Allogeneic Transplant Outcomes in Imatinib-Refractory Chronic Myeloid Leukaemia (CML) Are Similar to Transplant Outcomes in Imatinib-Responsive/Imatinib-Naïve CML and Appear To Be Predicted by the EBMT Risk Score

Aims: To review the outcome of allogeneic stem cell transplantation (SCT) in imatinib refractory chronic myeloid leukaemia (CML). Methods: Outcomes of all allogeneic transplants performed after January 2001 for CML at our institution were retrospectively reviewed. Imatinib-refractory CML was defined as either lack of any cytogenetic response (CGR) after at least 6mths of imatinib, loss of CGR or progression to a more advanced disease stage (accelerated or blast phase) during imatinib therapy. Using the EBMT risk score (Lancet1998; 352: 1087), transplant outcomes for imatinib refractory CML were compared with all other CML transplants performed during the same time period. Survival analysis was performed using the Kaplan-Meier product-limit and comparison of survival data via the log-rank test. Results: Of 31 allogeneic transplants (19M; 12F) performed for CML, 12 had been performed for imatinib refractory CML (no CGR to imatinib n=3; loss of CGR n=3; progression to AP n=3; progression to BC n=3), 5 in patients with imatinib responsive CML, and 14 in patients never exposed to imatinib. Median age at SCT was 40yrs (range 19–63yrs). Donor source included HLA-matched unrelated donors in 14 cases, HLA-identical siblings in 16 and other matched family donors in 1. Conditioning regimens included Cy/TBI (20 cases), Bu/Cy (8 cases), Flu/Mel (2 cases) and Flu/Cy (1 case); CsA + MTX was used as standard GVHD prophylaxis. EBMT risk scores were 1 (4 cases), 2 (6 cases), 3 (8 cases), 4 (5 cases), 5 (3 cases) and 6 (5 cases). At median follow-up post-SCT of 37mths (range 6–64mths), median PFS and OS are not reached; at 2yrs PFS, EFS and OS are 81%, 58% and 61% respectively. For patients with EBMT risk scores of 1–2 versus 3–4 versus 5–6, OS at 2 yrs post-SCT is 80%, 62% and 38% respectively (p=0.03). Based on EBMT risk score, no significant differences in PFS, EFS or OS were observed when comparing SCT for imatinib-refractory versus imatinib-responsive / imatinib-naïve CML. Conclusion: Our experience suggests that survival post-SCT for imatinib-refractory CML is similar to SCT for imatinib-responsive / imatinib-naïve CML. The EBMT risk score appears to remain useful in predicting survival post-SCT in imatinib-refractory CML.