Melanoma Risk Model Validated in Global Study

Key Takeaways

• A global cohort of more than 15,000 melanoma patients validated the MIA sentinel node risk calculator across diverse populations.

• Predictive accuracy was reported as strong, despite incomplete data.

• Version 2 of the tool offers tighter 95% confidence intervals.

The Melanoma Institute Australia (MIA) sentinel node (SN) metastasis risk calculator has been validated in a large, multinational cohort, reinforcing its applicability across diverse populations and clinical settings.

The MIA SN metastasis risk calculator provides estimates of positivity for individual patients based on 6 standard clinicopathological parameters and the full 6-parameter model has been externally validated previously using US data," the researchers wrote in the study, published in JAMA Dermatology. "However, given its geographically widespread use, further validation is required to ensure its applicability to other populations.

The retrospective multicenter study included 15,731 patients aged 18 years and older who underwent SN biopsy for invasive primary cutaneous melanoma between July 2021 and December 2023. Data were drawn from national registries and cancer centers in Denmark, the United Kingdom, the United States, New Zealand, Sweden, and Brazil. Eligible patients had available data for at least SN status, patient age at diagnosis, Breslow thickness, and melanoma subtype. Data on ulceration, lymphovascular invasion, and mitotic rate were also collected where possible.

The full 6-parameter model showed an area under the curve (AUC) of 73.0% (95% CI, 70.6% to 75.3%) in patients with complete data. Although anythere between one and three optional parameters were missing, the model continued to show high performance (AUCs of 70.1% to 71.5%). Model calibration remained 'excellent,' suggesting strong alignment between predicted and observed outcomes, the authors noted.

“The MIA SN-positivity calculator performed best with all 6 parameters and has been significantly improved (version 2), with the same risk point estimates but much tighter 95% CIs,” the authors wrote. “These results demonstrated that the calculator was robust, precise, and applicable to geographically widespread melanoma populations.”

Sources: Lo Serigne N, et al. JAMA Dermatol. 2025. doi:10.1001/jamadermatol.2025.0318