Periodic Reporting for period 1 - MULTIR (MULti-Tumour based prediction and manipulation of Immune Response)
Reporting period: 2024-01-01 to 2025-06-30
Cancer remains a leading cause of mortality and morbidity worldwide. Despite progress in immunotherapies and targeted treatments, outcomes remain highly variable, with relapse and resistance often unpredictable. There is an urgent need for earlier detection, better prediction of treatment response, and tools readily integrated into clinical practice.
The MULTIR project (MULti-Tumour based prediction and manipulation of Immune Response) addresses this by focusing on three challenging cancers—melanoma, lung and bladder—where predictive biomarkers are scarce. Through multi-omics analyses, clinical and imaging data, preclinical models and synthetic datasets, MULTIR seeks to unravel tumour–host interactions and deliver biomarkers, diagnostic tools and decision-support models for personalised treatment.
Aligned with the EU Cancer Mission to improve survival, reduce inequalities and save 3 million lives by 2030, MULTIR contributes by developing interoperable data infrastructures, generating privacy-preserving synthetic datasets, and collaborating within the Mission Cancer cluster on “Understanding Tumour–Host Interactions,” ensuring results inform both science and policy.
Expected impacts are scientific and societal. MULTIR will deliver validated biomarkers and predictive models tested in clinical cohorts, underpinned by AI integration. By engaging patients, healthcare professionals, regulators and the public, the project ensures outputs are relevant, acceptable and ready for uptake, with attention to ethical, legal and societal aspects.
In doing so, MULTIR paves the way for transformative cancer care—enabling earlier relapse detection, reducing unnecessary treatments, and supporting personalised decisions—bridging molecular research, clinical translation and societal adoption.
The MULTIR project (MULti-Tumour based prediction and manipulation of Immune Response) addresses this by focusing on three challenging cancers—melanoma, lung and bladder—where predictive biomarkers are scarce. Through multi-omics analyses, clinical and imaging data, preclinical models and synthetic datasets, MULTIR seeks to unravel tumour–host interactions and deliver biomarkers, diagnostic tools and decision-support models for personalised treatment.
Aligned with the EU Cancer Mission to improve survival, reduce inequalities and save 3 million lives by 2030, MULTIR contributes by developing interoperable data infrastructures, generating privacy-preserving synthetic datasets, and collaborating within the Mission Cancer cluster on “Understanding Tumour–Host Interactions,” ensuring results inform both science and policy.
Expected impacts are scientific and societal. MULTIR will deliver validated biomarkers and predictive models tested in clinical cohorts, underpinned by AI integration. By engaging patients, healthcare professionals, regulators and the public, the project ensures outputs are relevant, acceptable and ready for uptake, with attention to ethical, legal and societal aspects.
In doing so, MULTIR paves the way for transformative cancer care—enabling earlier relapse detection, reducing unnecessary treatments, and supporting personalised decisions—bridging molecular research, clinical translation and societal adoption.
During the first reporting period, MULTIR built a solid scientific and technical foundation, focusing on data collection, technological innovation, preclinical validation, and integrative analyses across melanoma, lung and bladder cancers.
Cohorts and data generation: Clinical cohorts were initiated and expanded, with biological samples and clinical/imaging data collected across the three cancer types. Harmonisation protocols for clinical endpoints were defined, and thousands of patient datasets curated for integration. External datasets were also mapped to strengthen analytical robustness.
Technological developments: High-resolution multi-omics analyses (WES/WGS, RNA-seq, proteomics, peptidomics, imaging mass cytometry) were implemented across partners. Computational pipelines were created for dataset processing and integration, including synthetic data generation and AI-based biomarker discovery. Tools such as Proteomics Eye (ProtE) support reliable and reproducible proteomics analysis.
Preclinical models: Organoids and humanised mouse models were established for mechanistic studies of tumour–host interactions, linking omics-derived hypotheses to experimental validation.
Preliminary findings: Integration of clinical and molecular data identified candidate biomarkers and molecular signatures linked to treatment response and relapse risk. Early analyses confirmed the feasibility of correlating omics profiles with clinical outcomes, paving the way for predictive and prognostic tools.
Integration and data infrastructure: A GDPR-compliant platform with anonymisation pipelines, synthetic data modules and role-based access ensures FAIR data management and interoperability with European initiatives such as EOSC and UNCAN.eu.
Overall, this first-period work delivered consolidated multi-cancer datasets, analytical pipelines and validated preclinical models, positioning MULTIR for biomarker validation, clinical translation and cross-cancer integration in line with EU Cancer Mission goals.
Cohorts and data generation: Clinical cohorts were initiated and expanded, with biological samples and clinical/imaging data collected across the three cancer types. Harmonisation protocols for clinical endpoints were defined, and thousands of patient datasets curated for integration. External datasets were also mapped to strengthen analytical robustness.
Technological developments: High-resolution multi-omics analyses (WES/WGS, RNA-seq, proteomics, peptidomics, imaging mass cytometry) were implemented across partners. Computational pipelines were created for dataset processing and integration, including synthetic data generation and AI-based biomarker discovery. Tools such as Proteomics Eye (ProtE) support reliable and reproducible proteomics analysis.
Preclinical models: Organoids and humanised mouse models were established for mechanistic studies of tumour–host interactions, linking omics-derived hypotheses to experimental validation.
Preliminary findings: Integration of clinical and molecular data identified candidate biomarkers and molecular signatures linked to treatment response and relapse risk. Early analyses confirmed the feasibility of correlating omics profiles with clinical outcomes, paving the way for predictive and prognostic tools.
Integration and data infrastructure: A GDPR-compliant platform with anonymisation pipelines, synthetic data modules and role-based access ensures FAIR data management and interoperability with European initiatives such as EOSC and UNCAN.eu.
Overall, this first-period work delivered consolidated multi-cancer datasets, analytical pipelines and validated preclinical models, positioning MULTIR for biomarker validation, clinical translation and cross-cancer integration in line with EU Cancer Mission goals.
During its first reporting period, MULTIR has delivered advances beyond the state of the art in studying tumour–host interactions and predicting therapeutic responses. The project established a unique multi-layered data integration framework, combining clinical, molecular, imaging and immunological datasets across melanoma, lung and bladder cancer. Unlike existing single-modality approaches, MULTIR’s GDPR-compliant federated infrastructure enables secure aggregation and use of highly heterogeneous datasets at scale.
The consortium also developed AI-based pipelines for synthetic data generation and federated learning, tackling a key barrier in cancer research: the difficulty of sharing sensitive health data. These pipelines generate realistic yet privacy-preserving datasets, enabling advanced computational models to be trained without compromising confidentiality. This methodological innovation both drives cancer research and sets a benchmark for responsible data use in health.
Another major achievement is the design of explainable predictive models for treatment response, bridging the gap between deep learning performance and clinical interpretability. As transparency and trust remain barriers to adoption, early involvement in clinical trials provides the basis for robust validation of predictive signatures in real-world settings.
The potential impact includes improved patient stratification, earlier detection of resistance and personalised treatment, supporting precision oncology in healthcare systems. To secure uptake and sustainability, MULTIR highlights the need for further validation in larger cohorts, continued clinical demonstration, regulatory alignment and standardisation, sustained access to finance and IPR protection, and internationalisation to maximise global relevance.
In sum, MULTIR advances fundamental knowledge while laying the foundations for clinical translation, regulatory convergence and societal uptake, fully aligned with the EU Cancer Mission. It thereby accelerates innovation and strengthens Europe’s capacity to deliver equitable, safe and effective cancer care.
The consortium also developed AI-based pipelines for synthetic data generation and federated learning, tackling a key barrier in cancer research: the difficulty of sharing sensitive health data. These pipelines generate realistic yet privacy-preserving datasets, enabling advanced computational models to be trained without compromising confidentiality. This methodological innovation both drives cancer research and sets a benchmark for responsible data use in health.
Another major achievement is the design of explainable predictive models for treatment response, bridging the gap between deep learning performance and clinical interpretability. As transparency and trust remain barriers to adoption, early involvement in clinical trials provides the basis for robust validation of predictive signatures in real-world settings.
The potential impact includes improved patient stratification, earlier detection of resistance and personalised treatment, supporting precision oncology in healthcare systems. To secure uptake and sustainability, MULTIR highlights the need for further validation in larger cohorts, continued clinical demonstration, regulatory alignment and standardisation, sustained access to finance and IPR protection, and internationalisation to maximise global relevance.
In sum, MULTIR advances fundamental knowledge while laying the foundations for clinical translation, regulatory convergence and societal uptake, fully aligned with the EU Cancer Mission. It thereby accelerates innovation and strengthens Europe’s capacity to deliver equitable, safe and effective cancer care.