Prognostic Importance of Functional Capacity for Pre-Transplant Risk Assessment Among Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Prognostic Importance of Functional Capacity for Pre-Transplant Risk Assessment Among Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Lee W. Jones,¹ Sean M. Devlin,¹ Molly A. Malroy,¹ Noel Espiritu,¹ Jennifer Aquino,¹ Ceceile Hall,¹ William A. Wood,² Meghan G. Michalski,¹ Boglarka Gyurkocza,¹ Sergio Giralt,¹ Ann A. Jakubowski¹

¹Memorial Sloan-Kettering Cancer Center, New York, NY, USA

²Lineberger Comprehensive Cancer Center, University of North Carolina, Chapel Hill, NC, USA

Background: Identification of patients at high risk of hematopoietic cell transplantation (HCT)-related complications is of major clinical importance. Exercise-based assessments provide a quantitative measure of physical functioning not captured by traditional resting parameters. We investigated whether functional capacity predicted mortality after HCT.

Patients and Methods: Using a prospective design, 409 patients completed a six-minute walk distance test (6MWD) prior to undergoing HCT. 6MWDs were obtained within 5 days of transplant admission; carbon monoxide diffusing capacity (D_LCO) and left ventricular ejection fraction (LVEF) were determined within 1 month of admission, Karnofsky performance status  (KPS) was reported immediately prior to admission and hemoglobin (Hgb) was obtained on the day of transplant. Cox proportional hazards regression were used to estimate the risk of mortality due to relapse or progression of disease (RM), non-relapse mortality (NRM), and overall survival (OS) according to 6MWD (<400 m versus ≥400 m) with adjustment for cardiovascular risk factors.

Results: Participant characteristics are presented in the Table. Mean 6MWD was 404 ± 113 m (range: 71 to 737 m). Median follow-up was 36 months (range 3 to 61 months). During this period, 43% patients died (n=104, NRM). In univariate analyses, 6MWD predicted NRM (p=0.025) and OS (p=0.028; Figure), but not RM (p=0.468). Compared with patients achieving a 6MWD <400 m, the unadjusted hazard ratio (HR) for NRM and OS was 0.64 (95% CI, 0.44 – 0.95) and 0.72 (95% CI, 0.53 – 0.96), respectively for those achieving ≥400 m. In multivariable analyses, the relationship between 6MWD and NRM and OS became non-significant after adjustment for KPS, D_LCO, and Hgb concentration. LVEF did not predict any post-HCT outcomes (p's>0.05).

Conclusions: Although 6MWD was not an independent predictor of post-HCT outcomes, the strong univariate association with NRM and OS suggests further investigation of alternative measures of physical functioning (e.g., exercise capacity) are warranted.