AIDS-related Burkitt's lymphoma versus diffuse large-cell lymphoma in the pre-highly active antiretroviral therapy (HAART) and HAART eras: significant differences in survival with standard chemotherapy.

PURPOSE: To compare outcomes of patients with HIV-Burkitt's lymphoma (HIV-BL) and HIV-diffuse large-cell lymphoma (HIV-DLCL) after treatment with CHOP (cyclophosphamide, doxorubicin, vincristine, prednisone) or M-BACOD (methotrexate, bleomycin, cyclophosphamide, etoposide) in pre-highly active antiretroviral therapy (HAART) versus HAART eras.

PATIENTS AND

METHODS

Three hundred sixty-three patients with AIDS-related lymphoma diagnosed from 1982 to 2003 were reviewed retrospectively, including 262 in the pre-HAART (HIV-BL, 117; HIV-DLCL, 145) and 101 in the HAART era (HIV-BL, 18; HIV-DLCL, 83). Pre-HAART included those who did not receive HAART, and HAART era included those diagnosed after January 1997 who received HAART.

RESULTS

There were no significant differences between groups in terms of age, sex, history of injection drug use, prior AIDS, lactate dehydrogenase level, and disease stage at diagnosis. Compared with HIV-BL, HIV-DLCL was associated with significantly lower CD4 counts in the pre-HAART but not the HAART era. Although the overall median survival was similar for both groups in the pre-HAART era (HIV-BL, 6.4 months v HIV-DLCL, 8.3 months; P = .43), survival was significantly worse in patients with HIV-BL in the HAART era (HIV-BL, 5.7 months v HIV-DLCL, 43.2 months; P = .0003). Failure to attain complete remission and CD4 count less than 100 cells/mm(3) independently predicted for poor survival in the pre-HAART era. In comparison, histology of HIV-BL and no attainment of complete remission were independent poor prognostic factors in the HAART era.

CONCLUSION

Survival of patients with HIV-DLCL has improved in the HAART era, along with CD4 count, whereas survival of similarly treated patients with HIV-BL remained poor. The current practice of using the same regimen for both groups of patients should be re-evaluated.