Outcome of HIV-associated lymphoma in a resource-limited setting of Jos, Nigeria

BACKGROUND: Lymphoma is a leading cause of cancer-related death among human immunodeficiency virus (HIV)-infected individuals in the current era of potent anti-retroviral therapy (ART). Globally, mortality after HIV-associated lymphoma has profound regional variation. Little is known about HIV-associated lymphoma mortality in Nigeria and other resource-limited setting in sub-Saharan Africa. Therefore, we evaluated the all-cause mortality after lymphoma and associated risk factors including HIV at the Jos University Teaching Hospital (JUTH) Nigeria. METHODS: We conducted a ten-year retrospective cohort study of lymphoma patients managed in JUTH. The main outcome measured was all-cause mortality and HIV infection was the main exposure variable. Overall death rate was estimated using the total number of death events and cumulative follow up time from lymphoma diagnosis to death. Cox proportional hazard regression was used to assess factors associated with mortality after lymphoma diagnosis. RESULTS: Out of 40 lymphoma patients evaluated, 8(20.0%) were HIV positive and 32(80.0%) were HIV negative. After 127.63 person- years of follow-up, there were 16 deaths leading to a crude mortality rate of 40.0 per 100 person-years. The 2-year probability of survival was 30% for HIV-infected patients and 74% for HIV-uninfected. Median survival probability for HIV-infected patients was 2.1 years and 7.6 years for those without HIV. Unadjusted hazard of death was associated with late stage, HR 11.33(95% CI 2.55, 50.26,p = 0.001); low cumulative cycles of chemotherapy, HR 6.43(95% CI 1.80, 22.89,p = 0.004); greater age, HR 5.12(95% CI 1.45,18.08,p = 0.01); presence of comorbidity, HR 3.43(95% CI 1.10,10.78,p = 0.03); and HIV-infection, HR 3.32(95% CI 1.05, 10.51,p = 0.04). In an adjusted model only stage was significantly associated with death, AHR 5.45(1.14–26.06, p = 0.03). CONCLUSION: Our findings suggest that HIV- infection accounted for three times probability of death in lymphoma patients compared to their HIV-uninfected counterparts due to late stage of lymphoma presentation in this population. Also initiation of chemotherapy was associated with lower probability of death among lymphoma patients managed at JUTH, Nigeria. Earlier stage at lymphoma diagnosis and prompt therapeutic intervention is likely to improve survival in these patients. Future research should undertake collaborative studies to obtain comprehensive regional data and identify unique risk factors of poor outcomes among HIV-infected patients with lymphoma in Nigeria.