Improving treatment selection and reducing adverse events in frail Veterans with multiple myeloma

The majority of U.S. Veterans with multiple myeloma (MM) are frail at diagnosis, and frail patients are at high risk of adverse events, treatment discontinuation, and mortality. Oncology care guided by frailty assessment has been shown to improve treatment outcomes, but limited resources in busy oncology settings prevent implementation. The research team has developed in Veterans with MM measures of frailty that are readily- derived from existing data in VA’s national electronic health record (EHR). They have validated the electronic VA-Frailty Index (VA-FI) against gold-standard clinical frailty assessments, and have shown its ability to not only predict mortality and hospitalizations, but also to evaluate treatment outcomes in Veterans with the highest levels of frailty. The long-term goal of this project is to integrate frailty into research and practice to (1) determine benefits and harms of novel regimens in frail Veterans and (2) to aid oncology teams in identifying and optimizing nononcologic, aging-related conditions that contribute to frailty. This goal addresses the central hypothesis that inferior outcomes in frail Veterans with MM are due to both suboptimal treatment selection and undertreatment of nononcologic health deficits, which in turn are due to the exclusion of frail patients in clinical trials. The interdisciplinary study team consists of experts in MM, data science, and frailty research. They have leveraged comprehensive VA data—including the VA Cancer Registry, national EHR, pharmacy, and laboratory data—to detail the care of 6,281 Veterans treated for MM. Aim 1 will determine the optimal type and intensity of treatment regimen in frail Veterans with newly-diagnosed (NDMM) and relapsed and refractory multiple myeloma (RRMM). All Veterans with NDMM and RRMM treated in VA will be identified, using clinician-supervised annotation to confirm line of therapy, number/type of drugs in each regimen, and other MM-specific variables. Balancing clinical characteristics across regimens, the benefits and harms of novel regimens will be evaluated in Veterans with increasing frailty. Aim 2 will characterize adverse events in frail Veterans with NDMM and RRMM undergoing treatment. Content analysis will be performed on hospitalization discharge summaries, coding adverse events according to NCI’s Common Terminology Criteria for Adverse Events (CTCAE). Adverse events will be evaluated by frailty severity, investigating the degree to which nononcologic factors of frailty contribute to complications while on treatment. Aim 3 will pilot a clinical application that automates frailty assessment using data refreshed nightly from VA’s national EHR. A convergent parallel mixed-methods pilot study will be conducted to determine the feasibility, adoption, and acceptability of the application among oncology clinicians who treat MM. The pilot will start at VA Boston Oncology and then expand to other clinics in New England VA. The findings of Aim 1 will provide critical evidence to inform treatment selection in frail Veterans with NDMM and RRMM. Aim 2 will demonstrate a need to address nononcologic causes of adverse events in frail Veterans. Aim 3 findings will aid oncologists to incorporate frailty assessment into clinical practice to optimize treatment selection and supportive care. The real-time translation of retrospective findings (Aims 1 and 2) to prospective interventions (Aim 3) can be scaled across VA and updated regularly. These contributions will build on prior work to establish a platform for continuous investigation into the benefits and harms of novel MM treatments in frail Veterans cared for in VA. The research is innovative because it leverages feasible and valid electronic measures of frailty to overcome (1) the lack of formal frailty measurement in retrospective analyses of VA data and (2) the time and resource barriers preventing prospective measurement of frailty in busy oncology clinic. The impact of this work will Project Number: 1I01CX002961-01 | Fiscal Year: 2026 | NIH Institute/Center: Veterans Affairs (VA) | Principal Investigator: Clark DuMontier (+1 co-PI) | Institution: VA BOSTON HEALTH CARE SYSTEM, BOSTON, MA | Activity Code: I01 | Study Section: Special Emphasis Panel[ZRD1 ONCE-A (01)] View on NIH RePORTER: https://reporter.nih.gov/project-details/11179606