ctDNA May Help to Predict Early Recurrence in Stage III Melanoma

ctDNA levels may help to predict early recurrence for patients with stage III melanoma before adjuvant therapy and during follow-up.

A recent study published in The Lancet Oncology highlighted that using digital droplet polymerase chain reaction (PCR) measurements of circulating tumor DNA (ctDNA) to assess minimal residual disease prior to adjuvant therapy, as well as during follow-up, could help detect patients with stage III melanoma who were at a high risk of recurrence.¹

CancerNetwork®, in an email correspondence, spoke with study author David Polsky, MD, PhD, professor in the Department of Pathology, Alfred W. Kopf, MD, Professor of Dermatologic Oncology and vice chair of research at the Ronald O. Perelman Department of Dermatology at NYU Grossman School of Medicine, and director of Pigmented Lesion Service at NYU Langone Health, about the results highlighted in the publication.

The phase 3 COMBI-AD trial (NCT01682083) assessed dabrafenib (Tafinlar) plus trametinib (Mekinist) vs 2 matched placebos in patients with resected BRAF V600-mutated stage III melanoma. A biomarker analysis was a prespecified exploratory end point, and it is where investigators assessed the droplet digital PCR assay. Of the 870 patients enrolled in the trial, 597 had baseline plasma samples available and ready for the ctDNA analysis.

Data from the follow-up time points at 3, 6, 9, and 12 months after treatment initiation were available for 94 patients. Within a 2-month timeframe, there were 118 samples available before or after clinical progression.

Overall, 13% of baseline samples had detectable ctDNA. Authors noted that for patients with higher disease stages, the ctDNA positivity rates and mutant copies per mL plasma were significantly higher.

Given that the study showed ctDNA levels can predict recurrence in stage III melanoma, how do you envision this test being integrated into the current standard of care for post-surgical surveillance, and what specific changes in management might it prompt?

Polsky: The blood test measures whether there is still melanoma in the patient. Although the test is not able to identify the presence of melanoma in all patients who have it, our data showed that when the test was positive during the follow-up period of the study, nearly 100% of patients recurred. We’re envisioning the test being used to monitor patients over time, perhaps every month or couple of months in the first 1 to 3 years after surgery, for an early indication that their melanoma is recurring. In a patient with a positive ctDNA test, the physician may choose to accelerate the schedule for a follow-up CT scan or get a PET-CT scan, or they may choose to escalate the patient’s treatment regimen based on a combination of the blood test results and scan results.

A press release mentioned that ctDNA levels were more predictive than other experimental tests.² Can you elaborate on the specific experimental tests compared? What were the key differences in sensitivity and specificity?

Polsky: The other experimental tests were performed on the melanoma tumor samples themselves. One test measured the number of mutations in the tumor, and the other looked at the expression of genes related to the body’s immune response to the tumor. Because of the way those tests were conducted, we cannot calculate the sensitivity and specificity of those tests to predict recurrence. We used statistical methods that compared the results of all the experimental tests for how well they predicted recurrence, and we found that the results of the ctDNA assays were better at predicting recurrence than the other tests.

Although the study showed a strong correlation between ctDNA and recurrence, it also noted cases where recurrence occurred despite a negative ctDNA test. What strategies are being explored to improve the test’s sensitivity and minimize false negatives?

Polsky: The test we employed measured the amount of a specific mutation in the DNA from the tumor. The mutation is a driver mutation, meaning it is one of the main causes of the cancer cells growing when ordinary cells wouldn’t. Even though that mutation is expected to be present in all the tumor cells, when the tumor recurrence is relatively small, there likely isn’t much tumor DNA in the circulation. Because the human body has about 5 liters of blood, and we only sampled a tiny fraction (0.2%) of the blood volume, there is a good chance that the mutated DNA molecules didn’t end up in the blood tube. To get around this problem, we and others are exploring other tests that simultaneously measure multiple mutations. The greater the number of mutations that we can measure, the greater the chance that one or more of those mutated DNA molecules will end up in the blood sample we receive in the lab.

For clinicians considering adopting this test, what is the optimal timing and frequency of ctDNA monitoring in patients with stage III melanoma post-surgery? Are there specific patient subgroups where this test would be particularly beneficial or less informative?

Polsky: The exact test that we described is not yet available for patients. When it does become available, we think it would be useful for all patients with stage III [disease]. It would probably be most useful for patients who are unsure about whether to undergo adjuvant therapy. They may wish to watch-and-wait until there is evidence of tumor recurrence before starting treatment. The blood test, in combination with CT scans, may provide useful information, and the blood test could be conducted more often than the scans.

The study included patients from various geographical locations. Were there any significant differences in ctDNA levels or recurrence rates observed across these diverse cohorts? How might this impact the test’s generalizability?

Polsky: We did not find any differences in the ctDNA levels or recurrence rates across the different geographies.

What are the current limitations of ctDNA testing in melanoma, and what future directions in research and development are most promising for overcoming these challenges?

Polsky: Aside from the lack of sensitivity to identify all patients with resected stage III [disease] who have minimal residual disease, a major limitation is that we haven’t yet conducted the necessary clinical studies to determine whether using the test results improves patient outcomes compared with not using the test results. This is called “clinical utility,” and demonstrating it would be a major advance for the treatment of patients with melanoma whose disease has spread beyond the skin.

References

1. Syeda MM, Long GV, Garrett J, et al. Clinical validation of droplet digital PCR assays in detecting BRAF^V600-mutant circulating tumour DNA as a prognostic biomarker in patients with resected stage III melanoma receiving adjuvant therapy (COMBI-AD): a biomarker analysis from a double-blind, randomised phase 3 trial. Lancet Oncol. Published online April 15, 2025. doi:10.1016/S1470-2045(25)00139-1
2. Gene-based blood test for melanoma may catch early signs of cancer’s return. News release. NYU Langone Health. April 16, 2025. Accessed April 21, 2025. https://tinyurl.com/3twjrxyu