Implementation of a Pancreatic Cancer Treatment Selection Program Based on a Real-time Biomarker Analysis in Available Biopsies

Background: Systemic chemotherapy is the mainstay for metastatic Pancreatic Ductal Adenocarcinoma (PDCA). There is a paucity of effective predictive markers of drug sensitivity or resistance in this setting, due in large part to marked difficulties in prospectively obtaining baseline tumor. In this study we describe all the clinical and histological limitations to implementing a real-time biomarker panel program for cancer in advanced PDCA patients.

Methods: A retrospective, non-interventional study was conducted using data from the medical records of eligible patients participating in an advanced PDCA treatment selection program based on a real-time biomarker analysis in available tumor tissue.   Biomarker Panel (BP) Program was implemented in a single center from 2007 to 2016. To be eligible, patients were required to meet the following inclusion criteria: (i) have histologically-confirmed advanced PDCA, (ii) have started systemic therapy (5´fluouracil-based or gemcitabine-based chemotherapy at single agent or in combination) and (iii) aged ≥18 years at the time of diagnosis of advanced PDCA. A        BP consists in a predefined set of 7 molecular targets, including: KRAS mutations, EGFR amplification (FISH), and Thymidylate Synthase (TS), Thymidine Phosphorylase (TP), Excision Repair Cross-Complementing 1 (ERCC.1), Topoisomerase I (Topo I), and SPARC expression by IHC. Patients treated as part of an ulterior phase II clinical trial were included.

Results: Between January 1^st 2007 and January 1^st 2017, 111 metastatic PDCA patients were identified as candidates.  In 65 patients (58.6%) it was possible to implement a BP.  A re-biopsy was performed in only 3 cases (2.7%) to obtain sufficient tumor for molecular analysis. In 31 (47.7%) patients it was feasible to study almost 5 of the 7 planned targets. In registered patients, poor performance status (13.5%) was the most frequent limitation to performing the BP.  Other limitations were anatomical limitations to the biopsies (4.5%), incomplete biomarker data (4.5%) or clinical deterioration during procedures (2.7%).

Conclusion: PDCA treatment based on real-time biomarker analysis in available biopsies presents significant limitations due to patient deterioration and sample processing.  New approaches are necessary to optimize results in real-time targeted therapy studies.