Extended Data Fig. 2: Immunogenomic context of tumors from POLAR.
a) Baseline genomic characteristics from WES (N = 35) stratified by POLAR cohort (A: HRD, B: ncHRD, C: HRP): (i) Tumor purity, (ii) WES-derived tumor mutational burden (WES_TMB) was significantly higher in Cohort A (HRD) p < 0.05, (iii) ploidy, and (iv) fraction genome altered (FGA). No statistically significant differences were observed across cohorts. (b) Mutation types and neoantigen burden: (i) Number of nsSNVs, (ii) total indel burden, (iii) frameshift indel burden, and (iv) predicted neoantigen burden. Indel and frameshift indel burden were significantly higher in Cohort A (HRD). Statistical comparisons were performed using Wilcoxon rank-sum tests. Significance is indicated by p values (p < 0.01). (c) Baseline genomic and immune features across platforms: IMPACT HRD score (N = 35), IMPACT TMB (N = 42), neoantigen burden from WES (N = 35), TIL density from H&E (N = 38), and baseline neutrophil-to-lymphocyte ratio (NLR) from blood (N = 63). No significant differences were observed between cohorts in these parameters. Each dot represents an individual patient. Boxes indicate median and interquartile range; whiskers represent 1.5× IQR. Cohorts: A (purple), B (teal), C (navy). Statistical significance was assessed by two-sided Wilcoxon test and annotated as: ns (not significant), p < 0.05, p < 0.01. Abbreviations: WES, whole exome sequencing; HRD, homologous recombination deficiency; ncHRD, non-core homologous recombination deficiency; nsSNVs, nonsynonymous single-nucelotide variants; OS, overall survival; PFS, progression-free survival.
