ctDNA or Merkel Virus Antibodies for Surveillance of Merkel Cell Carcinoma Recurrence
Menée sur 169 patients atteints d'un carcinome à cellules de Merkel (durée médiane de suivi : 483 jours ; 64 % d'hommes), cette étude multicentrique compare la performance de l'ADN tumoral circulant et d'un test par anticorps anti-oncoprotéine du polyomavirus de Merkel pour détecter une récidive
Importance : Merkel cell carcinoma (MCC) is associated with a high recurrence rate of 40%. It is important to know which of 2 blood-based tumor biomarkers available for use in clinical practice is more accurate in detecting MCC recurrence.
Objective : To compare the performance of circulating tumor DNA (ctDNA) and the MCC polyomavirus (MCPyV) oncoprotein antibody test for detecting MCC recurrence.
Design, Setting, and Participants : In this multicenter cohort study across 3 US academic centers (Stanford, Dana-Farber Brigham, and University of Washington), patients with stage I-IV MCC underwent serial paired ctDNA and antibody testing for disease surveillance between April 2020 and March 2024. ctDNA was considered positive if mean tumor molecules per milliliter of plasma values were greater than 0.00, while a rising antibody test result was defined as a 30% or greater increase in titer. Inclusion criteria were clinically disease-free at baseline, detectable MCPyV antibodies at diagnosis, and both ctDNA and antibody tests performed within 45 days of each other in at least 1 time point per patient. Predictive performance for recurrence was assessed and compared between tests.
Main Outcomes and Measures : The primary outcome was clinical recurrence. Predictive performance measures for each test included positive (PPV), negative predictive value (NPV), and hazard ratios (HRs).
Results : In this study among 169 patients with 703 paired tests (median testing interval, 96 days; 109 males [64%]; 60 females [36%]), there were 36 clinical recurrences from 32 patients over a median follow-up of 483 days (IQR, 274-714 days). Compared with antibody testing, ctDNA had a significantly higher PPV at 365 days (73% [95% CI, 58%-85%] vs 52% [95% CI, 32%-71%]; P = .02), higher NPV at 90 days (99% [95% CI, 98.8%-100%] vs 97% [95% CI, 95%-98%]; P = .001), and a greater HR (HR, 47.9 [95% CI, 16.5-139.0] vs 7.3 [95% CI, 3.8-13.9]; HR ratio, 6.6 [95% CI, 2.5-41.0]; P < .001). Use of both ctDNA and antibody tests together identified only 1 additional recurrence compared with the use of ctDNA testing alone.
Conclusions and Relevance : This study found that ctDNA was more accurate than MCPyV antibody testing in detecting recurrence of MCC. ctDNA appears to be a more reliable guide than the MCPyV antibody for surveillance practices and ongoing risk stratification in most patients with MCC.
JAMA Dermatology , résumé, 2026