Clinico-pathological diagnostic and prognostic biomarkers for duodenal adenocarcinoma from a UK retrospective Cohort study.

Duodenal adenocarcinoma (DA), a very rare malignancy which is difficult to diagnose and has poor outcome. With increasing incidence of disease, identification of diagnostic and prognostic biomarkers is of paramount clinic-pathological importance. Seven United Kingdom Duodenal Cancer Study Group (UKDCSG) centers contributed 98 quality assured DA cases undergoing surgical resection leading to creation of bespoke Tissue Micro-Array (TMA) which was interrogated with immunohistochemical and immunofluorescent stains for known diagnostic and prognostic markers following STROBE and REMARK guidelines on reporting. Variables with possible prognostic significance were assessed with Kaplan-Meier estimations (log-rank tests) and Cox proportional hazards univariate and multivariate models. The median estimated overall survival of 43 months (IQR: 38-84 months) after a median follow-up time of 37.5 (interquartile range: IQR 16-75) months. Lower tumor stage (T1-3 vs T4, hazard ratio (HR) 0.411 (95% confidence interval (CI) 0.208-0.812), absence of lymphovascular invasion (HR 0.457 (95%CI: 0.240-0.871), and lower CD20 (HR 0.450 (95%CI: 0.237-0.855)) immune infiltrate demonstrated better prognosis in multivariate analysis whilst Aurora-A, Beclin-1 expression and CD3 or CD4 or CD8 infiltration has no prognostic impact. Low CK20 expression may have pitfalls in diagnostic assessment. Our work presents the largest UK cohort of DA with long-term follow-up making it an excellent resource for future biomarker studies. Validation studies in prospective clinical studies are required for clinical application.