Background: Merkel cell carcinoma (MCC) is an aggressive skin cancer with a recurrence rate of 40%. Early detection of recurrence can improve outcomes, and effective surveillance is crucial for management of patients with MCC. While Merkel cell polyomavirus (MCPyV) oncoprotein serology is useful in surveillance for MCPyV-positive MCC tumors, patients with MCPyV-negative tumors have no available blood biomarkers and require frequent imaging. This prospective, multicenter study assessed whether circulating tumor DNA (ctDNA) can assess disease burden and detect recurrence regardless of virus status. Methods: A total of 328 blood samples were collected from 125 patients at various time points with a median follow-up of 6 months (range: 0-21 months) between April 2020 to January 2022. Whole-exome sequencing was performed on tumor tissue and matched normal blood to identify a set of somatic, clonal single nucleotide variants, which were tracked in subsequent blood (plasma) samples using a personalized and multiplex PCR-NGS based ctDNA assay (Signatera). Clinically evident disease was defined as MCC noted either by physical exam or by imaging, and molecular evidence of disease was defined as a positive ctDNA test. Surveillance phase began once there was no clinically evident or molecular evidence of disease. Results: Among 125 patients, 47 (38%) had clinically evident MCC and all were found to be ctDNA-positive at the first time point (sensitivity: 100%; 95% CI: 91-100%). Of the 47, 24 were newly diagnosed with MCC and had a median primary tumor size of 2.2 cm (range 0.5-8.5 cm) and a median ctDNA value of 26 mean tumor molecules (MTM)/mL (range: 0.08-1470 MTM/mL). Primary tumor diameter and ctDNA value were strongly correlated (Spearman’s r = 0.81, p < 0.001). Of the 125 patients, 73 (58%) patients were assessed in the surveillance setting and had a total of 152 plasma samples available for longitudinal ctDNA testing. Over this period, 7 ctDNA tests were positive while 145 were negative. After a positive test, 5/7 developed a clinically evident recurrence (4 within 60 days). Of the remaining 2 without clinical recurrence, one had < 60 days of follow-up at time of data analysis. The estimated risk of recurrence, accounting for incomplete follow-up, was 57% within 60 days of a positive ctDNA test (n = 7 tests). In contrast, after a negative ctDNA test (n = 145 tests), the risk of recurrence was 0% within 60 days and 3% between 60-90 days. Conclusions: To our knowledge, this is the largest study to explore ctDNA testing in MCC patients. This study demonstrates that ctDNA testing can detect MCC recurrence early and is a promising clinical surveillance tool regardless of tumor viral status.