prof. hartmut derendorf university of florida the role of pharmacological predictors in drug...

Prof. Hartmut Derendorf

University of Florida

The Role of Pharmacological

Predictors in Drug Development

Resistance Development

Approved Antibacterial Agents1983-2004

Pharmacokineticsconc. vs time

Co

nc.

Time0 250.0

0.4

PK/PDeffect vs time

Time

Eff

ect

0

1

0

Pharmacodynamicsconc. vs effect

10-3Conc. (log)

Eff

ect

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Cmax

Time (hours)

Cmax/MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Cmax

Time (hours)

Cmax/MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

t > MIC

Time (hours)

Time above MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

t > MIC

Time (hours)

Time above MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Time (hours)

AUC24/MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Time (hours)

AUC24/MIC

PK PD

Serum MIC

Pharmacokinetics

Problems:

• Protein Binding

• Tissue Distribution

vascular space extravascular space

plasma protein binding

blood cell binding,

diffusion into blood cells,

binding to intracellular biological material

tissue cell binding,

diffusion into tissue cells,

binding to intracellular biological material

binding to extracellular biological material

AzithromycinTissue Concentrations

tonsil (t)prostate (P)lung (L) serum (S)

500 mg p.o. from Foulds et al. (1990)

Tissue can be looked at as an aqueous dispersed system of biological material. It is the concentration in the water of the tissue that is responsible for pharmacological activity.

Total tissue concentrations need to be interpreted with great care since they reflect hybrid values of total amount of drug (free + bound) in a given tissue

‘Tissue-partition-coefficients’ are not appropriate since they imply homogenous tissue distribution

Tissue Concentrations

The free (unbound) concentration of

the drug at the receptor site should be

used in PK/PD correlations to make

prediction for pharmacological activity

Blister Fluid

• Blister fluid is a ‘homogenous tissue fluid’

• Protein binding in blister fluid needs to be considered

Ampicillin

Cloxacillin

Serum Free blister fluid

Interstitium

CapillaryCell

PerfusateDialysate

Microdialysis

S

NO

NH

CH

COOH

C

O

CS

NNH2 N O CH2COOH

CH2

S

NO

NH

CH2

COOH

C

O

CS

NNH2 N O CH3

O CH3

Cefixime (Protein binding 65%)

Cefpodoxime(Protein binding 17-30%)

M uscle

Lung

H um an S tud ies

M uscle Lung

Anim al S tud ies

Pharm acokinetics

i.v. 20 mg/kg in rats

Time (min)

Concentr

ation (

mg/L

)

0 50 100 150 200 250 300

100

101

102

103

iv dose of 20 mg/kg of cefpodoxime (n=6)

plasma lung muscle

Clinical study Cefpodoxime and

Cefixime

• To compare the soft tissue distribution of these two antibiotics after 400mg oral dose in healthy male volunteers by microdialysis

• Two way cross-over, single oral dose study

Microdialysis

0

1

2

3

4

5

6

0 2 4 6 8 10

Time (h)

Co

nce

ntr

ato

in (

mg

/L) plasma muscle free plasma

0

1

2

3

4

5

6

0 2 4 6 8 10

Time (h)

Co

nc

en

tra

tio

n (

mg

/L)

plasma muscle free plasma

Cefixime400 mg po

Cefpodoxime400 mg po

Clinical Microdialysis

Liu & Derendorf, JAC 50, 19 (2002)

Pharmacokinetics

Cefpodoxime Cefixime

AUCP [mg*h/L] 22.4 (8.7) 25.7 (8.4)

AUCT [mg*h/L] 15.4 (5.2) 7.4 (2.1)

Cmax, P [mg/L] 3.9 (1.2) 3.4 (1.1)

Cmax,T [mg/L] 2.1 (1.0) 0.9 (0.3)

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Cmax

Time (hours)

Cmax/MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Cmax

Time (hours)

Cmax/MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

t > MIC

Time (hours)

Time above MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

t > MIC

Time (hours)

Time above MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Time (hours)

AUC24/MIC

0 6 18 2412

Con

cent

ratio

n (µ

g/m

L)

0

8

12

16

4

MIC

Time (hours)

AUC24/MIC

PK PD

Serum MIC

CeftazidimeK. pneumoniae in neutropenic mice

Craig 2002

TemafloxacinS. pneumoniae in neutropenic mice

Craig 2002

Pharmacodynamics

Problems:

• MIC is imprecise

• MIC is monodimensional

• MIC is used as a threshold

• When MIC does not explain the data, patches are used(post-antibiotic effect, sub-MIC effect)

Auto-dilution system

flaskreservoir

tubingconnector

pump

waste

Kill Curves

H. influenzae ATCC10211

MIC: 5 ng/mLS. pneumoniae ATCC6303

MIC: 20 ng/mL

Kill Curves of Ceftriaxone

S. pneumoniae ATCC6303

MIC: 20 ng/mL H. influenzae ATCC10211

MIC: 5 ng/mL

Kill Curves of Ceftriaxone

NCEC

Ckk

dt

dN

f

f

50

max

Maximum Growth Rate Constant k

Maximum Killing Rate Constantk-kmax

PK-PD Model

Initially, bacteria are in log growth phase

Single DosePiperacillin vs. E. coli

0 2 4 6 8 10

Time (h)

100

101

102

103

104

105

106

107

108

109

1010

1011

1012

1013

1014C

FU

/mL

control

2g

8g

4g

0 5 10 15 20 25102

103

104

105

107

108

109

1010

1011

CF

U/m

L

106

Time (h)

50µg/mL q24h

0 5 25102

103

104

105

106

107

108

109

1011

10 2015

CF

U/m

L

1010

100µg/mL q24h

Time (h)

0 5 10 15 20 25102

103

104

105

106

107

108

109

1011

1010

CF

U/m

L

50µg/mL q8h

Time (h)

0 5 10 15 20 25102

103

104

105

106

107

108

109

1010

1011

CF

U/m

L

100µg/mL q8h

Time (h)

0 5 10 15 20 25102

103

104

105

106

107

108

109

1010

1011

CF

U/m

L

50µg/mL q4h

Time (h)

0 5 10 15 20 25102

103

104

105

106

107

108

109

1010

1011

CF

U/m

L

100µg/mL q4h

Time (h)

Dosing IntervalPiperacillin (2g and 4g) vs. E. coli

q24h q8h q4h

PK-PD ModelIn animals

Bacterial survival fraction of P. aeruginosa in a neutropenic mouse model at different doses (mg/kg) of piperacillin (Zhi et al., 1988)

FDA Draft-Guidance for Industry (1997)Providing Clinical Evidence of Effectiveness for Human Drug and Biological Products

New Dosage Form of a Previously Studied Drug

In some cases, modified release dosage forms may be approved on the basis of pharmacokinetic data linking the new dosage form from a previously studied immediate-release dosage form. Because the pharmacokinetic patterns of controlled-release and immediate release dosage forms are not identical, it is generally important to have some understanding of the relationship of blood concentration to response to extrapolate to the new dosage form.

Plasma and free tissue levels

n = 12 (means +/- S.D.) total plasma concentrations free tissue concentrations

500 mg IR

Plasma and free tissue levels

n = 12 (means +/- S.D.) total plasma concentrations free tissue concentrations

500 mg MR 750 mg MR

Cefaclor 750 mg MR bid vs 500 mg IR tid

time (h)

0 6 12 18 24

Streptococcus pneumoniae

CF

U/m

l

101

102

103

104

105

106

107

108

109

1010

1011

1012

1013

Resistance Modeling

• Pre-existing subpopulations

• Emerging subpopulations

• Adaptive resistance (transporters)

Drug (C)

fs(C)

Growth

( k0)fr(C)

Bacteria (R)

Bacteria (S)

Bacteria pool

Killing

Two sub-population model

OBS: same growth rate for sensitive (S) and resistant (R)

tzs

s

tz

r

r

eNsCEC

CKk

dt

dNs

eNrCEC

CKk

dt

dNr

1

1

50

max0

50

max0

NsNrN t

Two sub-population Emax model

tzr

r

eNCEC

CkCIC

Ck

kdt

dN

1

1

50

250

1

laglage ttttkr eeCC

0

Dose

ka

Cp ke

kill (-) Cr

k0 kecr

Bacteria2

501max 1 k

CrIC

CrkK

Comparing to Emax model:

Modified Emax Model:

control

0 4 8 12 16 20 24 28 32 36 40 44 48

Time (hours)

100

101

102

103

104

105

106

107

108

109

1010

1011

1012

CF

U/m

L

0.03 0.06 0.13 0.25 0.5

E. coli (MIC=0.013 mg/L)

t (h)

CF

U/m

L

0 12 24 36 4810-5

10-4

10-3

10-2

10-1

100

101

102

103

104

105

106

107control 0.03 0.25 0.50.130.06

E. coli

Two sub-population model

(simultaneous fit)

Modified Emax model

(simultaneous fit)

Model Comparison – E. coli

• A simple comparison of serum concentration and MIC is usually not sufficient to evaluate the PK/PD-relationships af anti-infective agents.

• Protein binding and tissue distribution are important pharmacokinetic parameters that need to be considered. Microdialysis can provide information on local exposure.

• PK-PD analysis based on MIC alone can be misleading.

• Microbiological kill curves provide more detailed information about the PK/PD-relationships than simple MIC values.

Summary

Proposal

Wild Card Patent Extension

A company that receives approval for a new antibiotic, or a new indication for an existing

antibiotic, that treats a targeted pathogen would be permitted to extend the market exclusivity period for

another of the company’s FDA-approved drugs.

ISAPInternational Society of Anti-

Infective Pharmacology• Workshops at ECCMID and ICAAC• Symposia at ECCMID and ICAAC• Joint Symposium with FDAand IDSA• Website with slides, presentations and tons of

information

www.isap.org

AcknowledgementsEdgar Schuck

Qi Liu

Ping Liu

Teresa Dalla Costa

Amparo de la Peña

Ariya Khunvichai

Arno Nolting

Wanchai Treeyaprasert

Stephan Schmidt

Elizabeth Potocka

Martin Brunner

Markus Müller

Kenneth Rand

Alistair Webb

Maria Grant

Andreas Kovar

Olaf Burkhardt

Vipul Kumar

Yanjun Li

Oliver Ghobrial

April Barbour