KRAS mutations and their prognostic implications in appendiceal cancers.

person Snigdha Nutalapati Department of Medical Oncology, University of Kentucky HealthCare, Lexington, KY info_outline Snigdha Nutalapati, Negar Sadeghipour, Joanne Xiu, Sharon Wu, Heinz-Josef Lenz, Ryan Schmocker, Matthew James Oberley, Anna Reagan, Hannah McDonald, Reema Anil Patel, Sanjay Goel, Prakash Pandalai, Joseph Kim

Authors person Snigdha Nutalapati Department of Medical Oncology, University of Kentucky HealthCare, Lexington, KY info_outline Snigdha Nutalapati, Negar Sadeghipour, Joanne Xiu, Sharon Wu, Heinz-Josef Lenz, Ryan Schmocker, Matthew James Oberley, Anna Reagan, Hannah McDonald, Reema Anil Patel, Sanjay Goel, Prakash Pandalai, Joseph Kim Organizations Department of Medical Oncology, University of Kentucky HealthCare, Lexington, KY, Caris Life Sciences, Phoenix, AZ, Norris Comprehensive Cancer Center, University of Southern California, Los Angeles, CA, University of Tennessee Health Science Center, Knoxville, TN, University of Kentucky, Lexington, KY, University of Kentucky Medical Center, Lexington, KY, Rutgers Cancer Institute of New Jersey, New Brunswick, NJ, Department of Surgery, University of Kentucky HealthCare, Lexington, KY Abstract Disclosures Research Funding No funding sources reported Background: Despite shared embryonic origin, appendiceal cancers (AC) have distinct clinical and molecular features compared to colorectal cancers (CRC). Detection of mutant KRAS has been associated with worse survival in CRC, whereas in AC the prognostic significance of mutant KRAS (55-65% prevalence) has yet to be characterized. Methods: AC tissues from 852 individual patients that underwent DNA (592, NextSeq, or whole exome sequencing) and whole transcriptome sequencing (NovaSeq) at Caris Life Sciences (Phoenix, AZ) were analyzed. Low-grade appendiceal mucinous neoplasms were excluded. Chi-square, Fishers-exact, and Mann Whitney U tests were used to determine statistical significance and were adjusted for multiple hypothesis testing (p< 0.05). Real-world overall survival (OS) was calculated using insurance claims of AC patients from time of sample collection to last contact. Hazard ratios (HRs) were calculated using the Cox proportional hazards model (log-rank test). Multivariate regression analysis was performed on age, sex, histology, site, comutations and KRAS mut vs. KRAS wt . Results: AC histological subtypes comprised mucinous adenocarcinoma (37.9%), goblet cell carcinoma (19.6%), signet ring cell carcinoma (12.1%), and adenocarcinoma- not otherwise specified (NOS, 30.4%). Among all AC specimens, KRAS was the most common mutation (49%). KRAS mut vs. KRAS wt tumors were more frequently associated with TP53 mut (51.2% vs. 36.8%, respectively; p< 0.01) and GNAS mut (41.1% vs. 5.7%, respectively; p < 0.00001). Most frequent co-mutant gene with KRAS was TP53 (61.2%) in adenocarcinoma-NOS, and GNAS (46.9%) in mucinous adenocarcinomas.Median OS among KRAS mut vs. KRAS wt was 33.3 vs. 25.1 months, respectively (HR 0.68, 95% CI: 0.58-0.80, p < 0.00001). KRAS mut was associated with improved survival in mucinous adenocarcinomas (HR 0.71; 95% CI: 0.52-0.96, p = 0.028) and adenocarcinoma-NOS (HR 0.77; 95% CI: 0.62-0.97, p = 0.025), but not with signet ring cell (HR 1.92; 95% CI: 0.90-4.10, p = 0.087) or goblet cell carcinoma (HR 1.39; 95% CI: 0.50-3.91, p = 0.536). Multivariate analysis showed KRAS mut was an independent prognostic factor for improved survival among all AC (HR 0.77, 95% CI: 0.64-0.93, p = 0.007) and mucinous adenocarcinoma (HR 0.60, 95% CI: 0.42-0.84, p<0.0001) . Notably, GNAS mut was associated with improved survival (HR 0.57, 95% CI: 0.47-0.70, p < 0.00001), while TP53 mut was associated with poorer survival (HR 1.58, 95% CI: 1.35-1.86, p < 0.00001) among all AC. Conclusions: KRAS was one of most frequently mutated genes in ACs. In contrast to CRC, KRAS mut was associated with significantly improved survival. This observed survival advantage remained consistent in the histologic subgroup of mucinous adenocarcinoma, but not in goblet and signet ring cell cancers. Further studies with population level data understanding the molecular profiles and their survival implications in histologically distinct AC are needed.