Highlight
The MERLIN_001 study demonstrated that a combined clinicopathological and gene expression profile (CP-GEP) test can reliably stratify melanoma patients by sentinel lymph node (SLN) metastasis risk, identifying those with less than 10% likelihood to safely avoid SLNB. The test showed a negative predictive value (NPV) of 92.9% in low-risk cases and successfully distinguished SLN positivity rates across tumor stages and patient demographics, supporting its clinical utility in personalized surgical decision-making.
Study Background
Sentinel lymph node biopsy (SLNB) is a standard staging procedure in patients with cutaneous melanoma to assess regional lymph node involvement, an important prognostic factor influencing treatment and follow-up. Current guidelines recommend SLNB for patients with a predicted risk of sentinel lymph node metastasis greater than 10%, and consideration when the risk is between 5% and 10%. However, SLNB is an invasive procedure with potential morbidity and healthcare resource implications. Better risk stratification tools are needed to identify patients with low risk of nodal metastasis who may safely forgo SLNB, optimizing patient care and healthcare efficiency.
Gene expression profile (GEP) testing, which assesses molecular characteristics of the primary tumor, has emerged as a promising adjunct to conventional clinicopathological parameters. The MERLIN_001 study was designed to evaluate the predictive accuracy of a combined clinicopathological and GEP (CP-GEP) test for SLN metastasis risk in invasive cutaneous melanoma patients deemed candidates for SLNB.
Study Design
The MERLIN_001 was a prospective, blinded, multicenter prognostic study conducted from September 2021 to June 2024 across 9 academic centers with expertise in melanoma surgery. Eligible participants had biopsy-proven invasive cutaneous melanoma classified as T1 to T3 tumors and clinically negative regional lymph nodes, all considered appropriate candidates for SLNB based on existing clinical criteria.
The intervention involved CP-GEP testing performed on formalin-fixed, paraffin-embedded tissue samples from the primary melanoma biopsy. The test classified patients into low-risk or high-risk groups for SLN metastasis. The primary outcome was the negative predictive value (NPV) of the test in low-risk patients, with subgroup analyses based on tumor category (T1-T3), primary site, histologic subtype, mitotic index, age, and clinical stage.
Key Results
The study enrolled 1761 patients with successful CP-GEP testing and SLNB performed, yielding a sentinel node positivity rate of 17.6% (310/1761). The CP-GEP test successfully completed in 97.7% of samples, reflecting robust assay feasibility.
Among the cohort, 651 patients (37.0%) were classified as low risk by CP-GEP. Within this low-risk group, 46 (7.1%) had a positive SLN, corresponding to an NPV of 92.9% (95% CI, 90.7%-94.8%), confirming that the test reliably identifies patients with a low likelihood of nodal metastasis. High-risk patients had a substantially higher SLN-positive rate of 23.8% (264/1110), indicating approximately a threefold increased risk relative to low-risk counterparts.
The proportion of patients deemed low risk decreased with advancing tumor category: 68.2% for T1, 32.9% for T2, and only 2.8% for T3 melanomas, demonstrating the CP-GEP test’s capability to integrate tumor burden into risk stratification. The test’s discriminative ability remained consistent across different primary tumor sites, histologic subtypes, and mitotic counts.
In clinically relevant subpopulations such as stage IB disease (n=1187) and patients aged 65 years or older (n=832), the CP-GEP test similarly stratified patients with roughly 6.5-6.6% SLN positivity in low-risk groups versus approximately 18-20% in high-risk groups, underscoring its broad applicability.
Expert Commentary
The MERLIN_001 study addresses a critical challenge in melanoma management by combining molecular profiling and clinicopathological features to refine patient selection for SLNB. Its prospective, multicentric design with blinded outcome assessment strengthens the validity and generalizability of findings.
The observed high NPV supports the potential of CP-GEP to safely reduce unnecessary SLNB in selected patients, particularly those with lower tumor burden and favorable molecular profiles. Given the morbidity and costs associated with SLNB, integrating CP-GEP into clinical pathways could enhance personalized care and resource utilization.
However, the test did not eliminate SLN positivity in the low-risk group entirely (7.1%), reflecting an inherent trade-off between sensitivity and specificity. Clinicians must weigh the risk of missing nodal metastases against surgical risks and patient preferences. Additional outcome data regarding prognosis post-negative SLNB with CP-GEP stratification, not yet reported, will provide further insight into long-term safety.
Limitations include the applicability primarily to T1-T3 tumors and the need for broader validation across diverse populations. Cost-effectiveness analyses and integration into existing clinical decision frameworks remain future challenges.
Conclusion
The MERLIN_001 study robustly validates a combined clinicopathological and gene expression profile test to predict sentinel lymph node metastasis risk in cutaneous melanoma patients. With a high negative predictive value and consistent performance across tumor features and patient demographics, CP-GEP testing offers a valuable tool to refine SLNB decision-making. In appropriately selected patients, this approach supports a more individualized, less invasive management strategy, facilitating shared decision-making between surgeons and patients while potentially minimizing surgical morbidity.
Ongoing research should focus on long-term outcomes, external validations, and health-economic impacts to fully establish CP-GEP’s role in standard melanoma care.
Funding and Clinical Trials
The MERLIN_001 study was conducted at multiple academic centers with funding sources not explicitly detailed in the reported article. The clinicaltrials.gov registration and detailed funding disclosures are pending publication.
References
Hieken TJ, Egger ME, Angeles CV, et al. Gene Expression Profile-Based Test to Predict Melanoma Sentinel Node Status: The MERLIN_001 Study. JAMA Surg. 2025 Oct 22:e254399. doi: 10.1001/jamasurg.2025.4399. Epub ahead of print. PMID: 41123931; PMCID: PMC12547680.