using mycobacterium tuberculosis single nucleotide...

Using Mycobacterium tuberculosis single nucleotide polymorphisms to predict fluoroquinolone treatment response

Marva Seifert, Edmund Capparelli, Donald Catanzaro, Timothy Rodwell

TB International Workshop On Clinical Pharmacology of Tuberculosis DrugsThe Hague, Netherlands October 23, 2018

A TB Patient Dies Every 18 SecondsBy the end of this talk ~50 more TB patients will have died

Introduction

• TB treatment success rates• 82% for all TB

• 55% for MDR-TB (2016 cohort)

• 34 % for XDR-TB (2015 cohort)

• Options for improvement of treatment success rates• Novel therapies

• Optimize current treatment

Global Tuberculosis Report 2018. Geneva: World Health Organization, 2018.

Low MICN

um

ber

of

Iso

late

s

Wild type(“susceptible”)

Clinical Breakpoint

High MIC

Mutant(“resistant”)

= “treatable” by WHO Standards= “treatable” by Dosing Optimization

MICs above safely achievable serum concentration in patients

Critical Concentration

Optimizing Current Treatment

• Based on phenotypic/genotypic resistance detection

• Critical Concentration vs. clinical breakpoint

• Lost opportunity to optimize dosing

Global Tuberculosis Report 2018. Geneva: World Health Organization, 2018.

Goals of Study

Proof of Concept• Mtb mutations can predict Fluoroquinolone MICs• Predicted MICs can be combined with population-level PK/PD

modeling to provide insights into individual dosing decisions

Methods: Isolate selection and categorization

• Clinical isolate selection

• Archived clinical isolates from repositories in Mumbai (India), Chisinau (Moldova), Manila (Philippines)

South Africa

• Isolates were sequenced at gyrA using Pyrosequencing, Sanger, or PacBio whole genome sequencing

• Randomly selected up to 20 isolates with the following gyrA SNPs: 88TCG, 90GTG, 91CCG, 94GGC, 94GCC,

94AAC, 94TAC, 94CAC, as well as gyrA wild types

• Isolate processing (n=138)

• Cultured isolates and did DST with serial dilutions GFX, MFX, LFX, and OFX (~5 dilutions above critical

concentration and ~3 dilutions below)

• SNP categorization into high or low MICs

• SNPs grouped by MIC modes (wildtype, low resistance, and high resistance)

• Differentiation was demonstrated statistically (Kruskal-Wallis test, Dunn’s test)

Rodwell TC, Valafar F, Douglas J, et al. Predicting extensively drug-resistant Mycobacterium tuberculosis phenotypes with genetic mutations. Journal of clinical microbiology 2014; 52(3): 781-9.Hillery N, Groessl EJ, Trollip A, et al. The Global Consortium for Drug-resistant Tuberculosis Diagnostics (GCDD): design of a multi-site, head-to-head study of three rapid tests to detect extensively drug-resistant tuberculosis. Trials 2014; 15: 434.

Methods: Population PK/PD modeling

• 10,000 patients were simulated using NONMEM

• AUC (for each FQ) calculated based on virtual subjects PK

parameters and standard PK equations

(AUC=bioavailability*dose/clearance)

• Probability of AUC/MIC target attainment

• Simulated using SAS 9.4 & published population PK/PD models for

each FQ, linked (randomly) to SNP estimated MIC distributions

• FQ AUC/MIC targets based on published literature Time

Seru

m C

on

cen

trat

ion

MIC

lower dose

higher dose

AUC

Peloquin CA, Hadad DJ, Molino LP, et al. Population pharmacokinetics of levofloxacin, gatifloxacin, and moxifloxacin in adults with pulmonary tuberculosis. Antimicrobial agents and chemotherapy 2008; 52(3): 852-7.Zvada SP, Denti P, Sirgel FA, et al. Moxifloxacin population pharmacokinetics and model-based comparison of efficacy between moxifloxacin and ofloxacin in African patients. Antimicrobial agents and chemotherapy 2014; 58(1): 503-10.Smythe W, Merle CS, Rustomjee R, et al. Evaluation of initial and steady-state gatifloxacin pharmacokinetics and dose in pulmonary tuberculosis patients by using monte carlo simulations. Chigutsa E, Meredith S, Wiesner L, et al. Population pharmacokinetics and pharmacodynamics of ofloxacin in South African patients with multidrug-resistant tuberculosis. Antimicrobial agents and chemotherapy 2012; 56(7): 3857-63.

Results: SNPs categorized as high or low level resistance

0

10

20

30

40

0,1875 0,375 0,75 1,5 2,5 3,5 4,5 10 15

Nu

mb

er o

f Is

ola

tes

mg/L

C) LFX

01020304050607080

0.0625 0.125 0.25 0.5 1.5 2.5 >10

Nu

mb

er o

f Is

ola

tes

mg/L

B) GFX

0

10

20

30

40

0.062 0.125 0.25 0.5 1 1.5 2 3 >3

Nu

mb

er o

f Is

ola

tes

mg/L

A) MFX WT LOW MIC HIGH MIC

0

10

20

30

40

50

60

70

80

0.5 1 2 2.5 3 3.5 4 5 >5.0N

um

ber

of

Iso

late

smg/L

D) OFX

Cumulative distribution plots of proportion of population likely to reach defined therapeutic targets for MFX

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350 400

Per

cen

t

AUC/MIC

Wild type (800 mg)

Wild type (400 mg)

Target (106)

Gumbo T, Louie A, Deziel MR, Parsons LM, Salfinger M, Drusano GL. Selection of a moxifloxacin dose that suppresses drug resistance in Mycobacterium tuberculosis, by use of an in vitro pharmacodynamic infection model and mathematical modeling. The Journal of infectious diseases 2004; 190(9): 1642-51.Companion handbook to the WHO guidelines for the programmatic management of drug-resistant tuberculosis.; 2014.


MFX (target 106)5% of low MIC pop @ 400 mg/day44% of low MIC pop @ 800 mg/day

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350 400

Per

cen

t

AUC/MIC

Low level resistance mutations (800 mg)


Target (106)



0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350 400

Per

cen

t

AUC/MIC

High level resistance mutations (800 mg)


Target (106)


Cumulative distribution plots of proportion of population likely to reach defined therapeutic targets for GFX

GFX (target 125)22% of low MIC population at 400 mg/day59% of low MIC population at 800 mg/day73% of low MIC population at 1200 mg/day

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400

Per

cen

t

AUC/MIC

Wild type (1200 mg)

Wild type (800 mg)





targets (125 & 184)

Wild type (400 mg)




Cumulative distribution plots of proportion of population likely to reach defined therapeutic targets for LFX

LFX (target 146)4% of low MIC pop @ 750 mg/day14% of low MIC pop @ 1250 mg/day

0

10

20

30

40

50

60

70

80

90

100

0 100 200 300 400

per

cen

t

AUC/MIC

Wild type (1250 mg)

Wild type (750 mg)





targets (100 & 146)

Lanoix JP, Chaisson RE, Nuermberger EL. Shortening Tuberculosis Treatment With Fluoroquinolones: Lost in Translation? Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2016; 62(4): 484-90.Companion handbook to the WHO guidelines for the programmatic management of drug-resistant tuberculosis.; 2014.

Cumulative distribution plots of proportion of population likely to reach defined therapeutic targets for OFX

OFX (target 100)0% of low MIC pop @ 600 mg/day0% of low MIC pop @ 800 mg/day

0

10

20

30

40

50

60

70

80

90

100

0 50 100 150 200 250 300 350 400

Per

cen

t

AUC/MIC

Wild type (800 mg)

Wild type (600 mg)





target (100)

Lanoix JP, Chaisson RE, Nuermberger EL. Shortening Tuberculosis Treatment With Fluoroquinolones: Lost in Translation? Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 2016; 62(4): 484-90.Companion handbook to the WHO guidelines for the programmatic management of drug-resistant tuberculosis.; 2014.

Conclusions from this Proof-of-Concept Study

• SNPs in gyrA gene of Mtb can be used to predict MIC distributions against individual fluoroquinolones

• SNP estimated MICs together with population PK/PD models could be used as decision support tools for rapid individualized TB treatment based on rapid molecular diagnostics

• Methods are generalizable to other antimicrobial agents which have reliable genetic markers of resistance in Mycobacterium tuberculosis

AcknowledgementsWe are grateful to Mark Pettigrove for his laboratory support for this study

Disclosure StatementDr. Rodwell reports grants from National Institutes of Health (NIH) and salary support from the not-for-profit organization Foundation for Innovative New Diagnostics (FIND). Dr. Capparelli reports personal fees from Celltrion, personal fees from Melinta, personal fees from Patara, personal fees from Atox Bio, personal fees from Ironshore, outside the submitted work. Drs. Seifert and Catanzaro have nothing to disclose.

FundingThis work was supported by NIH R01AI111435 and T32HL134632

using mycobacterium tuberculosis single nucleotide...

Documents