Frequent suboptimal thermocycler ramp rate usage negatively impacts MTBDRsl performance for second-line drug resistant tuberculosis diagnosis

Derendinger, B.; de Vos, M.; Pillay, S.; Venter, R.; Metcalfe, J.; Ghebrekristos, Y.; Minnies, S.; Dolby, T.; Beylis, N.; Warren, R.; Theron, G.

Strengthening the detection of second-line drug-resistance is a key tuberculosis (TB) control priority. The performance of MTBDRplus, a multidrug-resistant (MDR)-TB assay is reduced when suboptimal ramp rates are used. We investigated ramp rate's effect on MTBDRsl; the most widely-used molecular second-line drug-resistant TB assay. We tested 52 smear-negative Xpert MTB/RIF Ultra-positive sputa and a Mycobacterium tuberculosis (Mtb) dilution series at manufacturer recommended (2.2 {degrees} C/s) or most common suboptimal ramp rate (4.0 {degrees} C/s; identified via an earlier survey). Mtb-complex DNA (TUB-band)-positivity, indeterminate rates, fluoroquinolone- and second-line injectable-resistance accuracy, banding differences and, separately, inter-reader variability were assessed. 39% of re-surveyed laboratories (5/13) did not use the manufacturer-recommended MTBDRsl ramp rate. On sputum, this ramp rate improved indeterminates vs. 4.0 {degrees} C/s (0/52 vs. 7/51; p=0.006), false drug-resistance calls (0/104 vs. 6/102; p=0.013), and incorrect banding calls (0/1300 vs. 55/1275; p<0.001). Valid results (neither TUB negative, indeterminate, nor any false drug-resistance calls) (52/52 vs. 41/51; p=0.001) on sputa hence improved by +21% (95% CI: 8-34%) with optimal ramp rate usage. Suboptimal ramp rate increased banding call inter-reader variability [52/1300 (4%) vs. 34/1300 (3%); p=0.030] on sputa but not dilution series; highlighting the importance of using clinical specimens for assay performance evaluations. Suboptimal ramp rate contributes to poor MTBDRsl performance. Ramp rate correction will improve second line drug-resistant TB diagnoses. Laboratories must ensure the optimal manufacturer-recommended ramp rate is used.