Risk factors predicting desmoid occurrence in patients with familial adenomatous polyposis: a meta-analysis

Aim Desmoid tumours (DT) are myofibroblastic proliferations occurring in 15% of patients with familial adenomatous polyposis (FAP). Several small series have analysed the incidence of DT and predisposing risk factors. Using meta-analytical techniques, this study aimed to identify risk factors for DT development in patients with FAP. Method Studies of sporadic DT were excluded. The study end-points were the incidence of DT in FAP and DT development by gender, adenomatous polyposis coli (APC) mutation, family history of DT and previous abdominal surgery. A random effect Mantel-Haenszel model was used to calculate odds ratios for each risk factor and age group. Results Ten studies of 4625 patients with FAP fulfilled our inclusion criteria. A total of 559 (12%) patients developed DT. Cumulative analysis demonstrated that 80% of DT developed by age 40, the peak incidence rate being in the second and third decades. A positive family history of DT was the most significant risk factor (OR 7.02, 95% CI 4.15-11.9, P < 0.001). An APC mutation 3' to codon 1399 (OR 4.37, 95% CI 2.14-8.91, P < 0.001) and previous abdominal surgery (OR 3.35, 95% CI 1.33-8.41, P = 0.01) were also implicated. Women were more likely to develop DT (OR 1.57, 95% CI 1.13-2.18, P = 0.007). Conclusion There is consistency amongst polyposis registries in documenting the incidence and risk factors for DT development. Having a positive family history for DT is of greater significance than a 3' mutation, suggesting the existence of modifier genes, independent of the APC genotype-phenotype correlation. Few of these risk factors are modifiable. Delaying prophylactic surgery could be appropriate in female patients with a 3' APC mutation and attenuated polyposis.