Physiologic Frailty Among Hematopoietic Cell Transplantation (HCT) Survivors Suggests Accelerated Aging and Is a Predictor for Premature Mortality: A Report from the Bone Marrow Transplant Survivor Study (BMTSS)

Background: Frailty is common in older adults in the general population; ~10% of >65y-olds are frail. Frail individuals are vulnerable to adverse health outcomes and early mortality. The high intensity therapeutic exposures and post-HCT complications can injure normal tissues and could possibly increase risk of frailty even among non-geriatric HCT recipients.

Patients and Methods: We evaluated the risk of frailty among 18-64y-old HCT survivors (n=988) and siblings (n=297). Mean age at study was 42.5±11.6y (survivors) and 43.8±10.9y (siblings). HCT survivors had been transplanted at City of Hope or University of MN between 1974 and 1998; Stem cell source: allogeneic HCT (n=562, 56.3%); autologous (n=436, 43.7%); primary diagnoses: AML/ALL/CML (57.5%), NHL/HL (28.3%), and other (14.2%); Total body irradiation (TBI): 39.5%; chronic graft vs. host disease (cGvHD): 54% of allogeneic HCT recipients.  HCT survivors and siblings completed a frailty survey (Pialoux T et al. Geriatr Gerontol Int. 2012;12:189-97) after a median of 7.2y (range, 2-28) from HCT. Survey addressed the following unique domains: low lean muscle mass, exhaustion, low energy expenditure, slow walking speed, and muscle weakness (frailty phenotype: >=3 conditions).  Logistic regression was used to identify predictors of frailty. The “survivor-sibling” model included age at study, sex, race, SES, and CTCAE grade 3-4 chronic health conditions. The “within HCT survivor” comparison model included variables listed above, and cancer diagnosis, stem cell source, time since HCT, relapse risk at HCT, year of HCT, TBI, and cGvHD.

Results: Survivor-sibling comparison: The prevalence of frailty in HCT survivors and siblings was 8.4% and 0.7%, respectively, p<0.001; HCT survivors were 10.8-fold more likely to be frail when compared with siblings (p=0.001, Table).  Within HCT Survivor Comparisons:  Compared with autologous HCT recipients, related donor HCT recipients were 2.7-fold more likely (p<0.0001) and unrelated donor HCT recipients were 6.5-fold more likely (p<0.0001) to report frailty. Autologous HCT: exposure to TBI (OR=5.0, p=0.03) was associated with an increased risk of frailty (Table). Allogeneic HCT: exposure to TBI (OR=6.7, p=0.01), active cGvHD (OR=9.1, p<0.001), and grade 3-4 chronic health conditions (OR=2.4, p=0.01) were associated with an increased risk of frailty. Subsequent Mortality: After a median of 10.3y (0.04-12.9) of follow-up from survey completion, 182 (18%) participants had died;  10y cumulative incidence of all-cause mortality (from survey completion) was 41% and 16% for patients with and without frailty (p<0.0001, Figure).  Proportion of patients with relapse-related and non-relapse-related mortality did not differ by the frailty phenotype. Autologous HCT recipients: Cox regression analysis adjusted for existing chronic health conditions, year of HCT, TBI, sex and age at HCT revealed that frailty at survey completion was associated with a 5.4-fold increased risk of subsequent death (p<0.0001). Allogeneic HCT recipients: Adjusting for above variables and active cGvHD, frailty at survey completion was associated with a 2.1-fold increased risk of subsequent death (p=0.007). 

Conclusions:  Young HCT survivors are 10.8 times as likely to be frail as compared to their siblings. Prevalence of frailty approaches that seen in the elderly general population, suggesting evidence of premature aging among HCT recipients. Frail HCT survivors are at increased risk of subsequent mortality when compared with non-frail survivors.  The study identifies vulnerable populations among HCT survivors needing close monitoring to anticipate and manage morbidity and prevent mortality.

Table

	Survivors vs. Siblings		Autologous HCT recipients		Allogeneic HCT recipients
	OR (95% CI)	P-value	OR (95% CI)	P-value	OR (95% CI)	P-value
Siblings	1.0	0.001
Survivors	10.8 (2.6-44.2)
No Grade3-4 conditions	1.0		NA		1.0
Yes Grade 3-4 conditions	2.1 (1.3-3.5)	0.004	NA		2.4 (1.2-4.6)	0.01
Chemo-based conditioning			1.0		1.0
TBI-based conditioning			5.0 (1.2-21.2)	0.03	6.7 (1.5-30.5)	0.01
No cGvHD					1.0
Active cGvHD					9.1 (4.3-19.0)	<0.001
Resolved cGvHD					2.0 (0.9-4.9)	0.11