Table 1

Adjusted cut-offs calculated for TSO500 and OTML tests, equivalent to different cut-off points for the FO reference standard

Reference variable	FO≥10 muts/Mb		FO≥13 muts/Mb		FO≥16 muts/Mb		FO≥20 muts/Mb
Classification variable	TSO500	OTML	TSO500	OTML	TSO500	OTML	TSO500	OTML
Adjusted cut-off point with sensitivity >88%	7.847	8.380	9.434	9.240	10.995	10.900	13.842	11.730
Sensitivity at cut-off point (%)	89.90	89.90	88.24	88.24	88.89	88.89	88.24	88.24
Specificity at cut-off point (%)	74.47	65.96	74.19	61.29	82.61	72.46	87.34	65.82
Positive predictive value at cut-off point (%)	78.57	73.33	65.22	55.56	66.67	55.81	60.00	35.71
Negative predictive value at cut-off point (%)	87.50	86.11	92.00	90.48	95.00	94.34	97.18	96.30

Cut-off points of 10, 13, 16 and 20 muts/Mb were selected for FO, and the performance of the two next generation sequencing tests was evaluated by continuously interrogating the sensitivity and specificity at each point, as well as positive and negative predictive values, to discern high tumor mutational burden versus med/low tumors compared with FO. We selected as adjusted the points that offer maximum sensitivities (>88%) and negative predictive values for each test and for each equivalent cut-off value for the FO test. Adjusted cut-off values were lower than those for FO. The complete set of values is shown in online supplemental table S4.
FO, Foundation One; OTML, Oncomine Tumor Mutation Load; TSO500, TruSight Oncology 500.