Chemotherapy combined with zidovudine and GM-CSF in human immunodeficiency virus-related non-Hodgkin's lymphoma

Abstract

Objective: The treatment of patients with non-Hodgkin's lymphoma related to the human immunodeficiency virus (HIV-NHL) is complicated by the underlying acquired immunodeficiency syndrome (AIDS). Patients without adverse prognostic factors (no AIDS prior to lymphoma, CD4+ lymphocyte counts greater than 100 x 10(6)/l and good performance status) can be cured of lymphoma and experience long-term survival. Our previous study with the intensive chemotherapy LNH84 regimen yielded a 63% complete response (CR) rate but median survival was only nine months, half of the patients died of AIDS and the other half of their lymphoma. We report here the results of a phase II study combining the same chemotherapy with zidovudine and GM-CSF. Our goal was to improve the treatment outcome over that of our previous study; GM-CSF was expected to decrease the hematological toxicity of chemotherapy and thus permit a dose intensity increase, while zidovudine was supposed to slow down the evolution of AIDS.

Design and setting: A phase II non-randomized prospective clinical trial in 7 centres.

Patients and methods: Thirty-two consecutive adult patients presenting HIV-NHL and performance status of less than three without active opportunistic infection underwent three cycles of doxorubicin 75 mg/m2, cyclophosphamide 1,200 mg/m2, vindesine 2 mg/m2 for two days, bleomycin 10 mg for two days and prednisolone 60 mg/m2 for five days (ACVB). Chemotherapy was associated with zidovudine (5 mg/kg/d) and GM-CSF (5 mu g/kg/d). The induction phase was followed by a four-month consolidation phase.

Results: CR and PR > 75% were observed in 56% of patients; 25% of the patients died during the induction phase. These results were analogous to those of the previous study (63% and 14%, respectively). Neither hematological tolerance nor dose intensity were improved. With a mean follow-up of 23.5 months, median survival was 6.7 months. The rate of non-NHL AIDS-related death during CR was not reduced (22% in our study vs. 16% in our previous one).

Conclusions: GM-CSF failed to reduce significantly the cumulative hematological toxicity of chemotherapy and zidovudine. New antiviral agents without hematological toxicity would perhaps be useful in this setting.