We read with interest the results of the AIDS Malignancy Consortium phase 3 study of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) versus CHOP-rituximab in the treatment of patients with HIV infection and non-Hodgkin lymphoma.1 In this study, the authors observed increased infectious deaths and a trend toward increased opportunistic infections among patients randomized to receive CHOP-rituximab. Although rituximab therapy is known to cause prolonged B-lymphopenia and may decrease immunoglobulin levels in some patients,2 it has not been clearly associated with increased infections either as monotherapy2 or in combination with alkylating agents.3 In the AIDS Malignancy Consortium study, the risk of infectious death following CHOP-rituximab therapy was significant higher among patients with less than or equal to 50 CD4 lymphocytes/μL,1 and the authors postulated that the interaction of CD4 lymphopenia, CD20 lymphopenia, and neutropenia may have contributed to this increased risk. Moreover, the authors highlighted the association of fludarabine with CD4 lymphopenia and opportunistic infections, and implied that a similar mechanism of combined immunodeficiency may result from the addition of rituximab to fludarabine-containing regimens.
We agree that, in view of the above observations, a systematic assessment is warranted as to whether the addition of rituximab to fludarabine-containing therapy increases infection risk. In this regard, historic comparisons of the fludarabine, cyclophosphamide, and rituximab (FC-R) as well as the fludarabine-rituximab (F-R) regimens by investigators from the M.D. Anderson Cancer Center4 and the Cancer and Leukemia Group B (CALGB),5 respectively, did not show significant differences in early infection risk compared with those recorded in the preceding (non-rituximab-containing) regimens. Similarly, in randomized prospective comparisons of fludarabine versus F-R,6 fludarabine, cyclophosphamide and mitoxantrone (FCM) versus FCM-rituximab,7 and fludarabine, mitoxantrone, dexamethasone (FND) versus FND-rituximab,8 no excess in early infections was recorded in the rituximab-containing arms. We and others9 have previously examined the risk of early or late infections among 160 patients receiving FC or FC-R, and have found no significant differences in infections either during chemotherapy or in the first year of remission, even among patients at high predicted risk of infectious complications. In contradistinction to the results reported by the AIDS Malignancy Consortium, we did not observe significant increases in Herpes simplex, Varicella zoster, Pneumocytis jiroveci, or fungal infections among patients receiving FC-R, despite the lack of antimicrobial prophylaxis or growth-factor support in more than 70% of our patients.
In conclusion, despite theoretical concerns about combined immunosuppression leading to increased infections in patients receiving combination fludarabine and rituximab therapy, there is little evidence at present to support this concern. One explanation for this apparent discrepancy may be related to the observation that prolonged CD4 lymphopenia following purine analog therapy is associated with lower infection risk than when similar levels are observed among patients with HIV infection.10 Fludarabine and rituximab combinations are among the most potent regimens in the treatment of patients with indolent lymphoproliferative malignancies, and current safety data support the ongoing exploration of these promising regimens.