Breast cancer is a heterogeneous disease with multiple biological, molecular and histological subtypes. In such a context, metabolomics is a promising approach, since it is as close as possible to the patient's phenotype, paving the way towards the identification of signature biomarkers and consequently towards better patient stratification. Studying metabolites in breast cancer could help to understand how a defective metabolism can trigger cancerous processes, and alterations in the metabolome may also be used as potential indicators of cancer aggressiveness. According to the authors of the present study, there is no published work that specifically distinguishes high-grade breast tumors by metabolomics.
In this study, researchers performed an untargeted metabolomic analysis of breast tumors in 51 patients treated at the Nice Anti-Cancer Center between 2013 and 2016 (stage I to IIIB cancers) and 49 patients treated at another institution (Dijon Anti-Cancer Center) between 2007 and 2012 (stage IIA to IV cancers). The first series of patients was used to create a "training" database, to provide a discriminating model, and the second series provided "validation" data. The authors identified a metabolomic signature for high-grade invasive tumors with AUCs of 0.88, representing a nearly 90% probability of ability to discriminate high-grade versus lower-grade tumors. They identified several biomarkers of tumor aggressiveness, such as N1,N12-diacetylspermine and catabolites of tryptophan metabolism, both of which are involved in the inhibition of the immune response.
This work provides new information on the biological mechanisms underlying the aggressiveness of breast tumors. The biomarkers identified should make it possible to develop new strategies for better stratification of patients in the various immunotherapy clinical trials, and ultimately to obtain personalized medical care.
Contacts : Thierry Pourcher (thierry.pourcher@cea.fr ou thierry.pourcher@univ-cotedazur.fr) or Caroline Bailleux (caroline.bailleux@nice.unicancer.fr)
- The Scarff-Bloom Richardson (SBR) grade is a histopronostic classification of breast cancer into three grades of malignancy, based on the degree of tumor differentiation, nuclear atypia and mitotic activity.
- Metabolomics is the science of metabolites, small compounds (usually less than one kilodalton) formed during metabolism, whether end products or intermediates. Targeted metabolomics aims to identify a pathway or metabolite of interest on the basis of a previously identified relationship. The non-targeted approach aims to identify and quantify as many metabolites as possible in a sample.
- The chosen probabilistic model is the Area Under the Curve (AUC), which can be interpreted as the probability that, among two subjects chosen at random, for example, one sick and one not sick, the value of the marker is higher for the sick than for the not sick. Thus, an increase in AUC indicates an improvement in discriminatory abilities, with a maximum of 1.0 (100%).