Impact of Frailty on Hematopoietic Cell on Early Transplant Outcomes in Older Recipients

Introduction: With improvements in transplant strategy and supportive care, hematopoietic cell transplantation (HCT) is increasingly being used to treat older patients (60-80 years). Frailty is a well-accepted phenomenon in the geriatric population characterized by diminished physiological reserve and increased vulnerability to stress, with strong associations with falls, disability, hospitalization and mortality. Since HCT is a substantial stressor, we hypothesized that the prevalence of frailty and pre-frailty will increase in the immediate post-transplant period, and that frailty will be associated with early (1y) post-transplant non-relapse mortality (NRM).

Methods: We conducted a prospective, longitudinal study of 96 patients undergoing HCT at age 40y or older between February 2014 and April 2015, and serially performed a multi-domain geriatric assessment (incorporating function, comorbidity, cognition, psychological state, social activity/support, and nutritional status) in additional to demographic, transplant and disease related variables. Frailty was defined by presence of 3 or more of the following criteria: unintentional weight loss, exhaustion, weakness, slow walking speed and low physical activity; pre-frailty was defined by presence of 1 or 2 of the criteria. Geriatric assessments were performed pre-HCT and at 100 days, 6 months and 1 year post HCT. The prevalence of frailty and pre-frailty was evaluated at each time point in older (60-74y, n=48) vs. younger (40-59y, n=48) recipients. Longitudinal assessments (baseline, 100 days, 6 months, 1 year) of ordinal frailty measures were assessed using random effects and non-linear mixed logistic regression models.  Impact of frailty on 1 year NRM was estimated using cumulative incidence estimates.

Results: Older patients were more likely to undergo reduced intensity conditioning, and less likely to be employed at HCT. The prevalence of frailty increased with time since HCT:  The prevalence of frailty was 8% at baseline and increased to 39% by 6 months (<0.01). The prevalence of pre-frailty was 44% at baseline and remained at 45% by 6 months. The prevalence of frailty and pre-frailty did not differ in the younger vs. older age groups at any time point and was independent of the baseline HCT-comorbidity index (HCT-CI).

In the random effects model for frailty, increasing time from transplant was associated with a 3.7 times higher odds of frailty (OR: 3.7, 95% CI: 1.9-7.2, p< 0.01). Other variables associated with a higher odds of frailty were pre-transplant employment status (retired: OR: 7.3, 95% CI: 1.2-46.2, p= 0.03); patients on medical leave or disabled (OR: 11.2, 95% 1.8-67.7, p=0.01), limitations in social activities (OR: 1.04, 95% CI: 1.01-1.07, p< 0.01), and type of HCT (allogeneic vs. autologous: OR: 3.1, 95% CI: 0.9-10.2, P=0.06). Baseline pre-frailty was associated with a 3.1 times odds of subsequently developing frailty (OR: 3.1, 95% CI: 2.3-45.5, P< 0.01)

Despite only limited follow up to date in this recent cohort, we identified a trend towards increasing non-relapse mortality (NRM) in frail patients. The cumulative incidence of NRM at one year was 7% (95% CI: 0-14%) in the non-frail group, 23% (95% CI: 8-38%) in the pre-frail group and 28% (95% CI: 0-57%) in the frail group (p-value for trend= 0.07).

Conclusion: Independent of specific medical co-morbidities or the HCT-CI, frailty was noted in 8%, and pre-frailty in 44% of our HCT population prior to HCT, and was not dependent on age. Frailty is a transitional state and appears to reflect a dynamic progression from robustness to functional decline with time since HCT. It is associated with a higher mortality, indicating a need to identify vulnerable populations and define their need for specific interventions.