Abstract 7647: Identification of transcriptional plasticity biomarkers for patient stratification

Abstract Biomarker signatures associated with disease progression are important tools for the management of patient care. However, such signatures are frequently indication-specific, which limits their utility to stratify broader populations. Transcriptional plasticity, the ability of a transcriptome to change in response to external stimuli, is of interest due to its relation to disease progression and therapeutic response. To extend and generalize the measurement of a patient’s risk of disease progression to a pan-cancer, sequencing-based biomarker, Emax, the maximum cytotoxicity of a compound in a cell line, was explored as a potential universal marker of transcriptional plasticity. To explore correlation between Emax and transcriptional plasticity, cancer cell lines were ranked by mean Emax across multiple classes of cytotoxic compounds analyzed in GDSC (Yang et. al., 2013). Cell lines with lower Emax (more surviving cells at high drug concentrations) were filtered to exclude cell lines which tended to have higher IC50 values to remove cell lines that may be generally more resistant. With GDSC1 as the train set and GDSC2 as the test set, differentially expressed genes between high-Emax and low-Emax cell lines were selected for further analysis and possible inclusion in the plasticity biomarker signature. Test analysis in GDSC2 revealed an R2 of 0.88 suggesting transferability. For clinical validation, public repositories were queried for tumor RNA sequencing data with paired response data. To avoid batch effects between experiments, preliminary validation focused on one sufficiently large dataset with well-curated response data (n = 170; Barrett et. al., 2013, GSE91061; phs000452.v1.p1). Analysis of the expression of the genes identified in the in vitro signature revealed significant differential expression between patients who were reported as having an objective response and patients who achieved stable disease as best response. In silico pathway analysis of this signature revealed enrichment in cell-cell adhesion, signaling, and gene regulation, which relate to the hallmarks of phenotypic plasticity and epigenetic reprogramming (Hanahan, 2022). This analysis reveals similarity between gene expression patterns associated with Emax measured in vitro and transcriptional plasticity in the clinic. Additional experimentation is underway to prospectively analyze this set of biomarkers and to retrospectively validate this signature in additional datasets with the goal of patient stratification by the risk of disease progression following an initial, incomplete response to therapy. Citation Format: Jeremy Goldstein, Christina Gavazzi, Mikhail Grushko, Mahta Samizadeh, Zakary ElSeht, Katherine Arline. Identification of transcriptional plasticity biomarkers for patient stratification [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7647.