Feasibility and clinical utility of EUS guided biopsy of pancreatic cancer for next-generation genomic sequencing.

Abstract

e15755 Background: Next-generation sequencing (NGS) has made genomic profiling to guide therapy a reality for many cancer types. The aim of this study is to investigate the feasibility of genomic profiling using standard clinical endoscopic ultrasound (EUS) core biopsy samples of Pancreatic Cancer (PC) to allow personalised cancer care. Methods:Patients undergoing EUS and biopsy for suspicion of PC underwent additional biopsies which was snap frozen. En-face frozen section enabled targeted macro-dissection prior to DNA extraction, quantification and targeted sequencing using a commercially available 151 gene ClearSeq Comprehensive Cancer Panel. Matching formalin-fixed (FFPE) diagnostic EUS biopsy and fresh frozen surgical resection specimens underwent genomic profiling for comparison. Whole genome sequencing (WGS) was performed in 2 patients. RNA sequencing was performed in samples with sufficient RNA yield. Results: Known PC genes ( KRAS, GNAS, TP53, CDKN2A, SMAD4) were identified in 27 out of 30 (90%) patients with histological diagnosis of PC. Potentially actionable somatic mutations (BRCA1, BRCA2, ATM, BRAF, JAK3) were found in 6 (20%) patients. In the 2 samples selected, WGS of the EUS samples confirmed point mutations identified on panel sequencing and revealed relevant mutational signatures and structural variation patterns. Targeted panel sequencing was successful in all FFPE samples. In 1 chemotherapy naïve patient, sequencing of a matching trio of fresh frozen and FFPE EUS biopsies, and resection sample revealed evidence of spatial intra-tumoral heterogeneity. In another patient, pre-treatment biopsy revealed a somatic BRCA1 mutation, and patient had a near complete pathological response to platinum containing neoadjuvant therapy in the resected specimen. RNA sequencing segregated patients into key clinically relevant molecular PC subtypes based on transcriptome as recently described. Conclusions: We demonstrate here novel multi-omic analysis of pancreatic cancer using standard clinical EUS guided fine needle biopsies. Multi-omic analysis of EUS biopsies offers potential clinical utility to guide personalized therapy of PC in the neoadjuvant and advanced settings.