Abstract A070: Genomic profiling of pancreatic cancer by KRAS status

Abstract

Abstract Introduction: Ineffective treatments and late detection combine to give pancreatic cancer (PC) patients especially poor prognosis. However, as more targeted therapies are approved and are being investigated, there is hope for such therapies to play a pivotal role in improving PC patient outcomes. KRAS is an important oncogenic driver in most pancreatic cancers, making the RAS signaling pathway an obvious therapy target. In this study we examine the genomic profile of pancreatic cancers that do or do not harbor a RAS alteration to identify other oncogenic drivers that may represent therapy targets. Methods: Pancreatic tumor samples were profiled using the OncoExTraTM tumor-normal, whole exome, whole transcriptome sequencing assay. This assay detects somatic single base substitutions, indels, copy number alterations, gene fusions and alternative transcripts. Actionable altered genes, defined as those with associated FDA-approved targeted therapies in any cancer type, those in guidelines or with published evidence for matched therapies, or those required for eligibility for an active clinical trial, were identified. Results: A total of 384 pancreatic patients were profiled, 305 (79.4%) of which harbored 310 KRAS alterations. Most KRAS alterations were missense mutations (307 of 310; 99.0%). Most (283) missenses caused a substitution at amino acid position 12, with G12D (n=115), G12V (n=104) and G12R (n=54) being the most common. There were also 3 missense substitutions at position 13, and 21 at position 61. In addition, three patients harbored both a KRAS missense mutation and an amplification. In addition to KRAS, several other altered biomarkers were relatively prevalent, including TP53 (63.3%), CDKN2A (25.3%), ARID1A (9.1%), CDKN2B (7.0%), BRCA2 (4.4%), MEN1 (3.9%) and MTAP (3.6%). TP53 alterations were significantly overrepresented in KRAS-altered cancers. Alterations at MEN1, DAXX, ATRX, SETD2 and CREBBP, and high tumor mutational burden (TMB-high) were all overrepresented in cancers lacking KRAS alterations. In addition, all 19 (4.9%) actionable gene fusions involving oncogenic drivers (6 BRAF, 6 FGFR1/2, 3 NRG1, 2 RET, 1 RAF1, 1 ROS1) were found only in cancers lacking a KRAS alteration. Conclusions: Our study shows differences in the genomic profile based on KRAS status. We identified potentially targetable gene fusions and other alterations enriched in pancreatic tumors lacking KRAS alterations, suggesting several therapies may be reasonable options in such patients. Additionally, this analysis underscores the benefit of whole transcriptome profiling that identified actionable fusions in 4.9% of patient samples, all of which were unaltered at KRAS. Citation Format: Gargi D. Basu, Snehal G. Thakkar, David W. Hall, Cynthia A. Flannery, Satish Seerapu, Maen Abdelrahim. Genomic profiling of pancreatic cancer by KRAS status [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Pancreatic Cancer; 2023 Sep 27-30; Boston, Massachusetts. Philadelphia (PA): AACR; Cancer Res 2024;84(2 Suppl):Abstract nr A070.