Longitudinal PROMIS10 global health scores as independent prognostic values in advanced sarcoma.

Abstract

417 Background: Patients with advanced sarcoma experience significant symptom burden from disease and treatment, contributing to emergency department (ED) visits and unplanned hospitalization. There are no US-based studies evaluating longitudinal health-related quality of life (HRQoL) assessments using patient-reported outcomes (PROs) and their clinical impact during chemotherapy for sarcoma. PROs can be used in risk stratification and earlier intervention to improve clinical outcomes. Our objective was to assess the association between PRO scores and clinical outcomes in patients with advanced sarcoma. Methods: We performed a retrospective cohort registry-based longitudinal analysis of routinely collected PROMIS-10 surveys at The Ohio State University (2018–2024). The PROMIS Global Health v1.2 (PROMIS-10) survey, a validated tool to assess patients’ physical and mental health in patients with cancer, was implemented into the OSU electronic medical record. Serial PROMIS-10 physical and mental t-scores were collected alongside mortality, disease progression (RECIST), and ED visits/unplanned hospitalization. Time-to-event analyses were performed using marginal Cox models with time-varying covariates. Cause-specific hazard models were applied for progression and hospital utilization, with death as a competing event. Sensitivity analyses were performed by including the total number of PROMIS10 tests completed by each patient as an additional confounding variable. Results: Among the 241 patients, higher PROMIS-10 physical scores had a 13% decrease in hazard of death (unadjusted HR 0.87; P < 0.001), 3% decrease in hazard of progression (unadjusted HR 0.97; P = 0.024), and 6% decrease in hazard of ED visit/unplanned hospitalization (unadjusted HR 0.94; P < 0.001). Greater increase in physical PROMIS-10 score was associated with an 11% decrease in hazard of mortality (unadjusted HR 0.89; P < 0.001), 9% decrease in hazard of ED visit/hospitalization (unadjusted HR 0.91; P = 0.031), Greater increase in mental PROMIS-10 score was associated with a 7% decrease in hazard of progression (unadjusted HR 0.93; P = 0.009). In multivariate analyses, histology and chemotherapy were significant risk factors in mortality and hospital utilization. Conclusions: Longitudinal patient-reported global health scores are independently associated with mortality, disease progression, and hospital utilization in advanced sarcoma. Higher global health patient scores during treatment lead to decreases in mortality and hospitalization. These findings support integrating PRO HRQoL assessments into routine cancer care to enhance risk stratification and clinical decision-making. Patients with worsening PRO health scores may warrant early symptom management referrals and closer assessment of hospitalization risk.