Perspectives on the Science
The Evolving Role of ctDNA in Merkel Cell Carcinoma: Challenges and Opportunities
Akaike T, et al. Circulating Tumor DNA Assay Detects Merkel Cell Carcinoma Recurrence, Disease Progression, and Minimal Residual Disease: Surveillance and Prognostic Implications. Journal of Clinical Oncology. 2024;42:3151–3161.1
Introduction
On March 3, 2025, the Society of Cutaneous Oncology (SoCO) Journal Club convened to discuss a recent study published in the Journal of Clinical Oncology: “Circulating Tumor DNA Assay Detects Merkel Cell Carcinoma Recurrence, Disease Progression, and Minimal Residual Disease: Surveillance and Prognostic Implications1.” The session brought together a diverse group of participants, including non-clinician researchers, student trainees, and predominantly clinicians specializing in surgical oncology, medical dermatology, and medical oncology (Figure 1).
Attendees represented a range of institutions, including Massachusetts General Hospital, Mass Eye and Ear Infirmary, University of Washington, Moffitt Cancer Center, Dana-Farber Cancer Institute, Brigham and Women’s Hospital, Northwestern University, and the National Institutes of Health. Their experience managing Merkel Cell Carcinoma (MCC) varied (Figure 2).
This perspective piece summarizes the key takeaways from the Journal Club discussion, highlighting both the study’s findings and their broader implications for MCC surveillance and treatment decisions. The views expressed here reflect those of the authors and do not necessarily represent the positions of SoCO or its affiliated institutions.
Background
Merkel cell carcinoma (MCC) is a rare but potentially aggressive neuroendocrine skin cancer with a high propensity for recurrence and metastasis. Traditional surveillance strategies rely primarily on physical examinations and imaging modalities such as computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI). While effective, these approaches have limitations, including the potential for delayed detection of recurrence, high costs, radiation exposure, and logistical challenges associated with frequent imaging.
In recent years, circulating tumor DNA (ctDNA) has emerged as a promising biomarker in oncology, offering a non-invasive approach for detecting minimal residual disease (MRD), monitoring treatment response, and identifying disease recurrence earlier than conventional imaging2–5. Unlike protein-based serum tumor markers, ctDNA analysis detects tumor-derived genetic alterations in cell-free DNA, providing molecular insights into tumor burden and disease progression.
Early studies have demonstrated that ctDNA levels in MCC correlate with disease burden, treatment response, and overall prognosis6–8. Importantly, ctDNA monitoring is possible regardless of the etiology of MCC, whether virus-associated (Merkel cell polyomavirus, MCPyV-positive) or virus-negative. Despite its promise, challenges remain in integrating ctDNA into routine practice, particularly regarding assay sensitivity, standardized thresholds, and its role in guiding treatment decisions. In the following sections, we explore the findings of Akaike et al. (2024)1 and insights from our Journal Club discussion, including clinician perspectives on the potential benefits, limitations, and future applications of ctDNA in MCC management.
Study Design
Akaike et al. (2024) conducted a prospective, multicenter, observational study evaluating the role of ctDNA in MCC surveillance and prognosis. The discovery cohort (n=167) was based at the University of Washington and Stanford University, while the validation cohort included patients (n=152) from Dana-Farber Cancer Institute, Northwestern University, University of California San Francisco, and Moffitt Cancer Center. The discovery cohort patients were enrolled from April 2020 through August 2022 while the validation cohort patients were enrolled from February 2021 to July 2022. Patients with confirmed MCC provided plasma samples at predefined intervals, including during routine follow-up and at the time of recurrence. ctDNA detection used tumor-informed somatic sequence variants amplified by custom PCR primers. Somatic variants were identified using a next-generation sequencing of tumor DNA. The study compared ctDNA levels with imaging findings, physical examination findings, and tissue biopsies, assessing its sensitivity and specificity in detecting recurrence. Additionally, the authors evaluated ctDNA trends over time to determine their association with recurrence-free survival (RFS). Statistical analyses included Cox proportional hazards models and Kaplan-Meier survival estimates to quantify ctDNA’s prognostic significance.
Main Findings
Akaike et al. (2024) found that post-treatment ctDNA positivity was strongly associated with disease recurrence, often preceding radiographic progression. Among 84 patients with stage I-III MCC who underwent ctDNA testing within 4 months after curative-intent therapy, 14 were ctDNA-positive, and 70 were ctDNA-negative. At one year, recurrence was significantly higher among ctDNA-positive patients (74%) compared to ctDNA-negative patients (21%), with a hazard ratio of 7.6 (95% CI, 3.0–19; P < .001). This association remained significant after adjusting for disease stage, immunosuppression status, sex, and age (adjusted HR 7.4, 95% CI: 2.7–20, P < .001). Notably, the positive predictive value of a ctDNA-positive test for recurrence was 69% (95% CI, 32–91) after one year of follow-up in the discovery cohort and 94% (95% CI, 71–100) in the validation cohort. Further, a negative ctDNA test was highly predictive of recurrence-free survival, with a negative predictive value of 94% (95% CI, 90–97) at 135 days (4.5 months) and 90% (95% CI, 85–94) at 180 days. These findings suggest that early ctDNA positivity after curative-intent therapy is a strong predictor of recurrence, while serial negative ctDNA results are associated with prolonged recurrence-free survival.
Discussion
The integration of ctDNA into clinical practice remains an area of active discussion, particularly regarding its utility in guiding surveillance and treatment decisions. While the study by Akaike et al. demonstrated a strong association between ctDNA detection and disease recurrence, real-world implementation is influenced by clinician familiarity, existing practice patterns, patient preference, logistical constraints, and financial considerations. To better understand current perspectives on ctDNA use in MCC, we surveyed Journal Club participants prior to the session on their experiences, concerns, and expectations for the technology.
Among clinicians who manage MCC, the majority reported prior experience with ctDNA, highlighting a growing interest in its potential applications (Figure 3A). However, opinions varied regarding its influence on clinical decision-making. While some respondents indicated that ctDNA had shaped their approach to surveillance, far fewer reported using it to guide treatment decisions (Figure 3B). Among respondents who have incorporated ctDNA into their practice, surveillance emerged as the predominant application, with 56% indicating that ctDNA had influenced their monitoring strategies. However, treatment decision-making remained less affected, with only 22% stating that ctDNA guided systemic therapy choices. These findings underscore ongoing uncertainty regarding ctDNA’s role beyond surveillance.
To better understand how ctDNA is being used in clinical practice, we asked respondents about their specific applications of ctDNA for MCC surveillance (Figure 4). Among clinicians who manage MCC, 35% reported using ctDNA regularly for surveillance, while 20% indicated selective use in high-risk cases. Notably, 25% of respondents stated they would use ctDNA if it were formally recommended, reflecting an interest in adoption but hesitation in the absence of clear guidelines.
Despite its increasing use, perceptions of ctDNA’s value remain mixed. While most respondents acknowledged its potential, 26% remained uncertain about its clinical utility (Figure 5). The majority rated ctDNA as somewhat (43%) or extremely valuable (30%), with no negative ratings—suggesting broad recognition of its promise, though consensus on its role has yet to emerge.
Practical Considerations for Implementing ctDNA
While interest in ctDNA is rising, several factors limit its broader clinical adoption. In our survey, cost and coverage issues emerged as the most frequently cited concern (37%), followed by the lack of clear clinical guidelines (26%). Additionally, some clinicians expressed uncertainty about ctDNA’s impact on clinically meaningful patient outcomes (Figure 6).
Beyond these broad concerns, two practical challenges further complicate real-world adoption: logistical burdens and uncertainty surrounding cost and insurance coverage.
Logistical Challenges
Integrating ctDNA into clinical workflows adds another layer of complexity to cancer care, where clinicians and care teams already manage a growing array of diagnostic tests—including imaging, routine bloodwork, next-generation tumor gene sequencing, and now, blood-based ctDNA assays. Unlike standard lab tests, ctDNA analysis often requires informed consent and specialized handling for the initial test, including coordination with pathology teams to retrieve and submit tumor-informed reference samples. Furthermore, the lag time to obtain test results for a ctDNA order add to the logistical challenge. This additional workload typically falls on clinicians or clinical staff, often without dedicated administrative support or compensation. The logistical burden of tracking sample collection, shipping, and follow-up with commercial vendors further complicates widespread adoption. However, these challenges will likely diminish as tissue-based molecular send-out tests become more commonplace and standardized within oncology practices. At present, ctDNA is not routinely managed through local laboratories nor is it easily integrated into most electronic medical record platforms.
Cost and Coverage Issues
The financial aspects of ctDNA testing remain opaque, both at the patient and system level. As with many emerging technologies, initial costs are often absorbed by testing companies, particularly in cases where balance billing is not practiced. However, the long-term affordability, and cost-effectiveness, of ctDNA testing remains unclear. Many ctDNA assays are classified as laboratory-developed tests, meaning they lack specific FDA approval and thus fall into a regulatory gray zone regarding reimbursement. This uncertainty may contribute to clinician hesitancy, as patients frequently inquire about financial responsibility, yet clear answers are not always available. Despite these financial and logistical challenges, the perceived clinical value of ctDNA remains central to its adoption. Further, the potential to save healthcare expenditures by omitting specific imaging studies with the implementation of ctDNA assays, has not been evaluated. The study by Akaike et al. provides key insights into its role in MCC surveillance and disease monitoring.
Revisiting the Evidence: Akaike et al. on ctDNA in MCC
As interest in ctDNA grows, its clinical implementation remains a subject of careful evaluation. The study by Akaike et al. offers important insights into ctDNA’s potential role in MCC surveillance and disease monitoring. By leveraging separate discovery and validation cohorts, the study provides strong evidence linking ctDNA detection with disease burden and recurrence risk. However, certain aspects warrant further consideration, particularly in how ctDNA is evaluated against conventional clinical and radiological assessments.
One key factor is the study’s approach to sensitivity and specificity. The analysis compared ctDNA detection against clinical and radiographic assessments as the gold standard for recurrence. If ctDNA detects recurrence before imaging, cases labeled as false positives may actually be early true positives. Likewise, sensitivity could be underestimated if these cases are later confirmed as recurrence on imaging. This raises an important question: should ctDNA be considered a more sensitive diagnostic tool rather than one being measured against conventional modalities?
Additionally, the study reinforced that ctDNA levels correlate with disease burden, supporting its potential role in surveillance. This finding is particularly relevant given that ctDNA trends may offer a dynamic, real-time indicator of tumor progression or treatment response. However, this also introduces uncertainty—how should clinicians act on ctDNA positivity in the absence of radiographic progression?
Incorporation of ctDNA into surveillance
While imaging remains the standard for post-treatment monitoring, ctDNA could offer an earlier signal of recurrence. The National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology now recognize ctDNA as an option for MCC surveillance, noting that circulating tumor DNA can assess disease burden in both virus-positive and virus-negative MCC and typically becomes positive prior to or at the time of a clinically evident recurrence. For surveillance, this test is often obtained every 3 months9. The inclusion of ctDNA as a surveillance option for MCC in the NCCN guidelines highlights the potential role of this technology.
After reviewing the study, clinicians were asked whether they would consider incorporating ctDNA into surveillance in place of imaging. The responses reflected a clear interest in ctDNA’s potential, with 78% of respondents at least somewhat likely to integrate it into their surveillance strategies. Notably, 39% indicated they would favor ctDNA over imaging entirely (Figure 7). Still, 17% remained uncertain, emphasizing the need for continued clinical evaluation of ctDNA’s role in routine practice.
ctDNA Positivity and Treatment Decision-Making
While ctDNA is increasingly accepted for surveillance, its role in guiding treatment decisions remains unclear. A key unresolved question is whether ctDNA positivity alone, without imaging or biopsy confirmation, should warrant systemic therapy initiation. While ctDNA positivity suggests residual disease, initiating systemic therapy without imaging or biopsy confirmation remains highly uncertain, as it lacks the same level of clinical confidence. Although preemptive intervention is conceptually appealing, enthusiasm for acting on ctDNA positivity alone remains low, with no respondents initiating systemic therapy without additional clinical context (Figure 8).
Comfort Level with Initiating ICI in ctDNA-Positive, NED Patients
Currently, adjuvant therapy—including immune checkpoint inhibitors (ICI) and chemotherapy—is not FDA-approved and is not recommended in clinical guidelines for patients with clinically NED (no evidence of disease) MCC following definitive treatment. Despite this, some clinicians have considered whether ctDNA positivity could justify preemptive immunotherapy in high-risk patients.
Following the Journal Club discussion, we reassessed clinician comfort with using ICI for ctDNA-positive, clinically NED patients. While some perspectives evolved, the majority remained hesitant, with 47% stating they would not recommend ICI based on ctDNA alone and 29% remaining uncertain due to insufficient supporting evidence. Only a small subset felt comfortable using ICI in select high-risk cases (Figure 9).
This reluctance largely stems from concerns about overtreatment, as systemic immunotherapy carries risks of immune-related toxicity. Some local-regional recurrences may be effectively managed with surgery or radiation, making systemic therapy unnecessary in select cases. Unlike imaging- or biopsy-confirmed recurrence, where the need for systemic therapy is clearer, ctDNA positivity alone lacks the same level of clinical certainty.
While no patients have been initiated on ICI solely based on a low ctDNA level in the absence of clinical disease, many clinicians use ctDNA to guide therapy adjustments. Specifically, persistent ctDNA detection while on immune therapy may influence decisions against discontinuing treatment. This suggests that ctDNA may have a greater role in treatment de-escalation strategies rather than in initiating therapy, highlighting the need for more structured guidance on its role in long-term management.
What Would Increase Confidence in ICI Initiation?
To understand what might shift clinical practice, we asked clinicians what additional evidence would make them more confident in recommending ICI in this setting. The strongest support for ctDNA-driven intervention came from those emphasizing the need for rigorous prospective validation, with 61% citing clinical trial data demonstrating improved outcomes as a key factor. Additional priorities included further validation of ctDNA assays (22%) and inclusion of testing recommendations in consensus guidelines from expert panels (28%), reflecting the importance of clear standards for clinical use (Figure 10).
While ctDNA provides an early signal of recurrence, its role in guiding treatment decisions remains uncertain. Clinicians remain hesitant to act on ctDNA alone without additional corroborating evidence. There is also growing interest in whether ctDNA could help identify patients who might benefit from adjuvant ICI following loco-regional treatment with surgery and radiation. However, without prospective data, the optimal approach to integrating ctDNA into treatment decisions is still unclear.
Conclusion
The integration of ctDNA into MCC management remains an area of active investigation, with strong enthusiasm for its role in surveillance but persistent uncertainty regarding its use in treatment initiation. While the study by Akaike et al. provides compelling evidence linking ctDNA detection to disease recurrence, logistical, financial, and interpretive challenges limit its real-world application.
Survey responses from the Journal Club discussion reflect a growing willingness to incorporate ctDNA into routine monitoring. Consistent with this evolving interest, the NCCN Clinical Practice Guidelines in Oncology now acknowledge ctDNA as a tool that can assess disease burden in both virus-positive and virus-negative MCC. With a negative predictive value of approximately 94% at 135 days following a negative ctDNA test, this approach may support extending imaging intervals or reducing PET/CT scan frequency in appropriate patients.
Future research, most notably through prospective and, when possible, randomized clinical trials will be essential to better define the clinical value of ctDNA testing. Efforts to standardize assays, establish reimbursement pathways, and develop expert consensus will play a crucial role in shaping its use in clinical practice for MCC, and likely other cancers as well.
Materials and Methods
This Perspectives on the Science piece was published using Quarto®. The original survey figures were created using R (version 4.0.0) and the tidyverse suite of packages, including ggplot2. The reader-facing figures in this updated web presentation reproduce the published aggregate counts and percentages using HTML/CSS; they do not reanalyze individual-level survey data. The image on the “Perspectives on the Science” page was created by the authors (DMM) using the rosemary package. GPT-4, a language model developed by OpenAI, was employed in the drafting and editing of this manuscript. GPT-4 provided assistance in manuscript structuring, and generation of content, ensuring a comprehensive and cohesive presentation of the research and discussion points10. Two separate surveys were conducted using REDCap®. The pre-Journal Club survey was distributed to all Society of Cutaneous Oncology members, with 30 respondents. The post-Journal Club survey was only administered to attendees who remained through the session, yielding 21 respondents. This difference in total responses accounts for variations in figures.
Bibliography
Appendix
Share Your Thoughts
What did this article make you think about?
The Society of Cutaneous Oncology welcomes thoughtful reactions, questions, critiques, and ideas from readers. Share a brief note with the SoCO and Journal of Cutaneous Oncology team.