The 15 topics of HER-CARE

DC1 | Improving classification of pathogenic variants

DC1 focuses on reclassifying rare BRCA1/2 variants of uncertain significance (VUS) that currently complicate risk assessment and clinical management of patients and their families. By applying advanced statistical analyses and simulations on ENIGMA consortium data, this project aims to more accurately identify high- and moderate-risk variants, improving genetic counseling and personalized prevention strategies.

DC2 | Identifying novel regulatory variants

DC2 investigates novel rare variants in non-coding regions of the genome, such as enhancers and transcription factor binding sites, that may explain part of the missing heritability in breast cancer. Using publicly available large genomic datasets, data mining tools and advanced quality control, the project will uncover new genetic mechanisms contributing to breast cancer risk.

DC3 | Developing multi-ancestry risk prediction models

DC3 leverages the Confluence dataset to improve polygenic risk scores (PRS) of early onset breast cancer, for overall and major breast cancer subtypes, in populations of diverse ancestral populations. By combining Bayesian, machine learning, and transfer learning approaches, the project aims to develop a comprehensive, multi-ancestry risk prediction model for early-onset breast cancer.

DC4 | Understanding the impact of hormone replacement therapy (HRT) in BRCA1/2 carriers

DC4 evaluates how hormone replacement therapy (HRT) affects breast cancer risk in individuals with BRCA1/2 pathogenic variants and other non-BRCA pathogenic variants, particularly after risk-reducing salpingo-oophorectomy (RRSO, a surgery to remove both fallopian tubes and ovaries). The project will develop enhanced decision-making algorithms by integrating the CanRisk model, with both genetic and non-genetic factors and determine how obesity and other factors modify the effects of HRT on breast cancer risk post-RRSO. The project will provide more accurate recommendation for safe HRT use and personalized prevention strategies.

DC5 | Risk factors for contralateral breast cancer

DC5 investigates hereditary and non-hereditary risk factors for developing contralateral breast cancer, the most common second cancer in breast cancer survivors. By analyzing genetic, treatment, and mammographic data across large datasets, the project aims to improve risk prediction and clinical management for breast cancer survivors. The project will provide new insights into how genetic and immunological factors, breast tissue changes, and systemic treatment impact development of CBC.

DC6 | Linking environmental exposures to early-onset breast cancer

DC6 studies how environmental and biological exposures—captured through exposome datasets and chemical profiling—contribute to early-onset breast cancer risk. By combining geospatial data with large cohort studies, the project will identify key external environmental and internal metabolomic risk factors, advancing prevention and early detection strategies.

DC7 | Refining blood-based detection for hereditary breast cancer

DC7 investigates whether circulating tumor DNA (ctDNA) can be used to detect hereditary breast cancer at an earlier stage. By applying whole-genome sequencing and the MRD-EDGE algorithm to plasma samples from high-risk individuals, the project will refine and validate ctDNA analysis methods to increase its sensitivity and reliability. This work aims to assess ctDNA’s potential as a minimally invasive tool for early detection for high-risk, genetically predisposed women.

DC8 | Advancing minimally invasive screening with saliva DNA

DC8 explores innovative, non-invasive approaches to early breast cancer detection using saliva DNA samples from 200 women who have developed breast cancer during the follow-up of the MyPeBs trial. Combining long-read sequencing, methylation profiling, and AI-driven analysis, the project will facilitate development of a comprehensive breast cancer prediction assay.

DC9 | Personalizing breast cancer screening with AI and MRI

DC9 integrates artificial intelligence with MRI and mammography to improve personalized breast cancer risk prediction. By identifying imaging markers linked to short-term cancer risk, the project will help clinicians decide which women would benefit from short-term follow-up, and which women can safely prolong their surveillance interval, avoiding unnecessary additional screenings.

DC10 | Predicting contralateral breast cancer risk

DC10 enhances risk prediction for contralateral breast cancer (CBC) by integrating genetics, treatment data, and modifiers into advanced statistical models. Building on existing CanRisk and PredictCBC models, this project will deliver a clinical decision-support tool to guide clinicians for assessing second breast cancer risk alongside first breast cancer prognosis.

DC11 | Integrating AI-based risk models into clinical workflows

DC11 develops standardized protocols to integrate AI-based risk models into established health records systems used in clinical practice, overcoming technical, ethical, and regulatory barriers to clinical implementation of these models. By using real-world and synthetic health record data, the project will ensure reliable, interoperable, and continuously monitored AI tools that support clinicians in daily decision-making.

DC12 | Defining genetic signatures of early-onset breast cancer

DC12 investigates the unique genetic and molecular profiles of early-onset breast cancer to uncover their biological origins, improve detection, and identify high-risk groups and treatment targets. By integrating large-scale genomic, histological, and immunological data, the project aims to elucidate the genomic contribution to breast cancer subtypes, comparing early-onset and post-menopausal breast cancers and gain a clearer understanding of the molecular drivers of these early-onset tumors.

DC13 | Linking germline and somatic mutations

DC13 explores how inherited (germline) and acquired (somatic) mutations interact to shape tumor biology. Using international cancer genome datasets (TCGA, ICGC, Genomics England), the project will create a comprehensive framework that links germline variations to somatic mutation profiles, followed by an evaluation of how these changes affect protein structures. This will ultimately lead to new insights into tumor progression and potential therapeutic targets.

DC14 | Predicting tumor progression with protein biomarkers

DC14 aims to develop a predictive test to distinguish which early-stage breast cancers are likely to progress. By combining Protein Cross Section (PCS) biomarker data with genetic and clinical information, the project will build supervised machine learning models to generate PCS profile risk scores that distinguishes between early-stage breast cancer tumors with a high and low risk progressing.

DC15 | Building multimodal AI models for tumor classification

DC15 integrates imaging, genetic, and tumor data into advanced AI models to develop advanced multimodal models to predict breast cancer progression and aggressiveness in BRCA1 and BRCA2 pathogenic variant carriers. These multimodal models will be validated in clinical settings to ensure accuracy and personalized care for high-risk individuals and enhance clinical decision-making beyond current prediction methods.