The melanoma MEGA-study: Integrating proteogenomics, digital pathology, and AI-analytics for precision oncology

Abstract

Melanoma remains the most aggressive form of skin cancer, characterized by high metastatic potential, genetic heterogeneity, and resistance to conventional therapies. The Melanoma MEGA-Study is a multi-center initiative designed to address these clinical challenges by integrating advanced proteogenomic profiling, clinical metadata, with AI-driven digital pathology and machine learning analytics, aiming to enhance personalized treatment strategies and improve patient outcomes. Between 2013 and 2022, a cohort of 1653 melanoma patients each contributed a primary tumor sample, with 361 providing 819 metastatic tumor samples. Clinical data collection for this cohort continued until May 2023. Comprehensive analyses using high-resolution mass spectrometry, optimized workflows for formalin-fixed paraffin-embedded tissues, and advanced digital pathology platforms enabled precise mapping of the tumor microenvironment, identification of metabolic reprogramming, and characterization of immune evasion signatures. The European Cancer Moonshot Lund Center's MEGA-Study, under the academic umbrella of Lund and Szeged universities, marks a significant advancement in its collaborative efforts with the National Institutes of Health (NIH) under the Cancer Moonshot partnership. This initiative exemplifies the center's dedication to pioneering cancer research and underscores the strength of its international collaborations.

Significance

The significance of this study lies in its pioneering integration of high-resolution proteomics, AI-driven digital pathology, and comprehensive clinical annotation to unravel the complex molecular landscape of melanoma. By leveraging a robust, population-based cohort of 1653 patients, including extensive analyses of both primary and metastatic tumor specimens, our approach provides unprecedented insights into the proteogenomic alterations that underpin tumor progression, immune evasion, and therapeutic resistance. The preliminary application of advanced mass spectrometry techniques to formalin-fixed paraffin-embedded tissues, combined with state-of-the-art digital pathology and machine learning, has enabled the identification of novel protein biomarkers and metabolic signatures that hold promise for refining patient stratification and informing personalized treatment strategies. This integrative framework not only deepens our understanding of melanoma biology but also establishes a scalable model for precision oncology that can be extended to other complex malignancies. Ultimately, our findings have the potential to transform clinical practice by facilitating earlier risk stratification, improving prognostication, and guiding the development of targeted therapeutic interventions for this highly aggressive cancer.