Clinical utility of comprehensive transcriptome testing in advanced solid tumors.

Abstract

3072 Background: Despite breakthroughs in genomically-matched therapies, many patients lack actionable genomic alterations. Consequently, innovative new strategies to identify targets for effective therapies are imperative. To assess the potential of comprehensive whole transcriptomic sequencing (WTS) in clinical decision-making, we conducted a prospective trial to perform WTS in patients with metastatic or advanced solid tumors who had prior DNA-based panel testing (≥100 genes) with no reported AMP/ASCO/CAP Tier 1 actionable genomic alterations; along with whole exome sequencing (WES) to compare RNA expression to copy number alterations. We also developed an informed actionability classification scheme to determine actionability of transcriptomics findings. Methods: Between August 2022 and December 2024, 100 patients at MD Anderson Cancer Center with advanced cancers were enrolled and underwent comprehensive profiling. Alterations in RNA expression of 147 genes were considered as potentially actionable if the gene’s protein product is the direct target of a clinically available therapy (including cell surface targets of antibody-drug conjugates (ADCs)) and/or CNAs of the gene are predictive of response or resistance to a clinically available therapy. A clinical trial was considered a match if RNA expression was an enrollment criterion, or the gene alteration can be directly or indirectly targeted with a therapy utilized within the trial. Results: Actionable RNA expressions (AREs) were detected in all patients (100%). In total , 2,216 AREs were detected from 17,800 selected-reported RNA expressions; a median of 22 ARE changes were reported per patient [interquartile range (IQR), 17.0-26.0]. The majority (86.0%) of AREs were RNA overexpressions, defined as the distribution of tpm values of the gene expression > 83% of the pan-cancer reference cohort. A median of 13 RNA expression-matched trials were identified per patient [IQR, 10.0-16.0]. The most frequent drug class of actionable RNA overexpressions were ADCs, with a median of 10 ADC targets per patient. Out of 11 distinct genes of which at least 1 amplification is reported, a concordance rate of 61.5% and a Concordance Correlation Coefficient (CCC) value of 0.70 (95% CI, 0.56-0.84) between the actionable gene amplifications (n = 13) and the actionable RNA overexpressions detected in the same patient sample from the tumor profile, indicates a substantial concordance by transcriptional profiling with copy number gain. Conclusions: WTS identified actionable RNA expressions in all patients– including for novel ADC targets. These results underscore the utility of comprehensive transcriptional profiling to identify additional actionable targets beyond DNA-based comprehensive profiling. Clinical trial information: MDACC 2021-1049.