Abstract
The primary challenge in oncology research is the "valley of death" between the discovery of potential biomarkers and their successful implementation in clinical practice. The 1st International NTNU Symposium, held in Trondheim, Norway, brought together global experts to address this gap. This report synthesizes key developments in diagnostic, predictive, and prognostic biomarkers across diverse biofluids and tissues, emphasizing the transition from single-molecule discovery to integrated clinical tools.
The Clinical Need for Validated Biomarkers
Despite the identification of thousands of candidate molecules over decades, very few have transitioned into routine care. There is an urgent need for biomarkers that can detect cancer in its pre-symptomatic stages, improve patient stratification beyond traditional staging, and accurately predict responses to targeted therapies and immunotherapies, such as PD-1/PD-L1 inhibitors.
Multi-Modal Innovation and Liquid Biopsies
Modern biomarker research is shifting toward integrated, multi-modal approaches. Rather than relying on single markers, researchers are combining DNA methylation, mRNA, miRNA, and metabolic data to create robust "biosignatures." These signatures are designed to address unmet needs across the entire clinical pathway, from screening at-risk individuals to guiding adjuvant therapy.
A significant advancement discussed was the use of "liquid biopsies." High-sensitivity sequencing of circulating tumor DNA (ctDNA) allows for non-invasive monitoring of tumor burden and the early identification of resistance mutations. This longitudinal approach enables clinicians to adapt treatments in real-time as the molecular profile of the tumor evolves, identifying actionable resistance mechanisms as they emerge.
Leveraging Longitudinal Datasets
The symposium highlighted the power of large-scale, longitudinal cohorts. Data from the HUNT Biobank and the NOWAC Post-genome cohort have revealed that certain biomarkers, such as IL-8 and specific miRNA signatures, are elevated in the blood up to five years before a clinical diagnosis. This "pre-diagnostic window" offers a critical opportunity for early intervention before symptoms appear.
Key Clinical Insights and Results
- Context-Dependent Prognosis: The prognostic value of markers like PD-L1 is highly tissue-specific; it indicates a better prognosis in melanoma and lung cancer but a poorer one in gastric and kidney cancers.
- Technical Standardization: A "quality control crisis" exists in immunohistochemistry (IHC), with 20-30% of assays deemed insufficient due to poor calibration, necessitating a shift toward quantitative digital pathology.
- Predictive Accuracy: A 6-microRNA signature (miR-Score) demonstrated a 92.3% accuracy in predicting good prognosis in mesothelioma patients.
- Alternative Biofluids: Pleural effusions and urine are emerging as "biological gold mines" for molecular information, often providing faster diagnostic routes than traditional tissue biopsies.
Conclusion and Future Directions
While the potential for personalized oncology is vast, significant hurdles remain. Small sample sizes in pilot studies and technical variability in assays continue to impede clinical adoption. Future efforts must focus on developing accessible, automated bioinformatics pipelines and integrating biomarker validation into the earliest stages of clinical trial designs, such as umbrella trials like Lung-MAP, to bridge the implementation gap.
Summary updated