Clinical outcomes for patients treated with pembrolizumab weight-based dosing in the Enhancing Oncology Model.

27 Background: A multi-state community oncology network participating in CMS’s Enhancing Oncology Model (EOM) implemented a strategy to switch Medicare beneficiaries from fixed to weight-based dosing (WBD) of pembrolizumab as part of its participation in the EOM. Prior studies suggest therapeutic equivalence between fixed and WBD strategies, with WBD offering up to 25% cost savings. This single-arm observational study describes the clinical characteristics and outcomes of patients treated under a WBD protocol during EOM’s first performance period (PP1). Methods: We identified Medicare beneficiaries attributed to a multi-state community oncology network who received >2 doses of pembrolizumab WBD during PP1 (7/1/2023–12/31/2023). Patients were identified via EHR data and matched to CMS claims. Demographics, cancer type, and treatment regimen were derived from structured EHR chart review. Claims data provided information on treatment adherence, ED visits, inpatient admissions, and immunosuppressive medication use as signals for immune-related adverse events. Claims-based outcomes were limited to 6 months due to EOM episode duration; EHR follow-up continued through 5/9/2025. Results: 37 patients received pembrolizumab on a weight-based dose. The cohort was 73% female with a mean age of 73; 84% were White. NSCLC was the most common diagnosis (78%), followed by breast (n = 5), colorectal (n = 2), and prostate cancer (n = 1). Most patients (84%) received palliative-intent therapy. The majority (92%) were ongoing pembrolizumab users transitioned from flat dosing; 3 patients were new starts. Nearly half received monotherapy; half received immunotherapy with chemotherapy. Patients had a mean (SD) of 88 (39) days with episode claims coverage post-WBD. ED visits occurred in 5 patients (15%), and 1 patient (3%) had an inpatient admission post-WBD; none were due to immune-related events. Seven patients (19%) discontinued treatment during follow-up: 3 due to progression, 3 for functional decline, and 1 for a grade 2 immune-related adverse event (hypothyroidism). Four patients (11%) received steroids for immune-related adverse events. Two patients died during the follow-up period. Two were lost to follow-up. Conclusions: Weight-based dosing of pembrolizumab was feasibly implemented in a large, multi-state community oncology network participating in the EOM. Clinical outcomes and safety signals appeared consistent with historical flat-dose experiences, supporting the viability of this cost-saving strategy in value-based oncology care. We will continue to evaluate outcomes with this strategy as more CMS data becomes available.