Tumor mutational burden and prognosis across pan-cancers

Background: Tumor mutational burden (TMB) is associated with genome instability and immunogenicity, which has been reported to play an important role in predicting the efficacy of immune checkpoint inhibitors. However, the relationship of TMB and prognosis in solid tumors is not yet fully understood. In this study, we aimed to explore the association between TMB and prognosis across pan-cancers.Table: 57PDThe associations of TMB and OS across pan-cancerstumorfull namesTMB-median (interquartile)ALLStage I-IIStage III-IVNHR(95% CI)PNHR(95% CI)PNHR(95% CI)PBLCABladder Urothelial Carcinoma151(85-269)4080.51(0.33-0.78)<0.0011311.07(0.45-2.55)0.892750.41(0.24-0.68)<0.001BRCABreast invasive carcinoma39(26-69)9701.48(0.87-2.54)0.157200.98(0.49-1.94)0.952294.46(1.57-12.69)<0.001CHOLCholangiocarcinoma32(26-44)411.37(0.35-5.32)0.65282.33(0.44-12.49)0.32(-)(-)(-)COADColon adenocarcinoma111(82-167)3781.43(0.79-2.58)0.242023.24(0.93-11.37)0.071651.57(0.7-3.5)0.27ESCAEsophageal carcinoma100(77-144)1831.84(0.98-3.48)0.06973.01(1.18-7.68)0.02631.07(0.36-3.14)0.90GBMGlioblastoma multiforme47(37-64)3851.01(0.73-1.42)0.93(-)(-)(-)(-)(-)(-)HNSCHead and Neck squamous cell carcinoma92(60-140)5061.65(1.11-2.45)0.01963.94(1.34-11.59)0.013411.11(0.69-1.78)0.66KIRCKidney renal clear cell carcinoma48(35-63)3333.72(1.79-7.75)<0.0012228.23(1.9-35.6)<0.0011091.82(0.75-4.38)0.18KIRPKidney renal papillary cell carcinoma54(35-73)2770.39(0.13-1.15)0.091871.17(0.19-7.04)0.8663(-)(-)LGGBrain Lower Grade Glioma25(18-34)5015.03(2.87-8.8)<0.001(-)(-)(-)(-)(-)(-)LIHCLiver hepatocellular carcinoma77(55-107)3571.5(0.88-2.53)0.132501.87(0.87-3.98)0.11861.36(0.61-3.05)0.45LUADLung adenocarcinoma171(74-338)5010.83(0.54-1.28)0.393900.89(0.52-1.52)0.671040.69(0.33-1.46)0.33LUSCLung squamous cell carcinoma205(148-297)4830.75(0.52-1.09)0.133900.7(0.46-1.06)0.09890.92(0.38-2.21)0.84OVOvarian serous cystadenocarcinoma79(53-130)4290.73(0.52-1.03)0.08(-)(-)(-)(-)(-)(-)PAADPancreatic adenocarcinoma35(25-46)1731.62(0.88-3.02)0.12(-)(-)(-)(-)(-)(-)READRectum adenocarcinoma95(74-127)1270.68(0.21-2.27)0.53611.22(0.11-13.47)0.87580.6(0.1-3.72)0.59SARCSarcoma40(27-54)2351.25(0.63-2.49)0.52(-)(-)(-)(-)(-)(-)SKCMSkin Cutaneous Melanoma226(62-415)1030.45(0.15-1.3)0.14670.14(0.01-1.32)0.09311.51(0.36-6.32)0.58STADStomach adenocarcinoma106(66-216)4090.52(0.34-0.8)<0.0011780.45(0.2-0.98)0.052140.53(0.31-0.92)0.02THCAThyroid carcinoma9(6-13)4886.92(0.85-56.28)0.07325(-)(-)1611.34(0.16-11.19)0.79UCECUterine Corpus Endometrial Carcinoma89(47-562)5270.27(0.13-0.57)<0.001(-)(-)(-)(-)(-)(-)UVMUveal Melanoma12(9-15)682.5(0.45-13.88)0.2937(-)(-)302.4(0.19-30.98)0.50 Open table in a new tab Methods: Whole-exome sequencing from 7882 solid tumors, spanning 22 cancer types from The Cancer Genome Atlas were analyzed. TMB was defined by total non-silent somatic mutation counts in coding region. Patients were classified into four groups based on the quartiles of TMB in each tumor. All Hazard ratios (HR) were reported as the risk ratio of the top 25% to the bottom 25% group. Results: Among 22 tumors, the level of TMB of top five tumors (median, interquartile, sample size) were: SKCM (226, 62-415, 103), LUSC (205, 148-297, 483), LUAD (171, 74-338, 501), BLCA (151, 85-269, 408), COAD (111, 82-167, 378). In all patient analyses of each tumor, higher TMB was associated with longer overall survival (OS) in BLCA, STAD and UCEC, and was associated with shorter OS in HNSC, KIRC and LGG. In stage I-II of each tumor, higher TMB was significantly associated with prolonged OS in STAD, but with shorter OS in ESCA, HNSC and KIRC. In stage III-IV of each tumor, higher TMB was significantly associated with prolonged OS in BLCA and STAD, but with shorter OS in BRCA. In 5 tumors (including BLCA, HNSC, KIRC, LUSC and THCA), patients with higher pathological stage had significantly higher TMB (P < 0.05). However, only in COAD and READ, patients with higher stage had significantly lower TMB (P < 0.05). Conclusions: Higher TMB played different roles in various tumors, which may be attribute to different driver mutations or other factors. Further analyses of underlying mechanism were underway to confirm and explore the potential prognosis prediction of TMB across pan-cancers. Legal entity responsible for the study: Feng Qiu. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.