WEDNESDAY, March 12, 2025 -- In a review and meta-analysis published online March 12 in JAMA Dermatology, several risk prediction models are identified and externally validated for sentinel lymph node biopsy (SLNB) positivity in melanoma.

Bryan Ma, M.D., from the University of Calgary in Alberta, Canada, and colleagues reviewed the characteristics and discriminatory performance of existing risk prediction models for SLNB positivity in melanoma. Twenty-three articles describing development of 21 different risk prediction models were identified, in addition to 20 external validations of eight different risk prediction models; nine models included sufficient information to obtain individualized patient risk estimates in routine preprocedural clinical practice.

The researchers found that the pooled weighted C statistic was 0.78 among all risk prediction models, with significant heterogeneity (I2 = 97.4 percent) that was not explained in meta-regression. The most frequently externally validated models were the Memorial Sloan Kettering Cancer Center and Melanoma Institute of Australia models, both of which had strong and comparable discriminative performance (pooled weighted C-statistics, 0.73 and 0.70, respectively). There was no significant difference seen in discrimination between models that included gene expression profiles and those using only clinicopathologic features (pooled C statistics, 0.83 and 0.77, respectively).

"These findings may guide subsequent comparative studies to ensure that these models have adequate discrimination and calibration for recommendation as part of routine preoperative clinical assessments," the authors write.