Abstract 6467: Tumor mutation burden combined with tumor markers as predictive prognostic biomarkers for neoadjuvant chemotherapy in patients with gastric cancer

Abstract

Abstract Background: The ypTNM stage and tumor regression grading (TRG) is widely used to predict the efficacy of neoadjuvant chemotherapy (NAC). However, the potential prognostic role of molecular markers is unknown. There is growing evidence assessing the predictive impact of tumor mutational burden (TMB) on the clinical efficiency of immunotherapy, but few investigation has pursued the predictive value of TMB on the efficiency of neoadjuvant chemotherapy. In this study, the prognostic role of TMB and tumor markers for NAC in patients with GC were explored. Methods: Patients with Stage II/III gastric adenocarcinoma (n = 30) were enrolled in this study between Jan 2016 and Jun 2017 in Peking University Cancer Hospital. All patients were subjected to NAC before resection. Whole exome sequencing (WES) were implemented on the biopsy samples collected from the 30 patients before NAC .Tumor mutation burden (TMB) of total somatic substitutions and indels per megabase after fıltering known driver mutations were calculated. The five postoperative serum tumor markers: AFP,CEA,CA199,CA724 and CA242 were quantitatively measured before and after operation. Results: Of all the 30 patients with GC receiving neoadjuvant chemotherapy before resection, 12 patients (40.0%) with higher TMB (≥5 mutations per Mb) showed significantly better prognosis compared to the other 18 patients (P=0.019, HR=6.042). And all these 12 patients survived at follow-up. In patients with TMB<5, those who with tumor marker positive (TM+) postoperatively were at more risk of recurrence than patients with negative TM (TM-) in postoperative plasma. Compared to the patients with TMB≥5, the subgroup of both TMB<5 & TM- (P = 0.0631, HR= 6.449) and TMB<5 & TM+ (P <0.0001, HR=1303) had poorer survival. The results showed that TMB combined with TM (Log-rank test, P<0.0001) provide better predictive validity on prognostic value compared with ypTNM alone (Log-rank test, P=0.1534). In addition, whole exome sequencing data from 478 STAD patients (Stomach Adenocarcinoma (TCGA, Provisional)) obtained from TCGA were used to detect somatic mutations. The results also indicated that mutation count was the most relevant clinical attribute of overall survival (Kruskal Wallis Test, q-value<10−10). It is further suggested that TMB is related to prognosis. Conclusions: In patients with GC, higher TMB levels (TMB≥5) of biopsy tissue before NAC showed greater response rates for neoadjuvant chemotherapy and longer survival. TMB<5 together with TM+ was a strong and accessible biomarker of poor OS. This is the first report of TMB can be used as a predictive biomarker before NAC, and TMB combined with tumor markers (TM) were significantly associated with prognosis in patients with Gastric Cancer. Additional larger cohort study aimed at understanding relationship between response to therapy and TMB are further required. Citation Format: Yongning Jia, Honglin Zhu, Fei Pang, Fei Shan, Shuangxi Li, Lei Dong, Tonghui Ma, Ziyu Li. Tumor mutation burden combined with tumor markers as predictive prognostic biomarkers for neoadjuvant chemotherapy in patients with gastric cancer [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 6467.