see also C.2.3 and C.2.4). 28
29
WHO/BS/2013.2213 Page 55
From a clinical perspective, rDNA-derived biotherapeutics have particular challenges in 1
comparison to chemically-derived small molecule drugs and present special safety issues 2
to be addressed in the initial safety and tolerability studies: 3
• The nonclinical data may not be completely predictive of safety in humans. In 4
particular, since rDNA-derived biotherapeutics typically contain non-host proteins and 5
polysaccharides, nonclinical studies are usually not predictive for immunogenicity, i.e. 6
a test species may not react to a rDNA-derived biotherapeutic which could cause 7
serious adverse reactions in humans, or a test species may react when humans do not. 8
• Data from healthy volunteers may also not be fully predictive of safety/efficacy in 9
patients, such as target-mediated effect associated with monoclonal antibodies. 10
• Unlike many small molecule drugs, rDNA-derived biotherapeutics have long half-11
lives. 12
13
Predicting the potential for severe adverse drug reactions for first-in-human use of an 14
investigational medicinal product, involves the identification of risk factors, which may 15
be related to: 1) the mode of action; 2) the nature of the target; and/or 3) the relevance of 16
animal models. High-risk biologicals (e.g. TGN1412, an anti-CD28 superagonist which 17
caused an acute cytokine storm in humans that was not predicted from animal studies) 18
require extended safety measures, which may include strict sequential inclusion of trial 19
participants with clear stopping rules and extremely careful calculation of the first dose in 20
man (43). 21
22
The toxicity of most rDNA-derived biotherapeutics is related to their targeted mechanism 23
of action; therefore, relatively high doses can elicit adverse effects which are apparent as 24
exaggerated pharmacology. A rationale should be provided for dose selection taking into 25
account the characteristics of the dose-response relationship. PK-PD approaches (e.g. 26
simple exposure-response relationships or more complex modeling and simulation 27
approaches) can assist in high dose selection. Where in vivo/ex vivo PD endpoints are not 28
available, the high dose selection can be based on PK data and available in vitro binding 29
and/or pharmacology data. 30
31
WHO/BS/2013.2213 Page 56
C.2.2 Pharmacogenomics 1
Pharmacogenomic studies performed early during drug development can provide useful 2
information for the design of robust phase III trials, such as identifying receptor, genetic 3
or phenotypic characteristics and drug response in populations; using biomarkers to 4
identify dose response in individuals; and identifying patients with genetic 5
polymorphisms whose drug dosages should be adjusted for improved safety and/or 6
efficacy or for whom a particular treatment should not be used (44, 45). However, 7
pharmacogenomic effects are not commonly seen with rDNA-derived biotherapeutics. 8
The most recent guidance documents on this topic from appropriate regulatory agencies 9
should be consulted. 10
11
C.2.3 Pharmacokinetics 12
PK profile is an essential part of the basic description of a medicinal product and should 13
always be investigated. PK studies should be performed for the intended dose range and 14
routes of administration (11). In general, the PKs (absorption, distribution and 15
elimination) of rDNA-derived biotherapeutics should be characterized during single-dose 16
and steady-state conditions in relevant populations. However, historically, the PK 17
evaluation of peptide or protein products has suffered from limitations in the assay 18
methodology thus limiting the usefulness of such studies. Immunoassays and bioassays 19
are most frequently used for assaying therapeutic proteins in biological matrices. Special 20
emphasis should, therefore, be given to the analytical method selected and its capability 21
to detect and follow the time course of the protein (the parent molecule and/or 22
degradation and/or metabolic products) in a complex biological matrix that contains 23
many other proteins. The method should be optimized to have satisfactory specificity, 24
sensitivity and a range of quantification with adequate accuracy and precision (11). 25
26
The choice of the study population as well as the choice of single-dose and/or multiple-27
dose studies should be justified (11). If part of the PK information is gathered in healthy 28
volunteers, the validity of extrapolation of that information to the target population needs 29
to be addressed (46). A prospective plan for defining the dosing schedule based on 30
observed/calculated PK parameters should be developed and included in the PK study 31
WHO/BS/2013.2213 Page 57
protocol (47). It should be kept in mind that changes in the manufacturing process may 1
alter the PKs of rDNA-derived biotherapeuticss. 2
3
C.2.3.1 Absorption 4
The majority of biological products are administered parenterally through intravenous, 5
subcutaneous or intramuscular administration. Alternative routes proposed for delivery of 6
proteins may be considered, e.g. nasal and pulmonary administration, which bypass the 7
interstitial subcutaneous or intra-muscular environment. Oral delivery of proteins for 8
systemic effects is still rare due to low bioavailability (46). 9
10
Unless the intravenous route is exclusively used, appropriate in vivo studies should be 11
conducted in healthy volunteers or patients to describe the absorption characteristics of 12
the rDNA-derived biotherapeutics, i.e. the rate and extent of absorption. Single-dose 13
studies are generally sufficient to characterize absorption and to compare different 14
administration routes (48). It should be noted that the rate of absorption following 15
intramuscular or subcutaneous administration may vary according to the site and depth of 16
the injection, the concentration and volume of the solution injected, and may be 17
influenced by patient specific factors (46, 48). 18
19
Protein therapeutics administered by the subcutaneous route exhibit limited transport into 20
blood capillaries and enter the systemic circulation indirectly through the lymphatics. 21
Passage through the lymphatic system usually results in pre-systemic elimination and 22
consequently a bioavailability of less than 100% is obtained. In addition, small proteins 23
may undergo proteolytic degradation in tissues as a first-pass mechanism (46). Since 24
proteases can be affected by disease states and are reported to be upregulated with disease 25
progression, consideration should be given to patient specific circumstances (48). 26
27
C.2.3.2 Distribution 28
Tissue distribution studies should be undertaken unless otherwise justified. The volume 29
of distribution of a drug is determined largely by its physicochemical properties (e.g. 30
charge, lipophilicity) and its dependency on active transport processes. Because most 31
WHO/BS/2013.2213 Page 58
rDNA-derived biotherapeutics are large in size, their volume of distribution is usually 1
small and limited to the volume of the extracellular space due to their limited mobility 2
secondary to impaired passage through bio-membranes. Site-specific and target-oriented 3
receptor mediated tissue uptake and binding to intra- and extravascular proteins, however, 4
can substantially increase the volume of distribution of rDNA-derived biotherapeutics 5
(49). 6
7
The binding capacity to plasma proteins (albumin, α-acid glycoprotein) should be studied 8
when considered relevant (49). 9
PK calculations of steady-state volume of distribution may be problematic for some 10
rDNA-derived biotherapeutics. Noncompartmental determination using statistical 11
moment theory assumes first-order disposition processes with elimination occurring from 12
the rapidly equilibrating or central compartment. This basic assumption, however, is not 13
fulfilled for numerous recombinant peptide and protein products, as proteolysis in 14
peripheral tissues may constitute a substantial fraction of the overall elimination process 15
for such rDNA-derived biotherapeutics (49). There is an inverse correlation between 16
steady-state volume of distribution and molecular weights as well as for permeability and 17
molecular weight. Unlike small molecule chemical drugs, distribution to tissues (i.e. 18
cellular uptake) is often part of the elimination process and not part of the distribution 19
process as such, thus contributing to the small distribution volumes. Thus, a small steady-20
state volume of distribution should not necessarily be interpreted as low tissue 21
penetration and adequate concentrations may be reached in a single target organ due to 22
receptor mediated uptake (46). 23
24
C.2.3.3 Elimination 25
The main elimination pathway, including the major organs of elimination, should be 26
identified. Radiolabeled proteins can be used for this purpose (49). However, for 27
therapeutic proteins, the main elimination pathway in vivo can be predicted, to a large 28
extent, from the molecular size; thus, specific studies may not be necessary. 29
30
WHO/BS/2013.2213 Page 59
Breakdown products may have different PK profiles when compared with the parent 1
rDNA-derived biotherapeutics. However, in cases where measurement of separate active 2
peptide fragments is not technically feasible, the PKs of the active moiety could be 3
determined (46). 4
5
Catabolism of small proteins and peptides (molecular weight (MW) < 50,000 Da) appears 6
to occur mainly in the kidneys. The liver may also play a major role in the metabolism of 7
peptides and proteins, mediated by substance-specific enzymes such as for insulin, 8
glucagon, epidermal growth factor, antibodies, and tissue plasminogen activators (49). If 9
biliary excretion of peptides and proteins occurs, it generally results in subsequent 10
breakdown and metabolism of these compounds in the gastrointestinal tract (49). 11
Catabolism of proteins usually occurs by proteolysis via the same catabolic pathways as 12
for endogenous or dietary proteins. Proteolytic enzymes such as proteases and peptidases 13
are ubiquitously available throughout the body. Thus, locations of intensive peptide and 14
protein metabolism also include blood and various body tissues (49). 15
16
If elimination of the protein is largely dependent on target receptor uptake, differences in 17
receptor density between healthy volunteers and target populations, such as over-18
expression of receptors in tumors or inflamed tissues can create important 19
pharmacokinetic differences in half-life. These differences should be considered when 20
using healthy volunteer data for predictions to target population (46). 21
After subcutaneous administration of proteins with relatively rapid elimination, the rate 22
of absorption can be slower than the rate of elimination leading to longer apparent half-23
lives (flip-flop kinetics) and prolonged exposure when compared to intravenous 24
administration. As a consequence, dosing frequency may have to be reduced (50). 25
26
C.2.3.4 Subpopulations 27
The clinical development program should involve studies to support the approval in sub-28
populations such as patients with organ dysfunction. Whether such studies are necessary 29
depends on the elimination characteristics of the compound. If no study is conducted, this 30
should be justified by the applicant. An understanding of the influence of intrinsic factors, 31
WHO/BS/2013.2213 Page 60
such as age and body weight should be provided. Such information might arise from 1
dedicated studies in the respective population, or from population PK analyses of phase 2
II/III data (46). 3
4
Renal impairment 5
For proteins with MW lower than 50,000 Da, renal excretion is important for elimination 6
(increasing importance with lower MW) and consequently, for the half-life of the protein. 7
Thus, for these products, PK studies in patients with renal impairment are recommended. 8
It is also conceivable that renal impairment itself may affect functioning of other organs 9
and tissues (e.g. by up- or down-regulation of enzymes or receptors), thereby influencing 10
the PKs and/or PDs of the experimental compound. This should be taken into account in 11
the planning of the clinical pharmacology programme (46). 12
13
Hepatic impairment 14
Reduced hepatic function may decrease the elimination of a protein for which hepatic 15
degradation is an important elimination pathway. Where relevant, PK studies in patients 16
with different degrees of hepatic impairment are recommended (46). 17
18
C.2.3.5 Interaction studies 19
Therapeutic proteins may influence the pharmacokinetics of conventional drugs 20
metabolized by cytochrome P450 enzymes (CYPs) even if the proteins are not 21
metabolized by CYPs (51). Therefore it is important that drug interaction studies are also 22
conducted with therapeutic proteins. Additionally, since elimination of proteins may 23
involve capacity-limited steps such as receptor binding, the inhibition or induction of 24
receptors might impact pharmacokinetics. However, there is currently lack of knowledge 25
about suitable tools to explore such interactions. 26
27
Dose-and time dependency 28
The dose-concentration relationship may be non-proportional, depending on the relative 29
impact of capacity-limited barriers to distribution and elimination of the product. The 30
dose-proportionality should be evaluated in single- or multiple-dose studies and the 31
WHO/BS/2013.2213 Page 61
clinical consequences discussed. Time-dependent changes in PK parameters may occur 1
during multiple-dose treatment, e.g. due to down- or up-regulation of receptors 2
responsible for (part of) the elimination of the rDNA-derived biotherapeutics or to 3
formation of anti-drug antibodies. Using appropriate methods, soluble receptors may be 4
measured before treatment and during treatment, differentiating between free and bound 5
receptors. The effect on the PKs should be evaluated and the clinical relevance discussed 6
(34). 7
In special situations, apparent time-dependency may also originate from the fact that 8
immunologically active breakdown products may be slowly accumulating and have long 9
half-lives. It is recommended that PKs at several dose levels be determined on several 10
occasions during long-term studies. Population PK analysis of data from long-term trials 11
could be considered (46). 12
13
C.2.3.6 Pharmacokinetic data analysis 14
As for small molecule products the pharmacokinetics may be analyzed using 15
compartment- or non-compartment methods. The choice of the PK model used to derive 16
PK parameters should be justified. Mean (median) and individual results should always 17
be included in a license submission. The inter-subject variability should be estimated and, 18
if possible, the important sources of the variability, e.g. demographic factors such as 19
weight and age, should be identified. Potential additional sources of inter-subject 20
variability specific to therapeutic proteins are formation of antibodies, absorption 21
variability (e.g. differences in site of injection), variable levels of binding components in 22
blood, variability in target burden (e.g. tumor load), variability in degradation rate (e.g. of 23
de-pegylation) or in degradation pattern. Based on the results, individualized dosing 24
should be considered if necessary from safety and/or efficacy perspectives. For products 25
intended for multiple-dose administration, the variability within an individual should also 26
be quantified, since knowledge about the variability between occasions is valuable 27
especially for products for which titration is recommended. Population PK analysis of 28
phase II/III data using a sparse sample approach is recommended for characterizing the 29
pharmacokinetics, the variability of the PK parameters and possible covariate 30
relationships (46). Population analyses may thus support the individualization of doses. 31
WHO/BS/2013.2213 Page 62
1
C.2.4 Pharmacodynamics 2
In many cases, PD parameters are investigated in the context of combined PK/PD studies. 3
Such studies may provide useful information on the relationship between dose/exposure 4
and effect, particularly if performed at different dose levels. PD markers should be 5
selected based on their clinical relevance. 6
Studies in relevant animal models, if available, provide important information on the PD 7
properties of a biological medicinal product and may guide PD studies in humans. If no 8
animal model is available, a suitable human population must be chosen. In any case, 9
relevant PD effects should always be confirmed in human subjects with the disease that is 10
being targeted by the biological medicinal product. 11
Human PD studies are usually carried out during phase I or phase II studies. Phase II 12
studies can also be called proof of concept clinical studies and are important for the 13
subsequent development of the product by helping to determine the dose to be used in 14
further confirmatory trials, and providing some level of confidence that the biotherapeutic 15
is pharmacologically active and can do what it is intended to do. 16
17
C.2.5 Pharmacokinetic/pharmacodynamic relationship 18
The relationship between drug concentration and PD response (PK/PD relationship) 19
should be evaluated as part of drug development. If feasible, markers for both efficacy 20
and safety should be measured, preferably in the same study. It should be noted that PK 21
and PD for a biological medicinal product may not necessarily be entirely and fully 22
correlated (e.g. ceiling effect due saturation of target receptors) and both may be altered 23
by modifications to the molecule, binding to blood components, or formation of anti-drug 24
antibodies. Early pre-clinical and clinical data can be evaluated using appropriate models 25
for a mechanistic understanding of the disease and the PK/PD relationship. PK/PD 26
models may be developed accounting for the time-delay between plasma concentrations 27
and measured effect. The model may also need to take into account the presence or 28
absence of the therapeutic target (e.g. presence of antigen in case of anticancer 29
monoclonal antibodies). PK/PD models may allow extrapolation from volunteers to target 30
population given that suitable assumptions have been made, e.g. regarding the influence 31
WHO/BS/2013.2213 Page 63
of disease-related factors. These models may provide guidance for dose selection and are 1
helpful when interpreting changes in the PKs in important subpopulations or when 2
evaluating comparability in the context of a change in manufacturing process. Efforts to 3
explore relevant biomarkers and their link (surrogacy) to safety and efficacy endpoints 4
are encouraged (46). 5
6
C.2.6 Modifications of PK and PD profiles of therapeutic proteins 7
Many protein drugs display suboptimal therapeutic efficacies due to their inherent poor 8
molecular stability, low systemic bioavailability, and, as a consequence of their innate 9
susceptibility to various clearance mechanisms, short circulatory lifetimes. Higher protein 10
concentrations and increased dosing frequencies, are therefore, often employed to achieve 11
favourable therapeutic responses. Approaches to improve these factors, and thus in vivo 12
efficacy, include targeted mutations, generation of fusion proteins and conjugates, 13
glycosylation engineering, and pegylation (452). 14
Glycosylation may influence a variety of physiological processes at both the cellular (e.g. 15
intracellular targeting) and protein levels (e.g. protein-protein binding, protein molecular 16
stability, plasma persistence lifetimes). Since the glycosylation pattern of a biological 17
medicinal product may be influenced by even subtle changes in the manufacturing 18
process, the potential effects on PK and PD profiles need to be considered when 19
evaluating comparability of pre- and post-change product in the context of a change in 20
manufacturing process. Pegylation increases the size of a protein which prolongs its half-21
life by reducing renal clearance. Pegylation can also provide water solubility to 22
hydrophobic drugs and proteins. 23
24
C.3 Efficacy 25
C.3.1 Phase II 26
Phase II studies provide the first test of efficacy in patients with the disease targeted by 27
the rDNA-derived biotherapeutics. They aim at determining the correct dosage, 28
identifying common short-term side effects, and the best regimen to be used in pivotal 29
clinical trials. 30
WHO/BS/2013.2213 Page 64
Conventionally, the first step (frequently called phase IIa) is focused on an initial proof of 1
concept. This step is to demonstrate that the rDNA-derived biotherapeutics interacts 2
correctly with its molecular target and, in turn, alters the disease or its symptoms. 3
Subsequent trials (frequently called phase IIb trials) are larger and may use placebo 4
and/or active comparator agents and a broader dosage range to obtain a much more robust 5
proof of concept and additional guidance on dose selection. 6
For initial proof of concept, single-arm trials may be used with their results interpreted 7
relative to historical control subjects. However, this design could introduce bias since, for 8
example, current study participants may be different from historical control subjects in 9
ways that affect the outcome of interest or because of changes in supportive care that may 10
limit the validity of the conclusions. Therefore, comparative randomized phase II trials 11
are generally preferred. 12
13
Phase II trials usually explore a variety of possible endpoints (e.g. time-to-event 14
endpoints, change in a continuous endpoint of tumor size), and provide opportunities for 15
biomarker discovery. A variety of study designs can be used, including the randomized 16
parallel-group design, randomized discontinuation design, single-stage and two-stage 17
designs, delayed-start design, and adaptive (Bayesian) designs. In all cases, clear decision 18
rules should be in place. 19
Standard study designs for assessing dose-response have been described (53), such as 20
randomized parallel dose response studies. However, the approaches to selecting the 21
optimal dose may differ for rDNA-derived biotherapeutics compared to small chemical 22
molecules. For example, biological agents developed in oncology are usually cytostatic 23
and their maximal activities may occur at doses lower than their maximum-tolerated 24
doses. 25
Combination therapy is an important treatment modality in many disease settings such as 26
cancer. Increased understanding of the pathophysiological processes that underlie 27
complex diseases has provided further impetus for therapeutic approaches using 28
combinations of (new) products directed at multiple therapeutic targets to improve 29
treatment response, minimize development of resistance or improve tolerability. This 30
WHO/BS/2013.2213 Page 65
requires the use of flexible designs and new modeling approaches for the design of 1
clinical trials. 2
As observed for small chemical drugs, rDNA-derived biotherapeutics may affect cardiac 3
electrical activity either directly or indirectly. The amount and type of electrocardiogram 4
data considered appropriate should be individualized based on the type of product and the 5
nonclinical findings regarding its cardiotoxic potential. A Thorough QT/QTc (TQT) 6
Study (54) or a study that incorporates many of the key components of a TQT study 7
should be considered (54). However, this may not be necessary if electrocardiogram data 8
in at least a subset of patients are collected during clinical development and reviewed by 9
respective experts, preferably in a blinded manner. 10
11
C.3.2 Confirmatory phase III 12
Phase III clinical trials are designed to evaluate the benefit of the rDNA-derived 13
biotherapeutics in a carefully selected patient population with the disease. These trials are 14
to confirm efficacy at the chosen dose(s) and dosing regimen(s), to further evaluate safety 15
and monitor side effects, and sometimes to compare the candidate product to commonly 16
used treatments. For common conditions, phase III studies are usually conducted with 17
large populations consisting of several hundred to several thousand participants who have 18
the disease or the condition of interest. 19
20
Confirmatory trials should be prospective randomized trials comparing the test agent 21
against placebo (in addition to the best supportive care) or an active comparator, usually 22
the best available, evidence-based current standard. If no such comparator is available 23
(e.g. in patients who have failed several lines of therapies), the comparator may be the 24
investigator’s best choice. Ideally, trials should be double blinded, where neither the 25
patient nor the investigator knows the nature of the product received by the patient. 26
Blinding or masking is intended to limit the occurrence of conscious or unconscious bias 27
in the conduct and interpretation of a clinical trial (55). 28
29
The design of the trials depends on the hypothesis to be tested: superiority to 30
placebo/active comparator, equivalence or non-inferiority to an active comparator (56). 31
WHO/BS/2013.2213 Page 66
The choice of endpoints depends on the therapeutic indication; there should be sufficient 1
evidence that the primary endpoint can provide a valid and reliable measure of clinically 2
relevant and important treatment benefit in the targeted patient population. If a single 3
primary variable cannot be selected, a composite endpoint integrating or combining 4
multiple measurements into a single variable, using a pre-defined algorithm, can also be 5
used; such validated endpoints are commonly used in inflammatory diseases (e.g. ACR20 6
in rheumatoid arthritis, ASAS20 in ankylosing spondylitis, CDAI in Crohn’s disease, 7
PASI in psoriasis) or in oncology (disease progression, disease-free survival, or overall 8
survival). Patient reported outcomes and quality of life scales are also important 9
endpoints, which may already be included in some of these composite endpoints. 10
11
When direct assessment of the clinical benefit to the patient is not practical, a surrogate 12
endpoint may be considered if it is a reliable predictor of clinical benefit. The strength of 13
the evidence for surrogacy depends upon (i) the biological plausibility of the relationship, 14
(ii) the demonstration of the prognostic value of the surrogate for the clinical outcome in 15
epidemiological studies and (iii) evidence from clinical trials that treatment effects on the 16
surrogate correspond to effects on the clinical outcome. Most surrogate endpoints are not 17
formally validated, but such endpoints can be used if they are reasonably likely to predict 18
the desired clinical benefit, e.g. the effect on tumor size, as assessed by imaging, in 19
patients refractory to available treatments. 20
Specific decisions about the size of the study group will depend on such factors as the 21
magnitude of the effect of interest, characteristics of the study population, and study 22
design (see C.4). 23
24
Preferably, two confirmatory trials should be performed in order to show that the results 25
can be replicated. However, one controlled study with statistically compelling and 26
clinically relevant results may be sufficient, especially in life-threatening conditions or 27
rare disorders. If the biological medicinal product shows promising efficacy for a 28
serious/life-threatening condition where no other treatment option exists, licensing based 29
on a limited amount of data may be possible with further confirmatory efficacy data 30
being provided post-marketing. Because most rare diseases have a more homogeneous 31
WHO/BS/2013.2213 Page 67
genetic pattern than common diseases and because they are often characterized by similar 1
or identical genetic or epigenetic defects, patients with these diseases could be expected 2
to have a more uniform therapeutic response. This should reduce the size of phase III 3
studies required to demonstrate efficacy. The use of historical controls (or possibly no 4
controls) may also be justified if the rare disease has a defined course in the absence of 5
treatment that will permit comparisons with the results for the investigational rDNA-6
derived biotherapeutics. 7
8
C.3.3 Biomarkers 9
The identification of laboratory based disease biomarkers has the potential to enhance the 10
benefit-risk profile of rDNA-derived biotherapeutics by enabling the selection of patients 11
that are more likely to respond, especially with molecules that target serum or cell 12
markers. In such case, the treatment may only benefit a subset of patients defined by the 13
biomarker, e.g. patients positive for HER-2 or negative for KRAS tumors. The biomarker 14
evaluation process should consist of the following three steps: (i) analytical validation; 15
(ii) qualification, i.e. assessment of available evidence on associations between the 16
biomarker and disease states, including data showing effects of interventions on both the 17
biomarker and clinical outcomes; and (iii) utilization, i.e. contextual analysis based on the 18
specific use proposed and the applicability of available evidence to this use (56-58). 19
Biomarker qualification should not be part of pivotal phase III trials. 20
21
C.3.4 Manufacturing and formulation changes 22
While manufacturing and formulation changes may be expected during product 23
development, the phase III trials should be conducted with the test rDNA-derived 24
biotherapeutics manufactured according to the final manufacturing (commercial) process. 25
If this is not the case, a comparability exercise between the clinical and commercial 26
products is necessary to ensure that the change would not have an adverse impact on the 27
clinical performance of the product (27, 28). This comparability exercise should normally 28
follow a stepwise approach, starting with a comparison of quality attributes of the active 29
substance and relevant intermediates. A comparability exercise should not be limited to 30
release testing but should include more extensive characterization using a range of 31
WHO/BS/2013.2213 Page 68
suitable analytical methods, as appropriate to the product and process changes in question 1
(see A.9). If differences are detected that may have an impact on the clinical properties of 2
the product, nonclinical and/or clinical bridging studies may be needed such as PK/PD 3
studies, and possibly immunogenicity studies. 4
5
C.3.5 Special populations 6
As in any clinical development programme, studies in special populations would be 7
expected where relevant to the indications, e.g. in the elderly and in paediatric patients. 8
The elderly population is arbitrarily defined as comprising patients aged 65 years or 9
older. However, patients 75 years and above, should also be considered to the extent 10
possible (59). Recommended age categories for the paediatric population include preterm 11
and term newborn infants, infants to toddlers, children and adolescents (60). 12
13
Some rDNA-derived biotherapeutics may be of particular importance to elderly patients, 14
such as those developed for cancer, Parkinson’s disease, Alzheimer’s disease, coronary 15
heart disease or diabetes mellitus. It is important to determine whether the PK profile of a 16
rDNA-derived biotherapeutics is different in elderly compared to younger subjects since 17
impairment of organ function such as renal or hepatic function is more frequent in an 18
aged population. The elderly subpopulation should also be represented sufficiently in the 19
clinical trials to permit the comparison of treatment effects, dose response and safety 20
between older and younger patients. Where the disease to be treated is characteristically 21
associated with aging, it is expected that elderly patients will constitute the major portion 22
of the clinical database (59). 23
The extent of the studies needed in children depends on the possibility of extrapolation 24
from adults and children of other age groups. Some rDNA-derived biotherapeutics may 25
be used in children from the early stages of drug development, especially those targeting 26
genetic diseases where manifestations occur early in life. Evaluation should be made in 27
the appropriate age group and it is usually recommended to begin with older children 28
before extending the trial to younger children and then infants (60). Where justified, 29
extrapolation of efficacy data from adult to paediatric patients may be based on PK 30
and/or PD data, e.g. when a similar effect can be expected with similar exposure. 31
WHO/BS/2013.2213 Page 69
However, safety data usually cannot be extrapolated and need to be generated in children 1
(see C.5). 2
3
C.3.6 Post-marketing: Phase IV 4
Phase IV trials may be required to further evaluate an approved rDNA-derived 5
biotherapeutic and obtain more information about safety or effectiveness or both, 6
especially if it has been approved on the basis of a surrogate endpoint. 7
8
C.4 Statistical Considerations 9
C.4.1 General considerations 10
The application of sound statistical principles to the design, conduct, analysis and 11
interpretation of clinical trials should be considered an important and integral component 12
of the overall development of a rDNA-derived biotherapeutic. The success of a trial 13
depends on the appropriateness of the study design, trial conduct and analysis of trial 14
results. Statistical principles are relevant to all three aspects of the clinical trial. In general, 15
details regarding these aspects should be specified in the trial protocol, which should be 16
written and finalized prior to the start of the trial. Any subsequent amendments to the 17
protocol should be clearly justified and documented in a formal amendment to the 18
protocol, and should include the statistical consequences of the proposed changes. 19
The scientific integrity of the trial and the credibility of the data from the trial depend 20
substantially on the trial design (61). The study protocol should include a clear 21
description of the specific design selected for a particular trial. Additional details 22
regarding the primary endpoint, which is directly related to the primary objective of the 23
trial, should also be included. If multiple primary endpoints are defined, the criteria for 24
achieving study success should be clearly laid out in order to avoid potential problems 25
with the interpretation of the trial results. The protocol should also clearly define 26
secondary endpoints, and their role in the interpretation of the trial results should be 27
stated. Details regarding measures that have been put in place to avoid or minimize bias 28
in the trial should also be provided, for example, randomization and blinding. 29
30
WHO/BS/2013.2213 Page 70
With regards to the type of hypothesis to be tested in a specific trial, it should be clear in 1
the protocol whether the trial is designed to show superiority, non-inferiority, or 2
equivalence. The statistical issues involved in the design, conduct, analysis and 3
interpretation of equivalence and non-inferiority trials are complex and subtle, and 4
require that all aspects of these trials be carefully evaluated. Sample size and power are 5
important for the success of a clinical trial, and should be given careful consideration at 6
the trial design stage. In determining sample size, the specific hypothesis being tested 7
should be taken into consideration. 8
9
It is important to ensure that the protocol will provide good quality data that permit an 10
adequate evaluation of the efficacy (and safety) of the product under development. In 11
addition, if formal interim analyses are planned, then the details governing such analyses 12
should be pre-specified in the protocol. 13
14
In an era when it is recognized that improvements in the drug development process are 15
needed in order to increase the likelihood of trial success, decrease costs, and increase the 16
efficiency with which efficacious and safe medicines are brought to market, adaptive 17
clinical trial designs are increasingly considered as one tool through which these 18
improvements can be achieved. Adaptive design refers to a clinical study design that 19
uses accumulating data to decide how to modify aspects of the study as it continues, 20
without undermining the validity and integrity of the trial (62, 63). A key statistical issue 21
for adaptive designs is the preservation of the Type I error rate. The methods used to 22
properly control the Type I error rate should be described in the study protocol, with 23
additional details provided in the Statistical Analysis Plan (SAP). 24
25
Details regarding the statistical methodology to be applied to the clinical trial should be 26
provided in the protocol, with the more technical details being captured in the SAP. The 27
SAP should be prepared and finalized prior to un-blinding the clinical study. Any 28
amendments to the SAP must also be finalized prior to un-blinding. 29
30
C.4.2 Specific considerations for rDNA-derived biotherapeutics 31
WHO/BS/2013.2213 Page 71
Since rDNA-derived biotherapeutics are often indicated to treat severe and/or life 1
threatening diseases and chronic diseases, trials for rDNA-derived biotherapeutics present 2
unique statistical challenges. 3
4
Trials in small populations and single arm studies 5
Some rDNA-derived biotherapeutics are intended for the treatment of rare diseases for 6
which the target population is very small. Consequently, trials that are considered 7
confirmatory for rare disease indications are often based on a limited number of subjects 8
While such studies must still be designed with the rigor of traditional trials, and should be 9
conducted with high quality in order to provide reliable and valid data for assessing 10
efficacy and safety, some flexibility is needed with regards to the statistical methods that 11
will be utilized in such trials. Single arm studies with comparisons being made to an 12
external control can sometimes be justified. 13
14
Tumor-based endpoints in oncology trials and composite endpoints 15
In confirmatory oncology trials for rDNA-derived biotherapeutics, the use of tumor-based 16
endpoints such as disease-free survival and progression-free survival as the primary 17
endpoint is not uncommon (64). The use of tumor-based endpoints as the primary 18
endpoint creates several statistical challenges, and considerations for the collection and 19
analysis of such endpoints have been discussed (e.g. 65). Clinical trials may involve the 20
use of a composite primary endpoint, arising from the combination of multiple clinical 21
measurements or outcomes (e.g. major adverse cardiac events (MACE): the most 22
commonly used composite endpoint in cardiovascular studies). For such a composite 23
endpoint, it is important that the individual components are analyzed separately (usually 24
as secondary endpoints) in order to ensure that the treatment effect is shown across all 25
components, and is of similar magnitude. 26
27
Missing data 28
Missing data is a common problem in long-term trials of rDNA-derived biotherapeutics 29
targeting chronic diseases, for example, diabetes and rheumatoid arthritis, while it is 30
usually not a problem in short-term trials. The impact of missing data on the validity of 31
WHO/BS/2013.2213 Page 72
trial results should be carefully assessed using sensitivity analyses with appropriate 1
underlying assumptions. 2
3
C.5 Safety 4
Pre-licensing safety data should be obtained in a sufficient number of patients to 5
characterize and quantify the safety profile of the rDNA-derived biotherapeutics 6
including type, frequency and severity of adverse drug reactions (ADRs). The safety 7
evaluation should cover a reasonable duration of time, taking into account the intended 8
duration of use of the drug, to assess potential changes in the ADR profile over time and 9
to capture delayed ADRs. 10
11
For drugs intended for long-term treatment of non-life-threatening conditions, a 12-month 12
exposure of at least 100 patients to the investigational medicinal product at the intended 13
clinical dosage should be considered (66). When no serious ADR is observed in a one-14
year exposure period, this number of patients can provide reasonable assurance that the 15
true cumulative one year incidence is no greater than 3%. This estimate is based on the 16
statistical “rule of three”, which states that if no major ADR occurred in a group of n 17
people, there can be 95% confidence that the chance of a major ADR is less than one in n 18
/ 3 (or equivalently, less than 3 in n). This estimate is considered a good approximation 19
for n > 30. 20
The safety database may need to be larger or may require longer patient observation if a 21
safety signal is identified, if the drug is expected to cause late developing ADRs, or if 22
ADRs increase in severity or frequency over time. Concerns requiring a larger safety 23
database may arise from nonclinical or early clinical data or from experience with other 24
products of the same or related pharmacologic class. A smaller safety database may be 25
acceptable if the intended treatment population is small. 26
Safety data should be obtained from prospective and preferably controlled studies 27
including a placebo or active comparator arm since comparison with an external control 28
group (e.g. with published data) is usually hampered by differences in the investigated 29
patient population, concomitant therapy, observation period and/or reporting. Causality 30
WHO/BS/2013.2213 Page 73
assessment, i.e. whether the observed adverse event is causally related to the 1
investigational drug, is usually easiest in placebo-controlled studies. 2
Generally accepted definitions and terminology as well as procedures to harmonize the 3
way to gather and, if necessary, to take action on important clinical safety information 4
arising during clinical development are important (67). The term “adverse event” (AE) 5
describes any untoward medical occurrence developing with administration of a 6
pharmaceutical product irrespective of a causal relationship. The term “adverse drug 7
reaction” (ADR), on the other hand, should only be used for adverse events that have at 8
least a reasonable possible causal relationship to the pharmaceutical agent. 9
10
Standardized reporting is important for transmission of pre- or post-marketing safety 11
information, for example, between reporting source or pharmaceutical industry and 12
regulatory authorities or between regulatory authorities and WHO Collaborating Center 13
for International Drug Monitoring (68). Data elements to be included in individual case 14
safety reports should comprise all important information on the primary source, date, 15
sender and receiver of the information, the type, seriousness, duration and outcome of the 16
AE or ADR, detailed patient characteristics and drug information, actions taken with the 17
drug (e.g. dose reduction, discontinuation), and an assessment of the degree of suspected 18
relatedness of the drug to the AE (68). 19
To facilitate sharing of regulatory safety information internationally for medical products 20
used by humans, specific MedDRA terminology has been developed, a rich and highly 21
specific standardized medical terminology for accurate and consistent safety information 22
allowing for aggregation of reported terms in medically meaningful groupings 23
(69). Products covered by the scope of MedDRA include pharmaceuticals, vaccines and 24
drug-device combination products. 25
26
Since safety data obtained from pre-marketing clinical trials can be expected to detect 27
mainly common and shorter-term ADRs, further monitoring of clinical safety of the 28
biological product to detect rare but sometimes serious adverse effects and an ongoing 29
benefit-risk evaluation are necessary in the post-marketing phase (see C.7). 30
31
WHO/BS/2013.2213 Page 74
C.5.1 Special populations 1
Elderly population 2
The safety of rDNA-derived biotherapeutics should be investigated in elderly patients 3
during clinical drug development (70, 71), except where there is no intention to use the 4
rDNA-derived biotherapeutics in this age group. Elderly patients are more prone to 5
adverse effects since they often have co-morbidities and are taking concomitant 6
medication that could interact with the investigational drug. The adverse effects can be 7
more severe, or less tolerated, and may have more serious consequences than in the non- 8
elderly population. Depending on the mechanism of action of the drug and/or the 9
characteristics of the disease, specific effects on cognitive function, balance and falls, 10
urinary incontinence or retention, weight loss and sarcopenia should be investigated. 11
Elderly patients may be included in the main Phase III or Phase II/III studies or in 12
separate studies. Inclusion of younger and elderly patients in the same studies has the 13
advantage of allowing direct comparisons using data collected in similar ways. Certain 14
assessments, however, such as studies of cognitive function, require special planning and 15
can be best accomplished in separate studies. 16
Where enrolment of elderly patients has been insufficient despite the efforts of the 17
applicant, a specific plan to collect data post-marketing should be presented in the 18
marketing application. 19
20
Paediatric population 21
Data on the safety of medicinal products in the pediatric population should be generated 22
unless their use is clearly inappropriate (60). During clinical development, the timing of 23
pediatric studies will depend on the medicinal product, the type of disease being treated, 24
safety considerations, and the efficacy and safety of alternative treatments. Justification 25
for the timing and the approach to the clinical program needs to be clearly addressed with 26
regulatory authorities. 27
Medicinal products may affect physical and cognitive growth and development, and the 28
adverse event profile may differ in pediatric compared to adult patients. In addition, 29
adverse effects may not be apparent immediately, but only at a later stage of development. 30
Long-term studies or surveillance data while patients are on chronic therapy and/or 31
WHO/BS/2013.2213 Page 75
during the posttherapy period, may be needed to determine possible effects on skeletal, 1
behavioral, cognitive, sexual, and immune maturation and development. 2
3
C.6 Immunogenicity 4
rDNA-derived biotherapeutics may induce unwanted humoral and/or cellular immune 5
responses in recipients. Immunogenicity of rDNA-derived biotherapeutics should 6
therefore always be investigated pre-authorization (72). Since animal data are usually not 7
predictive of the immune response in humans, immunogenicity needs to be investigated 8
in the target population. Although in-silico modeling may help to identify T-cell epitopes 9
related to immunogenicity (i.e. T-helper epitopes) this does not predict whether 10
immunogenicity will occur. The frequency and type of anti-drug antibodies induced as 11
well as possible clinical consequences of the immune response should be thoroughly 12
assessed. 13
14
The immune response against a biotherapeutic is influenced by many factors such as the 15
nature of the drug substance, product- and process-related impurities (e.g. host-cell-16
proteins, aggregates), excipients and stability of the product, route of administration 17
(subcutaneous administration usually more immunogenic than intravenous 18
administration), dosing regimen (intermittent use usually more immunogenic than 19
continuous use), and patient-, disease- and/or therapy-related factors (e.g. antibody 20
development more likely in immune-competent than immunosuppressed state and 21
potentially enhanced in the presence of autoimmune disease). The consequences of 22
unwanted immunogenicity on safety may vary considerably, ranging from clinically 23
irrelevant to serious and life-threatening (e.g. serious infusion/anaphylactic) reactions. 24
Neutralizing antibodies directly alter the PD effect of a product (i.e. by blocking the 25
active site of the protein) leading to reduced or loss of efficacy. Binding antibodies often 26
affect pharmacokinetics and, thereby, may indirectly influence pharmacodynamics. Thus, 27
an altered effect of the product over time due to anti-product antibody formation might be 28
a composite of pharmacokinetic, PD and safety effects. 29
30
WHO/BS/2013.2213 Page 76
The proposed antibody testing strategy should be appropriately justified including the 1
selection, assessment, and characterization of assays, identification of appropriate 2
sampling time points (including baseline samples), sample storage and processing as well 3
as selection of statistical methods for analysis of data (72). The studies to be considered 4
for immunogenicity testing (e.g. short-term and / or long-term clinical trials or even 5
single-dose studies) and the sampling time points depend on the expected appearance of 6
antibodies: some rDNA-derived biotherapeutics are highly immunogenic and may elicit 7
an immune response already after the first dose, others may require prolonged or 8
intermittent exposure to mount an immune response and some may have a very low 9
immunogenic potential, rarely leading to any antibody development . 10
11
Antibody assays should be validated for their intended purpose. Validation studies need 12
to establish appropriately linear responses to relevant analytes as well as appropriate 13
accuracy, precision, sensitivity, specificity and robustness of the assay(s). A highly 14
sensitive screening assay should be used for antibody detection and a confirmatory assay 15
to confirm the presence of antibodies and eliminate false positive results. To achieve 16
confirmation of specificity, it is necessary to include an assay which evaluates specificity. 17
A neutralization assay should be available for further characterization of antibodies, if 18
present, unless the development of neutralizing antibodies is unlikely (e.g. based on 19
experience with the substance class). Possible interference of the circulating antigen with 20
the antibody assay(s) should be taken into account. 21
22
If the rDNA-derived biotherapeutics are a monoclonal antibody (mAb), the development 23
of assays to detect antibodies against this mAb can be technically challenging (73). Many 24
standard assay formats involve the use of anti-immunoglobulin reagents such as 25
antibodies against immunoglobulins, protein A or protein G, but these are inappropriate 26
for use in detecting antibodies against mAbs as they very often bind to the product itself. 27
Different assay approaches have been developed to overcome this problem such as the 28
‘bridging’ enzyme-linked immunosorbent assay (ELISA) format or Surface Plasmon 29
Resonance (SPR) procedures which do not require anti-immunoglobulin reagents but 30
may be less sensitive than other immunoassay methods (2). 31
WHO/BS/2013.2213 Page 77
Detected antibodies should be further characterized with regard to antibody content 1
(concentration/titre) and, depending on case-by-case considerations, possibly other 2
criteria such as antibody class and subclass (isotype), affinity and specificity. For 3
example, the isotype of the antibodies could be determined if this may be predictive of 4
safety (such as development of IgE antibodies causing allergic and anaphylactic 5
responses). Potential clinical implications of detected antibodies regarding safety, 6
efficacy and pharmacokinetics should always be evaluated. Special attention should be 7
paid to the possibility that the immune response seriously affects the endogenous protein 8
and its unique biological function (e.g. neutralizing anti-epoetin antibodies cross-reacting 9
with endogenous erythropoietin and causing pure red cell aplasia). 10
11
The required observation period for immunogenicity testing will depend on the intended 12
duration of therapy and the expected time of antibody development, if known, and should 13
be justified. In the case of chronic administration, one-year data will usually be 14
appropriate pre-licensing to assess antibody incidence and possible clinical implications. 15
If considered clinically relevant, development of antibody titers, their persistence over 16
time, potential changes in the character of the antibody response and the possible clinical 17
implications should be assessed pre- and post-marketing. 18
Since pre-licensing immunogenicity data are often limited, further characterization of the 19
immunogenicity profile may be necessary post-marketing, particularly, if rare but 20
clinically meaningful or even serious antibody-related ADRs have been encountered with 21
biological agents of the same or related substance class that are not likely to be detected 22
in the pre-marketing phase. 23
24
C.7 Pharmacovigilance and risk management planning 25
NRAs should be vigilant that the health of the public is protected. The aim is to ensure 26
that the risks associated with rDNA-derived biotherapeutics are actively minimized. 27
Patient safety is a key concern for all medicinal products that are on the market: rDNA-28
derived biotherapeutics are no exception. Due to the specific characteristics of rDNA-29
derived biotherapeutics already discussed in this Guideline, pharmacovigilance (PhV) 30
activities required for rDNA-derived biotherapeutics may differ in some respects from 31
WHO/BS/2013.2213 Page 78
those required for small molecule drugs. For example, biotherapeutic use may lead to 1
antibody formation with consequences for clinical efficacy and/or safety. 2
3
In some countries, a risk management plan (RMP) should be submitted and agreed to by 4
the NRA. The key components of a RMP may include: 5
(1) Safety Specifications, which summarize the known and potential safety issues and 6
missing information about the rDNA-derived biotherapeutics; 7
(2) A PhV plan to further evaluate important known or potential safety concerns and, 8
provide post-marketing data where relevant information is missing; and 9
(3) A risk minimization plan (RMinP), which provides proposals on how to minimize any 10
identified or potential safety risk. 11
12
In the RMP, the known or potential risks may be described with PhV and risk 13
minimization activities proposed to identify, characterize, prevent, or minimize risks 14
related to the use of the rDNA-derived biotherapeutics; to assess the effectiveness of 15
those interventions; and to communicate those risks to both patients and health care 16
providers. 17
PhV and risk minimization activities that might be included in an RMP usually fall into 18
two categories: routine activities, which would generally be conducted for any medicine 19
where no special safety concerns have arisen, and additional activities designed to 20
address identified and potential safety concerns that could have an impact on the benefit-21
risk balance of a product. Routine PhV activities would include the monitoring and 22
reporting of spontaneous adverse events post-approval and any safety evaluations 23
incorporated in clinical trials that may be initiated by the marketing authorization holder 24
following marketing authorization for a wide variety of reasons. In case of relevant safety 25
issues, NRAs may request additional PhV activities in the form of active surveillance (e.g. 26
registries), epidemiology studies, further clinical studies, and drug utilization studies. 27
Routine RMin activities would ensure that suitable contraindications and warnings are 28
included in the product information and that this information is updated on an ongoing 29
basis. A RMinP can further specify other risk minimization activities, as appropriate, 30
which could include: specific educational material about the product and its use, patient- 31
WHO/BS/2013.2213 Page 79
or physician-oriented training programs, restricted use of the product and registration 1
programs for patients, doctors and/or pharmacists. 2
Once on the market, manufacturers should monitor the effectiveness of their RMinPs, and 3
revise them if new safety and effectiveness concerns are identified. Changes in the 4
manufacturing processes introduced post-marketing could also influence the safety 5
profile (e.g. by enhancing immunogenicity) of rDNA-derived biotherapeutics and may 6
necessitate enhanced safety monitoring. 7
8
In case a relevant or even serious potentially drug-related adverse event occurs, it is 9
important to be able to identify the specific biological causing this event. Therefore, all 10
ADR reports should carry information unique to the product, including the proprietary 11
(brand) name, the INN, identification code (if there is one), and lot information of the 12
respective biological to quickly trace an ADR to a specific product and ascertain any 13
relation to causality. 14
An RMP will not reduce the scientific and clinical standards or the data requirements for 15
the market authorization of rDNA-derived biotherapeutics; nor will it replace the 16
precautionary approach that is taken to managing the risks associated with those products. 17
To the contrary, implementation of an RMP will further strengthen the rigor of post-18
market surveillance, allowing for earlier identification of risks associated with rDNA-19
derived biotherapeutics and earlier interventions to minimize those risks. 20
21
C.8 Additional guidance 22
Further guidance on various aspects of clinical trials is available from several other 23
bodies such as the ICH, the EMA, the US FDA as well as from several other NRAs. The 24
WHO guidelines are not intended to conflict with, but rather to complement these other 25
documents with respects to medicinal products prepared by rDNA technology. 26
27
28
WHO/BS/2013.2213 Page 80
Authors 1
The preliminary draft of this document was prepared after a WHO Drafting Group 2
meeting on the Guidelines to Assure the Quality, Safety and Efficacy of Biological 3
Products Prepared by Recombinant DNA Technology held at National Institute for 4
Biological Standards & Control (NIBSC), UK on 19-20 March 2012, by the following 5
members: Dr Marie-Christine Bielsky, Medicines and Healthcare products Regulatory 6
Agency (MHRA), London, UK; Dr Elwyn Griffiths, WHO consultant, Kingston-upon-7
Thames, UK; Dr Hans-Karl Heim, Federal Institute for Drugs and Medical Devices (in 8
Germany, Bundesinstitut für Arzneimittel und Medizinprodukte; BfArM), Bonn, 9
Germany; Dr Hye-Na Kang , HIS/EMP, WHO, Geneva, Switzerland; Dr Robin Thorpe, 10
NIBSC, Potters Bar, UK; Dr Meenu Wadhwa, NIBSC, Potters Bar, UK; Dr Martina 11
Weise, BfArM, Bonn, Germany. 12
13
The first draft was prepared by the following authors for the parts indicated: 14
Introduction - Dr Elwyn Griffiths, WHO consultant, Kingston-upon-Thames, UK; Part A 15
- Dr Kowid Ho, Agence nationale de sécurité du médicament et des produits de santé 16
(ANSM), Anatole, France; Dr Robin Thorpe, NIBSC, Potters Bar, UK; Dr Meenu 17
Wadhwa, NIBSC, Potters Bar, UK; Part B - Dr Laura Gomes Castanheira, National 18
Health Surveillance Agency (in Portuguese, Agência Nacional de Vigilância Sanitária; 19
ANVISA), Brazilia, Brazil; Dr Hans-Karl Heim, BfArM, Bonn, Germany; Part C - Dr 20
Marie-Christine Bielsky, MHRA, London, UK; Dr Agnes Klein, Health Canada, Ottawa, 21
Canada; Dr Catherine Njue, Health Canada, Ottawa, Canada; Dr Jian Wang, Health 22
Canada, Ottawa, Canada; Dr Martina Weise, BfArM, Bonn, Germany; with support from 23
the WHO Secretariat: Dr Hye-Na Kang and Dr Ivana Knezevic, HIS/EMP, WHO, 24
Geneva, Switzerland; taking into considerations of the discussion at a WHO Informal 25
Consultation on the Revision of the Guidelines on the Quality, Safety and Efficacy of 26
Biological Medicinal Products Prepared by Recombinant DNA Technology held in 27
Xiamen, China, on 31 May – 1 June 2012, attended by: 28
29
WHO/BS/2013.2213 Page 81
Mrs Arpah Abas, Ministry of Health Malaysia, Selangor, Malaysia; Dr Wesal Salem 1
Alhaqaish, Jordan Food and Drug Administration, Amman, Jordan; Ms Jennifer Archer, 2
Hospira, Thebarton, Australia, representative of the International Generic Pharmaceutical 3
Alliance (IGPA); Dr Boontarika Boonyapiwat, Ministry of Public Health, Nonthaburi, 4
Thailand; Dr Laura Gomes Castanheira, ANVISA, Brasilia, Brazil; Dr Weihong Chang, 5
State Food and Drug Administration (SFDA), Beijing, China; Dr Ranjan Chakrabarti, 6
United States Pharmacopeia-India, Shameerpet, India, representative of the United State 7
Pharmacopoeial Convention; Mr Dusheng Cheng, Beijing Four Rings Bio-8
Pharmaceutical Co., Ltd., Beijing, China; Dr Liang Chenggang, National Institutes for 9
Food and Drug Control (NIFDC), Beijing, China; Dr Youngju Choi, Korea Food and 10
Drug Administration (KFDA), Osong, Korea; Ms Juliati Dahlan, National Agency of 11
Drug and Food Control, Jakarta, Indonesia; Mr Geoffrey Eich, Amgen Inc.Corporate 12
Services / Global Regulatory Affairs & Safety, Thousand Oaks, USA, representative of 13
the International Federation of Pharmaceutical Manufacturers and Associations (IFPMA); 14
Dr Kai Gao, NIFDC, Beijing, China; Mr Thomas Go, Health Sciences Authority (HSA), 15
Helios, Singapore; Dr Elwyn Griffiths, Kingston-upon-Thames, UK; Dr Lawrence Gu, 16
Shenyang Sunshine Pharmacetical Co. LTD., Shenyang, China; Dr Zhongping Guo, 17
Chinese Pharmacopoeia Commission, Beijing, China, representative of the Chinese 18
Pharmacopoeia; Dr Nazila Hassannia, Biological Office Food and Drug Organization, 19
Tehran, Iran; Dr Kowid Ho, ANSM, France; Dr Simon Hufton, National Institute for 20
Biological Standards and Control, Potters Bar, UK; Mrs Wichuda Jariyapan, Ministry of 21
Public Health, Nonthaburi, Thailand; Mr Ren Jian, HSA, Helios, Singapore; Dr Jeewon 22
Joung, KFDA, Osong, Korea; Dr Hans-Karl Heim, BfArM, Bonn, Germany; Dr Hye-Na 23
Kang, HIS/EMP, WHO, Geneva, Switzerland; Dr Yasuhiro Kishioka, Pharmaceutical 24
and Medical Devices Agency (PMDA), Tokyo, Japan, representative of the Japanese 25
Pharmacopoeia; Dr Ivana Knezevic, HIS/EMP, WHO, Geneva, Switzerland; Mr James 26
Leong, HSA, Helios, Singapore; Dr Jing Li, Shanghai CP-Guojian Pharmaceutical Co., 27
Ltd., Shanghai, China; Dr Jianhui Luo, SFDA, Beijing, China; Mrs Vivian Madrigal, 28
Recepta Biopharma, Sao Paulo, Brazil; Dr Catherine Njue, Health Canada, Ottawa, 29
Canada; Mrs Yanet Hechavarria Nunez, Centro para el Control Estatal de la Calidad de 30
los Medicamentos (CECMED), Habana, Cuba; Dr Pan Huirong Pan, Innovax BIOTECH 31
WHO/BS/2013.2213 Page 82
CO. Ltd., Xiamen, China, representative of the Developing Country Vaccine 1
Manufacturing Network (DCVMN); Dr Rolando Perez, Biotech Pharmaceutical Co., Ltd., 2
Beijing, China; Dr Stefanie Pluschkell, Pfizer Inc., Groton, USA, representative of the 3
IFPMA; Prof Chunming Rao, NIFDC, Beijing, China; Dr Martin Schiestl, Sandoz GmbH, 4
Kundl/Tirol, Austria, representative of IGPA; Dr Thomas Schreitmueller, F. Hoffmann-5
La Roche, Ltd. Basel, Switzerland, representative of IFPMA; Dr Satyapal Shani, Ministry 6
of Health and Social Welfare, Government of India, New Delhi, India; Dr Qi Shen, 7
NIFDC Beijing, China; Dr Xinliang Shen, China Bio-Tech Group, Beijing,China; Dr G. 8
R. Soni, National Institute of Biologics, Ministry of Health and Family Welfare, 9
Government of India, Noida, India; Dr Li Sun, Xiamen Amoytop Biotech Co., LTD., 10
Xiamen, China; Dr Robin Thorpe, NIBSC, Potters Bar, UK; Mrs Cornelia Ulm, Mylan 11
GmbH, Zurich, Switzerland, representative of the European Generic medicines 12
Association (EGA); Dr Antonio Vallin, Centre of Molecular Immunology, Habana, Cuba; 13
Dr Jian Wang, Health Canada, Ottawa, Canada; Dr Junzhi Wang, NIFDC, Beijing, 14
China; Dr Martina Weise, BfArM, Bonn, Germany; Dr Miao Xu, NIFDC, Beijing, China. 15
16
The second draft of this document was prepared by the following authors for the parts 17
indicated: Introduction - Dr Elwyn Griffiths, WHO consultant, Kingston-upon-Thames, 18
UK; Part A - Dr Elwyn Griffiths, WHO consultant, Kingston-upon-Thames, UK; Dr 19
Kowid Ho, ANSM, Anatole, France; Dr Jeewon Joung, KFDA, Osong, Republic of 20
Korea; Dr Robin Thorpe, NIBSC, Potters Bar, UK; Dr Meenu Wadhwa, NIBSC, Potters 21
Bar, UK; Dr Junzhi Wang, NIFDC, Beijing, China; Part B - Dr Laura Gomes Castanheira, 22
ANVISA, Brazilia, Brazil; Dr Hans-Karl Heim, BfArM, Bonn, Germany; Part C - Dr 23
Agnes Klein, Health Canada, Ottawa, Canada; Dr Frederike Lentz, BfArM, Bonn, 24
Germany; Dr Catherine Njue, Health Canada, Ottawa, Canada; Dr Jian Wang, Health 25
Canada, Ottawa, Canada; Dr Martina Weise, BfArM, Bonn, Germany; with support from 26
the WHO Secretariat: Dr Hye-Na Kang, Dr Jong-Won Kim, and Dr Ivana Knezevic, 27
HIS/EMP, WHO, Geneva, Switzerland; taking into account comments received from: 28
29
Dr Jennifer Archer, Hospira, Thebarton, Australia; Dr Janis Bernat, IFPMA, Geneva, 30
Switzerland; Dr Brigitte Brake, BfArM, Bonn, Germany; Dr Thomas Go, HSA, 31
WHO/BS/2013.2213 Page 83
Singapore; Mrs Wichuda Jariyapan, Ministry of Public Health, Nonthaburi, Thailand; 1
Mrs Yanet Hechavarria Nunez, CECMED, Habana, Cuba; Dr Martin Schiestl, Sandoz, 2
Kundl/Tirol, Austria; Dr Yeowon Sohn, KFDA, Osong, Republic of Korea; Dr G. R. Soni, 3
National Institute of Biologics, Ministry of Health and Family Welfare, Government of 4
India, Noida, India; Dr Teruhide Yahaguchi, Natinal Institute of Health Sciences (NIHS), 5
Tokyo, Japan. 6
7
The draft guidelines were posted on the WHO Biologicals web site for public 8
consultation from 20 March to 19 April 2013. 9
10
The document WHO/BS/2013.2213 was prepared by Dr Marie-Christine Bielsky, 11
MHRA, London, UK; Dr Elwyn Griffiths, WHO consultant, Kingston-upon-Thames, 12
UK; Dr Hans-Karl Heim, BfArM, Bonn, Germany; Dr Kowid Ho, ANSM, Anatole, 13
France; Dr Jeewon Joung, Ministry of Food and Drug Safety (formerly KFDA), Osong, 14
Republic of Korea; Dr Hye-Na Kang, HIS/EMP, WHO, Geneva, Switzerland; Dr Agnes 15
Klein, Health Canada, Ottawa, Canada; Dr Ivana Knezevic, HIS/EMP, WHO, Geneva, 16
Switzerland; Dr Frederike Lentz, BfArM, Bonn, Germany; Dr Robin Thorpe, NIBSC, 17
Potters Bar, UK; Dr Meenu Wadhwa, NIBSC, Potters Bar, UK; Dr Martina Weise, 18
BfArM, Bonn, Germany, taking into account comments received from the following 19
reviewers: 20
21
Ms P. Agsiri, Ministry of Public Health, Nonthaburi, Thailand; Mrs J. Bernat on behalf of 22
of the International Federation of Pharmaceutical Manufacturers and Associations 23
(IFPMA) (reviewed by Dr C. Phillips, Eli Lilly and Company; and Dr S. Ramanan, 24
Amgen); Dr B. Brake, BfArM,Bonn, Germany; Mrs J. Dahlan, National Agency of Drug 25
and Food Control, Jakarta, Indonesia; Ms C. Dubeaux on behalf of GSK, Belgium 26
(reviewed by Dr C. Lecomte, Dr F. Mortiaux, and Dr C. Saillez); Dr T. Go, HSA, 27
Singapore; Dr S. Jadhav. SSI, India, representative of the DCVMN; Dr Y. Kishioka, 28
PMDA, Tokyo, Japan; Dr B. Lan, China National Biotec Group (CNBG), China, 29
representative of the DCVMN; Dr T. Morris, United State Pharmacopoeial Convention, 30
Rockville, USA; Dr C. Njue, Health Canada, Ottawa, Canada; Dr G. Raychaudhuri, on 31
WHO/BS/2013.2213 Page 84
behalf of the Center for Biologics Evaluation and Research, Food and Drug 1
Administration (reviewed by Dr I. Mahmood, Dr C. Kimchi-Sarfaty, and Dr M. 2
Serabian); Dr Y. Ren, Chinese Pharmacopoeia, Beijing, China; Dr I. Shin, on behalf of 3
the Ministry of Food and Drug Safety, Republic of Korea (reviewed by Mr D. Baek, Dr J. 4
Joung, Mrs Y. Kim, and Mr O. Kwon); Dr S. Sontakke, Health Canada, Ottawa, Canada; 5
Dr J. Southern, South Africa, representative of the Developing Country Vaccine 6
Regulator’s Network (DCVRN); Dr R. Thorpe, NIBSC, Potters Bar, UK; J. Wang, 7
Health Canada. Ottawa, Canada; Dr J. Wang, NIFDC, Beijing, China; Dr A. Womack, 8
Biotechnology Industry Organization (BIO), USA. 9
10
Acknowledgements 11
Dr Martin Schiestl, Sandoz GmbH, Kundl/Tirol, Austria and Dr Thomas Schreitmueller, 12
F. Hoffmann-La Roche, Ltd. Basel, Switzerland for providing expertise on the section for 13
drug product container closure system and delivery devices. 14
15
16
WHO/BS/2013.2213 Page 85
References 1
2
1. Guidelines for assuring the quality of pharmaceutical and biological products 3
prepared by recombinant DNA technology; WHO Technical Report Series 814, 4
Annex 3, 59 – 70, 1991. 5
2. WHO guidelines on nonclinical evaluation of vaccines (TRS 927, Annex1). 6
3. WHO guidelines on clinical evaluation of vaccines: regulatory expectations. Annex 7
1 in: WHO Expert Committee on Biological Standardization. Fifty-second report. 8
Geneva, World Health Organization, 2004 (WHO Technical Report Series, No. 924) 9
(http://www.who.int/biologicals/publications/trs/areas/vaccines/clinical_evaluation/10
035-101.pdf, accessed 11 April 2011). 11
4. Recommendations for the evaluation of animal cell cultures as substrates for the 12
manufacture of biological medicinal products and for the characterization of cell 13
banks. Adopted by the 61st meeting of the WHO Expert Committee on Biological 14
Standardization, 1822 October 2010 as a proposed replacement of Annex 1 of 15
WHO Technical Report Series, No. 878 16
(http://www.who.int/entity/biologicals/Cell_Substrates_clean_version_18_April.pd17
f, accessed 4 October 2012). 18
5. Ad hoc Working Party on Biotechnology/Pharmacy, Guidelines on the production 19
and quality control of medicinal products derived by recombinant DNA technology. 20
Trends in Biotechnology 1987, 5 G1-4. 21
6. Points to consider in the production and testing of new drugs and biologicals 22
produced by recombinant DNA technology. Food and Drug Administration (USA), 23
1985. 24
7. Requirements for human interferons made by recombinant DNA techniques : WHO 25
Technical Report Series 771, Annex 7, 1997. 26
8. Requirements for human interferons prepared from lymphoblastoid cells WHO 27
Technical Report Series 786, Annex 3, 1988. 28
9. Guidelines for assuring the quality of monoclonal antibodies for use in humans, 29
WHO Technical Report Series 822 Annex 3, 1991. 30
10. Committee for Medicinal Products for Human Use. Guidelines on similar 31
biological medicinal products. London, European Medicines Agency, 2005 (CHMP 32
/ 437/04). 33
11. Guidelines on evaluation of similar biotherapeutic products (SBPs) WHO, 2010. 34
12. WHO/KFDA joint workshop on implementing WHO guidelines on evaluating 35
similar biotherapeutic products, Seoul, Republic of Korea, 24 -26 August 2010, 36
Biologicals , 39, (2011),349 -357. 37
13. Recommendations to assure the quality, safety and efficacy of recombinant 38
Hepatitis B vaccines. Adopted by the 61st meeting of the WHO Expert Committee 39
on Biological Standardization, 2010. 40
WHO/BS/2013.2213 Page 86
14. Guidelines on the quality, safety and efficacy of recombinant malaria vaccines 1
targeting the pre-erythrocytic and blood stages of Plasmodium falciparum. Adopted 2
by the 63rd
meeting of the WHO Expert Committee on Biological Standardization, 3
2012. 4
15. Report of WHO Informal Consultation on Characterization and Quality Aspects of 5
Vaccines Based on Live Viral Vectors, WHO, Geneva, 4-5 December 2003. 6
16. WHO Guidelines for assuring the quality and non-clinical safety evaluation of 7
DNA vaccines, WHO Technical Report Series 941, Annex 1, 2007. 8
17. Guidelines on the use of international nonproprietary names (INNs) for 9
pharmaceutical substances, Annex 4. Specific groups of biological compounds, 10
WHO 1997. http://whqlibdoc.who.int/hq/1997/WHO_PHARM_S_NOM_1570.pdf. 11
18. WHO good manufacturing practices: main principles for pharmaceutical products. 12
Annex 3 in: WHO Expert Committee on Specifications for Pharmaceutical 13
Preparations. Forty-fifth report. Geneva, World Health Organization, 2011 (WHO 14
Technical Report Series, No. 961). 15
19. Good manufacturing practices for biological products. Annex 1 in: WHO Expert 16
Committee on Biological Standardization. Forty-second report. Geneva, World 17
Health Organization, 1992 (WHO Technical Report Series, No. 822). 18
20. ICH Q5A guideline. Viral Safety Evaluation of Biotechnology Products Derived 19
From Cell Lines of Human or Animal Origin, 1999. 20
21. Guidelines on transmissible spongiform encephalopathies in relation to biological 21
and pharmaceutical products. Geneva, World Health Organization, 2003 22
(Document WHO/BCT/QSD/2003.01) 23
(http://www.who.int/biologicals/publications/en/whotse2003.pdf, accessed 11 24
January 2013). 25
22. WHO guidelines on tissue infectivity distribution in transmissible spongiform 26
encephalopathies. Geneva, World Health Organization, 2006 27
(http://www.who.int/bloodproducts/tse/WHO%20TSE%20Guidelines%20FINAL-28
22%20JuneupdatedNL.pdf, accessed 11 January 2013). 29
23. WHO tables on tissue infectivity distribution in transmissible spongiform 30
encephalopathies. Geneva, World Health Organization, updated 2010 (Document 31
WHO/EMP/QSM/2010.1) 32
(http://www.who.int/bloodproducts/tablestissueinfectivity.pdf, accessed 11 January 33
2013). 34
24. General requirements for the sterility of biological substances (Requirements for 35
Biological Substances, No. 6). Annex 4 in: WHO Expert Committee on Biological 36
Standardization. Twenty-fifth report. Geneva, World Health Organization, 1973 37
(WHO Technical Report Series, No. 530). 38
25. Stability testing of active pharmaceutical ingredients and finished pharmaceutical 39
products. Annex 2 in: WHO Expert Committee on Sepcifications for 40
Pharmaceutical Preparations. Forty-third report. Geneva, World Health 41
Organization, 2009 (WHO Technical Report Series, No. 953). 42
WHO/BS/2013.2213 Page 87
26. Guidelines on stability evaluation of vaccines. Annex 3 in: WHO Expert Committee 1
on Biological Standardization. Fifty-seventh report. Geneva, World Health 2
Organization, 2011 (WHO Technical Report Series, No. 962). 3
27. ICH Q5E guideline. Comparability of biotechnological/biological products subject 4
to changes in their manufacturing process, 2004. 5
28. Committee for Medicinal Products for Human Use (CHMP). Guideline on 6
comparability of biotechnology-derived medicinal products after a change in the 7
manufacturing process – non-clinical and clinical issues, 2007 8
(EMEA/CHMP/BMWP/101695/2006). 9
29. FDA Guidance concerning demonstration of comparability of human biological 10
products, including therapeutic biotechnology-derived products. US FDA, 1996. 11
30. ICH S5(R2) guideline. Detection of toxicity to reproduction for medicinal products 12
& toxicity to male fertility, 2000. 13
31. ICH S1A guideline. Guideline on the need for carcinogenicity studies, 1995. 14
32. ICH S7A guideline. Safety pharmacology studies for human pharmaceuticals, 2000. 15
33. ICH M3(R2) guideline. Guidance on nonclinical safety studies for the conduct for 16
human clinical trials and marketing authorization for pharmaceuticals, 2009. 17
34. ICH S6(R1) guideline. Preclinical safety evaluation of biotechnology-derived 18
pharmaceuticals, 2011. 19
35. Shin J et al. International regulatory requirements for vaccine safety and potency 20
testing: a WHO perspective. Procedia in Vaccinology, 2011, 5(164170). 21
36. Chapman K et al. Preclinical development of monoclonal antibodies. 22
Considerations for the use of non-human primates. mAbs 1: 505-516 (2009) 23
37. Guideline on strategies to identify and mitigate risks for first-in-human clinical 24
trials with investigational medicinal products, 2007 (EMEA/CHMP/SWP/ 25
28367/07). 26
38. Hansel TT et al. The safety nd side effects of monoclonal antibodies. Nature 27
Reviews Drug Discovery 9: 325-338, 2010. 28
39. Sathish JG et al. Challenges and approaches for the development of safer 29
immunomodulatory biologics. Nature Reviews Drug Discovery 12: 306-324, 2013. 30
40. EMA Guideline on Development, Production, Characterization and Specifications 31
for Monoclonal Antibodies and Related Products, 2008. 32
41. U.S. FDA Points to Consider in the Manufacture and Testing of Monoclonal 33
Antibody Products for Human Use 1983/1987/1994/1997. 34
42. Guidelines for good clinical practice for trials on pharmaceutical product. World 35
Health Organization, WHO Technical Report Series, No. 850, 1995, Annex 3. 36
http://apps.who.int/medicinedocs/pdf/whozip13e/whozip13e.pdf. 37
43. Guideline on requirements for first-in-man clinical trials for potential hi-risk 38
medicinal products, 2007 (EMEA/CHMP/SWP/28367/2007 Corr.). 39
WHO/BS/2013.2213 Page 88
44. FDA Draft Guidance for Industry, Clinical Pharmacogenomics: Premarketing 1
Evaluation in Early Phase Clinical Studies, February 2011. 2
45. Hudson, KL. Genomics, Health Care, and Society. N Engl J Med 2011 (Sept 15) 3
365:1033-1041. 4
46. Committee for medicinal products for human use (CHMP): Guideline on the 5
clinical investigation of the pharmacokinetics of therapeutic proteins, 2004. 6
47. FDA Guidance for Industry. Safety, Efficacy, and Pharmacokinetic Studies to 7
Support Marketing of Immune Globulin Intravenous (Human) as Replacement 8
Therapy for Primary Humoral Immunodeficiency, 2008. 9
48. Richter et al. Mechanistic determinants of biotherapeutics absorption following SC 10
administration, The AAPS Journal 14(3): 559 – 570, 2012. 11
49. Tang et al. Pharmacokinetic aspects of biotechnology products, J Pharm Sci 93(9): 12
2184 – 2204, 2004. 13
50. Porter et al. Lymphatic transport of proteins after subcutaneous administration, J 14
Pharm Sci 89 (3): 297 – 310, 2000. 15
51. Mahmood et al. Drug interaction studies of therapeutic proteins or monoclonal 16
antibodies, J Clin Pharmacol 47: 1540-1554, 2007 17
52. Solá RJ, Griebenow K. Glycosylation of Therapeutic Proteins. BioDrugs. 2010; 18
24(1):9-21. 19
53. ICH E4 guideline. Dose Response Studies to Support Drug Registration, 1994. 20
54. ICH E14 guideline. The Clinical Evaluation of QT/QTc Interval Prolongation and 21
Proarrhythmic Potential for Non-Antiarrhythmic Drugs, 2005. 22
55. ICH E10 guideline. Choice of Control Group and Related Issues in Clinical Trials, 23
2000. 24
56. ICH E9 guideline. Statistical Principles for Clinical trials, 1998. 25
57. ICH E15 guideline. Definitions for Genomic Biomarkers, Pharmacogenomics, 26
Pharmacogenetics, Genomic Data and Sample Coding Categories, 2007. 27
58. ICH E16 guideline. Biomarkers Related to Drug or Biotechnology Product 28
Development: Context, Structure and Format of Qualification Submissions, 2010. 29
59. ICH E7 guideline. Studies in support of special populations: Geriatrics, Questions 30
& Answers, 2010. 31
60. ICH E11 guideline. Clinical investigation of medicinal products in the paediatric 32
population, 2000. 33
61. ICH E6 guideline. Guideline for Good Clinical Practice, 1996. 34
62. Adaptive Designs in Clinical Drug Development – An Executive Summary of the 35
PhRMA Working Group: Journal of Biopharmaceutical Statistics, 16: 275-283, 36
2006. 37
WHO/BS/2013.2213 Page 89
63. FDA Guidance for Industry. Adaptive Design Clinical Trials for Drugs and 1
Biologics, 2010. 2
64. Committee for Medicinal Products for Human Use (CHMP). Guideline on Missing 3
Data in Confirmatory Clinical Trials, 2011. 4
65. FDA Guidance for Industry. Clinical Trial Endpoints for the Approval of Cancer 5
Drugs and Biologics, 2007. 6
66. ICH E1 guideline. The extent of population exposure to assess clinical safety for 7
drugs intended for long-term treatment of non-life-threatening conditions, 1994. 8
67. ICH E2A guideline. Clinical safety data management: definitions and standards for 9
expedited reporting, 1994. 10
68. ICH E2B guideline. Maintenance of the ICH guideline on clinical safety data 11
management: data elements for transmission of individual case safety reports, 2001. 12
69. MedDRA terminology M1. Medical dictionary for regulatory activities. Points to 13
consider ICH-Endorsed. Guide for MedDRA Users on Data Output, release 3.3 14
based on MedDRA Version 15.0, 1 April 2012. 15
70. ICH E7 guideline. Studies in support of special populations: Geriatrics. 1993. 16
71. ICH E7 guideline. Studies in support of special populations: Geriatrics, Questions 17
& Answers, 2010. 18
72. Committee for Medicinal Products for Human Use (CHMP). Guideline on 19
immunogenicity assessment of biotechnology-derived therapeutic proteins. London, 20
European Medicine Evaluation Agency, 2007 (CHMP/BMWP/14327). 21
73. Guideline on immunogenicity assessment of monoclonal antibodies intended for in 22
vivo clinical use, 2012 (EMA/CHMP/BMWP/86289/2010). 23
24
25
WHO/BS/2013.2213 Page 90
Appendix 1. Manufacturing process validation 1
2
Process validation is the documented evidence that the process, operated within 3
established parameters, can perform effectively and reproducibly to produce a drug 4
product, drug substance or intermediate meeting its predetermined specifications and 5
quality attributes. 6
7
Process validation should include the collection and evaluation of data, throughout 8
production, to establish scientific evidence that a process is capable of consistently 9
delivering a quality drug substance. 10
The manufacturing process should be validated before commercial distribution of drug 11
product. It generally includes collection of data on an appropriate number of production 12
batches. The number of batches can depend on several factors including but not limited 13
to: (1) the complexity of the process being validated; (2) the level of process variability; 14
and (3) the amount of experimental data and/or process knowledge available on the 15
specific process. 16
17
Process validation studies should include appropriate evaluation of the commercial 18
process and process steps (e.g. cell culture, harvest, purification, mixing, sterilisation, 19
filling), providing evidence that they are capable of consistently delivering quality 20
product and intermediates (i.e. meeting their predetermined specifications and quality 21
attributes). 22
23
The capacity of the purification procedures to remove product and process-related 24
impurities (e.g. unwanted variants, host cell proteins, nucleic acids, resin leachates) 25
should be investigated thoroughly. Process conditions (e.g. column loading capacity, 26
column regeneration and sanitisation, height) should be appropriately evaluated. 27
Columns should also be evaluated throughout their expected life span of the column 28
regarding their purification ability (e.g. impurity clearance, collection of intended 29
variants), leaching of ligands (e.g. dye, affinity ligand) and/or chromatographic material 30
(e.g. resin). 31
WHO/BS/2013.2213 Page 91
Process validation activities should normally include the evaluation of resin lifetime, 1
including maximum cycles and/or maximum time duration, using small scale studies to 2
ensure proper performance and integrity of the columns. In addition, the results should 3
normally be verified at full scale through the lifecycle of the product. These studies 4
should also confirm the suitability of the column cleaning, storage and regeneration 5
procedures. 6
7
Where hold times are applied to intermediates (e.g. harvest, column eluate), the impact of 8
hold times and hold conditions on the product quality should be appropriately evaluated 9
(e.g. degradation). 10
11
Evaluation of selected step(s) (e.g. steps for which high impurity or viral clearance are 12
claimed) operating in worst case and/or challenging conditions (e.g. maximum hold 13
times, spiking challenge) could be performed to demonstrate the robustness of the 14
process. Depending on the relevance of experimental model with regards to the final 15
process (e.g. scale, materials, equipment, operating conditions), these studies could be 16
leveraged in support of process validation and/or quality control data requirements. 17
The information provided in the dossier in support of process validation usually contains 18
both commercial-scale process validation studies and small-scale studies. Process 19
validation batches should be representative of the commercial process, taking into 20
account the batch definition as detailed in the process description. 21
22
Process changes at the level of fermentation and/or purification during progression to full 23
scale commercial production may have considerable consequences for the quality of the 24
product, yield and/or in quantitative and qualitative differences in impurities. Therefore, 25
the contribution of data from small-scale studies to the overall validation package will 26
depend upon demonstration that the small-scale model is an appropriate representation of 27
the proposed commercial scale. Data demonstrating that the model is scalable and 28
representative of the proposed commercial process should be provided. Successful 29
demonstration of the suitability of the small-scale model can enable manufacturers to 30
propose process validation with reduced dependence on testing of commercial-scale 31
WHO/BS/2013.2213 Page 92
batches. Data derived from commercial-scale batches should confirm results obtained 1
from small scale studies used to generate data in support of process validation. Scientific 2
rationale or reference to guidelines can be an appropriate justification to conduct certain 3
studies only at small scale (e.g. viral removal). 4
5
WHO/BS/2013.2213 Page 93
Appendix 2. Characterization of rDNA-derived biotherapeutics 1
2
This Appendix provides details of suggested approaches that can be applied to the 3
characterization of a rDNA-derived biotherapeutic. It also provides examples of technical 4
approaches which might be considered for structural characterization and confirmation, 5
and evaluation of physicochemical and biological properties of the desired product, drug 6
substance and/or drug product. The methods should provide an understanding of the 7
product with sufficient level of detail, e.g. complete primary structure, properties for the 8
higher order structure, qualitative and quantitative analysis of product related substances 9
and product and process related impurities, assessment of biological functions. 10
A subset of the methods described in this Appendix can be used for routine batch release 11
testing. Others are subject to extended characterization of the desired product during 12
product and process development, and are also often used to support process 13
evaluation/validation and/or comparability studies, e.g. after making significant process 14
changes. The selection of release testing methods depends on the overall design of quality 15
control for which the release testing is only one element among others. For example, if a 16
certain quality attribute can be controlled by in-process tests, parametric controls, and/or 17
demonstrated manufacturing process capability (e.g. high impurity clearance), such 18
attribute may not need to be tested routinely on every batch. 19
20
1. Physicochemical characterization 21
1.1 Primary structure 22
The primary structure, i.e. amino acid sequence including the disulfide linkages, of the 23
desired product can be determined to the extent possible using combined approaches such 24
as those described in items a) through c) and then compared with the sequence of the 25
amino acids deduced from the gene sequence of the desired product. Attention should be 26
paid to the possible presence of N-terminal methionine (e.g. in Escherichia coli derived 27
products), signal or leader sequences and other possible N- and C- terminal modifications 28
(such as acetylation, amidation or partial degradation by exopetidases). The variability of 29
N- and C- terminal amino-acid sequences should be analysed (e.g. C-terminal lysine(s)). 30
31
WHO/BS/2013.2213 Page 94
Free sulphydryl groups and disulfide bridges should be determined. Disulfide bridge 1
integrity and mismatch should be analysed. Experimentally determined disulfide bonding 2
patterns should be compared to the predicted structure based on the class of the molecule. 3
4
a) Peptide map 5
Selective fragmentation of the product into discrete peptides is performed using suitable 6
enzymes or chemicals and the resulting peptide fragments are analysed by high-7
performance liquid chromatography (HPLC) or other appropriate analytical procedures. 8
The peptide fragments should be identified to the extent possible using appropriate 9
techniques such as mass spectrometry (MS) methods (e.g. electrospray ionization MS, 10
matrix-assisted laser-desorption ionization time-of flight MS). The use of MS/MS 11
coupling should also be considered, as it could reveal more detailed sequence information 12
of the analysed peptide fragment. If one fragmentation method does not deliver the 13
complete amino acid sequence, the use of an orthogonal enzyme or chemical cleavage 14
method can increase the sequence coverage. The correct formation of the disulfide 15
bridges may be characterized by the use of peptide mapping under reducing and non-16
reducing conditions. 17
18
b) Molecular weight determination by mass spectrometry 19
The molecular weight of the intact molecule as determined by MS serves as an additional 20
confirmation of the primary structure. For smaller peptides, MS/MS sequencing can 21
provide the complete amino acid sequence. MS can be performed under reduced and non-22
reduced conditions and deglycosylated and intact conditions for multi-subunit and 23
glycosylated protein molecules such as monoclonal antibodies. 24
25
c) Other methods 26
Methods such as amino acid analysis or Edman sequencing can also be used although 27
their importance has decreased due to the technical advances in MS. 28
29
1.2 Glycan structure 30
WHO/BS/2013.2213 Page 95
Post-translational modifications, such as glycosylation should be identified and 1
adequately characterized. The glycan content (neutral sugars, amino sugars and sialic 2
acids) should be determined if linked to clearance or activity. In addition, the structure of 3
the glycan chains, the glycan pattern (antennary profile native glycan profile and site 4
specific glycan analysis) and the glycosylation site(s) of the polypeptide chain is analysed, 5
as far as possible. This task can be achieved by the combination of enzymatic or chemical 6
hydrolytic cleavage with a variety of separation methods (HPLC, electrophoresis) and 7
detection/identification methods (MS including MS/MS, ultraviolet (UV), fluorescence 8
detection, electrochemical detection). 9
The quantitative oligosaccharide analysis (chemical or enzymatic cleavage followed by 10
HPLC) provides additional useful qualitative and quantitative information on the glycan 11
structure. 12
Measurement of the quantitative charge patterns of the intact glycoprotein, e.g. by 13
measuring the charge based isoforms using appropriate method (e.g. capillary 14
electrophoresis, isoelectric focusing) may be useful as an overall measure of the degree of 15
sialylation and antennary profile. 16
Particular attention should be paid to glycan structures that may be associated with 17
adverse effects, such as non-human structures or residues. Further tests to be conducted 18
include analysis of charge heterogeneity. 19
20
1.3 Higher-order structure 21
Higher-order structure should be characterised by appropriate physicochemical 22
methodologies and confirmed by biological function. The analysis of pegylated proteins 23
should include, but not be limited to the average rate of modification, the location of 24
modification, and analysis of site occupancy. 25
The complete assessment of the three dimensional chemical structure in the context of 26
product characterization is rarely achieved, because absolute methods such as X-ray 27
crystallography or nuclear magnetic resonance (NMR) with isotope labeled amino acids 28
deliver only an approximation to the structure of the product of interest. They measure 29
the product either in a non-relevant state or require a separate production of the isotope 30
labeled sample. However, the use of applicable but relative orthogonal methods as 31
WHO/BS/2013.2213 Page 96
described below enable the determination and characterization of discrete folding and 1
also the assessment of changes in the higher order structure, e.g. in the case of 2
comparability studies. 3
4
The higher-order structure of the product should be examined using appropriate 5
procedures such as circular dichroism (CD), Fourier transform infrared spectroscopy (FT-6
IR), fluorescence, differential scanning calorimetry, proton nuclear magnetic resonance 7
(1H-NMR) and/or other suitable techniques, for example hydrogen-deuterium exchange 8
MS. When using these methods, their capabilities and limitations need to be considered 9
(e.g. impact of protein concentration). For instance, FT-IR and CD in the far UV range 10
deliver information on the secondary structure, whereas CD in the near UV reflects to 11
some extent the tertiary and quaternary structure. 12
13
In vitro or in vivo assays that illustrate the functional activity of the therapeutic may also 14
serve as an additional confirmation of the higher order structure in addition to 15
demonstrating biological function. 16
17
2. Biological activity 18
Assessment of the biological properties of a product constitutes an essential step in 19
establishing a complete characterization profile. The biological activity describes the 20
specific ability or capacity of a product to achieve a defined biological effect. Description 21
of a relevant biological assay to measure the biological activity should be provided by the 22
manufacturer. 23
24
The biological activity should be assessed by in vitro, in vivo, biochemical (including 25
immunochemical assays) and/or physicochemical assays as appropriate. 26
For antibody products, where effector function may play a role in the mechanism of 27
action, and/or have an impact on the product safety and efficacy, a detailed analysis of 28
biological activity demonstrating the mechanism of action (e.g. antibody dependent 29
cellular cytotoxicity, complement dependent cytotoxicity, apoptosis), ability for 30
complement binding and activation and other effector functions, including Fc gamma 31
WHO/BS/2013.2213 Page 97
receptor binding activity, and neonatal Fc receptor binding activity should be provided, as 1
appropriate. 2
The mechanism of action should be discussed, and where relevant, the importance (or 3
consequences) of other functions (e.g. effector functions) with regards to the safety and 4
efficacy of the product should be included. 5
6
Potency (expressed in units) is the quantitative measure of biological activity based on 7
the attribute of the product which is linked to the relevant biological properties, whereas, 8
quantity (expressed in mass) is a physicochemical measure of protein content. For 9
assessing potency, use of bioassays that reflect the biological activity in the clinical 10
situation is preferable, but not always possible or necessary for lot release. For example, 11
bioassays which assess some functional aspect of the protein or mechanism of action 12
(rather than the intended clinical effect) can also be used as the basis for a potency 13
assay. 14
15
Examples of procedures used to measure biological activity include: 16
• Animal-based biological assays, which measure an organism's biological response 17
to the product; 18
• Cell-based biological assays, which measure biochemical or physiological 19
response at the cellular level; 20
• Biochemical assays, which measure biological activities such as receptor or ligand 21
binding, enzymatic reaction rates or biological responses induced by 22
immunological interactions. 23
24
3. Immunochemical properties 25
Where relevant (e.g. for monoclonal antibody products), the immunochemical properties 26
should be fully characterised. Binding assays using purified antigens and defined regions 27
of antigens should be performed, where feasible, to determine affinity, avidity and 28
immunoreactivity (including cross-reactivity with other structurally homologous proteins). 29
30
The part of the target molecule bearing the relevant epitope should be characterized to the 31
WHO/BS/2013.2213 Page 98
extent possible. This should include biochemical identification of these structures (e.g. 1
protein, oligosaccharide, glycoprotein, glycolipid), and relevant characterisation studies 2
(amino acid sequence, carbohydrate structure) as appropriate. 3
4
Unless otherwise justified, the ability for complement binding and activation, and/or 5
other effector functions should be evaluated, even if the intended biological activity does 6
not require such functions. 7
8
4. Purity, impurity and contaminant 9
Biotechnological products commonly display several sources of heterogeneity (e.g. C-10
terminal processing, N-terminal pyroglutamation, deamidation, oxidation, isomerisation, 11
fragmentation, disulfide bond mismatch, N-linked and O-linked oligosaccharide, 12
glycation, aggregation), which lead to a complex purity/impurity profile comprising 13
several molecular entities or variants. This purity/impurity profile should be assessed by a 14
combination of methods, and individual and/or collective acceptance criteria should be 15
established for relevant product-related substances and impurities. 16
17
These methods generally include the determination of physicochemical properties such as 18
molecular weight or size, isoform pattern, determination of hydrophobicity, 19
electrophoretic profiles, chromatographic data including peptide mapping and 20
spectroscopic profiles including mass spectroscopy. 21
Multimers and aggregates should also be appropriately characterised using a combination 22
of methods. Unless otherwise justified, the formation of aggregates, sub-visible and 23
visible particulates in the drug product is important and should be investigated and 24
closely monitored at the time of release and during stability studies. 25
26
Impurities may be either process- or product-related. These materials should be 27
characterized to the extent possible and, their impact on biological activity should be 28
evaluated if appropriate. 29
30
WHO/BS/2013.2213 Page 99
Potential process-related impurities (e.g. host cell protein, host cell DNA, cell culture 1
residues, downstream processing residues) should be identified, and evaluated 2
qualitatively and/or quantitatively, as appropriate. 3
4
Contaminants, which include all adventitiously introduced materials not intended to be 5
part of the manufacturing process (e.g. microbial species, endotoxins) should be strictly 6
avoided and/or suitably controlled. Where non-endotoxin pro-inflammatory contaminants, 7
such as peptidoglycan, are suspected, the use of additional testing should be considered. 8
9
4.1 Process-related impurities and contaminants 10
The process-related impurities are derived from the manufacturing process itself and 11
could be classified into three major categories: cell substrate-derived, cell culture-derived 12
and downstream-derived. Contaminants, on the other hand represent unwanted material 13
which are introduced by unintentional means into the manufacturing process such as 14
adventitious viruses. 15
16
a) Cell substrate-derived impurities include, but are not limited to, proteins derived from 17
the host organism, nucleic acid (host cell genomic, vector, or total DNA). For host cell 18
proteins, a sensitive assay e.g. immunoassay, capable of detecting a wide range of protein 19
impurities is generally utilized. In the case of an immunoassay, polyclonal antibodies 20
used in the test are typically generated by immunization of animals with an appropriate 21
preparation derived from the production cell minus the product-coding gene, which have 22
been cultured in conditions representative of the intended culture and appropriately 23
collected (e.g. filtered harvest, partial purification). 24
The level of DNA from the host cells can be detected by direct analysis on the product 25
(e.g. qPCR, immunoenzymatic techniques). Clearance studies, which could include 26
spiking experiments conducted at the small scale, to demonstrate the removal of cell 27
substrate-derived impurities such as nucleic acids and host cell proteins may sometimes 28
be used to eliminate the need for establishing acceptance criteria for these impurities. 29
30
WHO/BS/2013.2213 Page 100
b) Cell culture-derived impurities include, but are not limited to, inducers antibiotics, 1
serum, and other media components. These impurities need to be tested and evaluated on 2
a case-by-case basis using a risk-assessment and management approach. In the case of a 3
potential impact on the safety of the product, the removal of such impurities to acceptably 4
low levels during downstream purification may need to be validated, or end product 5
testing and specification limits established. 6
7
c) Downstream-derived impurities include, but are not limited to, enzymes, chemical and 8
biochemical processing reagents (e.g. guanidine, dyes, oxidizing and reducing agents), 9
inorganic salts (e.g. heavy metals, non-metallic ions), solvents, carriers, ligands (e.g. 10
protein A), and other leachables. As for cell-culture-derived impurities, these impurities 11
should be evaluated on a case-by-case basis using a risk-assessment and management 12
approach. Where appropriate, development of analytical methods for these impurities and 13
validation of their removal could be considered. 14
15
4.2 Product-related substances and impurities including degradation products 16
Molecular variants of the desired product may need considerable effort in isolation and 17
characterization in order to identify the type of modification(s). When the activity of 18
those variants is comparable to the desired product, these variants should be included in 19
the product purity profile. Degradation products arising during manufacture and/or 20
storage in significant amounts should be appropriately considered. The most frequently 21
encountered molecular variant of the desired product and relevant technology for their 22
assessment are listed below. 23
24
a) Truncated forms 25
Hydrolytic enzymes or chemicals may catalyse the cleavage of peptide bonds. This may 26
lead to terminal heterogeneity; e.g. for C-terminal Lys in monoclonal antibodies. These 27
may be detected by HPLC and/or electrophoretic methods and verified by mass 28
spectrometry. Peptide mapping may also be useful, depending on the property of the 29
variant. 30
31
WHO/BS/2013.2213 Page 101
b) Amino acid modifications 1
Individual amino acid modification may include deamidation (Asp/Gln to Asp, Glu), 2
oxidation (e.g. Met to Met-sulfoxide), spontaneous formation of pyroglutamate out of N-3
terminal Glu or Gln residues, glycation of Lys residues and others. These forms may be 4
detected and characterized by relevant analytical methods (e.g. HPLC, capillary 5
electrophoresis, mass spectrometry). In some cases peptide mapping is important to 6
clearly identify and localize the site and nature of the amino acid modification. 7
8
c) High molecular weight species (HMWS) and particles 9
HMWS includes dimers and higher oligomers of the desired product. Particles include 10
intrinsic visible particles of the desired product. HMWS are generally resolved from the 11
desired product and product-related substances, and quantitated by appropriate separation 12
procedures (e.g. size exclusion chromatography, field flow fractionation, analytical 13
ultracentrifugation) coupled with sensitive detection methods (e.g. UV, fluorescence, 14
light scattering). Using orthogonal methods and/or procedures with overlapping 15
analytical windows (e.g. light obscuration testing, micro-flow imaging (MFI) for testing 16
of sub visible particles) can greatly enhance the characterization of aggregates and 17
particles. Foreign particles are not intended to be part of the product and should be 18
minimized. 19
20
5. Quantity 21
Quantity should be determined using an appropriate physicochemical and/or 22
immunochemical assay. 23
The protein content (expressed in mass units) can be determined by measuring the sample 24
against an appropriate reference standard using a suitable method (e.g. HPLC). 25
The protein content can also be measured in an absolute way, e.g. by UV photometry 26
using an extinction coefficient e.g. at 280 nm. Although the calculated extinction 27
coefficient delivers a satisfactory accuracy for many cases, it is advisable to use a second 28
absolute method (e.g. amino acid analysis, Kjeldahl) for verification. If the deviation is 29
too large, re-determination by another method could be considered. 30
31
32
WHO/BS/2013.2213 Page 102
Appendix 3. Routine control of rDNA-derived biotherapeutics 1
2
This appendix discusses approaches which should be taken to routine control of a rDNA-3
derived biotherapeutic. 4
5
1. Specification 6
A specification is defined as a list of tests, references to analytical procedures, and 7
appropriate acceptance criteria which are numerical limits, ranges, or other criteria for the 8
tests described. It establishes the set of criteria to which a drug substance and drug 9
product or materials at other stages of its manufacture should conform to be considered 10
acceptable for its intended use. “Conformance to specification” means that the drug 11
substance and drug product, when tested according to the listed analytical procedures, 12
will meet the acceptance criteria. The justification of specification should take into 13
account relevant development data and data from nonclinical, clinical and stability 14
studies. The setting of acceptance ranges should also take into account the sensitivity of 15
the analytical method used. 16
17
The selection of tests to be included in the specifications is product specific, and should 18
take into account the quality attributes (e.g. potential influence on safety, efficacy or 19
stability), the process performance (e.g. clearance capability, content), the controls in 20
place through the manufacturing process (e.g. multiple testing points), the material used 21
in relevant nonclinical and clinical studies. These tests could include criteria such as 22
potency, the nature and quantity of product-related substances, product-related impurities, 23
process-related impurities, and absence of contaminants. Such attributes can be assessed 24
by multiple analytical procedures, each yielding different results. Since specifications are 25
chosen to confirm the quality rather than to characterize the product, the rationale and 26
justification for including and/or excluding testing for specific quality attributes should be 27
provided. 28
29
The rationale used to establish the acceptable range of acceptance criteria should be 30
described. Acceptance criteria should be established and justified based on data obtained 31
WHO/BS/2013.2213 Page 103
from lots used in nonclinical and/or clinical studies. Nevertheless, where appropriately 1
justified, data from lots used for stability studies, or relevant development data could 2
support limits beyond ranges used in clinical studies. 3
4
2. Identity 5
The identity test(s) should be highly specific and should be based on unique aspects of 6
the product’s molecular structure and/or other specific properties (e.g. peptide map, anti-7
idiotype immunoassay, or other appropriate method). Depending on the product, more 8
than one test (physicochemical, biological and/or immunochemical) may be necessary to 9
establish identity, and such test(s) should possess sufficient specificity that they can 10
discriminate other products that may be manufactured in the same facility. 11
12
3. Purity and impurities 13
As noted in the characterisation section, recombinant proteins may display a complex 14
purity/impurity profile that should be assessed by a combination of orthogonal methods, 15
and for which individual and/or collective acceptance criteria should be established for 16
relevant product-related variants. Chromatographic and/or electrophoretic methods 17
capable of detecting product truncation, dissociation and aggregation should be included, 18
and quantitative limits should be proposed for these, as appropriate. 19
Considering that glycosylation and pegylation may have an impact on the 20
pharmacokinetic of the product, and may modulate its immunogenic properties, 21
appropriate acceptance criteria should be considered for this attribute. In addition, as 22
appropriate, such control could further confirm the consistency of the product. 23
The control of relevant process-related impurities should be included in the plan for 24
quality control. Control of process related impurities (e.g. protein A, host cell protein, 25
DNA and other potential culture or purification residues) are typically part of the drug 26
substance specification, as appropriate. In some situations, and where appropriately 27
demonstrated, their control may be performed on an intermediate product, at an 28
appropriate process step. Routine testing may not be necessary for some impurities for 29
which the process has been demonstrated to achieve high reduction levels. 30
31
WHO/BS/2013.2213 Page 104
4. Potency 1
Potency is the quantitative measure of biological activity based on the attribute of the 2
product which is linked to the relevant biological properties. A relevant potency assay 3
should be part of the specifications for drug substance and/or drug product, should reflect 4
the presumed mechanism of action whenever possible. Specific activity (units of 5
biological activity per mg or product) is of considerable value to demonstrate consistency 6
of production. 7
8
The potency of each batch of the drug substance and the final dosage form should be 9
established using, wherever possible, an appropriate national or international reference 10
material (e.g. A.1.3) which is normally calibrated in units of biological activity, for 11
example International Units (IU). In the absence of such preparations, an approved in-12
house reference preparation may be used for assay standardization. 13
14
For biologicals with antagonist activity, it may be appropriate to calibrate the potency 15
assay using the standard/reference preparation for the agonist and express activity of the 16
antagonist in terms of inhibition of biological activity i.e. units of the agonist. For 17
example for tumor necrosis factor (TNF) antagonists, bioassays can be calibrated using 18
the IS for TNF-alpha and activity expressed as number of IU of TNF neutralized by the 19
amount of the antagonist. 20
21
5. Quantity 22
The quantity of the drug substance and drug product, usually based on protein content, 23
should be determined using an appropriate assay. 24
25
6. General tests 26
General tests should be performed in accordance to relevant monographs, which could 27
include appearance (e.g. form, colour), solubility, pH, osmolality, extractable volume, 28
sterility, bacterial endotoxins, stabiliser and water, visible and sub-visible particulate, as 29
appropriate. 30
31
WHO/BS/2013.2213 Page 105
Appendix 4. Product specific guidance in nonclinical evaluation 1
(examples) 2
3
1. Anticancer pharmaceuticals 4
For anticancer pharmaceuticals, nonclinical evaluations are intended to identify the 5
pharmacologic properties, establish a safe initial dose level for the first human exposure, 6
and understand the toxicological profile, e.g. identification of the target organ, estimation 7
of the safety margin, and reversibility. In the development of anticancer drugs, most often 8
the clinical studies involve cancer patients whose disease condition is often progressive 9
and fatal. In addition, the clinical dose levels often are close to or at the adverse effect 10
dose levels. For these reasons, the type and timing and flexibility called for in designing 11
of nonclinical studies of anticancer pharmaceuticals can have a different pattern from 12
those for other pharmaceuticals (1, 2). 13
14
2. Monoclonal antibodies 15
For monoclonal antibodies, the immunological properties of the antibody should be 16
described in detail, including its antigenic specificity, complement binding, and any 17
unintentional reactivity and/or cytotoxicity towards human tissues distinct from the 18
intended target. For monoclonal antibodies and other related antibody products directed 19
at foreign targets (i.e. bacterial, viral targets etc.), a short-term (i.e. 2 weeks duration) 20
safety study in one species (choice of species to be justified by the sponsor) can be 21
considered; no additional toxicity studies, including reproductive toxicity studies, are 22
needed When animal models of disease are used to obtain proof of principle, a safety 23
assessment can be included to provide information on potential target-associated safety 24
aspects. Where this is not feasible, appropriate risk mitigation strategies should be 25
adopted for clinical trials. 26
27
Antibody-drug/toxin conjugates 28
Species selection for an antibody-drug/toxin conjugate (ADC) incorporating a novel 29
toxin/toxicant should follow the same general principles as an unconjugated antibody. If 30
two species have been used to assess the safety of the ADC, an additional short-term 31
WHO/BS/2013.2213 Page 106
study or arm in a short-term study should be conducted in at least one species with the 1
unconjugated toxin. In these cases a rodent is preferred unless the toxin is not active in 2
the rodent. If only one pharmacologically relevant species is available, then the ADC 3
should be tested in this species. A novel toxicant calls for an approach to species 4
selection similar to that used for a new chemical entity on a case-by case approach (e.g. 5
for anticancer products as described in ICH S9 Guideline) (2). For toxins or toxicants 6
which are not novel and for which there is a sufficient body of scientific information 7
available, separate evaluation of the unconjugated toxin is not warranted. Data should be 8
provided to compare the metabolic stability of the ADC in animals with human. 9
10
References 11
1. ICH S6(R1) guideline. Preclinical safety evaluation of biotechnology-derived 12
pharmaceuticals, 2011. 13
2. ICH S9 guideline. Nonclinical evaluation for anticancer pharmaceuticals, 2009. 14
15
16
WHO/BS/2013.2213 Page 107
Appendix 5. Animal species/model selection 1
2
1. Species selection 3
The biological activity together with species and/or tissue specificity of many rDNA-4
derived biotherapeutics often preclude standard pharmacological/toxicity testing designs 5
in commonly used species (e.g. rats and dogs). Pharmacological and safety evaluation 6
programs should include the use of relevant species. A relevant species is one in which 7
the test material is pharmacologically active due to the expression of the receptor or an 8
epitope (in the case of monoclonal antibodies). In addition to receptor expression, the 9
cellular/tissue distribution of receptors is an important consideration in selection of 10
appropriate species. 11
A number of factors should be taken into account when determining species relevancy. 12
Comparisons of target sequence homology between species can be an appropriate starting 13
point, followed by in vitro assays to make qualitative and quantitative cross-species 14
comparisons of relative target binding affinities and receptor/ligand occupancy and 15
kinetics. Assessments of functional activity are also recommended. Functional activity 16
can be demonstrated in species-specific cell-based systems and/or in vivo pharmacology 17
or toxicology studies. Modulation of a known biologic response or of a PD marker can 18
provide evidence for functional activity to support species relevance. 19
Consideration of species differences in target binding and functional activity in the 20
context of the intended dosing regimens should provide confidence that a model is 21
capable of demonstrating potentially adverse consequences of target modulation. When 22
the target is expressed at very low levels in typical healthy preclinical species (e.g. 23
inflammatory cytokines or tumor antigens), binding affinity and activity in cell-based 24
systems can be sufficient to guide species selection. 25
Tissue cross reactivity in animal tissues is of limited value for species selection. However, 26
in specific cases (i.e. where the approaches described above cannot be used to 27
demonstrate a pharmacologically relevant species) TCR studies can be used to guide 28
selection of species to be used in toxicology studies by comparison of tissue binding 29
profiles in human and those animal tissues where target binding is expected (see also 30
B.3.3). An animal species which does not express the desired epitope may still be of 31
WHO/BS/2013.2213 Page 108
some relevance for assessing toxicity if comparable unintentional tissue cross-reactivity 1
to humans is demonstrated. 2
3
When no relevant species exists, the use of relevant transgenic animals expressing the 4
human receptor or the use of homologous proteins should be considered. 5
6
2. Number of species 7
Safety evaluation programs should normally include two relevant species. However, in 8
certain justified cases one relevant species may suffice (e.g. when only one relevant 9
species can be identified or where the biological activity of the biotherapeutic is well 10
understood). 11
In addition, even where two species may be necessary to characterize toxicity in short 12
term studies, it may be possible to justify the use of only one species for subsequent long 13
term toxicity studies. If there are two pharmacologically relevant species for the clinical 14
candidate (one rodent and one non-rodent), then both species should be used for short-15
term (up to 1 month duration) general toxicology studies. If the toxicological findings 16
from these studies are similar or the findings are understood from the mechanism of 17
action of the product, then longer-term general toxicity studies in one species are usually 18
considered sufficient. The rodent species should be considered unless there is a scientific 19
rationale for using non-rodents. Studies in two non-rodent species are not appropriate. 20
The use of one species for all general toxicity studies is justified when the clinical 21
candidate is pharmacologically active in only one species. Studies in a second species 22
with a homologous product (see below) are not considered to add further value for risk 23
assessment and are not recommended. 24
25
Transgenic animals 26
The information gained from use of a transgenic animal model expressing the human 27
receptor is optimized when the interaction of the product and the humanized receptor has 28
similar physiological consequences to those expected in humans. 29
30
Homologous proteins 31
WHO/BS/2013.2213 Page 109
While useful information may also be gained from the use of homologous proteins, it 1
should be noted that the production process, range of impurities/contaminants, 2
pharmacokinetics, and exact pharmacological mechanism(s) may differ between the 3
homologous form and the product intended for clinical use. 4
Studies with homologous proteins can be used for hazard detection and understanding the 5
potential for adverse effects due to exaggerated pharmacology, but are generally not 6
useful for quantitative risk assessment. Therefore, for the purposes of hazard 7
identification it can be possible to conduct safety evaluation studies using a control group 8
and one treatment group, provided there is a scientific justification for the study design 9
and the dose(s) selected (e.g. maximum pharmacological dose). 10
11
Nonclinical testing in a non-relevant species 12
Pharmacological/toxicity studies in non-relevant species may be misleading and are 13
generally discouraged. However, where it is not possible to identify a relevant species or 14
to use transgenic animal models or if it is not possible to use a homologous protein for 15
testing purposes, it may still be prudent to assess some aspects of potential toxicity in a 16
limited toxicity evaluation in a single species, e.g. a repeated dose toxicity study of < 14 17
days duration that includes an evaluation of important functional endpoints (e.g. 18
cardiovascular and respiratory). 19
20
3. Animal models of disease 21
In recent years, there has been much progress in the development of animal models that 22
are thought to be similar to the human disease. These animal models include induced and 23
spontaneous models of disease, gene knockout(s) or knockin(s), and transgenic animals. 24
These models may provide further insight, not only in determining the pharmacological 25
action of the product, pharmacokinetics, and dosimetry, but may also be useful in the 26
determination of safety (e.g. evaluation of undesirable promotion of disease progression). 27
In certain cases, studies performed in animal models of disease may be used as an 28
acceptable alternative to toxicity studies in normal animals. 29
Animal models of disease may be useful in defining toxicity endpoints, selection of 30
clinical indications, and determination of appropriate formulations, route of 31
WHO/BS/2013.2213 Page 110
administration, and treatment regimen. It should be noted that with these models of 1
disease there is often a paucity of historical data for use as a reference when evaluating 2
study results. Therefore, the collection of concurrent control and baseline data is critical 3
to optimize study design. 4
The scientific justification for the use of these animal models of disease to support safety 5
should be provided. 6
7
WHO/BS/2013.2213 Page 111
Appendix 6. Explanatory notes 1
2
Note 1: The species-specific profile of embryo-fetal exposure during gestation should be 3
considered in interpreting studies. High molecular weight proteins (>5,000 D) do not 4
cross the placenta by simple diffusion. For monoclonal antibodies with molecular weight 5
as high as 150,000 D, there exists a specific transport mechanism, the neonatal Fc 6
receptor which determines fetal exposure and varies across species. 7
In the NHPs and humans, IgG placental transfer is low in the period of organogenesis and 8
begins to increase in early second trimester, reaching highest levels late in the third 9
trimester. Therefore, standard embryo-fetal studies in NHPs, which are dosed from early 10
pregnancy up to gestation day 50, might not be of value to assess direct embryo-fetal 11
effects in the period of organogenesis, although effects on embryo-fetal development as 12
an indirect result of maternal effects can be evaluated. Furthermore, maternal dosing in 13
NHPs after delivery is generally without relevance as IgG is only excreted in the milk 14
initially (i.e. in the colostrum), and not later during the lactation and nursing phase. 15
Rodents differ from the NHPs and humans, as IgG crosses the yolk sac in rodents by 16
neonatal Fc receptor transport mechanisms and exposure can occur relatively earlier in 17
gestation than with NHPs and humans. In addition, delivery of rodents occurs at a stage 18
of development when the pups are not as mature as the NHP or the human neonate. 19
Therefore, rat/mouse dams should be dosed during lactation in order to expose pups via 20
the milk up to at least day 9 of lactation when the offspring are at an equivalent stage of 21
development as human neonates. 22
23
Note 2: The minimum duration of post-natal follow-up should be one month to cover 24
early functional testing (e.g. growth and behaviour). 25
In general, if there is evidence for adverse effects on the immune system (or immune 26
function) in the general toxicology studies, immune function testing in the offspring 27
during the post-partum phase of the ePPND study is warranted. When appropriate, 28
immunophenotyping can be obtained as early as post-natal day 28. The duration of post-29
natal follow-up for assessment of immune function can be 3-6 months depending on the 30
functional tests used. 31
WHO/BS/2013.2213 Page 112
Neurobehavioural assessment can be limited to clinical behavioural observations. 1
Instrumental learning calls for a training period, which would result in a post-natal 2
duration of at least 9 months and is not recommended. 3
4
Note 3: A detailed discussion of the approach to determine group sizes in cynomolgus 5
monkey ePPND studies can be found in Jarvis P et al.: “The Cynomolgous Monkey as a 6
model for Developmental Toxicity Studies: Variability of Pregnancy losses, Statistical 7
power estimates, and Group Size considerations.” Birth Defects Research (Part B) 2010, 8
89: 175-187. Group sizes in ePPND studies should yield a sufficient number of infants 9
(6-8 per group at post-natal day 7) in order to assess post-natal development and provide 10
the opportunity for specialist evaluation if necessary (e.g. immune system). 11
Most ePPND studies accrue pregnant animals over weeks and months. Consideration 12
should be given to terminating further accrual of pregnant animals into the study, and 13
adapting the study design (e.g. by Caesarian section) when pre-natal losses in a test item 14
group indicate a treatment-related effect. 15
Reuse of vehicle-control treated maternal animals is encouraged. 16
If there is some cause for concern that the mechanism of action might lead to an effect on 17
EFD or pregnancy loss, studies can be conducted in a limited number of animals in order 18
to confirm the hazard. 19
20
Note 4: An example of an appropriate scientific justification would be a monoclonal 21
antibody which binds a soluble target with a clinical dosing regimen intended to saturate 22
target binding. If such a saturation of target binding can be demonstrated in the animal 23
species selected and there is an up to 10-fold exposure multiple over therapeutic drug 24
levels, a single dose level and control group would provide adequate evidence of hazard 25
to embryo-fetal development. 26
27
Note 5: Endpoints to be included in an interim report of an ePPND study in NHPs: 28
- Dam data: survival, clinical observations, bodyweight, gestational exposure data (if 29
available), any specific PD endpoints; 30
WHO/BS/2013.2213 Page 113
- Pregnancy data: number of pregnant animals started on study, pregnancy status at both 1
the end of organogenesis (gestation day (GD) 50) and at GD100, occurrence of abortions 2
and timing of abortions. There is no need for ultrasound determinations of fetal size in the 3
interim report; these are not considered essential since actual birth weight will be 4
available; 5
- Pregnancy outcome data: number of live births/still births, infant birth weight, infant 6
survival and bodyweight at day 7 post-partum, qualitative external morphological 7
assessment (i.e. confirming appearance is within normal limits), infant exposure data (if 8
available), any specific PD endpoints in the infant if appropriate. 9
10
WHO/BS/2013.2213 Page 114
Appendix 7. List of abbreviations 1
2
ADA anti-drug antibodies 3
ADC antibody-drug/toxin conjugate 4
ADME absorption, distribution, metabolism, and elimination 5
ADRs adverse drug reactions 6
AE adverse event 7
CD circular dichroism 8
CRS cytokine release syndrome 9
DNA deoxyribonucleic acid 10
EFD embryo-fetal development 11
ELISA enzyme-linked immunosorbent assay 12
EMA European Medicines Agency 13
ePPND enhanced pre/post-natal development 14
FDA Food and Drug Administration 15
FISH fluorescence in situ hybridization 16
FT-IR Fourier transform infrared spectroscopy 17
GCP good clinical practice 18
GLP good laboratory practice 19
GMP good manufacturing practice 20
HMWS high molecular weight species 21
HPLC high-performance liquid chromatography 22
ICH International Conference on Harmonization 23
IHC immunohistochemical 24
INN International Non-Proprietary Name 25
IU International Units 26
mAb monoclonal antibody 27
MACE major adverse cardiac events 28
MCB master cell bank 29
MFI micro-flow imaging 30
MS mass spectrometry 31
WHO/BS/2013.2213 Page 115
MTD maximum tolerated dose 1
MW molecular weight 2
NCL national control laboratory 3
NHPs non-human primates 4
NMR nuclear magnetic resonance 5
NOAEL no observed adverse effect level 6
NRA national regulatory authority 7
PBMC peripheral blood mononuclear cell 8
PCR-SSCP polymerase chain reaction-single-strand conformation polymorphism 9
PD pharmacodynamics 10
PhV pharmacovigilance 11
PK pharmacokinetics 12
PPND pre/post-natal development 13
rcDNA residual cellular DNA 14
rDNA-derived prepared by recombinant DNA technology 15
RFLP restriction fragment length polymorphism 16
RMinP risk minimization plan 17
RMP risk management plan 18
SAP statistical analysis plan 19
SPR s surface plasmon resonance 20
TCA trichloracetic acid 21
TCR tissue cross-reactivity 22
TNF tumor necrosis factor 23
TQT study thorough QT/QTc study 24
UV ultraviolet 25
WCB working cell bank 26
27
= = = 28