Long-Term Outcomes after Allogeneic Stem Cell Transplantation Versus New Generation Tyrosine Kinase Inhibitors in Chronic Phase Chronic Myeloid Leukemia

Introduction: Allogeneic hematopoietic stem cell transplantation (HSCT) remains a potentially curative option in tyrosine kinase inhibitors (TKI) - resistant chronic phase (CP) chronic myeloid leukemia (CML). New generation (NG) TKIs (Ponatinib - third generation TKI, Asciminib - specifically targeting the ABL myristoyl pocket inhibitor) demonstrate high efficacy and survival outcomes even in multi-TKI-resistant CP-CML. The objective of this study is to compare overall survival (OS) between patients with CP-CML who received NG-TKIs versus HSCT.

Methods: A retrospective study with indirect comparison of OS between patients (age≥18 years at the time of index treatment) with CP-CML without history of advanced phases who received NG-TKIs versus HSCT was conducted in 2 Russian centers. Overall, 43 patients were transplanted in their first CP-CML. Of these, 11 patients who received NG-TKIs before HSCT were excluded. In total, 76 CP-CML patients treated with NG-TKIs, were in our database with all available study data, only 64 patients were included in the study by their first NG-TKI or HSCT treatment. Propensity score matching analysis was applied to adjust clinically significant characteristics that were unbalanced between the groups, including age at HSCT or NG-TKI start and time from diagnosis to intervention. OS was calculated by Kaplan-Meier estimator from the start date of intervention until death or last available visit date. Log-rank test was used to compare OS in HSCT group versus matched NG-TKI whole group or by baseline level of BCR::ABL (>10% and ≤ 10%).

Results: After adjustment a total of 64 patients were finally extracted (1:1, 32 in each group). The patients had following baseline characteristics in NG-TKI and HSCT groups, respectively: a) males were 14 (44%) vs 13 (41%); b) median age at diagnosis was 37 (14-67) vs 36 (17-58) years; c) median age at index date was 40 (20-71) vs 42 (19-61) years; d) time from diagnosis to index date was 39.0 (5.4-284.6) vs 44.3 (8.6-250.1) months. Up to 40% received ≥ 3 of TKIs before the intervention in both groups. Median time of NG-TKIs treatment was 49.3 (0.5-85.8) months. In NG-TKI group 5 (16%) patients were switched to HSCT. All of them were alive at the last visit. They were not censored at the time of HSCT. Median time of follow-up for whole cohort and for alive patients in NG-TKI and HSCT was 73.4 (1.4-87.4) months and 77.4 (1.4-87.4) vs 43.2 (1.1 - 322.2) and 44.6 (6-322.2) months, respectively. No deaths occurred after 2 years of follow-up in both groups. Patients in NG-TKI group had significantly better OS at 24/48 months compared with HSCT group: 90.0 (79.9-100.0) % vs 62.5 (47.2- 82.9) %, respectively (p = 0.010). Median OS was not reached (NR) in both groups. All 3/32 (9.4%) who died patients in NG-TKI group had BCR::ABL1 >10% at the start of therapy. Overall the baseline BCR::ABL1 level was >10% in 24/32 (75%) patients in NG-TKI group. In these subgroup of patients OS at 24/48 months was 85.7 (72.0-100.0) %. There were no statistical differences in OS between HSCT group and high BCR::ABL cohort patients in NG-TKI group (p=0.058).

Conclusion: In our study CP-CML patients had better survival outcomes compared to those who underwent a HSCT. It seems that HSCT should be carried out in case of failure of at least one NG-TKI in CP-CML patients.

Lomaia:Fusion Pharma: Speakers Bureau; Pfizer: Other: Travel, accommodation, and expenses, Speakers Bureau; Novartis: Other: Travel, accommodation, and expenses, Speakers Bureau. Kulemina:Novartis: Speakers Bureau. Vlasova:Novartis: Speakers Bureau. Badaev:Abbvie: Speakers Bureau. Sbityakova:Novartis: Speakers Bureau. Alexeeva:Roche: Other: travel expenses, accommodation, Speakers Bureau. Morozova:Novartis: Speakers Bureau.