pmda’s approach to pre-market review and pharmacovigilance · presenter and should not be...
Post on 15-Aug-2020
0 Views
Preview:
TRANSCRIPT
Real World Data Utilization:
PMDA’s Approach to Pre-market
Review and Pharmacovigilance
Daisaku SATO, Ph.D.Chief Management Officer
& Associate Centre Director for Regulatory Science,
Pharmaceuticals and Medical Devices Agency, JAPAN
Disclaimer
The views and opinions expressed in the following PowerPoint slides are those of the individual
presenter and should not be attributed to Drug Information Association, Inc. (“DIA”), its directors,
officers, employees, volunteers, members, chapters, councils, Special Interest Area Communities
or affiliates, or any organisation with which the presenter is employed or affiliated.
These PowerPoint slides are the intellectual property of the individual presenter and are protected
under the copyright laws of the United States of America and other countries. Used by
permission. All rights reserved. Drug Information Association, DIA and DIA logo are registered
trademarks or trademarks of Drug Information Association Inc. All other trademarks are the
property of their respective owners.
The contents of this presentation represent the view of this presenter only, and do not represent
the views and/or policies of the PMDA.
© 2018 DIA, Inc. All rights reserved Page 2
Disclosure StatementI have no real or apparent relevant financial relationships to disclose
■ I am employed by a regulatory agency, and have nothing to disclose
Please note that DIA is not requesting a numerical amount to be entered for any disclosure, please indicate by marking the check box, and then
providing the company name only for those disclosures you may have.
Will any of the relationships reported in the chart above impact your ability to present an unbiased presentation? Yes No
In accordance with the ACPE requirements, if the disclosure statement is not completed or returned, participation in this activity will be refused.
Type of Financial Interest within last 12 months Name of Commercial Interest
Grants/Research Funding
Stock Shareholder
Consulting Fees
Employee
Other (Receipt of Intellectual Property Rights/Patent
Holder, Speaker’s Bureau)
© 2018 DIA, Inc. All rights reserved Page 3
Center for Regulatory Science
2018/09/04
4
Center for Regulatory Science (Organization Structure)
5
Director of Center for Regulatory Science
Associate Center Director Associate Executive Director
Office of
Medical Informatics
and Epidemiology
Office of
Advanced
Evaluation with
Electronic Data
Office of
Research
Promotion
Coordination
Officer for
Evaluation of
Advanced
Science and
TechnologyEMRs database
(RWD) CDISC database
(Clinical trials)
Big-Data analysis in regulatory science
Closely working together with Office of New Drugs, Office of Safety etc.
Center for Regulatory Science established on Apr 1
Office of Research Promotion
Review Offices
Office of Medical
Informatics and
Epidemiology
Safety Offices
Office of Advanced
Evaluation with Electronic Data
(former Advanced Review with
Electronic Data Promotion Group)
2018/09/04
6
Regulatory Science Centere-study
database
MID-NET
etc.
PMDA initiatives for quantitative science
Several initiatives/activities for quantitative science are established and are in execution for new drug/device development and review in Japan.
We are considering how we can efficiently use those data that we will obtain in each stage of clinical development.
2018/09/04
7
Drug development
Regulatory review
Post-marketing Surveillance
Approval
Advanced Review with Electronic Study Data
(CDISC)
Use of electronic patient registry data (e.g. Clinical Innovation Network)
Electronic medical records,claim data(e.g. MID-NET®)
Use of data standards is the key for all the initiatives
Premarketing Electronic Data Utilization
2018/09/04
8
Accumulation and utilization of data
2018/09/04
9
Submission of electronic
clinical study data has
started since Oct 1st 2016!
Timeline for implementation of e-data submission
2018/09/04
10
TaskJ-FY
2014
J-FY
2015
J-FY
2016
J-FY
2017
J-FY
2018
J-FY
2019
J-FY
2020
Guidance
and related
documents
Review
Consultation
for e-study data
submission
System
Development
2014 Pilot
2015 Pilot
Pilot
System Development /
Pilot for data
submission
Issuance of “Basic
Principles”
3.5 years of Transitional
period
FAQ
Briefing/Workshop
Eng.
Mar 31
End of the transitional period
Regular Update
Preparation for the
end of transitional
period
(Revision of the
notifications, etc.)
Today
WS
Oct 1
Portal Site
Open
New Consultation
framework
Issuance of “Notification on the consultation for the clinical e-data
submission”
Issuance of “Notification on Practical
Operations ”
Issuance of “Technical Conformance Guide”
Data Standards Catalog
PMDA Validation Rules
Initiation of e-study data
submission
WS WS
Q&A
Consultation for clinical e-data submission
169 consultation meetings have been requested by 49 companies as of July 31, 2018.
Multiple meetings have been held for the some products.
The number of consultation for NDA after transitional period is increasing.
Various characteristics– With/without official minutes
– Japanese/foreign company
– Oncology and other therapeutic areas
2018/09/04
11
Year N of request
J-FY 2015 (May 15, 2015 – Mar 31, 2016) 13
J-FY 2016 (Apr 1, 2016 – Mar 31, 2017) 62
J-FY 2017 (Apr 1, 2017 – Mar 31, 2018) 65
J-FY 2018 (Apr 1, 2018 – Jun 30, 2018) 29
Total 169
Practical cases for utilization of submitted data in review process (as of now)
Supplementary analyses to the dataset by reviewers contributed to improvingthe efficiency of NDA review process (e.g. reducing enquiries to the sponsors)
Examples:
(1) Conduct subgroup analysis to check the sponsor’s idea of dose adjustment topatient weight. (and other factors which affect efficacy/safety of the product)(clinical data)
(2) Conduct subgroup analysis by baseline status to check the consistency of efficacy among subjects (clinical data)
(3) Perform supplementary analysis to check the robustness of the primary result(e.g. value transformation, assumption of distribution, model)(clinical data)
(4) Check whether PPK analysis result from multinational trial was valid for Japanese or not. (PK/PD data)
2018/09/04
12
Analyses of CDISC data in review team
2018/09/04
13
• Distribution of patient demographics
• Changes in laboratory data
• Adverse events rates
Common analyses to
many clinical trials
• Simple analyses depending on thecharacteristics of evaluation variables– continuous/categorical/time-to-event)
General analyses for efficacy and safety data
• Analyses with programing(innovative/complicated analyses)
• Simulations
Relatively complicated
analyses
Software: JMP Clinical, etc.Datasets: SDTM
Software: JMP, SAS etc.Datasets: ADaM
Software: SAS, etc.Datasets: SDTM, ADaM
Prospect of e-Study data utilization in Japan
J-FY2019 - 2021
• e-study data can bereceived andmanagedappropriately
• e-study data can beutilized in the review
• Industries’ workloadis reduced graduallywhile keeping thereview period
• More predictableefficacy/safety
• Consideration ofexpanding the scopeof e-data utilizationto toxicological studyand post-approvalclinical study
- J-FY2017
J-FY2018
J-FY2022 -
• Preparations ofguidelines and relateddocuments
• Earnest on cross-product analysis anddevelopment ofdisease models
• Establishment ofdisease models
• Publication ofdisease-specificguidelines
First-class review quality
Setup e-data management and utilization
Ordinary utilization of e-
data in the product review
Starting earnest cross-product
analysis
Publication of guidelines to
contribute to drug development
e.g. guidelines and disease
models based on data on Asian
populationPromotion of
paperless operation
Transitional period are taken until March 31st, 2020
Prospect As of Sep 2017(Subject to Change)
Start e-study data submission for NDA*
from Oct 1st, 2016
*NDA=New Drug Application
14
Utilization of study data in the future
2018/09/04
15
Utilization of study data for new drug review
- Improvement of predictability of efficacy and safety
- Reviewing M&S results
- Reviewing novel evaluation methods
- Swift and effective decision-making
Utilization of accumulated study data
- Information from cross-product analysis
- Active use of M&S
- Evaluation of innovative analysis methods based on the accumulated data
- Experiences of meta-analytic approach
Efficient new drug development
- Use of consultation meeting based on the cross-product information by PMDA
- Active use of M&S
- Use of innovative and appropriate methods for the purpose
- Consultation based on the cross-product
information
- Guidance for therapeutic areas
- Issuance of points to considers for methodology
Submission of standardized study data
- Data
accumulation
- Experiences
of data
evaluation
Use of various data sources in the future- Importance of study quality, data quality, and data
standardization
- Innovative methods for analyzing data from various data sources
- Consultation/guidance
about innovative analysis
methods
- Contribution to data
standardization
Post-marketing Real-world Electronic Data Utilization
2018/09/04
16
Clinical Safety
Data
CT start NDA Approval
Pre-approvalPost
-approval
Keep agreeable benefit-risk balance in the lifecycle
Real world, day to day
medicine
Clinical Trial
Phase
“Optimal use”, at each stage, from pre-marketing to post-marketing,
Pre-market Post-market
Consistent risk management
From rigorous CTs To complex real world after product
launch
Clinical
Effectiveness
Data
リスク最小化活動Risk Minimization Action Plan
安全性監視計画Pharmacovigilance Plan
Regular
Additional
安全性検討事項Safety Specification
Serious specified risk
Serious potential risk
Serious missing
information
Additional measure
No
Yes
Concept of structured RMPimplementation in line with ICH-E2E guideline
Spontaneous Rep.
Literature survey
Labelling
Commentary
EPPV
Medical experience survey
Controlled study
Pharmaco-epi study
Patient medication guide
Leaflet
Education programme
Distribution control
Labelling revision
Vigilance
and / or
minimisation?(evaluation)※
Additional vigilance
Additional minimisation
Regu
lar re
vie
w
Limited medical
institutions and limited
doctors
Patient selection
criteria
What will become possible by utilizing a large-scale medical information database?
Comparison with
other drugs
Comparison with
symptoms due to
underlying disease
Verification of effects
of safety measures
Can compare frequency of
ADR occurrence drugs in
the same class
Rate of occurrence of ADRs
(ADRs/number of patients who
used the drugs)
Treated with
Drug A
Treated with
Drug B
Rate of occurrence of
symptoms
(symptoms/number of patients
who used the drugs)
Treated with
Drug A Treated without
Drug A
Can ascertain whether
the occurrence of a
certain symptom is
increased by
administration of a drug
Rate of occurrence of ADRs
(symptoms/number of patients
who used the drugs)
Before safety
measures
After safety
measures
Can compare to see whether
implementation of safety
measures actually results in
a change in ADR frequency
Data sources for post-market safetyassessment of a drug
Spontaneous
ADR report
DBSafety
measure
DPC DB
Risk
communi
cation
Literatures
Overseas
regulatory
actions
Presentation in
Academic
Conference
etc
Claims
DB
Electronic
Healthcare Data
PMDA
Conventional
Information Sources
MHLWMedical
institutions
Launched in 2009
Collaboration
Expected Outcome:Prompt and precise safety actions
PMDAsafety information collection and
analysis
Researcher, MAHs
Prompt safety
action
site
sitesite
DB
DB
DBDB
Networking 10
sentinel sites of
23 hospitals
Data
analysis
Tohoku U, Tokyo U, Chiba U,
NTT Hosps、Kitasato Hosps,
Hamamatsu M U, Tokushukai
Hosp Group, Kagawa U,
Kyushu U, Saga U.
site
DBSentinel site
hospitals
EHRClaim Data
DPC
DataLab. test
Medical Information Database Network(MID-NET)
【History and way forward】●April 2010 :「Revision of pharmaceutical administration etc. to prevent recurrence of pharmaceutical
disasters (final recommendation) 」● April 2011 - : Start construction of MID-NET system
● April 2013 - : Start data quality validation to assure precision and comprehensiveness of the collected data
● April 2015 - : Start trial operations by PMDA and sentinel sites
● April 2015 - : Setting utilization rules for full-scale operation and framework of operation cost / user fees.
● April 2018- : Full scale operation, enable MAHs and researchers to use MID-NET
Promote safety measures by pharmaco-epidemiological method using medical information
database.
MHLW/PMDA have established a medical information database for collecting large-scale
medical data at sentinel site hospitals and have constructed analytical systems at PMDA
since FY 2011.
More than
4,000,000
patients included
Common Data Model of the MID-NET®
Database
Claims data
DPC data
HIS data
・Patient identifying data
・Medical examination history data
(including admission , discharge data)
・Disease order data
・Discharge summary data
・Prescription order/compiled data
・Injection order/compiled data
・Laboratory test data
・Radiographic inspection data
・Physiological laboratory data
・Therapeutic drug monitoring data
・Bacteriological test data
HIS data
Contents Standard Code
Disease ICD-10
Drug
YJ, HOT9
(JP specific codes)
Laboratory test
JLAC10
(JP specific codes)
Example of standard codes
Example: Data Consistency Check
Data Extraction
Storage Server based on HL-7(SS-MIX2)
data standard
MID-NET data server
Hospital Information System (HIS)
transfer
AnnonymizationData
extraction system
Primary data
calculation system
Data server
Data ExtractionCompare number of cases and
contents per data element per hospital for certain periods
At the beginning, approximately hundreds issues per site were identified for further investigation or consideration
Examples of data inconsistency
Lack of a unit
Wrong place of data storage
e.g.; single dose, daily dose vs total dose
transfer
Data Quality of MID-NET®
Disease order data
Prescription order data
EMR MID-NET
Laboratory test data
compare
99.1%
67.0%
55.8%
Disease order data
Prescription order data
EMR MID-NET
Laboratory test data
compare
99.9%
100%
100.0%
PMDA has worked with cooperative hospitals
for assuring data quality of MID-NET®.
Before quality
managementAfter quality
management
25
Standardized data coding process-Example; Laboratory test-
• Confirming appropriateness of a code for individual laboratory test bychecking a distribution of laboratory test results (Approximately 200 tests)
ALT, AST, BUN, K, Creatinine, LDH, Gamma-GT, Cl, ALP, MCHC, MCH, Uric Acid, cGFR, TG, Cholesterol, Amylase, Blood Glucose, LDL-C, Inorganic Phosphate, HDL-C, PT-INR, HbA1c, PT, APTT, CEA, Fe, FT4, IgG, TSH, Sedimentation rate, RPR, IgM, HbA1c(NGSP), TPHA, AFP, Ferritin, Hb, Reticulocyte, Blood Gases(TCO2), Blood Gases(pH),etc
Distribution of laboratory test results among hospitalsp
rop
ort
ion
Original data (local unit)
Hosp. A
Hosp. B
Hosp. C
pro
po
rtio
n
Standardized data
Hosp. A
Hosp. B
Hosp. C
After quality check
Before After
Confirmed
Examples of available laboratory test
25
Further investigation were conducted in case of different distributions
for understanding a reason and identifying an appropriate code
26
Launched(2012.4.17)
Spontaneous ADR reports
・32 serious cases of hypocalcemiaincluding 2 death cases
(~2012.8.31)
MID-NET® pilot: Case 1
denosumab and severe hypocalcemia
Warning letter(Dear healthcare
professionals letter)
(2012.9.12)
A) More laboratory test on serum calcium etc.
B) Co-administration of calcium/vitamin D
C) Special caution to patients with severe
impairment of renal function
D) Prepare for emergency situation
At the post-market, the label change and
warning letter were issued for awaking the risk
of hypocalcemia associated with denosmab
Monthly transition of the incidence of hypocalcemia
27
0.0
1.0
2.0
3.0
4.0
5.0
0.0%
20.0%
40.0%
60.0%
80.0%
100.0%
Ris
k r
ati
o
Inci
den
ce P
rop
ort
ion
(%
) ランマークゾレドロン酸水和物リスク比
Blue
letterLabelling
change
Denosumab
Zoledronate
Risk ratio
MID-NET® pilot:
denosumab and severe hypocalcemia
Pilot study
・Calculate the incidence of hypocalcemia during 28 days from a prescription date.
・Perform segment regression analysis based on the incidence of hypocalcemia / month.
■ObjectiveTo examine impacts of label change and warning letter in clinical practice for the risk of hypocalcemia
associated with denosmab
28
MID-NET® pilot:
Risk of Acute Myocardial Infarction Associated with
Anti-Diabetes Drugs
Pilot study
Incidence Rateper 1,000 Person-
years
Adjusted
Rate Ratio
(95%CI)
Adjusted
Hazard Ratio
(95%CI)Non-sulfonylurea insulin
secretagogues2.4 1[ref] 1[ref]
DPP-4 inhibitors 2.10.86
(0.25-2.90)
0.93
(0.08-10.80)
Table. Adjusted rate ratio and
adjusted hazard ratio for AMI in
the standardized population.
Objective
Cardiovascular events associated with anti-diabetes drugs
are common risk in post-marketing phase
To compare the risk of acute myocardial infarction (AMI)
associated with DPP-4 inhibitors monotherapy to other
anti-diabetes drugs monotherapy.
Exposed GroupDPP-4 inhibitors
(n=2,578)
MID-NET®
(2010~2015)
DPP-4 inhibitors(n=2,952)
Cohort:New users of anti-diabetes drugs monotherapy
Propensity score standardization
(SMRW)
■Outcome definition(AMI)
Definitive diagnosis of AMI, Admission* and Elevation of cardiac biomarker values*(CK or CK-MB:≧URL ×2 or Troponin T:≧0.1ng/mL)
*during 30 days before and after the diagnosis date of AMICases of AMI Cases of AMI
Occurrence of AMI
Control groupGlinides
(n=2,717.2)
Glinides(n=237)
29
Cohort Case (people) Users (people) frequency (%) 95% CI
Whole cohort 24 7,267 0.3 0.2-0.4
subgroup①(under 12y.o.)
-*2 209 -*2 0.0-1.0
subgroup②(12y.o.~18y.o.)
0 199 0 0.0-0.0
subgroup③(above 18y.o.)
-*2 6,859 -*2 0.2-0.5
※1The case where the respiratory depression has been developed is defined below.
1) prescription of therapeutics for respiratory depression (levarophan, naloxone), or,
2) Implementation of diagnosis related to respiratory depression (dyspnea, acute respiratory failure,
respiratory failure) and oxygen inhalation
※2 When the number of cases in the subgroup is less than 10 people, concrete numerical values are not
disclosed in accordance with personal information protection rules
Method:Among patients in MID-NET (976,859 patient records of 7 sites
from 2009 to 2015), evaluate those who were prescribed codeine-
containing products and hereinafter suspected to develop
respiratory depression in each age group.
呼吸抑制の発生割合
Using MID-NET®, to evaluate the quantity of prescriptions of codeine-
containing products and the risk of developing respiratory depression in
Japan.
Purpose
30
Various kinds of data including laboratory test
results
High data quality (confirmed consistency with
the original data source)
Real-time data update (every 1-4 weeks)
Advantages and Limitation of MID-NET®
Advantages
• May be not enough sample size (currently 4M)
• No linkage of a patient among hospitals
• Need to consider data generalizability due to
limited cooperative organizations (mainly mid-
large size hospitals like University hospitals)
Limitations
Characteristics of MID-NETⓇ and NDB
© 2018 DIA, Inc. All rights reserved
Data Type Electronic Medical Records Health Insurance Claims
Data Provider23 hospitals of 10 Medical
institutions
All health insurers in Japan
Covered patients
People provided medical
service by each institution (~4
Million)
Entire Japanese population
(120 Million)
Obtainable Health
Information
Detailed information in
medical practices by each
institution
Standardized information
relevant to reimbursement
Diagnosis YES YES
Medical procedure YES YES
Pharmacy Dispensing YES (on-site pharmacy) YES
Laboratory test result YES NO
OTC Drug NO NO
National Claims DB
Lessons Learned in utilizing RWD
High data quality is pre-requisite in utilizing real world data such as claims and electronic medical records database.
To obtain clinically meaningful results, it is important to – understand characteristics of the database in details
– validate outcome definitions
– utilize appropriate methods for controllingconfounding (e.g., propensity score matching)
– conduct sensitivity analysis
32
Clinical Innovation Network (CIN)
Study group for epidemiological methods and data quality standards
Study group for ethical issues for registries and relationships with industries
AMED
Advice,Cooperation
MHLW
Utilizing registry data for promoting cost effective clinical
studies, accelerating drug development, and B/R assessment
Output
PMDACIN-Working Group About 20 members
from New drugs & Safety Offices
Muscular dystrophyRegistry
by NCNP
ALS(Antilymphocyticserum) Registry
By Nagoya Univ.
Cancer registry
By National Cancer Center Japan
Cerebral surgery
By Japan neurosurgical
society
34
Revised GPSP
Good Post-marketing Study Practice(The Ministerial Ordinance, Implemented on April 1st 2018)
Intervention
Primary data
collection
“Post marketingclinical trial”
Observation
DB
“Post marketingdatabase study”
Observation
Primary data
collection
“Drug use result survey”
Study frames in GPSP
Newly created
Revised GPSP clearly mentions that safety study based on database is
acceptable for re-examination under the Japanese Pharmaceuticals and
Medical Devices Law
General steps for considering a plan of post-market studies (January 23, 2018)
35
Step 1. What is a concern to be clarified in post-
approval?
Step 2. What is a suitable approach(i.e.; routine or
additional PV)? If additional, what is the
research question and suitable data
source?
Step 3. If additional, which GPSP frame must be
complied with?(clinical trial, observational study
with primary data collection, database)
Step 4. If additional, creating a study protocol
Afte
r a
pp
rova
lN
DA
revie
w
Step 1~4 per each safety specification
Describes basic
principle on how to
plan a post-market
study under Japanese
pharmaceutical
regulation
• Four steps approach
to plan an appropriate
post-market study
Related guideline
Guideline on pharmacoepidemiological study for drugsafety assessment based on medical information database(March 2014)
Basic Principles on the utilization of health informationdatabases for Post-Marketing Surveillance of MedicalProducts (June 2017)
General steps for considering a plan of post-market studies(January 2018)
Points to consider for ensuring data reliability on post-marketing database study for drugs (February 2018)
https://www.pmda.go.jp/safety/surveillance-analysis/0011.html
Many related guidelines focusing on Real World Data utilization
were recently published in synchronization to the GPSP revision
Active utilization of EHR databases toward advanced medical care
37
RMP implementation utilizing EHR databases• Efficient risk management
• Better quality of safety information
Better quality of Medical Care
• Maximize benefit/risk ratio
Provide leading-edge Medical Therapy with
ensuring Safety
Regulatory decisions based on better scientific
evidences• Proper safety assessment utilizing HER databases in
addition to the traditional approaches
• Scientific and speedy safety measure
Medical Institutions
Public
Industries
MHLW/PMDA
Our Goals
Make new drugs and medical devices, developedaround the world, available to patients available in atimely manner (maximize benefits)
Detect unknown risk emerged throughout fromdevelopment to post-marketing as early as possible,to minimize the damage of the patient as a result oftimely action(minimize risk)
Encourage efficient R&D with less waste of cost, andimprove safety measures, so as to promote themedical innovations (cost optimization)
38
PMDA web site
http://www.pmda.go.jp/english/index.html
Thank you very much for your kind attention !!
Ask
top related