Autologous Stem Cell Transplantation (autoSCT) for HIV-Associated Lymphoma in the Era of Combination Antiretroviral Therapy (cART): A Retrospective Analysis of the EBMT Lymphoma Working Party

Introduction: Patients infected with HIV have an increased risk of developing aggressive B-cell non-Hodgkin lymphoma and Hodgkin lymphoma (HL). The continuous development of cART during the last decade has improved the prognosis of HIV-associated lymphoma considerably. However, a significant proportion of these patients will experience lymphoma relapse and may be candidates for autoSCT. The purpose of the present study was to investigate if recent advances in anti-lymphoma therapy and anti-infectious strategies have influenced the outcome of autoSCT for HIV-related lymphoma.

Patients and methods: For this retrospective study, all EBMT-registered patients aged 18 years or older with HIV-positive serostatus who were treated with a first autoSCT between 2007 and 2013 were eligible. Baseline patient, disease, and transplant data were collected from MED-A forms. Centers were requested to provide additional HIV and lymphoma treatment and follow-up information. Statistical analysis used log rank test to assess the impact of baseline characteristics on survival endpoints. In multivariate analysis, the relevance of prognostic factors was estimated using Cox regression models. Curves of cumulative incidence of relapse (IR) and non-relapse mortality (NRM) were compared by Gray's test.

Results: 138 patients from 25 European centers met the eligibility criteria and had the full data set required for this analysis available. 86% were male, median age was 44 years (range 24-69). Underlying diagnoses were diffuse large B cell lymphoma (DLBCL) in 46%, HL in 21%, Burkitt lymphoma in 14%, plasmablastic lymphoma (PBL) in 10%, and other lymphoma in 9% of the patients. Disease status at autoSCT was complete remission (CR) in 51%, partial remission (PR) in 33%, and less than PR in 16% of the patients, achieved after 1 (28%), 2 (58%), or more than 2 lines of chemotherapy (14%). With HIV load below the threshold of detection in 74% of the patients, the median CD4+ cell count was 187/µl (range 0-800) at transplant. 95% of the patients continued with cART during salvage and high-dose chemotherapy. BEAM was used as high-dose regimen in 77% of the patients. With a median follow-up of 4 years, 2-year NRM, IR, progression-free survival (PFS) and overall survival for the whole series were 9%, 23%, 68% and 70%, respectively. By multivariate analysis, diagnosis DLBCL or PBL (vs HL), increasing number of chemotherapy pretreatment lines, and less than PR at autoSCT were significant predictors of an unfavorable PFS; whilst age, high-dose regimen, performance status, and viral load had no significant impact. 2-year PFS in patients with 1^st-line CR, later CR, PR, or less than PR at autoSCT was 91%, 80%, 64%, and 23%, respectively.

Conclusions: This series, which is the largest ever on lymphoma transplants in HIV+ patients, suggests that in the cART / chemoimmunotherapy era, the outcome of autoSCT for HIV-related lymphoma is driven by lymphoma-dependent risk factors rather than by characteristics of HIV infection. AutoSCT under ongoing cART therapy remains the treatment of choice for HIV+ patients with PBL or recurrent DLBCL or HL.