Comparative Outcome of Myeloablative and Reduced Intensity Allogeneic Stem Cell Transplantation for Chronic Lymphocytic Leukemia.

Allogeneic stem cell transplantation (SCT) remains the only potentially curative modality for patients with CLL. Reduced intensity conditioning (RIC) approaches are commonly used because of lower initial toxicity and the belief that cure is primarily mediated by a GVL effect. However, the role of dose intensity in these patients has not been fully explored. We analyzed the outcomes of 88 CLL patients undergoing a first allogeneic SCT from a HLA-matched (6/6) donor at DFCI between 1998 and 2007. 26 patients receiving ablative transplants were conditioned either with standard doses of cyclophosphamide and total body irradiation (96%), or cyclophosphamide and oral busulfan (4%). 62 patients receiving RIC were conditioned with fludarabine 30 mg/m² and intravenous busulfan 0.8 mg/kg/d, both for 4 days. GVHD prophylaxis regimens consisted mostly of tacrolimus/ sirolimus +/− MTX (35% ablative, 63% RIC), cyclosporine with prednisone or MTX (15% ablative, 23% RIC) and T cell depletion (31% ablative, 0% RIC). 50% of the ablative SCTs were performed after 2002, compared to 90% of the RICs (p < 0.001). RIC patients were older, with a median age of 55 (36–55), compared to 47 (27–60) for the patients having ablative SCTs (p < 0.001). Virtually all RIC patients (98%) received PBSC, compared to 54% of patients undergoing ablative SCT (46% bone marrow) (p < 0.001). 66% of RIC patients had unrelated donors, compared to 42% of patients undergoing ablative SCT (p = 0.06). No differences between groups were observed for patient gender, donor-patient sex matching, or disease status at conditioning (56% CR/PR for RIC, 58% for ablative SCT). No significant difference in rates of acute GVHD (grade 2–4, 34% in RIC and 48% in ablative SCT; p = 0.23) and chronic GVHD (72% in RIC and 55% in ablative SCT; p = 0.18) was observed. The median follow-up of survivors was 3.3 years for the RIC patients and 5.1 years for the ablative SCT patients. No statistical difference between RIC and ablative SCT was noted in PFS or OS, with 3 year PFS 54% for RIC and 38% for ablative SCT (p=0.30 from stratified log-rank test, by year of transplant before or after 2002) and 3 year OS 65% for RIC and 50% for ablative SCT (p=0.19 from stratified log-rank test by transplant year). As anticipated, the 3 year cumulative incidence of transplant related mortality (TRM), reflecting relapse as a competing risk, was 15% for RIC, compared to 54% for ablative SCT (p < 0.001, Gray’s test). The 3 year cumulative incidence of relapse following RIC was 30%, compared to 8% after ablative SCT (p=0.038, Gray’s test). In Cox proportional hazards regression analysis considering type of conditioning regimen (RIC vs ablative), age, donor patient sex mismatch, PBSC vs BM, matched related vs unrelated donor, disease status at conditioning (CR/PR vs other), GVHD prophylaxis (sirolimus vs non-sirolimus), and transplant year (before, or 2002 and after), only disease status at conditioning was an independent predictor of OS (CR/PR, HR=0.30 (95% CI 0.13–0.66), p = 0.003) and PFS (CR/PR, HR=0.48 (95% CI 0.25–0.93), p = 0.03). The type of transplant conditioning did not reach statistical significance for OS (ablative SCT HR=2.36, p = 0.07) or PFS (HR=1.28, p = 0.58). In a competing risks regression model evaluating the same factors, significant predictors of TRM included disease status at conditioning (CR/PR, HR=0.27 (95% CI 0.09–0.78), p = 0.02), and conditioning type (ablative SCT, HR=7.69 (2.97–19.92), p < 0.001). The only significant predictor of relapse was type of conditioning (ablative SCT, HR=0.11 (95% CI 0.03–0.39), p < 0.001). Although the patients undergoing RIC likely differ systematically from the patients undergoing ablative SCT, these data suggest that, although TRM is, as expected, higher following ablative SCT in CLL, relapse is also significantly lower, suggesting that dose intensity may play a role in disease control. Given the counterbalancing factors of TRM and relapse, PFS and OS are similar following the two approaches. Future research could focus on efforts to enhance GVL or dose intensity without increasing TRM.