EU Week Against Cancer and the role of the EIC

The European Week Against Cancer runs annually from 25 May to 31 May and is coordinated by the Association of Cancer Leagues. The week is intended to concentrate awareness raising across the continent and provide a platform for public bodies, non profit organisations and private actors to promote prevention, research and patient support activities. In this context the European Innovation Council is spotlighting a small set of its beneficiaries that are developing diagnostic and predictive technologies for oncology.

The four projects profiled here represent distinct technical approaches. They range from physical separation of extracellular vesicles to functional proteomic assays and machine learning applied to multiomic data. Each project is at a different maturity level and is supported under different EIC instruments. All claim potential to improve early detection, patient stratification or therapy selection. Those claims merit cautious scrutiny because diagnostic tools must pass multiple stages of technical validation, clinical validation, regulatory approval and adoption by healthcare systems before producing the implied population level benefits.

Projects in focus

AcouSome project — AcouSort AB, Sweden (EIC Transition)

AcouSort AB is developing AcouSome, a platform that uses ultrasound within a polymer microfluidic chip to isolate exosomes from blood samples. Exosomes are tiny extracellular vesicles released by cells that carry proteins, lipids and nucleic acids. They are of growing interest as minimally invasive biomarkers because they reflect the state of the cell of origin. AcouSome aims to make exosome isolation faster and compatible with point of care workflows, potentially improving early detection and monitoring for cancers such as glioblastoma, melanoma and prostate cancer.

Possible advantages include reduced processing time and a simpler workflow than many lab based ultracentrifugation methods. Key open questions are the analytical sensitivity and specificity for clinically relevant exosomal markers, reproducibility across blood sample types and volumes, and how the output integrates with downstream molecular assays. Transitioning from a prototype chip to a regulated diagnostic device also requires thorough clinical validation and alignment with regulatory standards.

3DSecret project — International Iberian Nanotechnology Laboratory (INL), Portugal (EIC Pathfinder)

3DSecret targets the central clinical problem that metastatic disease is the leading cause of cancer mortality. The project focuses on circulating tumour cells derived from breast cancer patients. Scientists plan to isolate these cells with microfluidic methods, culture them into three dimensional spheroids, and run genomic and phenotypic analyses. The aim is to use these data to detect stochastic patterns in metastasis formation and to develop an AI tool capable of predicting metastatic behaviour.

3DSecret combines wet lab microfluidics, culture models and computational analysis. The technical hurdles include reliably capturing very rare CTCs, maintaining viability during culture, avoiding culture induced artefacts, and obtaining datasets large and diverse enough to train generalisable AI models. Clinical translation will require robust prospective validation linking model predictions to patient outcomes.

IOpener project — PamGene International BV, Netherlands (EIC Accelerator)

PamGene is developing IOpener, a microarray based platform that profiles kinase activity in patient blood samples. The stated goal is to predict which patients will respond to immune checkpoint inhibitors. ICIs have transformed oncology for some cancers but response rates vary and treatments are costly and potentially toxic. IOpener aims to provide a functional in vitro diagnostic that informs patient selection for immunotherapy.

If validated, a blood based kinome assay could be clinically useful. However there are several practical challenges. Peripheral blood kinase signatures may be influenced by non tumour factors such as infection, medication or comorbidities. It will be necessary to demonstrate that the assay predicts ICI response independently of established biomarkers and that it improves clinical decision making. Regulatory approval and payer reimbursement are further barriers that influence real world uptake.

Qlucore project — Qlucore AB, Sweden (EIC Accelerator)

Qlucore produces AI enabled software for analysis and visualization of gene expression and genomic fusion data. The company focuses on tools used in precision cancer diagnostics. The EIC backed work emphasises acute myeloid leukaemia and bladder cancer as priority applications. Qlucore's platform combines machine learning models, interactive visualisation and workflows intended for researchers and clinical laboratories.

AI driven diagnostics face well documented challenges. Model robustness across sequencing platforms, batch effects, population diversity and integration into laboratory information systems are all non trivial. Qlucore's emphasis on visualization to make complex data accessible is a practical strength. The next steps for uptake are prospective clinical studies, external validation, and alignment with clinical pathways and reimbursement systems.

Comparative snapshot of the four projects

Broader context and realistic timelines

The European Innovation Council plays an important role in supporting early stage to scaling innovations across the EU. Its instruments range from Pathfinder discovery funding to Transition grants and Accelerator equity or grant blended funding. That funding helps teams progress technical prototypes and begin regulatory work but it does not guarantee clinical deployment. Diagnostic and predictive tools typically require multi year clinical studies, alignment with regulatory frameworks such as IVDR in Europe, and engagement with payers and hospital procurement to reach patients at scale.

Investors, policymakers and clinicians should watch for independent validation studies, peer reviewed publications and transparent reporting of limitations. Promising technologies can plateau if they fail to address pre analytical variability, integration with clinical workflows or the regulatory and reimbursement pathways. Conversely successful navigation of these steps can deliver more precise patient selection, less overtreatment and in some cases earlier detection.

Conclusions

The four EIC-backed projects represent a cross section of technologies being pursued in European oncology diagnostics. They illustrate how hardware, functional proteomics and AI driven software are complementary avenues toward better detection and treatment selection. Optimism about their potential is warranted but should be tempered with awareness of the translational bottlenecks that separate laboratory promise from routine clinical benefit. Continued public funding and coordinated pathways for validation and regulation will determine how many of these innovations reach patients.

DISCLAIMER: This information is provided in the interest of knowledge sharing and should not be interpreted as the official view of the European Commission or any other organisation.