The management and life expectancy of patients with cystic fibrosis have improved substantially in the past three decades, which has resulted in an increased number of these patients being diagnosed with malignancies. Our aim was to assess the risk of gastrointestinal cancers in patients with cystic fibrosis.

In this systematic review and meta-analysis, we searched PubMed, MEDLINE, Google Scholar, Scopus, Embase, and Cochrane databases with no language restrictions for studies published from inception of the databases to Aug 1, 2017, assessing the risk of gastrointestinal cancers in patients with cystic fibrosis. We also searched abstracts from scientific meetings and the bibliographies of identified articles for additional references. Studies were included if they reported the standardised incidence ratio (SIR) or incidence ratio per person-years. No exclusion criteria with regard to patient characteristics (age, sex, comorbidities, cystic fibrosis mutation type), study setting (location and time period), or method of reporting cancer diagnoses were applied. The primary outcome was risk of gastrointestinal cancer and site-specific gastrointestinal cancers in patients with cystic fibrosis compared with the general population. Pooled summary estimates were calculated using a random-effects model, and subgroup analyses were done to establish whether risk of gastrointestinal cancer varied according to patient lung transplant status. The study is registered with PROSPERO, number CRD42017075396.

Our search identified 95 681 records, of which six cohort studies including 99 925 patients (544 695 person-years) were eligible for the meta-analysis. The overall risk of gastrointestinal cancer was significantly higher in patients with cystic fibrosis than in the general population (pooled SIR 8·13, 95% CI 6·48–10·21; p<0·0001; log SIR 2·10, 95% CI 1·87–2·32; p<0·0001, I²=93·93%). Subgroup analyses showed that the risk of gastrointestinal cancer among patients with cystic fibrosis who had a lung transplant was increased compared with that of patients who did not receive a transplant (pooled SIR 21·13, 95% CI 14·82–30·14; p<0·0001; log SIR 3·05, 95% CI 2·70–3·41; p<0·0001, I²=28·52% vs pooled SIR 4·18, 3·10–5·62; p<0·0001; log SIR 1·43, 1·13–1·73; p<0·0001, I²=22·66%). The risk for the following site-specific cancers was also significantly increased in patients with cystic fibrosis compared with the general population: small bowel cancer (pooled SIR 18·94, 95% CI 9·37–38·27; p<0·0001; log SIR 2·94, 95% CI 2·24–3·64; p<0·0001, I²=38·61%), colon cancer (10·91, 8·42–14·11; p<0·0001; log SIR 2·39, 2·13–2·65; p<0·0001, I²=88·09%), biliary tract cancer (17·87, 8·55–37·36; p<0·0001; log SIR 2·88, 2·15–3·62; p<0·0001, I²=10·16%), and pancreatic cancer (6·18, 1·31–29·27; p=0·022; log SIR 1·82, 0·27–3·38; p<0·0001, I²=62·57%).

Our study suggests that patients with cystic fibrosis had a significantly increased risk of gastrointestinal cancer compared with the general population, including small bowel, colon, biliary tract, and pancreatic cancers. These findings highlight the need to develop individualised screening strategies for site-specific gastrointestinal cancers in patients with cystic fibrosis.

None.