A tool that can integrate breast cancer transcriptomic datasets could help clinicians to more accurately diagnose the disease and better target treatment, a study has found.
High throughput transcriptomic profiling had refined subtyping of breast cancer and diagnostics, the researchers wrote in npj Breast Cancer, however its wider use has been hampered for a range of reasons including difficulties in integrating different sources of molecular data.
To develop a new integrated model, researchers initially focused on patients with early-stage breast cancer and used bulk ribonucleic acid (RNA)-sequencing and microarray data from the Cancer Genome Atlas Program and the Molecular Taxonomy of Breast Cancer International Consortium cohorts.
To integrate the datasets, they looked at the expression levels of 1,044 genes of interest, using various mathematical approaches to test different numerical schemes that best explained the data.
This unified space of breast cancer transcriptomes, known as EMBER, was able to predict phenotypes of transcriptomic profiles on a single sample basis, they explained, with key biological pathways, such as oestrogen receptor signalling and cell proliferation, determining sample position.
‘The localisation of a patient sample in the EMBER space provides a novel way to interpret the molecular subtypes as a continuum and find candidate biological pathways of resistance to endocrine therapy,’ they wrote.
Validating the EMBER model
The model was validated in four independent breast cancer patient cohorts and with samples from a window trial called POETIC, which showed the model could capture clinical responses to endocrine therapy.
‘We observed that high androgen receptor signalling and low TGFβ scores were associated with a poor response to endocrine therapy,’ the researchers wrote.
Furthermore, they said an oestrogen-receptor signalling score from the model was superior to the immunohistochemistry-based oestrogen-receptor index that clinicians currently use to select patients for endocrine therapy.
‘By not assigning a singular molecular subtype to a sample but rather a position in the EMBER space, we embrace the characteristics of a sample on a continuous scale thereby providing a high-resolution view of the entire neighbourhood,’ the researchers said.
‘We have shown here how all the data generated in large independent cohorts can be combined and used both in clinical practice and research, further enabling the use of genomics.’
The research was led by scientists from The Institute of Cancer Research (ICR), London, UK, and funded by the European Union’s Horizon 2020 Research and Innovation Programme Marie Skłodowska-Curie and the charity Breast Cancer Now.
The benefits of RNA sequencing in breast cancer
Commenting on the research in February 2025, Dr Syed Haider, second author of the study and group leader of the Breast Cancer Research Bioinformatics Group at the ICR, said there had ‘previously been many efforts to integrate big data in breast cancer datasets, but their application to clinical samples had been somewhat limited’.
This transcriptomic tool was different in that it provided ‘a formal basis for understanding the aggressivity of a new patient’s tumour against a large knowledge base created from retrospective patient cohorts’, Dr Haider said.
‘Now, in theory, any time a patient is diagnosed with breast cancer and RNA sequencing can be performed on their biopsy, the sample can be placed in the EMBER space, and different prognostic and predictive factors can be determined,’ he said.
‘Until now, this has not been feasible because large numbers of samples must be accumulated and run in a single batch in order to extract enough information about the tumour to guide clinical decision making.’
The University of Cambridge’s Professor Raj Jena, the UK’s first clinical professor of AI in radiotherapy, has predicted that AI tools could use genomic information to help personalise cancer treatment.
‘Personalised medicine is very interesting to us because we now get so much information when a cancer is diagnosed, including genomics, which can highlight mutations and indicate a patient may benefit from some kind of targeted drug,’ he said in a recent interview with Hospital Healthcare Europe.
He added: ‘I think the paradigm changes around AI in medicine will come within the areas of precision medicine or drug discovery.’
