Sulindac sulfone (Exisulind), a metabolite of the non-steroidal anti-inflammatory drug, sulindac, was evalauted for its effects on the development of rectal polyps in patients with familial adenomatous polyposis. Three cohorts of 6 patients each were given doses of 200, 300, or 400 mg Exisulind twice daily. Hepatotoxicity, shown by elevation in blood transaminase levels, was the dose-limiting toxicity and occurred at the 400 mg bid dose. Due to this toxicity, all patients treated with the 400 mg dose were subsequently reduced to the 200 mg dose level. Subsequently, 2 of the 6 patients were dose-escalated to 400 mg bid dose. The patients were treated with Exisulind for a period of six months. Sixteen of 18 patients had regression of small polyps (> or = 6 mm in diameter) characterized by a flattening of the polyps and a macular "halo" appearance. Histopathologic examination of the polyp biopsy specimens showed a marked increase in the proportion of mucin producing cells in the glands after treatment with Exisulind at all dose levels. Ki-67 staining, a measure of cell proliferation, was higher in the polyps than in normal mucosa. There was no significant change in the proliferation index between baseline and six month values in any of the groups treated with Exisulind or in normal tissues. The median apoptotic labeling index, as determined by the TUNEL technique, was higher in the polyps than in normal-appearing mucosa. Overall, there was no significant change in the apoptotic labeling index between base-line and 6 months in normal-appearing mucosa however, the index in polyps was increased. These results suggest that treatment of FAP patients with Exisulind for a period of six months may lead to regression of small polyps, and that the mechanisms of Exisulind--induced regression appear to be through stimulation of mucus differentiation and apoptosis in glandular epithelium.