Peripheral Blood Stem Cell Mobilization In Hiv Positive Patients with Lymphoma Candidates to Autologous Transplantation: Predictive Factors Analysis for Failure or Suboptimal Stem Cell Collection

Autologous stem cell (SC) transplantation (ASCT) is a potentially curative treatment for several hematologic malignancies and has been demostrated feasible and effective in HIV-related lymphoma (ARL). Peripheral blood SC collection could represent a major issue in the use of ASCT in HIV infected patients (pts).

To evaluate the feasibility and efficacy of SC mobilization in HIV positive (pos) pts with lymphoma and identify factors influencing harvest results. Potential “ongoing” predictors of collection were also assessed.

We retrospectively analysed 98 consecutive pts with ARL, candidates to ASCT, who underwent SC mobilization at 3 Italian and 2 Spanish centers from 2000 to 2010. A collection less than 2×10⁶ CD34+ cells/kg was defined as “mobilization failure”, between 2–5 as “suboptimal collection” and more than 5 as “good collection”. Several parameters were evaluated for correlation with outcome: age, sex, lymphoma histopathology, disease status, WBC and Plt count at start of mobilization, type of mobilizing therapy, marrow disease, previous mobilization failure, n° of previous chemotherapy (CT) lines, months from first detection of HIV positivity, CD4 count and HIV-viremia. Moreover, circulating CD34+ and WBC count on the first day of CD34+ monitoring and their ratio (SC ratio = CD34/WBC) were assessed as “ongoing” outcome predictors.

A total of 127 attempts of SC harvest in 98 pts were analysed. Median age was 41.5 ys (28-65). Lymphoma diagnosis was DLBCL in 42% of cases, Burkitt 10%, plasmablastic 10%, HL 31%, anaplastic 5%, follicular lymphoma 1% and PEL 1%. Disease status was complete remission in 36%, chemosensitive disease in 53% and refractory disease in 10% of cases. In 3 cases bone marrow was involved and mobilizations failed. In 18% of cases pts received mobilizing therapy after 1 previous CT line, in 67% after 2 and in 16% after 3 or more. All pts but 2 were on antiretroviral therapy. Median CD4 count was 231/mcl (50-1146) and HIV-viremia was detectable in 22%. Median time from first HIV detection was 79.5 ms (3-295). In 24% of cases G-CSF alone (10-20 mcg/Kg) was used as mobilizing treatment, while CT + G-CSF (5-10 mcg/Kg) in 76%, including single-agent Cyclophosphamide (CTX) 1.5 gr/ms (13%), CTX >3 gr/ms (27%), platinum containing regimens (20%), ifosfamide containing regimens (11%) and others (5%). Mobilization failure occurred in 40% of procedures, a collection between 2–5 × 10^6 CD34/Kg in 24% and > 5 in 35%. Finally, of 98 pts who underwent SC mobilization, 22% failed to collect enough cell to perform ASCT, 12 pts even after repeated attempts, 33% had a suboptimal and 45% a good collection (4 and 5 pts respectively after repeated mobilizations). At univariate analysis failure was significantly associated with refractory disease, Plt < 150.000/cmm, CTX 1.5 gr/ms as mobilizing treatment, previous mobilization failure and circulating CD34+ cell < 7.4/mcl on the first day of monitoring; whereas CTX > 3 gr/ms, CD4 count and SC ratio > 0.002 were associated with a reduced risk of failure. In multivariate analysis refractory disease (p<0.0001) and CTX 1.5 gr/ms (p=0.003) were indipendent predictors for failure and SC ratio > 0.002 (p<0.0001) a protective factor. A good collection was predicted at univariate analysis by Plt and CD4 count, age, months from first HIV detection, CT + G-CSF as mobilizing therapy, CTX > 3 gr/ms, WBC count and circulating CD34+ cells >29,7/mcl at the first day of monitoring and SC ratio > 0,002, whereas G-CSF alone and previous mobilization failure were negative predictive factors. Multivariate analysis confirmed CTX > 3 (p<0.0001), CD34+ cells > 29,7 (p=0.0003) and SC ratio > 0,002 (0.0036) as indipendent factors for good collection.

In this series of 98 ARL and 127 SC mobilization attempts, a substantial number of pts failed SC harvest (22%) whereas 33% had a suboptimal and 45% a good collection. Lymphoma status and mobilizing treatment seems the strongest predictors for outcome, with refractory disease and low CTX dose (1.5 gr/ms) significantly associated with failure and CTX > 3 gr/ms predictor for good collection. A high ratio between circulating CD34+ cells and WBC on the planned day of first apheresis might represent a useful “ongoing” parameter to predict the outcome. These data might help to decide the mobilizing strategy in ARL and could provide the framework to rationally explore the use of new mobilizing agents

No relevant conflicts of interest to declare.