Periodic Reporting for period 2 - EpiPredict (Epigenetic regulation of endocrine therapy resistance in breast cancer: A systems medicine approach to predict treatment outcome)
Reporting period: 2017-01-01 to 2018-12-31
ESR1/B1 developed a stochastic model to interpret the role of transcriptional regulation of hypervariable genes for resistance development (experimentally tested by ESR3/B1). A cell line-specific Genome-Scale Metabolic Model was created integrating transcriptomic data (from B2) to interpret metabolic-based resistance (collaboration ESR7/B4).
ESR2/B1 worked on elucidating pathways that change during resistance acquisition in breast cancer cell lines performing integrative analysis of multi-omics (epigenetic) data (from ESR5/B3). A novel method was developed to follow breast cancer resistance acquisition.
ESR3/B1 characterised transcription cell-to-cell variability in drug treated breast cancer cell lines using single molecule RNA FISH. Single molecule RNA FISH and single cell transcriptomic data (from B2) were compared allowing to calibrate transcriptomic data. A stochastic model predicting nascent gene expression for bursty/non-bursty genes is produced (with ESR1/B1) and tested.
ERS4/B2 profiled the epigenome of 47 luminal breast cancer patients characterising enhancer and promoter activity. Novel computational approaches identified YY1 and SLC9A3R1 as key players of oncogenesis and drug resistance. Evolutive principles of the activity of regulatory regions at different stages of disease highlighted the role of specific enhancers in driving cancer metastasis (tested by ESR5/B3).
ESR5/B3 characterized the transcriptome and epigenome of resistance development in breast cancer cell lines and identified targets involved in resistance acquisition. These targets are being validated with functional cell line assays.
ESR6/B3 performed RNA-seq, ATAC-seq, EPIC array on sensitive/resistant breast cancer cell lines. Combinatorial analysis revealed novel targets in therapy resistance validated with RNA interference. Epigenetic editing was used with different catalytic domains of epi-modifiers (collaboration ESR8/B5;ESR9/B6). Validation of epigenetic editing for instance on drug response, mimicking/reversing resistance, proliferation, is ongoing.
ESR7/B4 used untargeted metabolic profiling and stable isotope resolved flux omics on resistant/sensitive breast cancer cell lines. Differences in metabolic flux distribution and flux rates were identified using a computational approach (collaboration ESR1/B1). Key vital fluxes for cell survival and new putative targets were identified. Evaluating the effect of metformin, an inhibitor of mitochondrial respiration, indicated its potential to be used as adjuvant treatment inducing drug resensitization.
ESR8/B5 created epigenetic editing tools for resistance involved genes to modulate expression (collaboration ESR9/B6;ESR6/B3). The CRISPR-dCas9 epigenetic targeting in breast cancer cell lines stably expressing dCas constructs, is tested for long-term effects on gene reprogramming.
ESR9/B6 developed an experimental system to study kinetics of epigenetic editing in breast cancer cell lines. Plasmids expressing a variety of targetable epigenetic editors were created to investigate epigenetic regulation of genes implicated in resistance development (collaboration ESR8/B5;ESR6/B3). Targeting constructs are tested for their role in transcription noise (collaboration ESR3/B1;ESR1/B1).
ESR10/B7 developed a sensitive DNA methylation based qPCR method to detect immune cells in cancer tissues. An in silico platform for methylation specific immunoprofiling is developed. A statistical model is set-up to predict methylation status of a locus using biological features (collaboration B1/UvA).
ESR11/B2 developed a framework to generate predictive signatures using a genetic algorithm (GABi) in which prior-knowledge can be incorporated. The framework was proven to be of value for cancer bioinformatics research to predict relapse in ER+, Tamoxifen treated patients.
In EpiPredict, computational modeling guides experimental technologies uncovering systems behavior of breast cancer resistance development upon endocrine treatment. This powerful multidisciplinary approach makes full use of our excellent expertise of innovative technologies and crosses borders of the various disciplines involved. We leveraged fundamental systems properties of resistance development and medical patient data using an angle that goes far beyond state-of-the-art. Currently, we are writing a conceptual paper commissioned for Nature Communications outlining our EpiPredict strategy.
Impact of EpiPredict
Host-based and network-wide training allowed our ESRs to gain a very critical mind-set. During EpiPredict monthly video conferences, annual meetings and secondments, the ESRs reflected on each other's research, implementing knowledge in a broad context. This will benefit future EU research and innovation. Our EpiPredict research initiated a thorough collaboration between partner labs. The positive atmosphere stimulated epigenetic research in the field of breast cancer resistance and serves as a harbor for future research enabling research developments largely stimulating EU industry.
Outreach to the public at large
EpiPredict research-training led to high impact research dissemination. Magnani et al, Nature Genetics (2017) is the first example of the collaborative, innovative and multidisciplinary research within EpiPredict. Patten et al., Nature Medicine (2018) is a breakthrough paper using epigenetic mapping of luminal breast cancer tumors to uncover how intra-tumor and inter-patient heterogeneity drives tumor evolution. Several additional high impact collaborative studies are being finalized. The highly-valued EpiPredict-organized International conference (http://www.systemsepigenetics2018.eu/) provided ample dissemination of EpiPredict research. EpiPredict generated a strong international hub with excellent opportunities for a follow-up collaborative network project.
Three Science Cafés targeting different audiences (scientists in general, patients and industry) including video and blog reporting created full awareness of the impact of epigenetics in health care. This intense outreach is important to set the scene to stimulate socio-economic impact and wide-societal implications of EpiPredict data.