2025-07-04 12:34:19
Comparing Two Gene Expression Profile Tests to Standard of Care for Identifying Patients with Cutaneous Melanoma at Low Risk of Sentinel Lymph Node Positivity
Prieto et al1 compared the performance of two gene expression profile (GEP) tests, clinicopathological (CP)-GEP and 31-GEP, against standard American Joint Committee on Cancer (AJCC) staging in identifying patients with cutaneous melanoma who have a low risk of sentinel lymph node (SLN) positivity, with the aim of guiding decision-making regarding SLN biopsy (SLNB). SLNB is an important prognostic procedure for patients with cutaneous melanoma, but most patients who undergo SLNB have negative results. This discrepancy highlights a need for improved risk stratification tools to avoid unnecessary, invasive procedures. According to the National Comprehensive Cancer Network (NCCN) Guidelines for Cutaneous Melanoma, SLNB is not recommended for patients whose risk of SLN positivity is less than five percent, considered for those with a 5 to 10 percent risk, and recommended when the risk exceeds 10 percent. For patients with T1 to T2 tumors, the rate of SLN positivity is particularly low, and thus there is a specific clinical need for more accurate predictive tools in this population.
GEP tests analyze the tumor expression levels of selected genes to assess the molecular risk of adverse outcomes, such as SLN metastasis. The 31-GEP test has previously been validated for predicting recurrence, metastasis, and melanoma-specific mortality. The 31-GEP test has also been integrated with clinical and pathological factors to develop the i31-SLNB algorithm for predicting SLN positivity, using a neural network model based on ulceration, Breslow thickness, mitotic rate, and age. The CP-GEP test combines CP factors, including Breslow thickness, ulceration, mitotic rate, tumor location, lymphovascular invasion, and age, with the expression of eight genes to categorize patients as high or low risk for SLN positivity. Unlike the 31-GEP, the CP-GEP was not developed for recurrence risk prediction.
This analysis included five validation cohorts including a total of 2,785 patients for the CPGEP and four cohorts including a total of 2,078 patients for the 31-GEP/i31-SLNB. These studies focusing specifically on patients with T1 to T2 tumors, where SLNB decision-making is most challenging. True negative to false negative (TN:FN) ratios and false-negative rates were calculated for each study. A test was considered superior to AJCC staging if it achieved a TN:FN ratio greater than 19:1 or a false-negative rate below five percent, with statistical comparisons made using Chi-squared tests and a significance threshold of p-value less than 0.05.
For the CP-GEP, TN:FN ratios ranged from 12:1 to 27:1, indicating false-negative rates from 3.5 to 7.9 percent. The weighted average performance of CP-GEP across studies was a TN:FN ratio of 15:1 and a false-negative rate of 6.2 percent. This result is inferior to the 19:1 TN:FN ratio and five-percent false-negative rate standard set by AJCC staging. For the 31-GEP/i31-SLNB, TN:FN ratios ranged from 25:1 to 58:0, and false-negative rates ranged from 0 to 3.9 percent. The weighted average for the 31-GEP/i31-SLNB was a TN:FN ratio of 34:1 and a false-negative rate of 2.8 percent, indicating superior performance compared to AJCC staging. Chi-squared analysis revealed that the false-negative rate of 31-GEP/i31-SLNB was significantly lower than that of CP-GEP (p=0.012).
CP-GEP did not improve upon AJCC staging predictions for identifying patients with T1 to T2 melanoma who have a low risk of SLN positivity. By contrast, the 31-GEP and i31-SLNB tests demonstrated enhanced predictive accuracy. These tests were able to more accurately predict a patient’s risk of SLN positivity compared to AJCC staging alone.
Several tools based on various CP factors have been developed to enhance SLN prediction, but none have been shown to outperform AJCC staging. In particular, the CPGEP has not demonstrated efficacy in SLN risk stratification. Recent research demonstrated that a nomogram using CP factors alone provided better risk stratification than CPGEP;2 additionally, one study showed a high recurrence rate for patients with low-risk CPGEP results who had positive SLNs, indicating a potential risk of misclassification if CP-GEP were used in clinical practice.3
In contrast, one follow-up study found that the 31-GEP and i31-SLNB showed no recurrences among patients classified as low risk and who avoided SLNB, with a median follow-up of two years. Multiple prospective studies have demonstrated that the 31-GEP provides accurate prognostic stratification for recurrence, metastasis, and melanoma-specific death. In addition, two studies reported that patients who received 31-GEP testing had higher survival rates than those who did not, suggesting a potential clinical benefit of integrating this test into management strategies.
This study by Prieto et al adds to the growing body of evidence that supports the 31-GEP and i31-SLNB as valuable tools for guiding SLNB decisions and improving patient outcomes.
REFERENCES
1. Prieto PA, Ferris LK, Guenther MJ. Comparing two gene expression profile tests to standard of care for identifying patients with cutaneous melanoma at low risk of sentinel lymph node positivity. Cancer Diagn Progn. 2025;5(3):261–267.
2. Mulder EEAP, Dwarkasing JT, Tempel D, et al. Validation of a clinicopathological and gene expression profile model for sentinel lymph node metastasis in primary cutaneous melanoma. Br J Dermatol. 2021;184(5):944–951.
3. Eggermont AMM, Bellomo D, Arias-Mejias SM, et al. Identification of stage I/IIA melanoma patients at high risk for disease relapse using a clinicopathologic and gene expression model. Eur J Cancer. 2020;140:11–18.
4. Guenther JM, Ward A, Martin B, et al. Patients who forego sentinel lymph node biopsy after 31-GEP testing are not harmed: a prospective, multicenter analysis. EJC Skin Cancer. 2024;2:100175.
©Matrix Medical Communications. View All Articles.