The evolution of treatment strategies for patients with chronic myeloid leukemia relapsing after allogeneic bone marrow transplant: can tyrosine kinase inhibitors replace donor lymphocyte infusions?

Abstract

The optimal treatment for chronic myeloid leukemia (CML) relapsing following allogeneic bone marrow transplant (alloBMT) is unknown. We performed a single-center retrospective analysis of 71 consecutive patients undergoing alloBMT for CML from 1995 to 2008. A multi-state model was used to quantify cumulative incidences of complete molecular response (CMR) and death following alloBMT. The primary analysis was comparison of three treatment interventions (tyrosine kinase inhibitor: TKI, donor lymphocyte infusion: DLI, and TKI + DLI) for relapsed disease post-alloBMT. Forty-five (63%) patients relapsed post-alloBMT (molecular relapse: n = 16, cytogenetic relapse: n = 20, hematologic relapse: n = 2, advanced phase relapse: n = 7) and 40 patients underwent one of three treatments: TKI-only (n = 13), DLI-only (n = 11) or TKI + DLI (n = 16). Although not statistically significant, the TKI-only group had the highest cumulative incidence of CMR and lowest cumulative incidence of death compared to DLI and TKI + DLI. These data support the finding that TKI therapy is active in the post-alloBMT setting.

Keywords: Chronic myeloid leukemia; allogeneic transplant; relapse.

Figure I

This schema highlights the cohort of 71 CML patients that underwent an alloBMT. Each box represents the transition of patients into respective categories based on the multi-state model.

Figure II

Cumulative Incidence Analysis A) Cumulative incidence of being alive in CMR following alloBMT. The solid line on the cumulative incidence curves for each treatment group represents the actual probability, and the dotted lines represent 95% confidence intervals. Though not statistically significant, the TKI only group had the highest cumulative incidence of CMR compared to DLI and DLI + TKI; 0.50 (95% CI: 0.13, 0.86) at 104 months (8.7) years. B) Cumulative incidence of death following alloBMT. The solid line on the cumulative incidence curves for each treatment group represents the actual probability, and the dotted lines represent 95% confidence intervals. Though not statistically significant, the TKI only group had the lowest cumulative incidence of death compared with DLI and TKI+DLI; 0.33 (95% CI: 0.03, 0.64) at 104 months (8.7) years.