Search in the Abstract Database

Abstracts Search 2018

OP037 Risk of cancer in inflammatory bowel disease patients is associated with age and recent use of immunomodulators, while biologics and aminosalicylates are protective factors: a cross-sectional and follow-up analysis of the Swiss IBD cohort study

T. Greuther1*, S. Scharl1, C. Barthel1, J.-B. Rossel2, L. Biedermann1, B. Misselwitz1, S. Vavricka1, G. Rogler1, M. Scharl1, on behalf of the Swiss IBD Cohort Study Group

1University Hospital Zurich, Division of Gastroenterology and Hepatology, Zurich, Switzerland, 2University Hospital Lausanne - CHUV, Institute of Social and Preventive Medicine, Lausanne, Switzerland

Background

Cancer is a long-term complication in patients with inflammatory bowel disease (IBD). Immunomodulators might contribute to an increased risk despite their control of the underlying chronic inflammation. However, available data remain inconsistent. We aimed at comprehensively analysing our Swiss IBD cohort (SIBDCS) in a cross-sectional and longitudinal manner to elucidate prevalence and incidence of cancer as well as protective and risk factors.

Methods

Data on all IBD patients included between 2007 and 2013 in the nation-wide SIBDCS were first analysed in a cross-sectional manner. Patients with malignancies (=composite of cancer, dysplasia and lymphoma) were compared with controls. In a second step, only patients without malignancy at enrolment and a follow-up of ≥1 year were included to calculate risk of cancer development.

Results

We identified 122 malignancy cases (3.9%) in a total of 3119 patients. Most of the patients had gastrointestinal carcinoma (23.0%), dysplasia (22.1%) and skin cancer (9.0%). Patients with malignancies were more often males, older at the age of diagnosis and had longer IBD duration. Fistula, intestinal surgery and surgery for fistula were more frequently reported. Cancer patients were more often taking antibiotics and steroids, but medication with biologics was less frequently reported. In a multivariate logistic regression model, age (OR=1.04, p < 0.001), ulcerative colitis (1.68, p = 0.03), intestinal surgery (OR 4.51, p < 0.001), fistula (OR 1.74, p = 0.015) and treatment with steroids (OR 2.13, p = 0.001) were independent predictors for the presence of cancer, while treatment with 5-ASA (OR 0.61, p = 0.036) and biologics (OR 0.38, p < 0.001) were identified as protective factors. From a longitudinal perspective, 67 malignancy cases occurred in 2580 patients not previously diagnosed with cancer (2.6%). Based on a median follow-up of 4.9 years (IQR 2.7–7.0 years) and a total follow-up of 12 420.8 patient years, we calculated an incidence rate of 539/100 000. Colorectal carcinoma (17.9%), dysplasia (14.9%), skin malignancy (13.4%) and lymphoma (9.0%) were most frequently reported. In a univariate cox regression model, age, intestinal surgery, treatment with antibiotics, and recent use of immunomodulators were predictors for development of cancer, while treatment with 5-ASA and biologics were identified as protective factors. These factors remained significant predictors in a multivariate regression analysis.

Conclusion

In a large, nation-wide IBD cohort, cancer was found in 3.9% of patients, while development of cancer was observed in 2.6% over 5 years of follow-up. Age and recent use of immunomodulators were the main risk factors for cancer, while treatment with 5-ASA and biologics seemed to be protective.