PRIMAGE proposes a cloud-based platform to support decision making in the clinical management of malignant solid tumours, offering predictive tools to assist diagnosis, prognosis, therapies choice and treatment follow up, based on the use of novel imaging biomarkers, in-silico tumour growth simulation, advanced visualisation of predictions with weighted confidence scores and machine-learning based translation of this knowledge into predictors for the most relevant, disease-specific, Clinical End Points.

PRIMAGE implements a hybrid cloud model, comprising the of use of open public cloud (based on EOSC services) and private clouds, enabling use by the scientific community (facilitating reuse of de-identified clinical curated data in Open Science) and also suitable for future commercial exploitation.

The proposed data infrastructures, imaging biomarkers and models for in-silico medicine research will be validated in the application context of two paediatric cancers, Neuroblastoma (NB, the most frequent solid cancer of early childhood) and the Diffuse Intrinsic Pontine Glioma (DIPG, the leading cause of brain tumour-related death in children). These two paediatric cancers are relevant validation cases given their representativeness of cancer disease, and their high societal impact, as they affect the most vulnerable and loved family members.

The European Society for Paediatric Oncology, two Imaging Biobanks and three of the most prominent European Paediatric oncology units are partners in this project, making retrospective clinical data (imaging, clinical, molecular and genetics) registries accessible to PRIMAGE, for training of machine learning algorithms and testing of the insilico tools’ performance. Solutions to streamline and secure the data pseudonymisation, extraction, structuring, quality control and storage processes, will be implemented and validated also for use on prospective data, contributing
European shared data infrastructures.

PRIMAGE objectives are classified in 8 major outcomes, each one requiring of the achievement of specific scientific and technical objectives as enumerated in the related work packages:

Ojective 1: To design a Decision Support System (DSS) for cancer management with advanced functionality and usability, under a user-centric approach guided by our clinical partners (European Key Opinion Leaders in Paediatrics Oncology), aligned with their current work flows and aimed to conquer trust and gain acceptability by the clinical practitioners.

Objective 2: To implement PRIMAGE DDS as a cloud-based platform, with an hybrid model of use of (i) open public cloud (based on EOSC services) and (ii) private clouds in order to deliver a platform to (i) be used by the scientific community to promote use of de-identified clinical curated data in Open Science and also (ii) to be suitable for future commercial exploitation under PaaS business models, as an scalable, safe and cost-effective cloud infrastructure.

Objective 3: To establish a symbiosis with major European initiatives for shared clinical data repositories for Neuroblastoma (c.a. 2600 patients) and DIPG (c.a. 700 patients), both managed by SIOPE, where (i) retrospective clinical Big Data is used for biomarkers and in-silico model training and validation and (ii) PRIMAGE delivers methodologies to facilitate the extraction, de-identification and quality control of imaging and clinical data from hospital databases, thus contributing to feeding new clinical cases into the existing shared repositories. Additionally, this project will develop a strategy for embedding imaging biobanks into wider biobanks networks (e.g. BBMRI-ERIC) and facilitate data cross-linking with other biorepositories.

Objective 4: To validate new imaging biomarkers of MR, PET/CT and MIBG for NB and DIPG and develop diagnostic models based on imaging and cross-linked to established biological biomarker panels for each disease for supporting decision by clinical practitioners.

Objective 5: To deliver in-silico models of solid tumour growth for NB and DIPG using multiscale simulation frameworks to couple model at subcellular scale, to cell, to tissue and to complete organ models, enabling assessment of a radiotherapy and chemotherapy treatment on tumour progression for a given patient under.

Objective 6: To promote the usability of the new generated knowledge from the in-silico models for virtual diagnosis (O4) and tumour growth (O5), by (i) delivering visualisation methods for high dimensional data analysis, and (ii) implementing Artificial Intelligence methodologies to generate responses to the most relevant 5 Clinical End Points for NB and DIPG.

Objective 7: To integrate a functional prototype of PRIMAGE cloud-based platform, offering predictive tools to assist management of NB and DIPG paediatric cancers, from diagnosis to prognosis, therapies choice and treatment follow up, based on the use of novel imaging biomarkers (O4), tumour growth models (O5), advanced visualisation of predictions with weighted confidence scores and responses to a set of CEPs, obtained from the use of patient-specific in-silico models in combination with AI analytics on relevant patient clusters.

Objective 8: To validate PRIMAGE platform performance in multicentre prospective non-interventional studies for NB (approx. 150 patients) and for DIPG (approx. 75 patients) in Spain, Germany and Austria, involving clinical multidisciplinary clinical teams in the use of the Decision Support System and, its assessment according to the defined Key Performance Indicators.

Project Coordinator
Dr. Marti-Bonmati, Head of Medical Imaging Department and Dr. Cañete, Head of the Paediatric Oncology Unit, Hospital Universitario y Politécnnico La Fe, Valencia, Spain

Project Management Team
Mario Aznar, MATICAL Innovation, Madrid, Spain
Gracia Marti, Project Manager, GIBI230, Hospital Universitario y Politécnnico La Fe, Valencia, Spain

Dissemination Management Team
Anne Blondeel, The European Society for Paediatric Oncology (SIOPE)