Almost 15 years ago, TIME magazine mentioned biobanks as one of the ten transforming ideas that were about to change our world, envisioning a future where the latest medical advancements, cutting-edge diagnostic tools, and life-saving treatments would be accessible to everyone, regardless of their location, income or background. This vision aimed to democratize access to medical knowledge and holds the key to unlocking the full potential of personalized medicine and improving patient outcomes worldwide.
Tissue biobanks have been instrumental in advancing healthcare facilitating the identification and validation of disease indicators and promoting the development of innovative treatment approaches. Population-based biobanks collect data from the general population to identify risk factors, develop prediction models, or identify markers for early detection of diseases. Disease-oriented biobanks collect multi-omics data from patients with cancer or neurodegenerative diseases to generate digital models that predict risk, prognosis, and treatment responses1.
Unlike conventional biobanks in which there are significant resources for samples management, imaging biobanks are fully digital. Their growth is closely linked to high-throughput computing’s ability to extract numerous quantitative features, known as imaging biomarkers, captured using advanced acquisition technologies1. The European Society of Radiology (ESR) defines imaging biobanks as systematic collections of medical images and corresponding imaging biomarkers which are shared among numerous researchers and interconnected to additional biorepositories2.
With the proliferation of high-resolution diagnostic imaging techniques and the rapid adoption of electronic health records, the volume of generated data is staggering, and data sharing among institutions has become more prevalent. These advances present substantial challenges for storing data and annotating images. Many physicians advocate for integrating medical images with health record data. Likewise, an increasing number of biopharma companies pursue forming partnerships with imaging experts, such as Quibim, which can extract valuable, previously unseen data from the vast array of imaging files alongside their clinical information.
The rationale behind this is clear and simple: medical imaging repositories hold a treasure trove of meticulously structured and organized information accumulated through years of clinical studies.
Imaging biomarkers are vital components of imaging biobanks, relying on large-scale data analysis through advanced bioinformatics tools. These biomarkers are critical in oncology and can be categorized by function including examples like predictive, diagnostic, morphologic, staging, and monitoring biomarkers3.
The intersection of medical imaging, biomarkers, and sophisticated analytics is set to enhance biobank capabilities significantly, leading to improved healthcare outcomes. By incorporating AI algorithms and cutting-edge bioinformatics tools, imaging biomarkers can be developed, validated, and implemented in clinical settings. These biomarkers should be objective, reproducible, and effective in disease detection, diagnosis, and treatment response evaluation. The main limitation in incorporating quantitative imaging biomarkers into clinical practice are standardization of technical acquisition, analysis processing, and clinical validation. To address these issues, a structured, step-by-step approach is essential for developing, validating, and implementing these biomarkers. This method ultimately produces a wealth of radiomic data for storage in imaging biobanks, facilitating the integration of imaging biomarkers into clinical practice4.
Successful biobank governance requires maintaining an operational network actively pursuing long-term objectives of knowledge creation, health improvement, and economic value. Cooperation is essential to optimize their value as drivers of translational change, involving scientists from various disciplines and the public, healthcare providers, policymakers, and industry representatives5.
The PRIMAGE project is revolutionizing cancer treatment with a cloud-based platform that supports decision-making in diagnosis, prognosis, and therapy for malignant solid tumors. PRIMAGE fosters collaboration between the scientific community and commercial partners by utilizing novel imaging biomarkers, in-silico tumor growth simulation, advanced visualization, and machine learning. Focusing on pediatric cancers Neuroblastoma and Diffuse Intrinsic Pontine Glioma, the platform validates data infrastructures and research models while streamlining and securing data processes. PRIMAGE represents a significant advancement in cancer diagnostics and treatment, offering a real solution for patients and their families. Both cancers not only represent the complexity of cancer disease but also carry significant societal impact, affecting some of the most vulnerable members of our families.
It is crucial to understand that the majority of cancer patients will not receive treatment at top-tier clinical centers but within local communities. This fact underscores the importance of empowering these communities with the necessary knowledge and resources to provide the best possible care. Democratizing medical knowledge goes beyond the traditional concept of diagnostics, evolving into a more comprehensive understanding of information sharing. By dismantling barriers between institutions and connecting data sources, we can start to grasp the immense complexity and variability of cancer diagnoses.
To genuinely revolutionize patient care, we must learn from the information age. Collaboration is vital to ensure that valuable insights are shared across the medical community. Biobanks play a critical role in this democratization process. They serve as a source of high-quality data and samples, enabling the transformation of raw biomedical evidence into practical applications and personalized patient care. As biobanks continue to evolve and expand, all stakeholders must collaborate to ensure their successful integration into modern biomedical research and development. By emphasizing interoperability, collaboration, and the democratization of information, biobanks can significantly contribute to evidence-based health policy decision-making and implement new initiatives in practice, ultimately improving public health and well-being.
Each patient’s diagnosis is unique, and the only way to understand the nuances is by combining learning from different institutions and regions. This collaborative approach will enable us to develop more effective treatments and strategies for the diverse range of cancer diagnoses that patients face.
Medical knowledge and information are the driving force behind transforming personalized medicine forever.
