Prevention of Emergence of Resistance: A Pharmacodynamic Solution

Prevention of Emergence of Resistance:

A Pharmacodynamic Solution

G.L. Drusano, M.D.Professor and Director

Division of Clinical PharmacologyClinical Research InstituteAlbany Medical College &

New York State Department of Health

Prevention of Emergence of Resistance: A Pharmacodynamic Solution


• Resistance to antimicrobial agents often occur as a function of single point mutations

• Other mechanisms include spread of plasmids with multiple resistance determinants

• Horizontal transmission also confuses the issue• Examples of a point mutation providing drug

resistance are stable derepression of AMP C beta lactamases for 3rd generation cephalosporins and target mutations or pump upregulation for fluoroquinolones


• As these occur at a frequency of 1/108 or less frequently, infection site populations exceed this frequency, often by multiple logs

• Consequently, such total populations do not behave as a single, sensitive population, but as a mixture of two populations of differing drug susceptibility

• This raises an important question:


Can a drug exposure be identified that will prevent the resistant subpopulation from

taking over the total population?

The Team

N. L. Jumbe, A. Louie, W. Liu,V. Tam, T. Fazili, R. Leary, C. Lowry, M.H. Miller and

G. L. Drusano

S. pneumoniae outcome studies

P. aeruginosa outcome studies

Rf in vitro Rfin vivo MIC (g/mL) MBC (g/mL)

2.35x10-6 2.2x10-6 0.8 1.6


• Clearly, Pseudomonas and Pneumococcus differ in their response

• Pneumococcus has no inoculum effect; Pseudomonas has a major inoculum effect

• The explanation probably rests in the mutational frequency to resistance

• Pseudomonas has a high frequency, while Pneumococcus has a frequency that is not measurable at the bacterial densities used in these experiments with this fluoroquinolone

Peripheral (thigh)Compartment (Cp)

Central Blood Compartment (Cc)

IPinjection

kcp kpc

+ Bacteria(XT/R)

f(c)

dCc= kaCa+kpcCp-kcpCc-keCc

dt

ke

dXS=KGS x XS x L - fKS(CcH ) x XS

dtdXR= KGR x XR x L- fKR(Cc

H ) x XR

dt

Kmax CcH

C H

50+CcH

f(CcH)=

Y1=XT=XS+XR

Y2=XR

[3][4]

[5]

[6]

[7]

, =K and = S,R[1]

L = (1-(XS+XR)/POPMAX)

[8]

dCp = kcpCc - kpc Cp

dt[2]

KmaxGS

0.117

KmaxGR

0.163

KmaxKS

94.01

KmaxKR

12.16

HKS

6.26

HKR

2.37

C50KS

123.5

C50KR

129.8

KmaxG -maximum growth rate (hr-1) in the presence of drug

KmaxK -maximum kill rate (hr-1)

C50K -drug concentration (g/mL) to decrease kill rate by half

HK -rate of concentration dependent kill

Popmax -maximal population size

Mean Parameter Estimates of the Model.

Popmax = 3.6 x 1010


• All regimens were simultaneously fit in a large population model

• The displayed graph is the predicted-observed plot for the total population after the Maximum A-posteriori Probability (MAP) Bayesian step



• The displayed graph is the predicted-observed plot for the resistant population after the Maximum A-posteriori Probability (MAP) Bayesian step



• In this experiment, a dose was selected to generate an exposure that would prevent emergence of resistance

• As this was at the limit of detection, the measured population sometimes had “less than assay detectable” for the colony count

• These were plotted at the detection limit


• We were able to determine how the overall (sensitive plus resistant) population responds to pressure from this fluoroquinolone

• More importantly, we were able to model the resistant subpopulation and choose a dose based on simulation to suppress the resistant mutants

• The prospective validation demonstrated that the doses chosen to encourage and suppress the resistant mutants did, indeed, work


• Now, for Pneumococcus• We were unable to recover resistant mutants

with levofloxacin as the selecting pressure in the mouse thigh model

• However, we then examined ciprofloxacin as the selecting agent

• Now, selecting mutants was straightforward

Study Design: Mouse Thigh Infection Model- Ciprofloxacin Studies [50mg/kg BID ~

AUC/MIC 100:1]

Begin therapy

Sacrifice, harvest,homogenize muscle

-2 hr 0 hr1. Microbial eradication

2. Selection of resistance

Infect

24 hr

BID

+ 2xMIC Cipro - Drug + 4xMIC Cipro + 3xMIC Levo

Drug #58 RC2

Cipro/±Reserpine 0.6/0.6 3.5/1.0

Levo/±Reserpine 0.6/0.6 0.6/0.6



• Strain 58, the RC2 and RC4 mutants were sequenced through Gyr A, Gyr B, Par C & Par E.

• The entire open reading frames were sequenced.• No differences were seen between parent and the RC2

daughter strain.• This, coupled with the decrement in ciprofloxacin MIC with

reserpine exposure (3.5 mg/L 1.0 mg/L), implies RC2 is a pump mutant.

• For RC4, a mutation was found in parC (aa 79, sertyr) and this strain also decreased its MIC with addition of reserpine.


• We have examined other new fluoroquinolones in this system or in our hollow fiber pharmacodynamic system

• All resemble levofloxacin and do not allow emergence of resistance for wild type isolates

• Why is ciprofloxacin different?• Likely because it is the most hydrophilic drug

and is most efficiently pumped by the PMRA pump


• Are there other factors that can alter the probability of emergence of resistance?

• The most likely is duration of therapy• Fluoroquinolones induce an SOS response• This resembles a “hypermutator phenotype”• Therapy intensity and therapy duration

should influence the probability of having the resistant population becoming ascendant


• Hollow fiber System allows simulation of human PK in vitro

• Useful for dose ranging and schedule dependency determinations

• Allows examination of different classes (beta lactams, fluroquinolones, etc.)

The original hollow fiber system was used by Blaser & Zinner


• A 10 day hollow fiber experiment was performed for MSSA and MRSA (CS) for 6 regimens

• The time to complete replacement of the population with resistant organisms was recorded

• CART was employed to look for a breakpoint in the exposure

• > 200/1 AUC/MIC ratio was identified


• A stratified Kaplan-Meier analysis was performed with this breakpoint

• The breakpoint was significant (Mantel test p = 0.0007); Tarone-Ware and Breslow Gahan tests were also significant

• To prevent resistance, hit hard (> 200 AUC/MIC) and stop early (< 7 days)


• The intensity of therapy and the duration of therapy have an impact upon the probability of emergence of resistance

• Short duration therapy trials should examine an endpoint of resistance frequency


Placebo

0

2

4

6

8

10

12

0 6 12 18 24 30 36 42 48T ime (h)

Total

ToyamaresistantCiproresistant

Tam et al ICAAC 2001


Cipro (AUC/MIC 65.6)

0

1

2

3

4

5

6

7

8

9

10

0 6 12 18 24 30 36 42 48

Time (h)

Total




T600 (AUC/MIC 3.2)

0

2

4

6

8

10

0 6 12 18 24 30 36 42 48 T ime (h)

Log1

0 CFU

/mL

Total

Toyamaresistant

Ciproresistant



T1800 (AUC/MIC 10.4)

01

23

456

78

910

0 6 12 18 24 30 36 42 48 Time (h)

Log1

0 CFU

/mL

Total




T13500 (AUC/MIC 88.6)

0

1

2

3

4

5

6

7

8

9

0 6 12 18 24 30 36 42 48

Time (h)

Total




T18000 (AUC/MIC 108.3)

0

1

2

3

4

5

6

7

8

9

0 6 12 18 24 30 36 42 48

Time (h)

Total




T36000 (AUC/MIC 200.8)

0

1

2

3

45

6

7

8

9

0 6 12 18 24 30 36 42 48

Time (h)

Total



Central Compartment (Cc)Infusion + Bacteria

(XT/R)

f(c)

dCc=Infusion-(SCl/V)xCc

dt

SCl

dXS=KGS x XS x L - fKS(CcH ) x XS

dtdXR= KGR x XR x L- fKR(Cc

H ) x XR

dt

Kmax CcH

C H 50 +Cc

H f(Cc

H)=

Y1=XT=XS+XR, IC(1)=2.4x108

Y2=XR , IC(2)= 30

[2][3]

[4]

[5]

[6]

, =K and = S,R

[1]

L = (1-X/POPMAX)

[7]

KmaxGS

0.745

KmaxGR

0.614

KmaxKS

27.85

KmaxKR

31.72

HKS

2.24

HKR

3.50

C50KS

16.94

C50KR

107.0

KmaxG -maximum growth rate (hr-1) in the presence of drug

KmaxK -maximum kill rate (hr-1)

C50K -drug concentration (g/mL) to decrease kill rate by half

HK -rate of concentration dependent kill

Popmax -maximal population size

Mean Parameter Estimates of the Bacterial Growth/Kill Model.

Popmax = 3.3 x 1010



• The displayed graph is the predicted-observed plot for the drug concentrations after the Maximum A-posteriori Probability (MAP) Bayesian step



• The displayed graph is the predicted-observed plot for the total bacterial counts after the Maximum A-posteriori Probability (MAP) Bayesian step



• The displayed graph is the predicted-observed plot for the resistant bacterial counts after the Maximum A-posteriori Probability (MAP) Bayesian step


•‘Inverted-U’ Phenomenon

– Resistant sub-populationis are initially amplified & then decline with increasing drug exposure 0

1

2

3

4

5

6

0 1 2 3 4 5 6 7

Therapeutic Intensity

Log1

0 C

FU/m

L

ResistantSub-Population


P. aeruginosa - Prevention of Amplification of Resistant Subpopulation

• The amplification of the resistant sub-population is a function of the AUC/MIC ratio

• The response curve is an inverted “U”.

• The AUC/MIC ratio for resistant organism stasis is circa 187/1



P. aeruginosa - Prevention of Amplification of Resistant Subpopulation

Placebo

0

5

10

15

0 12 24 36 48 60 72 Time (h)

Log1

0 C

FU/m

L

Total

Resistant

AUC/MIC 136.7

02468

10

0 12 24 36 48 60 72 Time (h)

Log1

0 C

FU/m

L

Total

Resistant

AUC/MIC 199.7

02468

10

0 12 24 36 48 60 72 Time (h)

Log1

0 C

FU/m

LTotal

Resistant

AUC/MIC 165.8

02468

10

0 12 24 36 48 60 72 Time (h)

Log1

0 C

FU/m

L

Total

Resistant

Prospective Validation


• This was the same strain as employed in the mouse model, but a different fluoroquinolone

• The mouse model contained granulocytes, while the hollow fiber system does not

• The total drug target for the mouse model was 157 which is a free drug target of 110

• The hollow fiber system target is 187 (1.7 fold )• Craig found that targets increase by 1.5 -2.0 fold

when granulocytes are removed• These results are concordant with this finding


Multiple Bacterial Populations Do Make a Difference!

• In Vitro pharmacodynamic model investigations frequently only examine the total bacterial population

• The presence of a small pre-existent population more resistant to the selecting drug pressure has major implications, particularly as the bacterial population size increases to (near) clinical infection size

0

2

4

6

8

10

0 50 100 150 200


P aeruginosa

Log1

0 C

FU/m

L

Daily AUC/MIC

Breakpoint = 187

0

2

4

6

8

10

0 50 100 150 200


K. pneumoniae

Log1

0 C

FU/m

L

Daily AUC/MIC

Breakpoint = 93

0

1

2

3

4

5

6

0 20 40 60 80 100 120 140 160


MSSA

Log1

0 C

FU/m

L

Daily AUC/MIC

Breakpoint = 66

0

2

4

6

8

0 20 40 60 80 100 120 140 160


MRSA-CS

Log1

0 C

FU/m

L

Daily AUC/MIC

Breakpoint = 143

0

2

4

6

8

10

12

0 100 200 300 400 500


MRSA-CR

Log1

0 C

FU/m

L

Daily AUC/MIC

Breakpoint = 484


• Some drug exposures allow amplification of the resistant subpopulations

• Exposures can be identified that will prevent this amplification and, functionally suppress emergence of resistance

Prevention of Emergence of Resistance: A Pharmacodynamic Solution

Documents

emergence of resistance

drug resistance

kgs x xs x

pharmacodynamic solutioncan

pharmacodynamic solutiong

pharmacodynamic solutionas

resistant population

kgr x xr x