Prognostic implications of tumor mutation burden in postoperative patients with colorectal cancer who received capecitabine-based adjuvant chemotherapy: a retrospective exploratory study

Abstract Background: This study aimed to investigate the prognostic implications of tumor mutation burden (TMB) in patients with colorectal cancer (CRC) who underwent surgical resection and capecitabine-based adjuvant chemotherapy. Methods: A total of 82 patients with CRC who underwent surgical resection and capecitabine-based adjuvant chemotherapy were included in this study retrospectively. Tumor tissue specimens were collected for DNA extraction after surgical resection. Somatic mutation detection and TMB analysis were performed using next-generation sequencing (NGS) of tumor-related genes. The recurrence status of the patients was assessed in the hospital during the adjuvant chemotherapy period, and the long-term survival data of patients were obtained by telephone follow-up. The univariate analysis between TMB status and prognosis was carried out by Kaplan-Meier survival analysis and adjusted by multivariate Cox regression analysis subsequently. Results: The median follow-up period of this study was 5.3 years (range: 0.25-9.6 years). The median disease-free survival (DFS) of the 82 patients was 4.5 years, the median overall survival (OS) was 5.7 years. The results of NGS analysis demonstrated that the most common mutated somatic genes among the 82 patients were TP53, APC, RAS, PIK3CA and EGFR, and the prevalence was 62.2%, 58.5%, 47.6%, 34.1% and 30.4%, respectively. Other somatic mutant genes were of relatively low frequency (<30%). Regarding the TMB analysis, the overall somatic mutation burden of the 82 patients was comparatively low [median: 3.9/Mb (range: 1.6-48.6/Mb)]. TMB status was divided into TMB-L (≤3.9/Mb) and TMB-H (>3.9/Mb) according to the median TMB threshold. And the patients with TMB-L and TMB-H were observed in 42 cases and 40 cases, respectively. Prognostic analysis according to TMB status demonstrated that the median OS of patients with TMB-L and TMB-H was 6.5 and 4.5 years, respectively (P=0.009). Additionally, in order to adjust the confounding factors that might influence OS, a multivariate Cox regression analysis was introduced and the results exhibited that TMB status was an independent factor for OS (HR=0.71, P=0.011). Conclusion: TMB might be considered as a potential biomarker for predicting the prognosis of patients with CRC who underwent surgical resection and capecitabine-based adjuvant chemotherapy. Results of this study should be elucidated in large-scale prospective clinical trial subsequently.