Genomic profiling of KRAS wide-type pancreatic ductal adenocarcinomas identifies targetable genetic alterations.

Abstract

4121 Background: Pancreatic ductal adenocarcinoma (PDAC) is one of the most lethal cancers and the 5-year survival of PDAC patients is below 10%. The oncogenic KRASmutations account for about 90% of PDAC cases. Unfortunately, there is no FDA-approved targeted therapy for KRAS mutations. Therefore, the genomic profiling of KRAS wide-type PDACs can provide invaluable sights to the etiology of these patients and offer them the opportunity of precision therapy trials. Methods: To characterize actionable targets in 521 Chinese PDAC patients, deep sequencing of a 831-gene panel (OncoPanscan, Genetronhealth) was applied to assess somatic mutations of their tumor tissues including SNV, insertions/deletions, CNV and re-arrangements, as well as possible pathogenic germline variants of paired genomic DNA sample. Results: There were 89% (463/521) of patients in our PDAC cohort harbored KRAS mutations. Among the remaining 58 patients in KRAS wild-type subgroup, 33% (19/58) carried activating mutations in the RTK/Ras/MAPK pathway. Targetable BRAF mutations were seen in five (9%) patients: V600E (1/5), G464V (1/5), N486_P490del (2/5) and BRAF fusion (1/5). The frequency of BRAF mutations was 9% (5/58) in KRAS wild-type PDACs but only 0.4% (2/463) in KRAS-mutated PDACs (P < 0.001). We found one classic EGFR activing mutation (L747_A750delinsP) and one MAP2K1 activating mutation (F53_Q58delinsL), which can be targeted by EGFR-TKIs and MEK inhibitor trametinib, respectively. An oncogenic ERBB3 mutation (V104L) was seen in one patient, who was eligible for HER2-targeted therapy clinical trials. We also found STK11/TSC2 inactivating mutations and a dominant-negative mutation of PTEN (R130Q) which could be targeted by mTOR inhibitor everolimus and AKT inhibitor capivasertib, respectively. Additionally, we observed a patient with high level amplification of MET and another patient with the NCOA4-RET fusion gene which can be targeted by MET inhibitor carbozantinib. Interestingly, we also identified two patients with inactivating mutations in ELF3 (one frameshift and one in splicing-site), which is a driver gene of ampullary carcinoma. Lastly, two patients carried deleterious germline mutations in BRCA1 and PALB2, which may be targeted with PARP inhibitors. Overall, at least 29% (17/58) KRAS wide-type patients harbored potentially actionable genomic alterations to currently used anticancer drugs. Conclusions: The mutational landscape of our PDAC cohort provided compelling evidence that targetable driver mutations accounted for a significant portion of KRAS wide-type tumors. Our findings demonstrated that genomic profiling of PDAC patients can enable physicians to optimize their clinical management and enroll them into genomically matched clinical trials.