Defining early epigenetic drivers of breast cancer metastasis through development and validation of a methylation-Based biomarker

Primary supervisor: Nnenna Kanu, UCL

Secondary supervisor: Lucia Cottone, UCL

Project

Triple-negative breast cancer (TNBC) remains one of the most aggressive subtypes of breast cancer, accounting for 10–15% of cases but a disproportionately high number of metastatic relapses and breast cancer-related deaths [1]. Unlike hormone receptor-positive or HER2-positive tumours, TNBC lacks targeted therapies and reliable prognostic biomarkers, making it difficult to predict which patients are at highest risk of early recurrence [2]. Moreover, current stratification tools largely rely on genomic alterations or histopathological features, which may not capture the early epigenetic reprogramming events that drive tumour evolution and metastasis. There is an urgent need to develop molecular tools that can identify high-risk tumours early and guide personalised surveillance or therapeutic strategies.

This PhD project builds on the PROMISE LC methylation signature, previously developed by the Kanu laboratory and validated in lung cancer using data from the TRACERx [3] and TCGA cohorts. PROMISE uses the ORACLE approach [4] to capture early, clonally stable hypermethylation events in treatment naïve lung cancers which may offer a promising avenue for biomarker discovery as early biomarkers of advanced cancer progression. This PhD will apply a similar strategy to TNBC within Breast TRACERx, aiming to uncover early epigenetic lesions predictive of relapse, and ultimately, to inform patient stratification and clinical decision-making.

Project Overview:

The student will perform reduced representation bisulfite sequencing (RRBS) on multi-region tumour biopsies sampled pre-neoadjuvant chemotherapy from Breast TRACERx patients enabling analysis of intratumour heterogeneity. The RRBS data will be integrated with matched RNA-sequencing and clinical outcome data to identify clonally conserved, hypermethylated loci that may represent early epigenetic events driving relapse.

The project will employ a combination of bioinformatics, machine learning, and molecular biology approaches, offering training across wet-lab and computational disciplines.

Key Aims of the Project:

1) To generate and analyse RRBS data from primary TNBC tumours to identify clonally stable DNA methylation events associated with metastatic relapse.

The student will process and analyse RRBS data from multi-region tumour samples, using bioinformatic pipelines to identify consistently altered methylation loci. The analysis will focus on early, stable hypermethylation patterns linked to relapse/metastasis, with specific attention to alterations that differentiate between extra-thoracic vs. intra-thoracic dissemination.

2) To develop and validate a breast cancer-specific PROMISE-Breast Cancer methylation signature predictive of metastatic potential.

The student will apply supervised and unsupervised machine learning models to derive a robust predictive signature. This will be validated across patients using cross-validation techniques and compared to clinical features and patterns of metastatic spread.

3) To functionally validate candidate methylation events using in vitro breast cancer models.

Top candidate hypermethylated loci will be investigated in patient-derived or established TNBC organoid models. The student will employ CRISPR-dCas9 epigenetic editing, epigenetic drugs (including demethylating agents) screening and phenotypic assays (e.g., invasion, proliferation, drug sensitivity), previously developed [5], to assess the functional relevance of the identified methylation changes.

Additionally, the student will explore metastatic biopsies from these TNBC patients and matched plasma-derived circulating tumour DNA (ctDNA), where available, to evaluate the detectability of the methylation signature in liquid biopsies. This will inform the potential for non-invasive monitoring of relapse risk or minimal residual disease in future clinical applications.

References

1. Dent R, Trudeau M, Pritchard KI, Hanna WM, Kahn HK, Sawka CA, et al. Triple‑negative breast cancer: clinical features and patterns of recurrence. Clin Cancer Res. 2007;13(15 Pt 1):4429–4434. doi: 10.1158/1078-0432.CCR-06-3045
2. Radosevic-Robin N, Selenica P, Zhu Y, Won HH, Berger MF, Ferrando L, et al. Recurrence biomarkers of triple negative breast cancer treated with neoadjuvant chemotherapy and anti-EGFR antibodies. npj Breast Cancer. 2021;7:124. doi:10.1038/s41523-021-00334-5.
3. Frankell AM, Dietzen M, Al Bakir M, Lim E, Karasaki T, Ward S….TRACERx Consortium; Jamal-Hanjani M, McGranahan N, Swanton C. Nature. 2024 Jul;631(8022):E15. doi: 10.1038/s41586-024-07738-w.
4. Biswas D, Liu YH, Herrero J, Wu Y, Moore DA, Karasaki T …, Jamal-Hanjani M, Kanu N, Birkbak NJ, Swanton C. Prospective validation of ORACLE, a clonal-expression biomarker associated with survival of patients with lung adenocarcinoma. Nat Cancer. 2025 Jan;6(1):86–101. doi:10.1038/s43018-024-00883-1.
5. Cottone L, Cribbs AP, Khandelwal G, Wells G…… Pillay N, Jenner RG, Oppermann U, Flanagan AM. Inhibition of histone H3K27 demethylases inactivates brachyury (TBXT) and promotes chordoma cell death. Cancer Res. 2020 Oct 15;80(20):4540-51. doi:10.1158/0008-5472.CAN-20-1387.