Evaluation of cycloserine dose regimens in an Indian cohort with multidrug-resistant tuberculosis: a population pharmacokinetic analysis

Abstract

Cycloserine is recommended for inclusion in regimens for multidrug-resistant tuberculosis (MDR-TB). Its efficacy is time dependent and relies on the concentration remaining above the minimum inhibitory concentration (MIC); however, there is a concentration-dependent risk of neurotoxicity. Limited pharmacokinetic (PK) data are available in individuals of Indian origin, despite the high burden of MDR-TB in India. We enrolled adults and adolescents receiving cycloserine for MDR-TB at a tertiary hospital in Mumbai, India, in a prospective cohort. Total daily doses ranged from 500 to 750 mg, and participants underwent serial PK sampling on multiple visits starting 1 month after treatment initiation. PK data were analyzed using non-linear mixed-effect modeling. A total of 180 participants (117 females) were enrolled, with a median age of 27 years (interquartile range [IQR] 21-35), weight of 56.0 kg (IQR 46.0-65.9), and fat-free mass of 38.6 kg (IQR 32.3-47.1). Cycloserine PK was best described by a one-compartment model with first-order elimination and transit compartment absorption. Allometric scaling by fat-free mass provided the best adjusment for body size. Serum creatinine improved the model fit and allowed separate estimation of renal and non-renal clearances, whose typical values were 0.589 and 0.901 L/h, respectively. Simulations showed median exposure of 308 mg·h/L after 250 mg twice daily (BID), which is lower than reported in literature. Monte Carlo simulations suggested that doses of 500 or 750 mg BID are required to reach efficacy targets of ≥30% and ≥64% time within dose interval above MIC. The reasons behind the low exposure identified in this Indian population require further investigation.

Keywords: MDR-TB; NONMEM; cycloserine; pharmacokinetics; pharmacometrics.

Fig 1

Duration of cycloserine treatment prior to drug-level testing. Each horizontal line represents a single participant’s cycloserine treatment duration, with the color coding denoting the total daily cycloserine dose prescribed over that period of time. Filled circles indicate intensive sampling visits, and open circles indicate sparse sampling visits.

Fig 2

Visual predictive check from the final model. The solid and dashed lines represent the 5th, 50th, and 95th percentiles from the original observations (blue circles), while the shaded areas correspond to the 95% confidence intervals for the same percentiles based on model predictions for the intensively sampled data (left panel) and the sparsely sampled data (right panel). The horizontal dotted line represents the lower limit of quantification (0.782 mg/L). The bins for sampling points are shown as vertical yellow ticks on top of the plot area.

Fig 3

Probability of target attainment by dose and minimum inhibitory concentration. Solid lines indicate the probability of target attainment on the primary y-axis (on left) based on the percentage of time during the dose interval in which the concentration exceeds the minimum inhibitory concentration (%T_>MIC) on the x-axis. The horizontal dashed red lines indicate the 90% attainment. The top panel shows (a) %T_>MIC of ≥ 64% (the preferred literature-derived target indicating 80% maximum kill effect) and the bottom panel (b) or %T_>MIC of ≥ 30% (bactericidal effect). Solid lines indicate simulated cycloserine doses of 250 mg QD, BID, or TID; 500 mg QD or BID; 250/500 mg a.m./p.m.; 750 mg BID; and WHO weight-based regimen (250 mg BID for body weights ≤45 kg and 250/500 mg a.m./p.m. for weights >45 kg, as indicated by color and shape in the legend. The dot-dashed curve and shaded area underneath indicate the distribution of cycloserine MICs for Mycobacterium tuberculosis isolates cultured from 171 study participants (the proportion of MIC in tested samples can be read on the right y-axis).

Fig 4

Observed and simulated cycloserine area under the curve (AUC₀₋₂₄) at different dose regimens. Boxes represent the median and interquartile range of the cycloserine area under the curve (AUC_0–24), while the whiskers show the range of values. Observed exposures are shown as boxes with divot pattern fill, while the simulated are shown as solid boxes. The horizontal red dotted line indicates the literature-derived toxicity threshold AUC_0–24 of 700 mg·h/L associated with toxicity (11). The horizontal dotted blue lines on top of each group of boxes represent the reported exposure for the specific dose regimen (12, 14).