Abstract 5470: Identification of gene amplification based signature as predictors for chemotherapy in squamous cell carcinoma of the lung

Abstract

Abstract Background: Lung squamous cell carcinoma (LUSC) is the second most popular histologic subtype accounting for 25%~30% of lung cancer. Due to the lack of targeted therapies for LUSC, the prognosis is poor and chemotherapy still plays a full role in the treatment of LUSC patients currently. However, a significant proportion of patients showed a bad response to chemotherapy. The development of predictive biomarkers is urgent. We aimed to identify new chemotherapy biomarkers based on genomic alterations. Methods: We retrospectively analyzed DNA sequencing data of 317 LUSC patients (pts) at West China hospital. We collected the clinical features and progression-free survival (PFS) of 35 pts with chemotherapy only. Pts were assigned to different subgroups based on genomic alterations using non-negative matrix factorization (NMF) clustering. Based on the characteristics of the cluster, we then explored clear gene signatures to predict the prognosis of chemotherapy. The clinical information and sequencing data of TCGA LUSC with chemotherapy were used to validate the prognostic value of the signature. Results: The 317 LUSC pts were grouped into 4 clusters characterized by different genomic alterations. Cluster 1 (C1) had 129 pts and was characterized by TP53 alterations. There were 116 pts in cluster 2 (C2) characterized by PIK3CA amplification (amp). Cluster 3 (C3) had 33 pts and was characterized by gene amp of CCND1, FGF3, FGF4, and FGF19. Cluster 4 (C4) was characterized by gene amp of FGFR1, KDR, KIT, and PDGFRA. The 35 pts treated with chemotherapy only were also classified into 4 clusters. C1 was associated with the shortest PFS (hazard ratio (HR), 2.87; 95% confidence interval (CI), 1.15-7.13; p = 0.018) independently of the clinical stage, but C2, C3, and C4 showed no significant difference. In addition, we also found that pts with TP53 loss of function (LOF) alteration had significantly shorter PFS than those without TP53 LOF (wildtype or not LOF, p = 0.029). So we further modified the clustering by TP53 alteration from C1 and gene amp from C2, C3, and C4. Therefore, 35 pts were then grouped into 3 subtypes based on TP53 LOF and gene amp characteristics, and the PFS was significantly different. Pts (n = 8, 23%) with at least one amp of 9 genes (PIK3CA, CCND1, FGF3, FGF4, FGF19, FGFR1, KDR, KIT, and PDGFRA) but no TP53 LOF had the longest PFS (HR, 0.12; 95% CI, 0.02-0.66; p = 0.011). Pts (n = 6) with TP53 LOF but no 9-gene amp had poor survival compared with pts in other subtypes. This gene signature was validated in TCGA pts. The signature of 9-gene amp without TP53 LOF also indicated the best PFS and TP53 LOF without 9-gene amp indicated the worst PFS (p = 0.081). Conclusion: We develop a biomarker signature that consists of 9-gene amp and TP53 LOF to indicate the prognosis of chemotherapy. Our results suggest that 9-gene amp without TP53 LOF in LUSC is a favorable prognostic marker for patients taking chemotherapy. Citation Format: Yuan Tang, Yuli Li, Ting Hou, Lili Jiang, Hongjie Liu, Chunxiao Pan, Weiya Wang, Li Qiu, Yajing Zhang, Guiping Zhang, Ke Zheng. Identification of gene amplification based signature as predictors for chemotherapy in squamous cell carcinoma of the lung. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5470.