Stem Cell Dose and Time To Engraftment Does Not Predict Outcome After Autologous Stem Cell Transplantation For Relapsed Aggressive NHL

Autologous stem cell transplantation (ASCT) is the standard of care for relapsed aggressive lymphomas. Time to neutrophil and platelet engraftment is strongly correlated with CD34+ cell number infused but data are conflicting as to whether patients who receive greater numbers of CD34+ stem cells have improved outcomes. We sought to determine whether short term engraftment predicts progression-free survival (PFS) independent of other disease-specific prognostic factors.

From the Princess Margaret Cancer Centre transplant database, we identified patients undergoing ASCT for relapsed aggressive lymphoma between 2007-2011. Data were extracted on prognostic features at relapse/progression, stem cell collection, engraftment, and time to progression and death.

All patients received platinum-based salvage chemotherapy and those with chemosensitivity were mobilised with cyclophosphamide, etoposide and filgrastim (minimum threshold 2x10⁶ CD34 cells/kg) and proceeded to ASCT. Patients who failed initial mobilisation were remobilised using plerixafor. High-dose therapy consisted of etoposide 60mg/kg Day -4 and melphalan 180mg/m² Day -3 with stem cells infused on Day 0. Filgrastim 300µg daily was started from Day +7 until neutrophil recovery to >1.0 x10⁶/uL. Platelet engraftment was defined as an unsupported platelet count >20 x 10⁹/L.

97 patients with DLBCL (n=66), transformed (n=24) and T-cell lymphoma (n=7) were reviewed. Median age was 54 (range 20-67), 61% were male and median IPI score on relapse was 2. Fifty one percent relapsed within 12 months of last therapy, and of the patients with B-cell lymphoma, 81% received rituximab prior to salvage chemotherapy. Median stem cell dose was 5.7x10⁶ CD34+ cells/kg (range 1.69-17.82) with a median number of apheresis sessions to achieve this of 2 (range 1-4). Median time to neutrophils >0.5x10⁶/uL was 11 days (range 9-14) and platelets >20x10⁶/uL was 14 days (range 10-23). The Spearman correlation test confirmed a higher stem cell dose was significantly associated with shorter time to neutrophil (p=0.0014) and platelet engraftment (p=0.0003).

From date of ASCT, median follow-up was 3.1 years in progression-free patients. For the entire cohort, PFS was 50% and overall survival (OS) was 74% at 3 years. On univariable analysis, patients with B-cell lymphoma with IPI score of 0-2 had a 3-year PFS of 59%, v 28% for those with IPI of 3-4 (p=0.03) (n=90). Patients with early relapse within 12 months of last therapy had inferior 3-year PFS, 42% v 59% for those with initial PFS > 1 year (p=0.08). Patients with B-cell lymphoma who received rituximab with primary chemotherapy had worse 3-year PFS: 47% v 69% (p=0.11). There were no associations of PFS with lymphoma subtype, dose of stem cells infused, number of apheresis sessions and neutrophil or platelet engraftment. Similarly, on univariable analysis of OS for the entire cohort, only secondary IPI (0-2 v 3-4) was significant, with 3-year OS 82% v 48%, respectively (p=0.01).

Multivariable Cox regression analysis of outcomes for patients with B-cell lymphoma in a model including IPI score, time to relapse, prior rituximab, CD34+ cell dose and neutrophil and platelet engraftment times confirmed IPI was the only significant variable predicting PFS (HR 1.99, p=0.03) and OS (HR 3.2, p=0.006).

In this cohort of patients with aggressive lymphomas, CD34+ cell dose was correlated with time to neutrophil and platelet engraftment but was not predictive of PFS or OS. Secondary IPI score, relapse within 12 months, and for B-cell lymphomas previous use of rituximab were predictive of outcomes post ASCT.

Kuruvilla:Roche: Honoraria. Kukreti:Millennium Pharmaceuticals: Research Funding; Onyx: Research Funding. Keating:Roche: Honoraria. Crump:Roche: Honoraria; Jansen-Ortho: Honoraria; Celgene: Honoraria; Lundbeck: Honoraria; Novartis: Research Funding; Seattle Genetics: Honoraria.