TumorScope

The reuse of clinical health data holds immense promise for advancing medical research, yet remains constrained by complex legal, technical, and organisational barriers. This article examines these challenges through the case study of TumorScope, a Belgian interdisciplinary initiative developing a secure, multimodal data environment for glioblastoma research. Drawing on five years of practical experience integrating imaging, genetic, tissue-based, and clinical datasets, the study identifies key legal, ethical, technical, and operational obstacles to effective data access, linkage, and reuse.

Technical issues included fragmented data flows, pseudonymisation complexities, and limited interoperability, while legal and ethical barriers arose from strict interpretations of the General Data Protection Regulation, medical secrecy obligations, and intellectual property constraints. These were compounded by operational challenges such as unclear governance structures, resource limitations, and the limited capacity of Medical Research Ethics Committees to assess data-driven research. The analysis further considers the European Health Data Space Regulation (EHDS) as a potential enabler of responsible secondary data use, while noting uncertainties in its national implementation.

Overall, the study demonstrates that meaningful health data reuse requires more than regulatory compliance, it depends on robust governance frameworks, institutional coordination, and sustained investment in infrastructure and expertise. The findings contribute to ongoing debates in healthcare informatics on how to translate the vision of the EHDS into practical, ethically grounded data reuse for patient benefit.

The mutations driving cancer are being increasingly exposed through tumor-specific genomic data. However, differentiating between cancer-causing driver mutations and random passenger mutations remains challenging. State-of-the-art homology-based predictors contain built-in biases and are often ill-suited to the intricacies of cancer biology. Protein language models have successfully addressed various biological problems but have not yet been tested on the challenging task of cancer driver mutation prediction at a large scale. Additionally, they often fail to offer result interpretation, hindering their effective use in clinical settings.

The AI-based D2Deep method we introduce here addresses these challenges by combining two powerful elements:

a nonspecialized protein language model that captures the makeup of all protein sequences
protein-specific evolutionary information that encompasses functional requirements for a particular protein.

D2Deep relies exclusively on sequence information, outperforms state-of-the-art predictors, and captures intricate epistatic changes throughout the protein caused by mutations. These epistatic changes correlate with known mutations in the clinical setting and can be used for the interpretation of results. The model is trained on a balanced, somatic training set and so effectively mitigates biases related to hotspot mutations compared to state-of-the-art techniques. The versatility of D2Deep is illustrated by its performance on non-cancer mutation prediction, where most variants still lack known consequences. D2Deep predictions and confidence scores are available via https://tumorscope.be/d2deep to help with clinical interpretation and mutation prioritization.