Evaluation and enhancement of the prognostic ability of the eighth edition of TNM staging in cutaneous malignant melanoma: a population-based study of 111,871 cases using machine learning

Mortagy, Mohamed, Cliff-Patel, Nikita, Askary, Regina, Bologan, Ana-Maria, Burns, Dan, Ramage, John and Akhras, Victoria (2026) Evaluation and enhancement of the prognostic ability of the eighth edition of TNM staging in cutaneous malignant melanoma: a population-based study of 111,871 cases using machine learning. British Journal of Dermatology. (doi:10.1093/bjd/ljag018).

Abstract

Background: melanoma is the fifth most common cancer in the United States (US) and the United Kingdom (UK), with global incidence on the rise. The TNM staging system guides treatment decisions and predicts patients’ outcomes.

Objectives: this study aims to evaluate the prognostic ability of the 8th edition TNM (TNM-8) staging system to predict overall and melanoma-specific survival (OS and MSS) in cutaneous malignant melanoma (CMM), and to explore the potential of machine learning (ML) methods to enhance melanoma prognostication.

Methods: adult patients diagnosed with CMM from 2018 to 2022 were extracted from the Surveillance, Epidemiology, and End Results (SEER) database. TNM-8 was evaluated using Kaplan-Meier (KM) plots and OS and MSS probabilities. Fine Gray (FG) competing-risk and accelerated failure time (AFT) models evaluated the factors affecting MSS and OS. ML models were used to evaluate the predictive ability of TNM stages for OS and MSS, compared with that of multiple clinically important prognostic variables.

Results: 111,871 patients with CMM were included in the analysis, with OS and MSS as survival endpoints. Most T, N, M, and TNM-8 stage categories had distinct MSS and OS probabilities, except in the N1 and N2 stages at 12 months and in TNM-8 stages 2 and 3 at 12, 36, 48, and 59 months. FG and AFT models showed that age, sex, race/origin, tumour site, histologic type, ulceration, Breslow thickness, mitotic rate, number of positive lymph nodes, and M stage were important prognostic variables for MSS and OS. ML models showed that TNM had higher predictive prognostic ability for MSS than OS and that including clinically important prognostic variables in addition to TNM stage has higher discriminative prognostic ability for OS and MSS (testing C-indices range: 0.84–0.85 and 0.89–0.92, respectively) than using TNM staging alone (testing C-indices range: 0.67–0.72 and 0.82–0.87, respectively).

Conclusion: this proof-of-concept study demonstrated that although TNM staging retains prognostic value, adding important prognostic variables and using ML could improve prognostication for melanoma patients. ML could be used to develop interactive clinical decision-support tools to improve the prognostication of melanoma patients.

This record has no associated files available for download.

Contributors

Author: Dan Burns

Download statistics