Cancer risk in inflammatory bowel disease according to patient phenotype and treatment: a Danish population-based cohort study

Abstract

Objectives: Population-based studies of site-specific cancer risk in patients with inflammatory bowel disease (IBD) according to IBD phenotype and treatment are lacking. We studied cancer risk in a well-characterized population-based IBD cohort from North Jutland County, Denmark.

Methods: A total of 1,515 patients were diagnosed with ulcerative colitis (UC) and 810 with Crohn's disease (CD) during 1978-2002. Patients were followed until 31 December 2010 for occurrence of incident cancer, identified in the Danish Cancer Registry. Observed numbers of cancer were compared with expected numbers (based on age- and sex-specific background rates) and presented as standardized incidence ratios (SIRs) with 95% confidence intervals (CIs).

Results: Patients with UC were not at increased risk of cancer overall (SIR, 1.12; 95% CI, 0.97-1.28) despite increased risk of prostate cancer (SIR, 1.82; 95% CI, 1.17-2.71). Patients with CD had a 55% increased risk of cancer overall (SIR, 1.55; 95% CI, 1.29-1.84) related to young age, colonic disease, smoking, and thiopurine exposure. Patients were at increased risk of small bowel cancer (SIR, 15.18; 95% CI, 1.84-54.78), lung cancer (SIR, 2.13; 95% CI, 1.19-3.52 (associated with female gender and smoking)), colorectal cancer in males (SIR, 2.43; 95% CI, 1.05-4.78), cervical dysplasia (SIR, 1.65; 95% CI, 1.10-2.37 (associated with young age at diagnosis, smoking, 5-aminosalicylic acid, and thiopurine exposure)), and non-Hodgkin lymphoma (SIR, 3.43; 95% CI, 1.38-7.07 (unrelated to thiopurine exposure)).

Conclusions: Patients with CD, but not UC, have an overall excess risk of cancer. Clinical characteristics of IBD patients at excess risk differ by cancer subtype.