Early, “pre-resistant” genetic signals predate the emergence of phenotypic Mycobacterium tuberculosis resistance. Mycobacterium tuberculosis

Tuberculosis (TB) is the leading global cause of infectious disease death, and drug-resistant forms of Mycobacterium tuberculosis (M.tb) are an increasingly ominous public health threat. Acquisition of second-line drug resistance during treatment strongly predicts poor outcome.1 Yet, conventional culture-based phenotypic drug-susceptibility testing (DST) takes weeks to months because of the slow generation time of M.tb, is unreliable for many second-line drugs,2 and requires expensive biosafety laboratory infrastructure. Existing commercial molecular TB diagnostic tests provide results within days, but can only detect resistant sub-populations greater than 5-65% of the total M.tb population.3-5 Here, we show that micro-heteroresistance (resistant sub-populations 0.1-5% of total M.tb) may be an early “pre-resistant” indicator of subsequent phenotypic drug susceptibility resistance. We applied a recent advance in targeted deep sequencing (Single Molecule-Overlapping Reads, or SMOR)6 to longitudinally quantify mutations conferring resistance to key second-line anti-tuberculosis drugs in patients progressing from multi-drug (MDR-) to extensively drug resistant TB (XDR-TB). Surpassing a micro-heteroresistance threshold of 0.1% (10-fold better than the theoretical limit of detection of routine phenotypic DST) was strongly associated with subsequent phenotypic amikacin resistance an average of 8.25 months later. This study supports the hypothesis that in vivo acquired drug resistance in M.tb is defined by very small, previously undetectable mycobacterial sub-populations (<1%), the significance of which will vary by drug, specific resistance mutation, efficacy of companion drugs, and patient predictors. Novel NGS procedures able to detect “pre-resistant” signals many months prior to onset of phenotypic resistance could significantly impact clinical management of MDR-TB.