Polishing the crystal ball: knowing genotype improves ability to predict desmoid disease in patients with familial adenomatous polyposis.

INTRODUCTION

Desmoid disease occurs in one third of patients with familial adenomatous polyposis. Patients may be protected by changing surgical strategy. We designed a formula to predict desmoid risk and tested the value of adding genotype to the formula.

METHODS

A desmoid risk factor was calculated by summing points awarded for gender (male = 1, female = 3), extracolonic manifestations (nil = 1, one = 2, >one = 3), and family history of desmoids (negative = 1, one relative = 2, more than one relative = 3). Performance of the score with and without genotype (5' 1309 = 1, 1309-1900 = 2, 3' 1900 = 3) was analyzed.

RESULTS

There were 839 patients (138 desmoids) without genotype and 154 (30 desmoids) with genotype. The mean desmoid risk factor score of patients without desmoids (no genotype) was 4.7 (+/-1.4 SD) and for patients with desmoid the desmoid risk factor was 6.0 (+/-1.7, P < 0.001). Corresponding data for patients with genotype was 6.1 +/- 1.3 (no desmoids) and 8.4 +/- 1.8 with desmoids (P < 0.001). Of patients without genotype, 648 patients were at low risk and 9.9% had desmoid disease, 178 patients were at medium risk and 34% had desmoids, and 10 patients were at high risk and all had desmoids. Of those with genotype information, 83 patients were at low risk and 5% had desmoids, 52 patients were at medium risk and 21% had desmoids, and 18 patients were at high risk and 83% had desmoids.

CONCLUSION

The desmoid risk factor identifies patients with various levels of risk for developing desmoid disease, and can be used to plan surgical strategies designed to minimize desmoid risk.