Antibiotic dose-response curves can measure antibiotic activity against Mycobacterium abscessus and Mycobacterium peregrinum

Abstract

Mycobacterium abscessus is a drug-resistant pathogen associated with poor clinical outcomes despite treatment with multidrug antibiotic regimens. Apart from clarithromycin, antimicrobial susceptibility testing (AST) results for Mycobacterium abscessus cannot guide antibiotic selection. AST involves measuring the minimum inhibitory concentration (MIC), which is allowed to span a four-fold range in concentration. This accepted variability of the MIC limits the clinical utility of AST. Antimicrobial dose-response curves, obtained by measuring the growth inhibition of a given organism to increasing concentrations of an antibiotic, can yield metrics of antibiotic activity that are less variable than the MIC. We used Clinical and Laboratory Standards Institute growth conditions for rapid-grower nontuberculous mycobacteria to generate 990 dose-response curves across three time points (72 hours, 96 hours, and 120 hours) for six guideline-recommended (clarithromycin, amikacin, cefoxitin, linezolid, tigecycline, and clofazimine) and five new (omadacycline, tedizolid, SPR719, SQ109, and bedaquiline) antibiotics against Mycobacterium abscessus subspecies abscessus ATCC 19977 and Mycobacterium peregrinum ATCC 700686. We established the fit of the dose-response curve (R2) as a quality control metric. Using the geometric standard deviation and median coefficient of variation, we demonstrated that the IC50 and IC75 (antibiotic concentrations corresponding to 50% and 75% growth inhibition, respectively) are less variable than the MIC. We identified time-dependent changes in dose-response curve metrics that allow the detection of inducible clarithromycin resistance with only five days of incubation. This study demonstrates the potential of dose-response curves in measuring antibiotic activity against Mycobacterium abscessus.