pharmacodynamics and the dosing of antibacterials

Pharmacodynamics and theDosing of Antibacterials

William A. Craig, MD

University of Wisconsin and

Wm S. Middleton Memorial VA Hospital

Madison, WI USA

PRESENTATION OUTLINE• History of pharmacodynamics• Different patterns of antimicrobial activity• Pharmacodynamic target setting in animal

models• Identify major factors that alter the

pharmacodynamic target• Correlation between animal and human

studies• Uses of pharmacodynamic modeling • Application to susceptibility breakpoints for

gram-negative bacilli

What is Pharmacodynamics ?

• Relationship between concentration and pharmacologic and toxicologic effects of a drug

• Relationship between measures of in vivo drug exposure and microbiologic (and clinical) efficacy

Why Interest in Pharmacodynamics?

• Always occurs when there is narrow difference between MICs and obtainable drug concentrations

- early days of penicillin in 1940s (low doses used)

- appearance of pseudomonas infections in 1960s and 1970s (high MICs)

- appearance of resistant Streptococcus pneumoniae and other bacteria

in 1990s

Bactericidal Activity of Tobramycin

and Ticarcillin againstPseudomonas aeruginosa

Time (hours)

0 2 4 6 8 12

Log

CF

U p

er T

high

og

4

5

6

7

8

9

4 mg/kg 12 mg/kg20 mg/kg

0 2 4

300 mg/kg 800 mg/kg2400 mg/kg

Tobramycin Ticarcillin

Vogelman et al. J Infect Dis 157, 1988

1st Pattern of Antimicrobial Activity

Concentration-dependent killing and prolonged persistent effects

Seen with quinolones, aminoglycosides, ketolides, and daptomycin

Goal of dosing regimen: maximize concentrations

2nd Pattern of Antimicrobial Activity

Time-dependent killing and minimal or no persistent effects (except with staphylococci)

Seen with all beta-lactams

Goal of dosing regimen: optimize duration of exposure; maximum killing when levels constantly above 4-5 times MIC

3rd Pattern of Antimicrobial Activity

Time-dependent killing and moderate to prolonged persistent effects

Seen with macrolides, azithromycin, clindamycin, tetracyclines, glycylcyclines, streptogramins, glycopeptides, oxazolidinones, deformylase inhibitors

Goal of dosing regimen: optimize amount of drug; maximum killing when T>MIC 100%

Major Goal of Pharmacodynamics

Establish the PK/PD TARGET required for effective antimicrobial therapy

- identify which PK/PD indice (T>MIC, AUC/MIC, peak/MIC) best predicts in vivo antimicrobial activity

- determine the magnitude of the PK/PD parameter required for in vivo efficacy (static effect, 1 or 2 log kill)

Neutropenic Mouse Thigh-Infection Model

2. Bacteria injected into thighs on day 0 (106-7)

1. Neutropenia induced by 2injections of cyclophosphamideon days -4 and -1

3. Treatment (usually given SQ)started 2 hr after infection and continued for 1-5 days

4. Thighs removed, homogenized, serially diluted and plated forCFU determinations

Time Course of Antimicrobial Activity of Ciprofloxacin Against Klebsiella ATCC

43816in Lungs of Neutropenic Mice

Days0 1 2 3 4 5

Lo

g10

CF

U/L

un

g

2

4

6

8

10

Saline0.29 q 6h1.17 q 6h4.69 q 6h18.8 q 6h75.0 q 6h

DeathDeath Death

Correlation of PK/PD Parameters with Efficacy Levofloxacin against Streptococcus

pneumoniae in Thighs of Neutropenic Mice

24-Hr AUC/MIC

10 100 1000

Log

10 C

FU /

Thig

h at

24

Hrs

0

2

4

6

8

10

Peak/MIC

1 10 100 1000

Time Above MIC

0 25 50 75 100

Andes & Craig, Int J Antimicrob Agents, 2002

Relationship Between PK/PD Parameters and Efficacy for Cefpirome against Klebsiella

pneumoniae in Lungs of Neutropenic Mice

24-Hr AUC/MIC

10 100 1000

Log

10 C

FU/T

high

at 2

4 H

rs

2

3

4

5

6

7

8

9

10

Peak/MIC

1 10 100 1000

Time Above MIC

0 25 50 75 100

Craig Antimicrobial Pharmacodynamics in Theory and Practive 2002

Mathematical Analysis of Dose-Response Data from Animal Models

after 24 Hours of Therapy

Nonlinear regression and Hill equation to estimate Emax (difference from untreated control), P50 (dose giving 50% of Emax) and slope (N) of the dose-response relationship

Dose (mg/kg/6 hrs)10 30 100 300

Lo

g1

0 C

FU

pe

r T

hig

h a

t 2

4 H

rs

5

6

7

8

9

Static Dose

1 Log K ill

P50

Emax

CFU=(Emax) DoseN

DoseN + P50N

Craig Diagn Microbiol Infect Dis 1995

Factors That Affect the Magnitude of PK/PD Parameters

- dosing regimen

- drug class

- protein binding

- infecting pathogen

- presence or absence of neutrophils

- site of infection

Magnitude of PK/PD Parameters

Does the magnitude of the parameter vary with:

1. different dosing regimens? NO, unless the clearance is too rapid in the animal model

Impact of Dosing Interval on Static Dose for Amikacin against K. pneumoniae and E. coli

in Mice with Normal and Impaired Renal Function

Dosing Interval (Hours)

3 6 12 24Sta

tic D

ose

(mg/

kg/2

4 hr

s)

1

10

100

1000

K. pneumoniaeE. coliNormal T1/2=17 min K. pneumoniae E. coliRenal Impaired T1/2=104 min

Reddington and Craig, JAC 1995



1. different dosing regimens? NO

2. different drugs within the same class? NO, if free drug levels are used

Protein binding is important !!

24-Hr AUC/MIC with Total and Free Drug for the Static Dose of Different Fluoroquinolones

with S. pneumoniae ATCC 10813

0

40

80

120

160

Gati Sita Moxi Gemi Garen Levo Cipro

24-H

r A

UC

/MIC

Total

Free

Andes & Craig 40th and 41st ICAAC, 2000 and 2001

Time Above MIC for Total and Free Drug for

the Static Dose of Different Cephalosporins with Klebsiella pneumoniae ATCC 43816

0

10

20

30

40

50

60

70

Total

Free

Craig Infect Dis Clin N Amer 2003



1. different dosing regimens? NO

2. different drugs within the same class? NO, if free drug levels are used

3. different organisms ? Similar for most organisms in neutropenic mice; less T>MIC with

staphylococci with ß-lactams

Drug GNB S. pneumoniae S.aureus

Ceftazidime 36 (27-42) 39 (35-42) 22 (19-24)

Cefpirome 35 (29-40) 37 (33-39) 22 (20-25)

Cefotaxime 38 (36-40) 38 (36-40) 24 (20-28)

Ceftriaxone 38 (34-42) 39 (37-41) 24 (21-27)

Time Above MIC Required for a Static Effect with 4 Cephalosporins

Time Above MIC (% of Dosing Interval)

Craig Diagn Microbiol Infect Dis 22:89, 1995



4. resistant organisms

No

T>MIC for Free Drug for Static Doses with Cephalosporins, Penicillins and

Carbapenems against Multiple Strains of S. pneumoniae with Various Penicillin MICs

MIC (mg/L)0.008 0.03 0.12 0.5 2 8

T>

MIC

(%

)

0

10

20

30

40

50

Cephalosporins

Penicillins

Carbapenems

24-Hr AUC/MIC of Gemifloxacin (Free Drug) for the Static Doses with 61 Strains of

S. pneumoniae in Thighs of Neutropenic Mice

MIC (mg/L)

0.008 0.03 0.12 0.5 2

Fre

e D

rug

AU

C/M

IC

2.5

5

10

20

40

80

Wild typemutants

Efflux



4. resistant organisms

NO

5. different sites of infection (e.g. blood, lung, peritoneum, soft tissue)?

NO for most drugs; vancomycin lesspotent in lung infection than thigh

Factors That Do Not Affect the Magnitude of PK/PD Parameters

- resistant organisms

- animal species

Can values obtained in animals be predictive of efficacy in humans ?

T>MIC for ß-Lactams Versus Mortalityin Animal Models: Literature Review

0 20 40 60 80 100

Mo

rta

lity

(%)

0

20

40

60

80

100 Cephalosporins Penicillins

Time Above MIC (% of Interval)Craig Antimicrobial Pharmacodynamics in Theory and Practive 2002

Double Typanocentesis Studies in Acute Otitis Media

First tap to identify the causative pathogen

Second tap at 3-5 days to determine if organisms had been eradicated

Techniques first performed by Howie. Most recent studies with resistant organisms performed by Dagan.

Time above MIC calculated from serum levels and MICs

Craig & Andes, Pediatr Infect Dis J 15:255, 1996; Multiple studies by Dagan et al

Relationship Between T>MIC and Bacterial Eradication with

ß-Lactams in Acute Otitis Media

Time Above MIC (%)

0 20 40 60 80 100

Bac

teri

al E

rad

icat

ion

(%

)

0

20

40

60

80

100

PSSP PISP-PRSPH. fluBest Fit

Craig Infect Dis Clin N Amer 2003

Uses of Pharmacodynamic Studies

• Drug development- new formulations active against organisms with high MICS (e.g. high dose amoxicillin/clavunate)- dosage regimens for phase II and III clinical trials- drug selection for clinical studies

• Optimize dosing regimens- longer infusions and continuous infusion

of beta-lactams - once-daily dosing of aminoglycosides

Uses of Pharmacodynamic Studies

• Guidelines for antimicrobial usage

• Reduction of emergence of resistance organisms

• Modifications of susceptibility and resistance breakpoints - oral ß-lactams for S. pneumoniae - parenteral cephalosporins for S.

pneumoniae - fluoroquinolones for S. aureus

ESBLs and Clinical Failures• Emergence of Enterobacteriaceae with

increased MICs due to novel ß-lactamases (ESBLs)

• Associated with clinical failures and existing breakpoints fail to detect all these enzymes

• Development by NCCLS of reference methods for screening and confirmation of ESBLs in some Enterobacteriaceae; does not work for all Enterobacteriaceae.

• Elevated MICs in may also arise from mechanisms not targeted by the current ESBL test

Pharmacodynamic Studies of Extended-Spectrum Cephalosporins against

Enterobacteriaceae with and without ESBLs in Thighs of Neutropenic Mice

Drugs: Cefotaxime, Ceftriaxone, Ceftazidime and Cefepime

ESBLs: TEM-3, TEM-7, TEM-10, TEM-12, TEM-26, SHV-2, SHV-4, SHV-5, SHV-7, CTX-M2, CTX-M3 alone or in combination with OXA-1, OXA-2, TEM-1, and/or AMP-C

Non-ESBLs: TEM-1, TEM-1 and OXA-1, AMP-C, wild-type

High inocula – 107-8

Activity of 4 Cephalosporins against Various Enterobacteriacea with and without ESBLs

inMurine Thigh-Infection Model

Time Above MIC (percent)

0 20 40 60 80 100

Cha

nge

in L

og10

CF

U/T

high

over

24

Hou

rs

-3

-2

-1

0

1

2

3

0 20 40 60 80 100

-3

-2

-1

0

1

2

3ESBLs Non-ESBLs

Monte Carlo Simulation

PK Variation In Normal Volunteersor Patients

PK Variation in10,000 Patients

Simulate

Determine Percentage of Patientsthat would meet the PK/PD Targetrequired for efficacy

Drusano et al

Monte Carlo Simulation: CefotaximePercent of 10,000 Patients Attaining Indicated PK/PD Exposure Target

T>MIC with 2g every 8 hr

MIC 30% 40% 50% 60% 70%

0.5 100 100 99 96 88

1 100 99 97 89 73

2 99 98 89 71 49

4 98 91 67 41 22

8 92 62 29 11 4

16 65 15 3 0 0

Ambrose & Dudley, ICAAC 2002

Clinical Outcome in 42 Patients with ESBL-Producing Klebsiella/E. coli

Bacteremia and Treated with Cephalosporin Monotherapy

Outcome

MIC

<1 ųg/L

MIC

2 ųg/L

MIC

4 ųg/L

MIC

8 ųg/L

Success 13 (81%) 4 (67%) 3 (27%) 1 (11%)

Failure 3 (19%) 2 (33%) 8 (73%) 8 (89%)

Paterson et al J Clin Micro 39:2206, 2001; Kim et al AAC 46:1481,2002; Wong-Beringer et al Clin Infect Dis 34:135, 2002; Kang et alAAC In press 2004; Bhavani et al 44rd ICAAC, Abstract K-1588, 2004

Distribution of Enterobacteriaceawith MICs of 4 and 8 mg/L

Total Strains

E. coli Kleb spp Other

Ceftazidime 464 89

(19%)

158

(34%)

217

(47%)

Ceftriaxone 532 46

(9%)

169

(32%)

317

(59%)

2002-2003 Sentry Program

Total 0f 11, 913 Isolates from ICU Patients

Jones, R

Acknowledgements

Pharmacodyamics

Andes DR

Bundtzen R

Ebert S

Fantin B

Gerber A

Gudmundsson S

Kiem S

Leggett J

MacDonald

Mouton J

Redington J

Totska K

Turnidge JVan Ogtrop MVogelman BWalker RWatanabe Y

OtitisDagan R

ESBLsAmbrose PDudley MJones R

It all started with a mouse.

-Walt Disney

pharmacodynamics and the dosing of antibacterials

Documents