Abstract
BACKGROUND: AIDS-related Kaposi sarcoma (AIDS-KS), a common malignancy in Kenya is associated with high morbidity and mortality. AIDS-KS is treated using bleomycin and vincristine (BV) plus or minus doxorubicin in most low resource settings, with response rates ranging from 24.8 to 87%. Survival in low resource settings has not been well documented. We report the three-year survival in a cohort of seventy patients referred to Moi Teaching and Referral Hospital (MTRH). METHODS: Study participants are part of a randomized phase IIA trial on the use of gemcitabine compared to bleomycin plus vincristine for the treatment of Kaposi sarcoma after combination antiretroviral therapy (cART) in Western Kenya. All patients were followed for three years in MTRH. Survival was determined by three monthly physical examination and analysed using Kaplan-Meier method, while possible determinants of survival such as baseline characteristics, type of chemotherapy, initial CD4 counts, age at enrolment, gender and early response to chemotherapy were analysed using univariate and multivariate Cox regression. RESULTS: Participants were aged between 19 and 70 years with 56% being male. The median CD4 count was 224 cells/μl, median duration of HIV diagnosis was 12.0 months and median duration of KS lesions after histology diagnosis before initiating chemotherapy was 4.8 weeks. At three years, 60 (85.7%) patients were alive. Six of those who died were under treatment with BV while four with gemcitabine. There was no difference in the probability of survival between the patients on either treatment arm (HR = 0.573 [95% C. I 0.143, 2.292; p = 0.4311]). Additionally, the hazard ratio (HR) for response after six weeks, age at enrolment and gender indicated that they were not significant determinants of survival. Patients with normal CD4 cell counts (> = 500/μl), had a HR of 0.401(0.05,3.23; p = 0.391), suggesting better survival. CONCLUSIONS: Patients with AIDS-KS treated with combined antiretroviral drugs had excellent three-year survival regardless of whether they were treated with BV or gemcitabine as first line therapy. An initial CD4 cell count of > = 500/μl appeared to improve survival while gender, age and early response to chemotherapy were not predictors of survival after three years. TRIAL REGISTRATION: Number PACTR201510001.