Long-term outcomes of CML patients undergoing hematopoietic stem cell transplantation in the era of tyrosine kinase inhibitors

Abstract Introduction: With the success of tyrosine kinase inhibitors (TKIs) in the treatment of chronic myeloid leukemia (CML), since 2005 far fewer patients routinely undergo allogeneic hematopoietic stem cell transplantation (HCT). Those that do undergo HCT are typically referred in chronic phase (CP) after resistance or intolerance of multiple TKIs, after progression to accelerated phase (AP) or blast phase (BP), or, rarely, after de novo BP presentation. We retrospectively analyzed the outcomes of a single-center cohort of CML patients who underwent HCT to evaluate risk factors affecting overall survival (OS) and relapse-free survival (RFS). Methods: We retrospectively evaluated patients who received HCT for CML in all phases at Fred Hutchinson Cancer Center between 2005 and 2023, following IRB approval. Relapse was defined per 2025 NCCN guidelines. RFS was defined as time from transplant to relapse or death; OS was defined as time from transplant to death. Hazard ratios (HR) were estimated using Cox regression. Results: Over a 19-year period, we identified 135 patients with a median age of 45 years (range 18-74). Fifty-one were in first CP; 26 and 58 had AP or BP prior to HCT, respectively. Fifty-five of the 58 patients with prior BP were in CP on pre-transplant evaluation. Forty-one patients received only imatinib prior to HCT, while 94 had exposure to at least one second-generation TKI prior to HCT. The plurality had a matched-unrelated donor (58; 43%); other donor sources were: matched-related (41), mismatched-unrelated (21), cord blood (10), haploidentical (4), and mismatched-related (1). One hundred twenty-four patients received myeloablative conditioning and 11 had either reduced-intensity or nonmyeloablative preparative regimens. Sixty-four patients received a maintenance TKI after HCT. Thirty-one patients (23%) relapsed after HCT; median time to relapse was not reached. The hazard ratios (HR) for relapse among patients transplanted after AP and those transplanted after BP relative to first CP were 2.34 (95% CI 0.75-7.28) and 3.80 (95% CI 1.51-9.56), respectively. Median RFS was 100 months (95% CI 30-179 months). In univariate analyses of RFS, the HR for failure for recipient positive versus negative CMV serostatus was 1.93 (95% CI 1.14-3.26); BP prior to transplant vs CP HR=1.95 (95% CI 1.14-3.35); and reduced-intensity/non-myeloablative versus ablative conditioning HR=2.28 (95% CI 1.08-4.78). Estimated 10-year OS for all patients was 58% (95% CI 49 – 67%). Median OS was 179 months (CI 100 months – not reached). Estimated 10-year OS for first CP, AP, and BP was 64, 61, and 52%, respectively. In univariate analysis, recipient positive CMV serostatus HR=1.82 (95% CI 1.01-3.29) and reduced intensity or non-myelablative conditioning HR=2.38 (95% CI 1.12-5.05) were associated with higher mortality. Discussion: Long-term outcomes for patients receiving HCT for CML were excellent, with a median RFS of 100 months and median OS of 179 months. Notable risk factors associated with worse RFS or OS were recipient positive CMV serostatus, BP prior to HCT, and reduced-intensity conditioning. Patients who underwent reduced-intensity conditioning were either older or had underlying comorbidities that precluded myeloablative conditioning, which may partially explain poorer outcomes. With the advancement of TKI therapy for CML, patients that do undergo HCT have received multiple TKIs and likely have more aggressive disease biology. It is reassuring that HCT remains an effective potentially curative option for this patient population.