Virtual Physiological Human: Personalized Predictive Breast Cancer Therapy Through Integrated Tissue Micro-Structure Modeling

Breast cancer is frequent and life threatening, but curable if detected early. Early detection and comprehensive characterisation of findings require optimized imaging and image understanding to maximise detection of significant disease while preventing overdiagnosis. Personalised predictive modelling of breast cancer allows treatment stratification, preventing unnecessary and unsuccessful treatments. VPH-PRISM addresses these key topics with integrated multidisciplinary, multi-scale ICT modelling of breast tissue microstructure in the context of environmental, genetic, and clinical factors.

Key challenges include establishment of combined biomarkers from the automated analysis and spatial correlation of digital pathology and advanced breast imaging. Tissue characterisation includes the peritumoural stroma, a key in tumour progression and therapy response. Comprehensive clinical breast cancer phenotypes are extracted from prospectively collected multidisciplinary data. Interactions of environmental and genetic factors with specific breast tissue patterns are analysed in three large ongoing population-based imaging cohorts. A standard breast model enables efficient, combined statistical modelling of sparsely sampled and heterogeneous, large-scale data across disciplines, scales, structures, time and patients.

Using the developed tools and models, and the data collected, we will:
- improve estimates of tumour spread to aid surgery and assess chemo- and radiotherapeutic response
- optimise multi-modal imaging methods through biophysical forward modelling of image formation for more efficient phenotyping and imaging biomarkers
- predict personal risks for cancer progression and select optimal treatment strategies

VPH-PRISM will provide a proof of concept for multidisciplinary model based discovery of environment-tissue interactions, quantitative drug efficacy assessment, surgery planning, and treatment outcome prediction at early and advanced stages of breast cancer.