Circulating biomarkers and outcomes from a randomised phase 2 trial of gemcitabine versus capecitabine-based chemoradiotherapy for pancreatic cancer.
Willenbrock F., Cox CM., Parkes EE., Wilhelm-Benartzi CS., Abraham AG., Owens R., Sabbagh A., Jones CM., Hughes DLI., Maughan T., Hurt CN., O'Neill EE., Mukherjee S.
BACKGROUND: The Phase 2 SCALOP trial compared gemcitabine with capecitabine-based consolidation chemoradiotherapy (CRT) in locally advanced pancreatic cancer (LAPC). METHODS: Thirty-five systematically identified circulating biomarkers were analysed in plasma samples from 60 patients enroled in SCALOP. Each was measured in triplicate at baseline (prior to three cycles of gemcitabine-capecitabine induction chemotherapy) and, for a subset, prior to CRT. Association with overall survival (OS) was determined using univariable Cox regression and optimal thresholds delineating low to high values identified using time-dependent ROC curves. Independence from known prognostic factors was assessed using Spearman correlation and the Wilcoxon rank sum test prior to multivariable Cox regression modelling including independent biomarkers and known prognostic factors. RESULTS: Baseline circulating levels of C-C motif chemokine ligand 5 (CCL5) were significantly associated with OS, independent of other clinicopathological characteristics. Patients with low circulating CCL5 (CCL5low) had a median OS of 18.5 (95% CI 11.76-21.32) months compared to 11.3 (95% CI 9.86-15.51) months in CCL5high; hazard ratio 1.95 (95% CI 1.04-8.65; p = 0.037). CONCLUSIONS: CCL5 is an independent prognostic biomarker in LAPC. Given the known role of CCL5 in tumour invasion, metastasis and the induction of an immunosuppressive micro-environment, targeting of CCL5-mediated pathways may offer therapeutic potential in pancreatic cancer. CLINICAL TRIAL REGISTRATION: The SCALOP trial was registered with ISRCTN, number 96169987 (registered 29 May 2008).