P350 Comparative risk of incident cancer in patients with Inflammatory Bowel Disease with prior non-digestive malignancy according to immunomodulator: a multicenter cohort study

Abstract

Abstract Background Knowledge about the cancer risk when initiating a biologic in Inflammatory Bowel Disease (IBD) patients with prior malignancy remains scarce especially for vedolizumab. Our aim was to evaluate the rate of incident cancer in a cohort of IBD patients with prior non-digestive malignancy according to the subsequent treatment given. Methods Consecutive cases of patients with IBD and a malignancy were retrospectively collected in, 33 centers in two countries. Patients with a biliary- or digestive tract cancer were excluded. Inclusion date corresponded to diagnosis of malignancy. Patients were categorized into different cohorts according to the first treatment they were exposed to after the malignancy diagnosis and before incident cancer. Follow-up time was calculated from the first administration of therapy to last follow-up visit. For patients receiving no therapy, follow-up started at diagnosis of cancer. Incident cancer during follow-up was defined as recurrence of the known cancer or occurrence of a new cancer. Crude cancer incidence rates were compared between the cohorts using chi-square test. Patients from the anti-TNF, vedolizumab, and no treatment cohorts were matched on age, lymph node and metastasis extension and malignancy recurrence risk according to the Penn classification using a propensity-score analysis with a, 1:1 ratio. Results Among the, 538 patients (58% female; mean [standard deviation (SD)] age inclusion:, 52 [15] years) analyzed, the most frequent malignancy was breast cancer (25%). First immunomodulator given after inclusion was a conventional immunosuppressant in, 27% of patients, anti-TNF in, 21% or vedolizumab in, 9%. With a median (inter-quartile range (IQR)) follow-up duration of, 55 (23–100) months, 100 incident cancers were observed. Crude cancer incidence rates per, 1000 person-years were, 47.0 for patients receiving no immunomodulator, 36.6 in the anti-TNF cohort and, 33.6 in the vedolizumab cohort (p=0.23). Incident-cancer free survival rates were not different between patients receiving anti-TNF and those receiving vedolizumab (p=0.56). After adjustment, incidence rates were not different between patients receiving no immunomodulator, anti-TNF or vedolizumab. On multivariate analysis, age ≥, 52 years at inclusion, follow-up after index cancer <55 months, and receiving immunomodulatory treatment within, 12 months after index cancer, were associated with an increased risk of incident cancer. Conclusion In this large multicenter cohort study, we found no difference in terms of incident cancer rates in patients with IBD and a prior non-digestive malignancy treated by vedolizumab compared to those treated by other immunomodulatory therapies including anti-TNF.