Phase II trial of CHOP plus rituximab in patients with HIV-associated non-Hodgkin's lymphoma.

TO THE EDITOR: Recently Boue et al 1 published the results of a French phase II trial with rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) in 61 patients with HIVassociated non-Hodgkin’s lymphoma (HIV-NHL). They reported a complete remission rate of 77% and a 2-year survival of 75% without any increase of both toxic deaths and life-threatening infections. These results are similar to that of our pooled analysis of three international phase two studies using rituximab plus infusional cyclophosphamide, doxorucin and etoposide (R-CDE) 2 in 74 patients with HIV-NHL with a complete remission rate of 70% and a 2-year survival of 64%. However, a significant increase in hematologic toxicity and infections were found in the R-CDE study, in particular, grade 3 to 4 neutropenia in 78% of patients and infection in 31% of patients. These differences could be related to several factors. First, in the R-CDE study all patients were treated, while in the R-CHOP study only patients with good or intermediate prognosis (ie, no more than one of the following prognostic factors: CD4 cell count 100/dL, prior history of opportunistic infections, and Eastern Cooperative Oncology Group performance status 2) were enrolled. Second, due to the continuous infusion of R-CDE all patients used a central venous catheter, which is a risk factor for bacterial infections. Third, in the R-CDE study all antiretroviral drugs were allowed while in the R-CHOP study, the use of mielotoxic antiretroviral drugs, such as zidovudine or ritonavir, was avoided, leading to a low rate of grade 3 or 4 neutropenia (33% in R-CHOP v 78% in R-CDE). Last, the use of etoposide could have contributed to the increase of bone marrow toxicity. Overall, the results of the French study confirm the results of R-CDE with a high efficacy of the addition of rituximab to chemotherapy in the treatment of patients with HIV-NHL without any overall increase in severe toxicity. These results are in contrast with the previous US data published by Kaplan et al 3 where rituximab was associated to an unacceptable rate of treatment-related infectious deaths; although, the patient selection could be the reason for this toxicity. In conclusion, rituximab plus chemotherapy should be considered the standard treatment of patients with HIV-NHL. Randomized studies comparing rituximab plus CHOP versus rituximab plus infusional chemotherapy (CDE or etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin [EPOCH]) could be warranted in order to improve both response rate and overall survival.