Prognostic Significance of Quality of Life in Patients Undergoing Allogeneic Hematopoietic Cell Transplantation

Patient-reported outcomes and quality of life (QOL) are becoming increasingly recognized as important factors in cancer clinical trials and patient care. Many studies have described the predictive value of QOL in cancer survival beyond the standard clinical measures. Allogeneic hematopoietic cell transplant (HCT) is the only curative option for many hematologic malignancies, however is often complicated by a high incidence of treatment related morbidity and mortality. While the HCT comorbidity index (HCT-CI) has been developed to assess the impact of co-morbidities and risk of death in patients undergoing allogeneic HCT, we hypothesized that QOL prior to transplant may also add to the assessment of patient risk for post-transplant survival. We thus undertook an analysis to determine the prognostic impact of pre-transplant QOL scores, as captured by the FACT-BMT, on post-HCT non-relapse (NRM) and overall mortality.

From 2003-2012, 512 adult patients underwent a first allogeneic transplant at our institution. Prospective psychometric instruments (FACT-BMT) were administered to 409 patients. No significant differences were found between the study cohort and the 103 patients who did not participate in a baseline assessment during the same time period in regards to transplant outcomes or baseline characteristics, except for year of transplant, higher proportion of females, and a shorter time from diagnosis to transplant in the study cohort. Patients with QOL assessments underwent HCT for malignant (n=395) and non-malignant (n=14) diseases. Most patients (n=314, 77%) underwent a myeloablative HCT, while the remaining received a reduced-intensity (RIC) preparative regimen (n=95, 23%). At the time of analysis, median time to follow up was 49 months, range (1.2-128). 41% of patients remained alive, 23% of patients died from relapse and 36% from NRM. We examined the association of overall mortality and NRM with: (1) individual QOL domains captured by the FACT-BMT instrument (physical wellbeing (PWB) functional wellbeing (FWB), social wellbeing (SWB), emotional wellbeing (EWB) and additional concerns (AC), (2) trial outcome index (TOI; PWB+FWB+AC), and (3) total score. In univariable analysis, poor performance status, high risk disease, and high (≥3) HCT-CI score were found to be significant factors for overall mortality but not NRM; umbilical cord, unrelated and mismatched donor transplants were risk factors for both overall mortality and NRM. Additionally, worse FACT-BMT PWB scores (HR 0.97, 95% CI 0.94-0.99, p=0.004), TOI score (HR 0.94, 95% CI 0.88-0.99, p=0.043), and total score (HR 0.95, 95% CI 0.90-0.99, p=0.043) were significant variables associated with overall mortality, but not NRM. In multivariable analyses evaluating individual FACT-BMT QOL domains and adjusted for patient and disease factors, including HCT-CI score, only PWB score was associated with overall mortality, but not NRM (Table). TOI scores which is representative of changes in physical and functional outcomes were associated with overall mortality but not NRM (Table). Total score was not associated with overall or NRM (Table). High (≥3) HCT-CI score remained prognostic for adverse overall mortality but not NRM in all multivariable models.

In summary, QOL assessments, particularly PWB and the TOI, may provide independent prognostic information beyond standard clinical measures. While the magnitude of difference of these measures is small, further study on how QOL may be reflected by disease burden or co-morbidities and how patient-reported outcomes and behaviors may further impact survival is warranted.