Analyses of Prognostic Factors for Survival in East African Patients with AIDS-Related Non-Hodgkin’s Lymphoma (AR-NHL) Treated with Dose Modified Oral Chemotherapy.

Background: The results of our feasibility trial of dose-modified oral chemotherapy for AR-NHL in East Africa are encouraging to frame / sustain alternative non-myelotoxic and pragmatic therapeutic strategies suitable for evaluation in the resource-constrained setting. To define a group of patients who may have a better outcome univariate and multivariate analyses of prognostic factors were performed.

Methods: Forty nine patients with AR-NHL received dose-modified oral chemotherapy. The following baseline variables were included as potential predictors of survival: age (≤60 vs. >60 yrs.); sex; stage (≤2 vs. >2); ECOG performance status (0/1 vs. 2/3); LDH (≤270 vs. >270 U/L); extranodal sites of involvement (≤1 vs. >1 site); prior thrush; prior AIDS [wasting or opportunistic infection(s)]; A/B symptoms; m-CHOP salvage chemotherapy (n=10); antiretroviral (ARV) therapy (n=18); and baseline albumin. International Prognostic Index (IPI) excluding LDH since not available for all patients (

Results: By univariate analysis, only LDH (p=0.0013) was significantly related to overall survival (OS). Hypoalbuminemia and LDH affected event free survival (EFS). By multiple regression, LDH and IPI were significantly related to OS (p=0.012). Similarly, the hazard ratio of patients without and with ARV therapy was 17.45 (p=0.0007); and the hazard ratio of patients IPI 2 comparing to those with IPI 0 was 6.62 (p = 0.028). LDH, ARV status and IPI were significantly related to EFS. The complete response rate was 56% and objective response rate was 77%. Median OS and EFS were 12.3 and 7.9 months respectively.

Conclusions: For patients with AR-NHL in East Africa and treated with dose-modified oral chemotherapy IPI and not ACTG 142 prognostic covariates best predicts survival. Albumin (perhaps substituting for LDH since more readily tested in East Africa) and ARV status provide further independent prognostic information. These factors may further guide therapeutic options in future trials in resource poor settings.