Imatinib Followed by Reduced Intensity Stem Cell Transplantation and Donor Lymphocyte Infusions Is Effective in Chronic Phase CML with Less Toxicity Than Standard Allogeneic Transplantation.

Despite recent advances in understanding the molecular biology of CML and the development of new therapies, standard allogeneic stem cell transplant (alloSCT) remains the only curative option for chronic myeloid leukaemia (CML). It was hoped the introduction of imatinib mesylate (IM) would provide long term remission or even cure, but although IM induces a complete cytogenetic response (CCR) in the majority of patients with chronic phase, it is unlikely to be curative due to molecular persistence. To overcome this, we developed a strategy of remission induction with IM followed by either reduced intensity stem cell transplant (RISCT; fludarabine/melphalan/Campath-1H) for those achieving CCR after 6–12 months or aged >50 years, or alloSCT (cyclophosphamide/TBI) for those who either progressed on IM or did not achieve CCR after 12 months. Patients received IM 400mg/d, increasing to 600mg/d if they did not achieve MCR after 3 months or CCR after 6–9 months. GVHD prophylaxis was either cyclosporine alone (RISCT) or cyclosporine/methotrexate (alloSCT). Quantitative RT-PCR for Bcr-Abl and donor chimerism were performed 6 weekly following engraftment. Incremental donor lymphocyte infusion (DLI; 5x10⁵, 10⁶, 5x10⁶, 10⁷, 5x10⁷ at 12 week intervals) was given at >6 months post RISCT to patients with <100% donor chimerism or remaining Bcr-Abl⁺ at >0.02% on 2 repeat analyses 2 weeks apart.

17 patients (aged 20–55 years) with Ph⁺ CML have been recruited and 13 have received a transplant (10 RISCT, 2 alloSCT, and 1 MUD alloSCT), all with satisfactory engraftment. Pre-transplant, 8 patients achieved CCR, 2 MCR, 2 had no response to IM and 2 progressed to blast crisis on IM therapy. 1/10 RISCT patients developed chronic GVHD and died of sepsis 12 months post transplant (no DLI). 6/8 patients >6 months post RISCT have required incremental DLI and one of these patients has developed GVHD (grade I). As yet, no patients are consistently PCR negative (median Bcr-Abl/Abl ratio 0.02%; range 0.002 to 0.07%). The results for this group of RISCT patients was compared with a set of historical controls that received standard alloSCT for CML. The RISCT group was significantly older than the alloSCT group (mean age 42 vs. 31 years; P=0.017). Mean neutrophil engraftment (11 vs. 19 days), platelet recovery (11 vs. 18 days) and days to discharge (17 vs. 25 days) were significantly shorter in the RISCT group (P<0.001, P=0.005 and P<0.001 respectively). 1/9 alloSCT controls died on day 129 post transplant after failing to engraft. 2/9 RISCT recipients required readmission in the first year post-transplant compared to 6/8 alloSCT patients. In comparison to the RISCT group (where DLI was part of the protocol), no patients in the control cohort required DLI for relapse. Despite this, the number of patients developing GVHD in the control group was significantly increased (7/9 vs. 2/10; P=0.012). The combination of IM with RISCT is much less toxic, with more rapid engraftment and fewer readmissions than alloSCT. It is likely that the majority of CML patients receiving IM followed by RISCT will require DLI to induce molecular “cure”. Prolonged follow-up is required to establish long-term efficacy.