(P 207) PATIENT-REPORTED GLOBAL HEALTH PREDICTS ADVERSE HEALTH OUTCOMES IN PATIENTS WITH ADVANCED SARCOMA

Objective: Given the rare and heterogeneous nature of sarcoma cancers and the sometimes-severe treatment-related effects of chemotherapy, many patients with sarcoma face quality of life issues during treatment regardless of histology. Unplanned hospital admissions and emergency room (ER) visits occur from uncontrolled symptoms and ultimately delay treatment regimens and negatively impact quality of life. The PROMIS Global Health v1.2 (PROMIS-10) surveys are validated tools to assess patients’ physical and mental well-being at the time of the survey. However, there are limited studies evaluating serial assessments during treatment, especially in sarcoma. We expanded our cohort of patients with advanced sarcoma to better understand the relationship between PROMIS-10 scores and their temporal relationship with adverse outcomes of interest: unplanned admissions/ED visits and patient mortality.

Methods: Patients seen in the Sarcoma Clinics at The Ohio State University routinely complete the PROMIS-10 survey at each clinic visit. For this expanded cohort, we collected the raw and normalized t-scores for PROMIS-10 surveys for all patients consented to The Ohio State University Sarcoma Registry (OSU-14242) from 6/1/2018 through 12/31/2022. We collected the physical and mental t-scores for each patient at different timepoints during their treatment period and collected data on clinical characteristics (age, gender, histology), unplanned hospital admission/ED visits and mortality for each patient. Survival analyses and generalized linear mixed model were performed for statistical analysis.

Results: A total of 2340 surveys for 446 patients were collected for the study period. Our prior cohort was 110 patients. Most patients were female (54%), and the median age was 58 years. 217 of 446 patients experienced an unplanned admission or ED visit during the study. Patients with a lower PROMIS-10 physical and mental scores were more likely to have an unplanned admission/ED visit (p = 0.000). We demonstrated a stronger correlation compared to our initial cohort, where we initially did not find a correlation with Mental PROMIS-10 scores and unplanned admission/ED visit. Physical and Mental PROMIS10 scores were independently analyzed as they were highly collinear. Physical and mental PROMIS-10 scores continued to correlate with unplanned admissions/ED visits up to 3 months after taking the survey (p = 0.000). A total of 94 patients died during the study period. Patients with a lower PROMIS-10 physical and mental score were more likely to die (p = 0.000). Physical and mental PROMIS-10 scores continued to correlate with mortality up to 1 year after taking the survey (p = 0.000). We had previously seen a signal that female patients had higher mortality rates, but with a larger sample size, there was no correlation between female gender, age or histology with unplanned hospital admission/ED visit and mortality.

Conclusion: We continue to demonstrate that obtaining patient-reported outcomes during treatment at serial timepoints to evaluate patients’ physical and mental well-being can predict adverse outcomes.