Background & Aims: In familial adenomatous polyposis, colorectal cance
r prevention is by prophylactic colectomy, but dietary or chemoprevent
ative strategies have been recently proposed in low-risk individuals.
The aim of this study was to define predictive risk factors for colore
ctal cancer in familial adenomatous polyposis. Methods: Between 1918 a
nd 1993, 317 patients underwent colectomy. A multivariate analysis was
performed to assess the relationship between colorectal cancer risk a
nd polyp count, sex, and age at colectomy. Results: The median polyp c
ount was 842 (range, 78-7500), and cancer was found in 22% of patients
. Polyp count and age, but not sex, predicted cancer risk. Patients wi
th >1000 polyps had 2.3 times greater risk of cancer than those with <
1000 polyps (P = 0.006). Synchronous cancers increased with polyp coun
t (P < 0.05). Each 10-year age group had a 2.4-fold difference in canc
er risk (95% confidence interval, 1.9-3.2; P < 0.001). Four cases of c
ancer occurred in patients at low risk (younger than 30 years of age,
(1000 polyps; 3.3%). Conclusions: More adenomas and older age are asso
ciated with a higher risk of colorectal cancer. However, cancer does o
ccur in low-risk individuals and may be missed by surveillance, making
alternatives to prophylactic surgery inadvisable.