High-Risk GEP Results Can Trigger SLNB and Imaging in Early Melanoma

About a quarter of US melanoma deaths occur in early-stage disease, and that statistic drives Dr. Singh's use of gene expression profiling (GEP, a molecular assay that classifies tumors by metastatic risk) in stage I and II melanoma. In an Expert Perspectives episode titled "Using GEP Results to Guide SLNB and Imaging in Stage I-II Melanoma," he detailed where the test genuinely alters management — and, implicitly, where clinicians must still weigh its limits.

Dr. Singh said he finds GEP testing most helpful in patients with stage I and stage II melanoma. Many stage I patients are not eligible for sentinel lymph node biopsy (SLNB, the surgical staging procedure that removes and examines the first draining lymph node). Others decline the procedure, often out of fear of surgery. When a stage I patient returns a high-risk GEP result, he will likely offer an SLNB that otherwise would not be performed.

Why offer SLNB after a molecular result?

He acknowledged this approach may surprise some clinicians. His justification is epidemiological, not cosmetic: roughly 25% of melanoma deaths occur in early-stage disease. A high-risk molecular signature identifies patients within that seemingly favorable group whose tumors carry behavior normally associated with thicker, more advanced lesions.

Stage II patients with a negative node biopsy concern him even more. This group, he noted, accounts for about two-thirds of US melanoma deaths — a figure that dwarfs the mortality contribution of most other melanoma cohorts. When these patients have a Class 2A or Class 2B GEP result, he will likely pursue additional imaging.

What does imaging add after a negative biopsy?

Imaging often detects disease before it becomes clinically evident, and patients whose spread is caught earlier tend to fare significantly better. The strength of the signal matters here: Dr. Singh cited data showing that about 25% of Class 2 patients eventually develop metastasis. In his view, that figure is large enough to warrant closer follow-up than a negative SLNB alone would typically command.

A caveat follows the promise. A negative node biopsy plus a high-risk molecular score creates an ambiguous clinical picture, and earlier imaging does not guarantee earlier cure. The source does not specify imaging modality, frequency, or what findings would trigger intervention — gaps that matter for anyone attempting to replicate this strategy in practice.

What did the 2025 prospective data show?

Dr. Singh discussed prospective evidence on the integrated 31-GEP algorithm, a 31-gene expression profile that reports a continuous risk score. He described a 2025 paper by Dr. Guenther and colleagues showing the 31-GEP score was the most accurate predictor of nodal melanoma.

In T1 and T2 tumors (lesions up to 2.0 mm in Breslow thickness, the standard histologic measure of tumor invasion depth), the score was independently the strongest predictor of metastasis. It correlated more closely with outcomes than Breslow depth or ulceration — the two anchors of conventional AJCC (American Joint Committee on Cancer) staging. That is a notable claim, and clinicians should note what it does not establish: the cited data address predictive accuracy, not yet survival benefit from acting on the score.

For thin melanomas, Dr. Singh called the i31-GEP test critical because it uncovers what he described as bad actors hiding within a seemingly favorable cohort. His metaphor was blunt: these tumors are "a wolf in sheep's clothing." Finding them, he said, can significantly improve survival. That survival claim rests on the logic of earlier detection rather than on a randomized trial cited in the discussion.

Where does the test add little?

Dr. Singh will address the test's limitations in a forthcoming segment, "Interpreting GEP Results in Melanoma: Limitations and Patient-Centered Care." There he plans to cover when GEP testing adds little value and how to keep results in proportion to each patient's goals and overall health. The framing concedes a real boundary: not every high-risk score justifies surgery or imaging, particularly in patients whose comorbidities or preferences argue against aggressive surveillance.

The practical takeaway is selective, not universal. GEP results appear to carry the most weight at two decision points — offering SLNB to ineligible or reluctant stage I patients, and intensifying imaging for node-negative stage II patients with Class 2 signatures. Whether prospective outcome data will validate survival gains from those management changes remains the question his next installment, and the field, must answer.