A Step Toward Personalized Surgical Decision Making: Machine Learning Predicts 1 Versus Numerous Melanoma Lymph Node Metastases Using RNA-sequencing

Abstract

Objective: Develop a predictive model to identify patients with 1 pathologic lymph node (pLN) versus >1 pLN using machine learning applied to gene expression profiles and clinical data as input variables.

Background: Standard management for clinically detected melanoma lymph node metastases is complete therapeutic LN dissection (TLND). However, >40% of patients with a clinically detected melanoma lymph node will only have 1 pLN on final review. Recent data suggest that targeted excision of just the single enlarged LN may provide excellent regional control, with less morbidity than TLND. The selection of patients for less morbid surgery requires accurate identification of those with only 1 pLN.

Methods: The Cancer Genome Atlas database was used to identify patients who underwent TLND for melanoma. Pathology reports in The Cancer Genome Atlas were reviewed to identify the number of pLNs. Patients were included for machine learning analyses if RNA sequencing data were available from a pLN. After feature selection, the top 20 gene expression and clinical input features were used to train a ridge logistic regression model to predict patients with 1 pLN versus >1 pLN using 10-fold cross-validation on 80% of samples. The model was then tested on the remaining holdout samples.

Results: A total of 153 patients met inclusion criteria: 64 with one pLN (42%) and 89 with >1 pLNs (58%). Feature selection identified 1 clinical (extranodal extension) and 19 gene expression variables used to predict patients with 1 pLN versus >1 pLN. The ridge logistic regression model identified patient groups with an accuracy of 90% and an area under the receiver operating characteristic curve of 0.97.

Conclusions: Gene expression profiles together with clinical variables can distinguish melanoma metastasis patients with 1 pLN versus >1 pLN. Future models trained using positron emission tomography/computed tomography imaging, gene expression, and relevant clinical variables may further improve accuracy and may predict patients who can be managed with a targeted LN excision rather than a complete TLND.