The response of symptomatic gastrointestinal Kaposi's sarcoma to chemotherapy: a prospective evaluation using an endoscopic method of disease quantification.

We used an endoscopic method of quantification to evaluate the response of symptomatic gastrointestinal Kaposi's sarcoma (KS) prospectively in seven patients with acquired immune deficiency syndrome (AIDS) who were participating in chemotherapy trials for extensive cutaneous KS. The sums of the diameters of KS lesions in the esophagus, stomach, duodenum, and distal colon were used as a measure of extent of disease. Intravenous therapy [adriamycin/bleomycin/vincristine (N = 5), adriamycin (N = 1), or bleomycin/vincristine (N = 1)] was given every 2 wk for a mean of six cycles. Five of seven patients (71%) had a cutaneous response, whereas 9/15 (60%) gastrointestinal sites showed a remission. Sites of complete response all had an initial sum of lesion diameters less than or equal to 30 mm. Five of five patients with duodenal KS responded (3/5 complete), whereas just two partial responses were seen in five patients with gastric KS (4/5 with duodenal KS had sums of diameters less than or equal to 20 mm, whereas 4/5 with gastric KS had sums greater than or equal to 150 mm). Symptoms (abdominal pain, nausea/vomiting, hematemesis, diarrhea) resolved in all patients within two cycles of therapy. In summary: 1) the response rate of gastrointestinal KS to chemotherapy is similar to that of cutaneous KS; 2) the best response is seen in patients with less extensive disease and duodenal involvement; and 3) symptoms of gastrointestinal KS respond to chemotherapy even if the KS lesions do not resolve.