Skip to main content

Table 7 (a) Correlations between \(\gamma_{mel}\) and clinical measures, with multiple potential values. (b) Correlations between \(\gamma_{mel}\) and binary covariates

From: Predicting response to pembrolizumab in metastatic melanoma by a new personalization algorithm

(a) \(\gamma_{mel}\) and clinical measures (b) \(\gamma_{mel}\) and binary covariates
Biomarker ROC AUC Maximal Cohen’s kappa (κ) Biomarker Cohen’s kappa (κ)
Age 0.56 0.219 Malignant melanoma status 0.185
Breslow thickness 0.63 0.304 Nodular melanoma status 0.258
LDH 0.52 0.128 BRAF V600 status 0.124
LY% 0.52 0.156   
SOD at the baseline 0.61 0.251   
  1. The maximal AUC and Cohen’s kappa (κ) values, obtained from ROC analysis (in italics) were obtained for Breslow thickness (see “Methods”)