Tumor immunogenomic signatures improve a prognostic model of melanoma survival

Table 1 The different thresholds of the number of somatic mutations used to define high and low TMB, the corresponding number of low TMB patients per each threshold (N < Threshold) in the discovery phase, and associations with OS in discovery, validation, and pooled cohorts

TMB Threshold	N < Threshold	Discovery (N = 139)		Validation 1 (N = 70)		Pooled (Discovery + Validation 1) (N = 209)
TMB Threshold	N < Threshold	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)
50	17 (12.2%)	8.64E−05	4.28 (2.07–8.84)
75	23 (16.5%)	4.57E−05	3.69 (1.97–6.92)
100	28 (20.1%)	1.11E−04	3.18 (1.77–5.71)
125	34 (24.5%)	1.30E−05	3.52 (2.00–6.20)	0.01	2.86 (1.23–6.62)	4.01E−07	3.36 (2.10–5.36)
150	36 (25.9%)	3.36E−05	3.15 (1.83–5.41)
175	40 (28.8%)	2.24E−04	2.78 (1.62–4.78)
200	46 (33.1%)	2.22E−03	2.28 (1.34–3.86)
323	71 (51.1%)	8.69E−02	1.59 (0.94–2.70)

The p-values (P), hazard ratios and 95% confidence intervals (HR 95% CI) for each threshold in the discovery phase were derived from univariate Cox PH models. The 125-mutation threshold (italic), which had the most significant p-value in the discovery phase, was moved forward to the validation stage. The p-value (P), hazard ratio, and 95% confidence interval (HR 95% CI) from the univariate Cox PH model from the validation phase are also included. The final two columns present the p-value (P) from the meta-analysis of the entire population, with accompanying hazard ratio and 95% confidence interval (HR 95% CI) from a univariate Cox PH model

ISSN: 1479-5876