Significance of alterations in DNA damage repair (DDR) genes in advanced biliary cancers (ABCs) treated with second-line active-symptom-control (ASC) alone or ASC with oxaliplatin/5-FU chemotherapy (ASC plus FOLFOX) in the randomised phase III, multicentre, openlabel ABC-06 trial.

593 Background: The ABC-06 clinical trial established ASC+mFOLFOX as the standard of care treatment after Cisplatin and Gemcitabine (CisGem) for ABC. Translational research explored the role of alterations in DDR-related genes in the context of the ABC-06 clinical trial. Methods: Translational research using samples collected from the ABC-06 trial included all recruited patients with tissue sample containing >20% tumour cellularity with sufficient DNA extracted for analysis. Primary objective: assess the prevalence of somatic mutations within DDR genes in ABC. Secondary objectives: explore the impact of somatic mutations in DDR genes on: A) Patient’s response to platinum-based chemotherapy (predictive biomarker): 1) Retrospectively, using progression-free survival (PFS) data from 1 st -line chemotherapy and 2) Prospectively, by assessing the impact on PFS/radiological response to 2 nd -line FOLFOX (vs. control) and B) Overall survival (prognostic biomarker). Survival analysis was performed with Kaplan-Meier and Cox Regression. Results: Of 162 pts randomised, 83 had a sample available for analysis; of these, analysis failed in 24 despite >20% tumour content. Thus, a total of 59 patients (30 ASC arm, 29 ASC+FOLFOX arm) were eligible for this translational analysis: male 50.85%, metastatic 77.97%, intrahepatic cholangiocarcinoma 47.46%, adenocarcinoma 91.53%, median age 65.84 years (95% CI 63.36-68.91). Pathogenic mutations in DDR genes were identified in 22 patients (37.29%). PFS did not vary depending on the DDR-gene alterations either in the first-line setting with CisGem (n=59; 8.73 months vs 8.18 months; p-value 0.155) or with second-line FOLFOX (n=29; 3.19 months vs 3.45 months; p-value 0.098). Median OS for DDR-altered patients was 4.59 months (95% CI 2.17-5.88) (vs 7.23 months (95% CI 5.45-8.28) for DDR-wild-type); HR 2.63 (95% CI 1.48-4.67); p-value 0.001. This prognostic impact was confirmed when the prognostic model was adjusted for treatment arm and stratification factors (HR 3.75 (95% CI 1.99-7.09); p-value <0.001). Conclusions: For ABC patients, the presence of DDR-related gene pathogenic mutations are present in around one third of patients. Despite presence of DDR-mutations having a negative prognostic impact, their predictive role is not confirmed either for first-line CisGem or second-line FOLFOX. Clinical trial information: NCT01926236 .