A novel approach identifies the potential biomarkers of targeted drug sensitivity in esophageal squamous cell carcinoma

Purpose: Esophageal squamous cell carcinoma (ESCC) has a high mortality with no effective targeted therapies. Previous studies from Cancer Cell Line Encyclopedia (CCLE) project adopted a commercial human pan-cancer cell line model incorporating genomic landscape to systematically identify drug sensitivity biomarkers. However, the biomarker of drug sensitivity in ESCC is still lack of widely exploring. Here, we established a novel approach combined patient-derived models, targeted deep sequencing and a drug sensitivity evaluation system to investigate potential biomarkers of targeted drug sensitivity in ESCC. Materials and Methods: Deep sequencing of 365 tumor drug-related genes was performed in tumor tissues and matched blood from an ESCC cohort (n = 161). In order to explore the potential biomarkers of drug sensitivity, we systematically established patient-derived cell lines (PDCs) from an independent cohort including 123 operable ESCC patients. Targeted deep sequencing, RNA sequencing and high-through drug sensitivity were integrated to identify the potential biomarkers of targeted drug sensitivity in ESCC PDCs. Molecular characterization methods and animal models were used to validate the potential biomarker in vitro and in vivo. In addition, patient-derived xenograft (PDX) was further confirmed the results. Results: The mutational profile of tumor drug-related genes indicated a mean of 9 non-silent somatic mutations (mutation allele frequency >= 0.05) per patient and a high recurrence rate of copy number variation (CNV) were discovered. To explore potential biomarkers of targeted drug sensitivity, we established eight PDCs derived from 123 ESCC patients were successfully established and used for molecular characterization and drug screening. Drug sensitivity evaluation and pharmacogenomics analyses of the eight PDCs revealed prevalent CDKN2A or CDKN2B loss as potentially sensitive biomarkers of the CDK4/6 inhibitors palbociclib and ribociclib, which was neither found in the previous study from CCLE models nor in the present study using commercial ESCC cell lines. Importantly, patient-derived models integrated with DNA and RNA sequencing validated this result in vitro. Moreover, ESCC patient-derived cell line xenografts (PDCX) models with CDKN2A/2B loss are more sensitive to the CDK4/6 inhibitor palbociclib than that in PDCX models with wild-type CDKN2A/2B. Furthermore, mouse model with PDX further confirmed CDKN2A/2B loss as a biomarker of the CDK4/6 inhibitor sensitivity. Conclusions: These results suggested that patient-derived models combined with deep sequencing and a drug sensitivity evaluation system, incorporating in vitro and in vivo validation platform, could be used as a novel and effective approach for exploring biomarkers of drug sensitivity in ESCC. Citation Format: Dan Su, Dadong Zhang, Jiaoyue Jin, Lisha Ying, Miao Han, Kaiyan Chen, Bin Li, Junzhou Wu, Zhenghua Xie, Fanrong Zhang, Yihui Lin, Guoping Cheng, Jinchao Wang, Minran Huang, Jing-Yu Li, Jianjun Zhang, Fugen Li, Lei Xiong, Andrew Futreal, Weimin Mao. A novel approach identifies the potential biomarkers of targeted drug sensitivity in esophageal squamous cell carcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4896.