This article is more than 5 years old. Information may no longer be current.
Patient-reported outcomes collected prior to HSCT predict clinical outcomes
Click Here to Manage Email Alerts
SALT LAKE CITY — Collecting patient-reported outcomes prior to hematopoietic stem cell transplantation may help predict clinical outcomes for patients after transplant, according to a presentation at the BMT Tandem Meetings.
Patient-reported outcomes provide health information directly from the patient. The information may include results from multi-item measures used to assess different aspects of quality of life, including quality-of-life measures, PROMIS measures and the 36-item short form survey (SF-36). Self-reports on financial hardship or pain diaries, as well as health metrics tracked through wearable devices like heart rate or sleep patterns, also are commonly used.
Research has increasingly shown the value of measuring patient-reported outcomes in disease and treatment settings.
“They are believed to be the most accurate measure of the patient’s experience with disease and treatment,” Bronwen E. Shaw, MD, PhD, professor of medicine at Medical College of Wisconsin and scientific director of CIBMTR, said during her presentation. “They are increasingly being used as primary and secondary outcomes in clinical trials, as biomarkers, and as a method to assess quality of providers or institutions. In many settings, [among] patients with cancer, they have been used as predictors of outcome.”
Despite the value of patient-reported outcomes in cancer care, they are not routinely collected or considered for a pre-HSCT patient-reported outcome score in large databases or even at many individual transplant centers.
According to Shaw’s presentation, pre-HSCT patient-reported outcomes are not collected and used due to the following reasons:
- lack of evidence to support the impact of patient-related outcomes on clinical outcomes;
- absence of consistency in which standardized patient-reported outcomes measure to collect;
- technological, logistic and financial barriers to patient-reported outcome collection; and
- respondent burden.
Positive impact of patient-reported outcomes
Studies published in Cancer and Bone Marrow Transplantation have shown positive evidence to support the impact of patient-reported outcomes on transplant outcomes.
In the secondary analysis of BMT CTN 0902, Wood and colleagues sought to determine whether pre-HSCT patient-reported outcomes predicted survival among patients undergoing autologous (n = 337) or allogeneic (n = 310) transplantation.
The health-related quality-of-life measures included SF-36 physical component and mental component summary scores. Researchers divided patients into four quartiles based on pretransplant SF-36 score. The median physical component score was 44 and median mental component score was 52, according to Shaw.
Among allogeneic HSCT recipients, the pre-HSCT SF-36 physical component summary score independently predicted overall mortality (HR per 10-point decrease = 1.4; P < .001) and performed at least as well as currently used, nonpatient-reported outcome risk indices.
Survival probability estimates at 1 year were 50% for the first quartile of the baseline physical component score, 65% for the second, 75% for the third and 83% for the fourth.
Early post-HSCT decreases in physical component score were associated with higher overall and treatment-related mortality. When adjusted for patient variables included in the U.S. Stem Cell Therapeutic Outcomes Database model for transplant center-specific reporting, the SF-36 physical component score held independent prognostic value.
“In the context of a clinical trial, when using this patient-reported measure, we do add value to the other routine clinical data we collect and objective scores we assign,” Shaw said.
Real-world outcomes
The question remained on whether this remains true outside of a clinical trial setting.
In a real-world, single-center setting, Hamilton and colleagues evaluated 409 patients who underwent allogeneic transplant and routinely completed the Functional Assessment of Cancer Therapy-Bone Marrow Transplant quality-of-life assessment.
Multivariable models showed functional well-being (HR = 0.95; P = .025) and additional concerns (HR 1.3; P = .002) were associated with risk for relapse, and better physical well-being was associated with reduced risk for overall mortality (HR 0.97; P = .04). Trial outcome index also was associated with reduced risk for overall mortality (HR 0.93; P = .05).
“Using these scores in the context of other important predictors sharpens prediction of the outcome,” Shaw said. “We think it is important to attempt to weigh these scores against other predictive factors, but currently we don’t have any algorithms ... which would incorporate this.”
Further work is needed, Shaw added.
“The really important thing is an effort to standardize HSCT patient-related outcome measures across transplantation, whether it is pre- or posttransplant,” Shaw said, adding that progress is being made.
“As a community, it is for us to decide what the most important measures are and have some harmonization,” she added. “It’s important to attempt to develop predictive scores where we can include [patient-reported outcomes].”
Easing the logistic and respondent burdens and challenges are important for the success of patient-related outcomes.
“Most methods used in the studies were pen and paper; that can be time consuming and is logistically expensive with a lot of follow-up needed,” Shaw said.
Electronic record systems are beginning to be used more frequently and give patients the opportunity to answer questions more quickly and easily; automatic mechanisms can remind patients to do reports and follow-up on concerns; and present forms allow patients to see their patient-reported outcomes over time and return these results to transplant center or coordinator. – by Melinda Stevens
Reference:
Shaw BE. The impact of QOL on transplant outcome. Presented at: BMT Tandem Meetings; Feb. 21-25, 2018; Salt Lake City.
Disclosures: Shaw reports no relevant financial disclosures.