Baseline Patient-Reported Outcomes and Their Association with Hematopoietic Stem Cell Transplant Outcomes

Background: Hematopoietic stem cell transplantation (HCT) is critical for the management of many hematologic malignancies but often results in serious toxicities and impaired quality of life. The value of obtaining patient-reported outcomes (PROs) is increasingly recognized in HCT. Yet, data are lacking regarding the association of baseline PROs with important transplant outcomes.

Methods: We conducted a secondary data analysis of 250 patients who were hospitalized for autologous or allogeneic stem cell transplant at Massachusetts General Hospital from 4/2011-2/2016, and were enrolled in two supportive care trials. We assessed quality of life (QOL) (using the Functional Assessment of Cancer Therapy-General), mood (using the Hospital Anxiety and Depression Scale), and fatigue (using the FACT-Fatigue) at baseline. We abstracted transplant outcomes from the Electronic Health Record (EHR): 1) days alive and out of the hospital in the first 100 days after HCT; 2) hospitalization during the first 6 months after HCT; and 3) acute and chronic GVHD among allogeneic HCT recipients. Using multivariable models adjusting for covariates (age, sex, race, marital status, education level, type of transplant, and receipt of supportive care intervention), we assessed the association of baseline PROs with transplant outcomes.

Results: Patients had a mean age of 56.3 years (range: 18-76) and the majority were female (128/250, 51.2%), white (220/250, 88.0%), and married (178/250, 71.2%). The most common diagnosis was acute myeloid leukemia (AML) (63/250, 25.2%). 44.4% (111/250) of patients underwent an autologous HCT, 25.2% (63/250) underwent myeloablative allogeneic HCT, and 30.4% (76/250) underwent a reduced intensity allogeneic HCT. The median length of stay for HCT was 21 days (range: 6-102), and 30.8% (77/250) of patients had a hospital admission during the first 6 months after HCT. The median days alive and out of the hospital in the first 100 days after HCT was 84 (0-99). Among allogeneic HCT recipients (n=139), 43.2% (60/139) of patients had acute GVHD after HCT, and 56.3% (76/135) had chronic GVHD at 1 year after HCT. In multivariable Poisson regression, higher baseline quality of life (β=0.002, p=0.033) and lower fatigue scores (β=0.004, p=0.006) were associated with days alive and out of the hospital in the first 100 days after HCT. In multivariable logistic regression, lower baseline QOL (OR=0.97, p=0.006), higher depression (OR=1.11, p=0.014), and anxiety (OR=1.15, p=0.001) scores were associated with likelihood of hospitalization within 6 months after HCT. In multivariable logistic regression, lower baseline QOL (OR=0.97, p=0.021), higher depression (OR=1.15, p=0.015), and lower fatigue (OR=0.94, p=0.002) were associated with likelihood of acute GVHD.

Conclusions: Baseline PROs are associated with important transplant outcomes, including days alive and out of the hospital in the first 100 days after HCT, hospitalization during the first 6 months after HCT, and acute GVHD. These findings underscore the ability of pre-transplant PROs to provide critical prognostic information for HCT outcomes and the need for interventions to optimize PROs in this population.