Source: Nmr in Biomedicine, by Irene Garcia‐Bocanegra; Jesus David Urbano‐Gamez; Antonio Perez‐Cardona; Laura Barrios; Maria Isabel Somoza‐Ramírez; Luis Vicioso; Elena Pardo‐Susacasa; Gema Diaz‐Cordoba; Maria Mercedes Acebal‐Blanco; Francisco Sendra‐Portero; Maria Luisa Garcia‐Martin (September 2, 2026). AI-generated summary by biochip.com, published . Not independently reviewed.
Key takeaways
- Researchers used HR-MAS NMR, LCModel and ERETIC2 to absolutely quantify metabolites in intact breast cancer tissue.
- Generalized linear models and transcriptomic correlations examined links between metabolite levels and immunohistochemical breast cancer subtypes.
- The approach does not yet establish whether metabolic profiling improves patient outcomes or clinical treatment selection.
Researchers in Málaga and Seville combined high-resolution magic-angle-spinning nuclear magnetic resonance, or HR-MAS NMR, with gene-expression analysis to examine metabolic differences among immunohistochemical breast cancer subtypes. The approach measures small molecules in intact tumor tissue while preserving much of the tissue's original biochemical setting. Rather than grinding a sample into an extract, HR-MAS NMR spins it at a carefully chosen angle that sharpens the spectral signals produced by metabolites. Irene Garcia-Bocanegra and colleagues used LCModel software and an ERETIC2 quantitative reference to calculate absolute metabolite concentrations in breast cancer specimens. They then applied generalized linear models to test how those measured metabolites related to tumor classification by immunohistochemistry, the laboratory staining method used to identify tumor markers. The team also correlated metabolic measurements with transcriptomic data, meaning information about which genes are actively being expressed. Together, these layers of measurement are intended to clarify why breast cancers that appear similar under one classification system can still behave differently at the molecular level.
Reading Tumor Chemistry Without Destroying the Sample
Metabolites are the small chemical products and ingredients of life, including molecules involved in energy production, membrane building, and nutrient use. They can offer a close-up view of what cells are doing at a particular time. In cancer, that chemistry can shift as tumor cells grow, adapt to limited nutrients, or interact with neighboring cells.
HR-MAS NMR provides a way to measure those molecules directly in intact tissue. It is a little like spinning a textured object until the visual blur becomes smooth enough to inspect: rapid spinning at the so-called magic angle reduces distortions that would otherwise broaden NMR signals in solid or semi-solid samples. The resulting spectrum contains peaks that can be matched to individual metabolites.
A Quantitative Workflow for Complex Spectra
Breast tumor spectra are difficult to interpret because many metabolites produce overlapping signals. Garcia-Bocanegra's team used LCModel, a computational fitting method that separates a complex spectrum into contributions from known compounds. This makes it possible to estimate individual metabolite levels even when their signals are not cleanly isolated.
The researchers paired LCModel with the Electronic REference To access In vivo Concentrations, or ERETIC2, method. ERETIC2 supplies a quantitative reference signal, allowing the analysis to move beyond relative peak comparisons toward absolute metabolite quantification. That distinction matters because a peak that looks larger in one sample may reflect measurement conditions as well as biology unless it is properly referenced.
Connecting Metabolites With Breast Cancer Subtypes
Breast cancers are commonly divided into immunohistochemical subtypes based on markers detected by staining tumor tissue. Those markers help guide clinical decisions, but they do not capture every biological difference between tumors. Metabolic profiling offers another layer of information by asking what chemical pathways are active within the tissue.
The team used generalized linear models to evaluate associations between quantified metabolites and immunohistochemical classification. These statistical models are designed to test relationships between variables while accommodating different kinds of biological data. The design therefore treats subtype not simply as a label, but as a feature that may be connected to measurable changes in tumor chemistry.
Adding the Genetic Activity Layer
Metabolites tell researchers about biochemical activity, while transcriptomics measures RNA molecules that reflect which genes a cell is using. Combining the two is comparable to looking at both a factory's inventory and its operating instructions: one view reveals the materials and products present, while the other indicates which production programs are switched on.
The transcriptomic correlation analysis was intended to identify links between metabolite concentrations and gene expression. Such links can help researchers develop hypotheses about the biological processes behind metabolic variation across tumors. Importantly, a correlation does not by itself prove that a gene-expression change causes a metabolic change, but it can identify patterns worth testing in future experiments.
Why Standardization Is Central
The investigators highlighted a broader practical problem in HR-MAS metabolomics: results can be hard to compare across studies when laboratories use different collection, processing, measurement, and quantification procedures. A metabolite signature is only useful if it can be measured reliably across samples and research settings. Standardized workflows are especially important for breast cancer, where tissue composition and tumor biology can vary substantially.
Using a defined spectral-analysis strategy and an external quantitative reference addresses part of that challenge. The combination of LCModel and ERETIC2 aims to make metabolite estimates more robust and reproducible than simple visual comparisons of spectral peaks. It also creates a clearer basis for integrating NMR results with transcriptomic datasets.
Why This Matters
Breast cancer is not one disease, and molecular classifications already recognize major differences among tumors. Yet tumors within the same broad subtype can still differ in growth patterns, treatment response, and underlying biology. Metabolic measurements may help expose some of that hidden diversity by capturing the chemical consequences of gene activity and cellular behavior.
This work positions intact-tissue HR-MAS NMR as a complement to established pathology and genomic methods, not as a replacement for them. Its value lies in bringing a direct biochemical readout into a multi-omics framework. Future studies will need to determine how consistently these metabolic patterns reproduce in additional tumor collections and whether they can contribute useful information for clinical decision-making.