Researchers at The University of Texas MD Anderson Cancer Center have developed a new computational approach designed to better account for changes in gene expression within tumors relative to their unique microenvironments. This approach outperformed current methods for predicting chemotherapy response in patients with triple-negative breast cancer (TNBC).
TNBC, which according to the American Cancer Society (ACS) accounts for 10%-15% of all breast cancers, lacks expression of estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2). The disease tends to be more common in women younger than age 40, who are Black, or who have a BRCA1 mutation.[1][2]
While immunotherapy and PARP inhibitors have been integrated into standard of care, their effectiveness varies considerably across patient subgroups due to TNBC’s remarkable molecular and cellular heterogeneity.[3] This unique tumour heterogeneity and the lack of effective therapies beyond chemotherapy result in TNBC, being a breast cancer subtype with the least favourable patient outcomes.[3]
A new tool, developed by Wenyi Wang, Ph.D., a professor of Bioinformatics and Computational Biology, and colleagues, will help with clinical decision-making.
Their study, which led to the development of the new tool, was supported by the National Cancer Institute, the Department of Defense, the Cancer Prevention and Research Institute of Texas (CPRIT), the American Cancer Society, and Lyda Hill Philanthropies and was published in the February 17, 2026, edition of Cell Reports Medicine.
Wang’s approach aims to improve upon similar methods for predicting treatment responses using deconvolution, which involves breaking down, quantifying, and interpreting cellular differences. This approach also revealed novel insights into population-level characteristics of TNBC.
“Deconvolution strategies are not one size fits all,” Wang said.
“We’re focused on making these methods more accessible to researchers without extensive computational backgrounds, with the goal of translating these powerful analytical approaches into practical tools that the broader cancer research community can readily apply to advance precision medicine,” she added.
Current Approach
Given the many computational tools available, Wang and colleagues recently published a comprehensive guide detailing 43 deconvolution methods. Their goal was to help researchers without extensive computational backgrounds understand which method might work best for their study-specific goals.
However, while existing classification strategies measure cell composition, they do not account for changes in gene expression within tumors relative to their unique microenvironments.
To address this, the researchers collaborated with MD Anderson’s Institute for Data Science in Oncology (IDSO) and the Department of Breast Medical Oncology to develop an integrative bulk analysis that also accounts for tumor-specific total mRNA expression (TmS), a pathway-agnostic deconvolution metric derived from matched RNA/DNA sequencing, as a robust stratification tool. This approach uses the tumor-to-non-tumor cell ratio to identify cancer-specific mechanisms.
While normal cells have mRNA expression that is directly proportional to chromosome number, cancer cells have an abnormal number of chromosomes. The tumor-specific total mRNA expression (TmS) biomarker factors this in, accounting for gene-expression changes relative to chromosome number in cancer cells. This biomarker also accounts for changes in RNA activity in tumor microenvironment cells compared to tumor cells.
By calculating the ratio of tumor cell to non-tumor cell (stromal plus immune cells) total mRNA expression levels per cell per haploid genome, the new approach simultaneously quantifies both tumor cell transcriptional output and the collective transcriptional activity of the surrounding microenvironment.
Outcomes
In a dataset of 575 patients with TNBC across ethnically diverse cohorts, the TmS biomarker accurately sorted patients into high-TmS (favorable prognosis) and low-TmS (poor prognosis).
The biomarker outperformed current methods to predict chemotherapy response, highlighting its potential as an effective starting point for patient stratification to optimize treatment selection.
Importantly, this prognostic biomarker applies across populations, while also highlighting key differences in the tumor microenvironments of high-TmS Western and Asian ethnic groups, which could help clinicians match additional treatments likely to work more effectively for each population.
While further validation is needed to advance this tool in the clinic, these results suggest that the TmS biomarker is a promising approach to optimize treatment selection across diverse populations.
Reference
[1] Triple-negative Breast Cancer. American Cancer Society. Online. Last accessed on February 18, 2026.
[2] Bianchini G, Balko JM, Mayer IA, Sanders ME, Gianni L. Triple-negative breast cancer: challenges and opportunities of a heterogeneous disease. Nat Rev Clin Oncol. 2016 Nov;13(11):674-690. doi: 10.1038/nrclinonc.2016.66. Epub 2016 May 17. PMID: 27184417; PMCID: PMC5461122.
[3] Bianchini G, De Angelis C, Licata L, Gianni L. Treatment landscape of triple-negative breast cancer – expanded options, evolving needs. Nat Rev Clin Oncol. 2022 Feb;19(2):91-113. doi: 10.1038/s41571-021-00565-2. Epub 2021 Nov 9. PMID: 34754128.
[4]Dai Y, Pan X, Guo S, Ji S, Cao S, Montierth MD, Jiang Y, Chang JT, Shi L, Shalapour S, Echeverria GV, Yates L, Staaf J, Lim B, Yuan Y, Wang W. Tumor microenvironment transcriptional activity enables robust stratification of chemotherapy response in triple-negative breast cancer. Cell Reports Medicine, Volume 7, Issue 2, 102610 [Article]
Featured image: Watching a mammogram-the result of x-ray examination of the mammary glands for the prevention of breast cancer. Courtesy: Photo courtesy: 20218 – 2026. Fotolia/Adobe/Evgeniy Kalinovskiy. Used with permission.
DOI
