Influence of Bacterial Growth Rate on Dose Optimization of Linezolid for Treatment of Tuberculosis Kristina Bigelow Johns Hopkins University School of Medicine 1/15/217
Linezolid Recently added to WHOs 2 nd line approved anti-tb agents Demonstrated efficacy as a salvage agent (Lee et. al, NEJM, 212) XDR-TB: LZD 3-6mg added to failing regimen 87% sputum conversion after 6 months of LZD 31 out of 38 patients had severe adverse events Myelosuppression, peripheral neuropathy Component of novel pan-tb regimen (Conradie et al, CROI, 217) Pretomanid 2mg + Bedaquiline 2mg + LZD 12mg x 6 months Of first 34 pts completing treatment, 74% culture-negative at 8 wks (all negative at 16 wks) Of first 2 pts completing 6 months of follow-up after treatment, only 1 relapse reported 24 out of 33 had severe adverse events Myelosuppression, peripheral neuropathy
LZD dose optimization What PD parameter correlates best w/lzd bactericidal effect? In vitro hollow fiber system, normal media (LZD, SZD) T >MIC 1 In vitro hollow fiber system, acidified media (LZD) AUC/MIC 2 Mouse model (LZD, SZD, AZD) AUC/MIC 3,4 What TD parameter correlates best w/lzd mitochondrial toxicity? In vitro hollow fiber system (LZD) C min 1 Clinic (LZD) C min 5-8 1. Brown et al, mbio 215; 6:e1741 and Louie et al, ICAAC 2. Srivastava et al, AAC 215;61:e751-17. 3. Williams et al, AAC 29;53:1314. 4. Balasubramanian et al, AAC 214;58:4185. 5. Cattaneo et al, IJAA 213; 41:586 6. Pea et al, JAC 212; 67:234 7. Matsumoto et al, IJAA 214;44:242 8. Song et al, EBiomed 215; 2:1627
Hypothesis Under certain conditions less frequent administration of the same total dose will maximize bactericidal effects while minimizing toxicity. If this is true then: Dosing LZD at 12 mg q48 hr will preserve efficacy (while reducing toxicity) compared to 6 mg q24 hr and 3 mg q12 hr.
Questions to be answered What is the effect of different linezolid dosing schedules on bacterial killing and selection of drug-resistant mutants? Does the PK/PD parameter that best correlates with efficacy differ against different growth states of Mtb? In vitro In vivo Log phase growth Slow growth (induced by companion drug) (induced by companion drug) (induced by other constraints*) *Constraints in the in vitro system are imposed by achieving the stationary phase of culture Constraints in the in vivo system are imposed by the onset of adaptive immune response
I. In-vitro Pharmacodynamics/Pharmacokinetics (IVPD) system
LZD Concentration (µg/ml) Linezolid PK/PD Parameters in IVPDS expts. Cmax (ug/ml) AUC (ug*h/ml) %Time/MIC 3mg q12hr 7.95 18 92 6mg q24hr 13.71 173 67 12mg q48hr 27.42 176 29 3 25 2 15 1 5 1 2 3 4 5 6 7 8 Time (hours) MIC -5
Δ log cfu/ml Effect of linezolid dose fractionation against log phase growth in the IVPD system Untreated 3mg q12hr 6mg q24hr 12mg q48hr Log cfu/ml Day Control = 6.68 ±.64 Δ log cfu/ml (Day Day 14) 12mg q48hr = -2.9 ± 1.14 6mg q24hr = -2.62 ±.85 3mg q12hr = -3.58 ±.49
Δ log cfu/ml D iffe re n c e lo g c fu /m L Effect of linezolid dose fractionation against a no-growth state imposed by companion drugs in the IVPD system - 1-2 - 3 Untreated C o n tro l 3mg q12hr 6mg q24hr 3 m g /k g q 1 2 h r 6 m g /k g q 2 4 h r 12mg q48hr 1 2 m g /k g q 4 8 h r - 4-5 Log cfu/ml Day Control = 7.78-6 2 4 7 9 1 1 1 4 Difference log cfu/ml Day 14 D a y 12mg q48hr = 4.72 ±.33 6mg q24hr = 4.74 ±.36 3mg q12hr = 5.5 ±.1
Questions to be answered What is the effect of different linezolid dosing schedules on bacterial killing and selection of drug-resistant mutants? Does the PK/PD parameter that best correlates with efficacy differ against different growth states of Mtb? Log phase growth Slow growth (induced by companion drug) (induced by companion drug) (induced by other constraints*) In vitro Time above MIC AUC/MIC In vivo *Constraints in the in vitro system are imposed by achieving the stationary phase of culture Constraints in the in vivo system are imposed by the onset of adaptive immune response
II. In-vivo Mouse Studies
No Treatment Log 1 CFU/lung Effect of linezolid dose fractionation in a log phase growth TB infection model in BALB/c mice D a ta 3 DDa ta a ta 3 3 D a ta 3 1 1 1 9 8 7 6 5 4 3 2 1 d (W 3 ) 7 2 b id (7 /7 ) 1 b id 2 q d 3 3 5 q d (3 /7 ) 2 1 4 b id (7 /7 ) 3 b id 9 9 8 8 7 7 6 6 5 5 4 4 3 3 2 2 1 1 1 mg/kg 3 mg/kg 1 mg/kg 6 q d 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 b id 2 q d 4 q d (3 /7 ) 9 8 7 6 5 4 3 2 1 Total dose per week U n tr e ta te d (W 3 ) 7 2 b id (7 /7 ) 1 b id 2 q d 3 3 5 q d (3 /7 ) U n tr e ta te d (W 3 ) 7 2 b id (7 /7 ) 1 b id 2 q d 3 3 5 q d (3 /7 ) 2 1 4 b id (7 /7 ) 3 b id 6 q d 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 b id 2 q d 4 q d (3 /7 ) U n tr e ta te d (W 3 ) 7 2 b id (7 /7 ) 1 b id 2 q d 3 3 5 q d (3 /7 ) 2 1 4 b id (7 /7 ) 3 b id 6 q d 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 b id 2 q d 4 q d (3 /7 ) Dosing frequency and # of weekly doses 3 doses/week 5 doses/week 1 doses/week 14 doses/week No treatment Day 2 1 4 b id (7 /7 ) 3 b id 6 q d 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 b id 2 q d 4 q d (3 /7 )
No Treatment Log 1 CFU/lung Effect of linezolid dose fractionation in a no net growth TB infection model in BALB/c mice 8 7 6 5 4 C h r o n ic Dosing frequency and # of weekly doses 3 doses/week 5 doses/week 7 doses/week 14 doses/week No treatment Day 3 2 1 U n treated 7 2 b id (7 /7 ) 1 4 3 q d (7 /7 ) 1 mg/kg 3 mg/kg 1 mg/kg 3 3 5 q d (3 /7 ) 2 1.4 b id (7 /7 ) 2 q d (5 /7 ) Total dose per week 4 3 q d (7 /7 ) 6 q d (5 /7 ) 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 4 q d (7 /7 ) 2 q d (5 /7 ) 3 3 q d (3 /7 ) IN H (C o n tro l) R ifa m p in (C o n tro l) L Z D (C o n tro l) a m p in + E th a m b u to l
Questions to be answered What is the effect of different linezolid dosing schedules on bacterial killing and selection of drug-resistant mutants? Does the PK/PD parameter that best correlates with efficacy differ against different growth states of Mtb? Log phase growth Slow growth (induced by companion drug) (induced by companion drug) (induced by other constraints*) In vitro Time above MIC AUC/MIC In vivo Time above MIC AUC/MIC *Constraints in the in vitro system are imposed by achieving the stationary phase of culture Constraints in the in vivo system are imposed by the onset of adaptive immune response
No Treatment PA Log 1 CFU/lung Effect of linezolid dose fractionation in a slow growth TB infection model induced by pretomanid (Pa) 12.5 mpk/day 8 7 P a 1 2.5 6 5 4 3 2 1 Control 1 mg/kg 3 mg/kg 1 mg/kg Total dose per week U n tre a te d 7 2 b id (7 /7 ) 1 4 3 q d (7 /7 ) 2 q d (5 /7 ) 3 3 5 q d (3 /7 ) 2 1.4 b id (7 /7 ) 4 3 q d (7 /7 ) 6 q d (5 /7 ) 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 4 q d (7 /7 ) 2 q d (5 /7 ) 3 3 q d (3 /7 ) IN H (C o n tro l) R ifa m p in (C o n tro l) L Z D (C o n tro l) R ifa m p in + P a 1 2.5 P a 1 2.5 U n tre a te d 7 2 b id (7 /7 ) 1 4 3 q d (7 /7 ) 2 q d (5 /7 ) 3 3 5 q d (3 /7 ) 2 1.4 b id (7 /7 ) 4 3 q d (7 /7 ) 6 q d (5 /7 ) 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 4 q d (7 /7 ) 2 q d (5 /7 ) 3 3 q d (3 /7 ) IN H (C o n tro l) R ifa m p in (C o n tro l) L Z D (C o n tro l) R ifa m p in + P a 1 2.5 P a 1 2.5
No Treatment PA Log 1 CFU/lung Effect of linezolid dose fractionation in a no net growth TB infection model induced by Pa 5 mpk/day P a 5 8 7 6 5 4 3 Dosing frequency and # of weekly doses 3 doses/week 5 doses/week 7 doses/week 14 doses/week No treatment Day 2 1 Control U n treated 7 2 b id (7 /7 ) 1 4 3 q d (7 /7 ) 2 q d (5 /7 ) 1 mg/kg 3 mg/kg 1 mg/kg 3 3 5 q d (3 /7 ) 2 1.4 b id (7 /7 ) 4 3 q d (7 /7 ) Total dose per week 6 q d (5 /7 ) 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 4 q d (7 /7 ) 2 q d (5 /7 ) 3 3 q d (3 /7 ) IN H (C o n tro l) R ifa m p in (C o n tro l) L Z D (C o n tro l) P a 5
No Treatment No Treatment PA Log 1 CFU/lung Log 1 CFU/lung No Treatment No Treatment PA Log 1 CFU/lung Log 1 CFU/lung 8 7 6 5 4 3 2 1 U n tr e ta te d (W 3 ) 7 2 b id (7 /7 ) 1 b id 8 7 6 5 4 3 2 1 Comparison of dose fractionation studies in mice 2 q d 3 3 5 q d (3 /7 ) 2 1 4 b id (7 /7 ) D a ta 3 Log phase growth 1 mg/kg 3 mg/kg 1 mg/kg Control Total dose per week 3 b id 6 q d 1 q d (3 /7 ) 7.2 b id (7 /7 ) P a 5 1 b id 2 q d 4 q d (3 /7 ) No growth due to companion agent 1 mg/kg 3 mg/kg 1 mg/kg U n treated 7 2 b id (7/7) 1 4 3 q d (7/7) 2 q d (5/7) 3 3 5 q d (3/7) 2 1.4 b id (7/7) 4 3 q d (7/7) Total dose per week 6 q d (5 /7 ) 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 4 q d (7 /7 ) 2 q d (5 /7 ) 3 3 q d (3 /7 ) IN H (C o n tro l) R ifa m p in (C o n tro l) L Z D (C o n tro l) P a 5 8 7 6 5 4 3 2 1 8 7 6 5 4 3 2 1 Control 1 mg/kg 3 mg/kg 1 mg/kg U n treated 7 2 b id (7/7) 1 4 3 q d (7/7) 2 q d (5/7) 3 3 5 q d (3/7) 2 1.4 b id (7/7) P a 1 2.5 Slow growth Total dose per week U n treated 7 2 b id (7/7) 1 4 3 q d (7/7) 2 q d (5/7) 3 3 5 q d (3/7) 2 1.4 b id (7/7) 4 3 q d (7/7) 6 q d (5/7) 1 q d (3/7) 7.2 b id (7/7) 1 4 q d (7/7) 2 q d (5/7) 3 3 q d (3/7) IN H (C o n trol) R ifam p in (C o n trol) L Z D (C o n trol) R ifam p in + P a12.5 P a12.5 U n treated 7 2 b id (7/7) 1 4 3 q d (7/7) 2 q d (5/7) 3 3 5 q d (3/7) 2 1.4 b id (7/7) 4 3 q d (7/7) 6 q d (5/7) 1 q d (3/7) 7.2 b id (7/7) 1 4 q d (7/7) 2 q d (5/7) 3 3 q d (3/7) IN H (C o n trol) R ifam p in (C o n trol) L Z D (C o n trol) R ifam p in + P a12.5 P a12.5 C h r o n ic 1 mg/kg 3 mg/kg 1 mg/kg Total dose per week 4 3 q d (7 /7 ) 6 q d (5 /7 ) 1 q d (3 /7 ) 7.2 b id (7 /7 ) 1 4 q d (7 /7 ) 2 q d (5 /7 ) 3 3 q d (3 /7 ) IN H (C o n tro l) R ifa m p in (C o n tro l) L Z D (C o n tro l) a m p in + E th a m b u to l Dosing frequency and # of weekly doses 3 doses/week 5 doses/week 7 doses/week 14 doses/week No treatment Day
Questions to be answered What is the effect of different linezolid dosing schedules on bacterial killing and selection of drug-resistant mutants? Does the PK/PD parameter that best correlates with efficacy differ against different growth states of Mtb? Log phase growth Slow growth (induced by companion drug) (induced by companion drug) (induced by other constraints*) In vitro Time above MIC AUC/MIC In vivo Time above MIC Time above MIC AUC/MIC AUC/MIC Log phase growth *Constraints in the in vitro system are imposed by achieving the stationary phase of culture Constraints in the in vivo system are imposed by the onset of adaptive immune response
P r o p o rtio n o f L Z D -r e s is ta n t C F U Effect of linezolid dose fractionation on selective amplification of resistance during log phase growth in the IVPD system P r o p o r tio n o f L Z D -re s is ta n t C F U a t D 1 4 1-2 1-3 1-4 1-5 1-6 1-7 U n tr e a te d q 4 8 h q 2 4 h q 1 2 h D o s in g fre q u e n c y
Interim Conclusions Log phase growth Slow growth (induced by companion drug) (induced by companion drug) (induced by other constraints) In vitro Time above MIC AUC/MIC In vivo Time above MIC Time above MIC AUC/MIC AUC/MIC Under net growth conditions, efficacy driver is likely Time above MIC Under no net growth conditions efficacy driver is likely AUC/MIC C max appears to drive LZD resistance suppression Taken together, these data suggest that: When Mtb is actively replicating (eg, early in treatment, w/weak companion drugs), more frequent dosing to maintain adequate T>MIC may be beneficial When growth is restrained (eg, w/active companion drugs), the same total dosage delivered in higher, less frequent doses (eg, q48 hrs) should achieve similar anti-tb effect while minimizing risk of trough-driven toxicity and selection of LZD resistance
Future Aims Replicate IVPDS experiments, using higher and lower drug exposures to account for population-level PK variability Perform PK/PD analyses to confirm target parameters for dose optimization Adopt toxicodynamic model using K562 cells to study mitochondrial toxicity in IVPD system Extend studies to novel TB-focused oxazolidinones in development with reduced MPS inhibition potential
Acknowledgements Dr. Eric Nuermberger Dr. Kelly Dooley Jin Lee Sandeep Tyagi Dr. Heena Soni Dr. Rokeya Tasneen Si-Yang Li Kala Barnes-Boyle Funding R1-AI-111992