Abstract 6637: Transcriptomic biomarkers of immunotherapy response in solid tumors

Abstract Background. Immune-checkpoint inhibitors can be administered in multiple solid tumor types. Currently available biomarkers, like the expression of PD-L1 or tumor mutational burden, miss a significant proportion of patients who could benefit from the therapy. Our goal was to set up a database consisting of both gene expression and clinical response data and to uncover the most significant biomarkers of response to anti-PD-1, anti-PD-L1, and anti-CTLA-4 immunotherapies. Methods. The GEO repository was screened to identify datasets with simultaneously available transcriptomic and clinical response data across different solid tumor types. The screening focused on studies involving the administration of nivolumab (anti-PD-1), pembrolizumab (anti-PD-1), atezolizumab (anti-PD-L1), durvalumab (anti-PD-L1), and ipilimumab (anti-CTLA-4). Mann-Whitney U-test and receiver operating characteristic analysis were performed across all genes to identify features related to therapy response. False discovery rate was computed to correct for multiple hypothesis testing. Results. The database that we established consists of 1,439 unique tumor samples from nineteen datasets with gastric-, lung-, esophageal-, head, neck-, and urothelial cancers, and malignant melanoma. The strongest upregulated candidates linked to resistance in anti-PD-1 treated tumors were PRXL2A (AUC=0.771), ODR4 (AUC=0.734), and RBM15B (AUC=0.734). The most robust hits in the anti-PD-L1 treated cohort were NOTCH3 (AUC=0.639), NECTIN1 (AUC=0.631) and ITGAV (AUC=0.63). In the anti-CTLA-4 treatment cohort C5orf24 (AUC=0.76) was the most promising candidate. We have also established a portal for additional analysis and ranking of further biomarker candidates by extending our previously established ROC plotter. The immunotherapy-treated transcriptomic database is accessible at the website www.rocplot.com/immune. Discussion. An integrated database and an online platform were set up to validate potential biomarkers of immunotherapy response in a sizeable cohort of cancer samples from multiple tumor types. By using the direction of expression change as a filter, we have identified the most promising biomarker candidates. Citation Format: Balazs Gyorffy, Szonja Anna Kovacs, János Tibor Fekete. Transcriptomic biomarkers of immunotherapy response in solid tumors. [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 6637.