Limitations to Receiving Allogeneic Hematopoietic Cell Transplantation for Treatment of Acute Myeloid Leukemia: A Large Multi-Center Prospective Longitudinal Observational Study

Abstract

Introduction Older pts with AML often have significant comorbidities and other geriatric health problems, and the effect of these on the probability of receiving HCT is unknown. We designed a prospective, longitudinal, observational study of adult pts with AML dating from first presentation at one of 13 centers. We examined the effects of different variables on the probability to 1) survive long enough to receive HCT and 2) to receive HCT if such survival occurred. Methods We enrolled 695 pts (Table 1). Data on demographics, AML status, cytogenetic risks per European Leukemia Network (ELN), response, age, comorbidities per the HCT-comorbidity index (CI), function, frailty, geriatric assessment, cognition, Karnofsky performance status (KPS), QOL, social support, and depression were collected at enrollment and at 1, 3, 6, 9, 12, 18, and 24 months thereafter. We used competing risk Cox regression analyses, treating HCT as the event of interest and death without HCT as a competing risk, with staggered entry (left truncation) at time of consent. Associations between variables were assessed both at enrollment and over time. Results The overall rate of HCT at 9 months after enrollment was 43% (Figure 1) and 92% who received HCT did so by 9 months. In multivariate analyses, death without HCT (Table 2) was associated with augmented HCT-CI scores ≥5, age ≥50 years with those aged ≥70 years having the highest association, ELN intermediate or unfavorable risks, receiving low-intensity induction regimens, relapsed/refractory disease at enrollment, dependent status per ADL scores Among survivors (Table 3), low likelihood to receive HCT was associated with age ≥70 years, low ELN risk, low-intensity induction, poor KPS, and relapse after initial complete remission (CR); while pts with high-risk MDS, relapsed/refractory disease at enrollment, and CR after induction were more likely to receive HCT. Among pts aged ≥60 years, and after considering previous factors, impaired cognition and hearing had a lower likelihood to receive HCT. Conclusions Increasing age, comorbidity burden, ELN risk, low-intensity initial AML induction regimen, depression, and functional dependence increase risks of early mortality without HCT. In those who survived long enough to potentially receive HCT, age up to 69 years and/or multiple comorbidities were not found to be barriers to HCT, likely reflecting the widespread use of reduced-intensity conditioning regimens. The adverse impact of impairments in psychological health and function on survival and of impairments of cognition, geriatric health, and performance status on receipt of HCT emphasize the need for interventions that target these health limitations in conjunction with AML treatment to improve outcomes. Finally, the benefit of intensive vs. less-intensive induction therapies should be addressed with a randomized trial.