Real-world evidence is increasingly used to support treatment effectiveness in oncology research, but endpoint validation in early stage melanoma is limited. This study evaluated the concordance between real-world recurrence-free survival (rwRFS) and distant metastasis-free survival (rwDMFS) with KEYNOTE-716 trial control arm estimates. This was a retrospective observational cohort study using electronic health record data from US community oncology practices within The US Oncology Network. Eligible patients were diagnosed with stage IIB-IIC cutaneous melanoma between 1 January 2018 and 5 December 2023 and had not initiated adjuvant therapy 12 weeks postresection with clear margins. Trial eligibility criteria were applied to define a real-world cohort aligned with that of KEYNOTE-716. Available baseline characteristics were balanced between the two populations using a matching-adjusted indirect comparison. Hazard ratios, confidence intervals (CIs), and P-values were calculated for rwRFS and rwDMFS with the KEYNOTE-716 control arm as the reference. After applying eligibility criteria, 75 patients were included in the real-world cohort. Adjusted hazard ratios for rwRFS and rwDMFS were: 1.17 (95% CI: 0.71-1.94; P = 0.527) and 0.83 (95% CI: 0.51-1.37; P = 0.462), respectively. Sensitivity analyses that censored real-world patients at the time of immunotherapy treatment gave adjusted hazard ratios of 0.98 (95% CI: 0.59-1.64; P = 0.938) for rwRFS, and 0.70 (95% CI: 0.41-1.17; P = 0.171) for rwDMFS. rwRFS and rwDMFS were concordant with estimates from the KEYNOTE-716 control arm. Results of real-world endpoints can reliably reflect clinical trial outcomes when key study design elements are aligned. These findings increase confidence in the validity of real-world outcomes for early stage melanoma.
Keywords: melanoma; observational data; real-world data; survival analysis; target trial; trial.
Copyright © 2026 Wolters Kluwer Health, Inc. All rights reserved.