Virtual Patient Models That Help Doctors Pick the Best Cancer Treatment for Each Child

Imagine a doctor could create a digital twin of a sick child — a computer copy that lets them test different cancer treatments before giving real medicine. That's what this project built: cloud-based virtual patient models for children with cancer. Instead of trial-and-error with harsh treatments, doctors can simulate which drug combinations would work best for that specific child. The team pulled together data from hospitals, labs, and clinical trials across 11 countries to make these simulations as accurate as possible.

What needed solving

Children with cancer often receive treatments based on protocols designed for adults or broad patient groups, not tailored to their individual biology. Doctors lack computational tools to simulate how a specific child's tumour will respond to different drug combinations before starting treatment. This means precious time and health can be lost on ineffective therapies, especially for rare paediatric cancers where clinical experience is limited.

What was built

The project built a cloud-based platform with virtual patient models for paediatric oncology, including a functioning in silico prototype of hepatocyte metabolism, a Data Access Committee portal for managing clinical data credentials, standardized metadata and analytical workflows for harmonizing multi-omics cancer data, and an initial cloud infrastructure demonstrator for running computational simulations at scale.

Who needs this

Who can put this to work

Quick answers

Who built it

The iPC consortium is large and research-heavy: 23 partners from 11 countries, with 8 universities and 9 research organizations forming the backbone. The industry presence is modest at 17% (4 partners, 3 of which are SMEs), which is typical for health research at this stage — the science needs to be proven before larger industry players commit. The coordinator is TECHNIKON, an Austrian SME specializing in research management, which means commercialization decisions may be distributed across the technical partners rather than centralized. For a business looking to adopt this technology, the key contacts would be the clinical partners running the trials and the IT partners who built the cloud infrastructure. The international spread (Austria, Germany, France, Italy, Netherlands, Belgium, Spain, Slovenia, Switzerland, Australia, and the US) signals broad applicability but also complex IP arrangements.

• TECHNIKON FORSCHUNGS- UND PLANUNGSGESELLSCHAFT MBHCoordinator · AT
• LUDWIG-MAXIMILIANS-UNIVERSITAET MUENCHENparticipant · DE
• UNIVERSITEIT GENTparticipant · BE
• XLAB RAZVOJ PROGRAMSKE OPREME IN SVETOVANJE DOOparticipant · SI
• IBM RESEARCH GMBHparticipant · CH
• BAYLOR COLLEGE OF MEDICINEparticipant · US
• UNIVERSITA DEGLI STUDI DI NAPOLI FEDERICO IIparticipant · IT
• DEUTSCHES KREBSFORSCHUNGSZENTRUM HEIDELBERGparticipant · DE
• ALACRIS THERANOSTICS GMBHparticipant · DE
• INSTITUT DE INVESTIGACIO EN CIENCIES DE LA SALUT GERMANS TRIAS I PUJOLparticipant · ES
• CONSIGLIO NAZIONALE DELLE RICERCHEparticipant · IT
• INSTITUT CURIEparticipant · FR
• STICHTING AMSTERDAM UMCparticipant · NL
• CHARITE - UNIVERSITAETSMEDIZIN BERLINparticipant · DE
• PRINSES MAXIMA CENTRUM VOOR KINDERONCOLOGIE BVparticipant · NL
• UNIVERSITATSKLINIKUM HEIDELBERGparticipant · DE
• UNIVERSITAT ZURICHparticipant · CH
• THE CHILDREN'S HOSPITAL OF PHILADELPHIA NON PROFIT ORGparticipant · US
• TECHNISCHE UNIVERSITAT DARMSTADTparticipant · DE
• MAX-PLANCK-GESELLSCHAFT ZUR FORDERUNG DER WISSENSCHAFTEN EVparticipant · DE
• FUNDACIO CENTRE DE REGULACIO GENOMICAthirdparty · ES
• BARCELONA SUPERCOMPUTING CENTER CENTRO NACIONAL DE SUPERCOMPUTACIONparticipant · ES