Validation of epigenetic markers to identify colitis associated cancer: Results of module 1 of the ENDCAP-C study

BACKGROUND: Chronic inflammation caused by ulcerative colitis (UC) causes a pro-neoplastic drive in the inflamed colon, leading to a markedly greater risk of invasive malignancy compared to the general population. Despite surveillance protocols, 50% of cases proceed to cancer before neoplasia is detected. The Enhanced Neoplasia Detection and Cancer Prevention in Chronic Colitis (ENDCaP-C) trial is an observational multi-centre test accuracy study to ascertain the role of molecular markers in improving the detection of dysplasia. We aimed to validate previously identified biomarkers of neoplasia in a retrospective cohort and create predictive models for later validation in a prospective cohort. METHODS: A retrospective analysis using bisulphite pyrosequencing of an 11 marker panel (SFRP1, SFRP2, SRP4, SRP5, WIF1, TUBB6, SOX7, APC1A, APC2, MINT1, RUNX3) in samples from 35 patients with cancer, 78 with dysplasia and 343 without neoplasia undergoing surveillance for UC associated neoplasia across 6 medical centres. Predictive models for UC associated cancer/dysplasia were created in the setting of neoplastic and non-neoplastic mucosa. FINDINGS: For neoplastic mucosa a five marker panel (SFRP2, SFRP4, WIF1, APC1A, APC2) was accurate in detecting pre-cancerous and invasive neoplasia (AUC = 0.83; 95% CI: 0.79, 0.88), and dysplasia (AUC = 0.88; (0.84, 0.91). For non-neoplastic mucosa a four marker panel (APC1A, SFRP4, SFRP5, SOX7) had modest accuracy (AUC = 0.68; 95% CI: 0.62,0.73) in predicting associated bowel neoplasia through the methylation signature of distant non-neoplastic colonic mucosa. INTERPRETATION: This multiplex methylation marker panel is accurate in the detection of ulcerative colitis associated dysplasia and neoplasia and is currently being validated in a prospective clinical trial. FUNDING: The ENDCAP-C study was funded by the National Institute for Health Research Efficacy and Mechanism Evaluation (EME) Programme (11/100/29).