Introduction: Metabolic adaptation occurs in transformed cells and can be exploited clinically for diagnosis and disease monitoring. Metabolic profiles can be influenced by genetic heterogeneity, both within and between tumors. Breast tumors show considerable intra and inter tumor genomic heterogeneity, which partly accounts for the variable clinical course of the disease and response to treatment. Imaging tumor metabolism may provide a metabolic signature that can be used to stratify patients.

Methods: We have investigated eight patient-derived xenograft (PDX) models of breast cancer to determine whether metabolic imaging can be linked to different “integrative clusters” (Int-Cluster), which are defined by their distinctive genomic profiles and clinical outcomes. Metabolic imaging was performed using PET measurements of the uptake, phosphorylation, and trapping of the glucose analog, 2-deoxy-2-[18F]fluorodeoxyglucose ([18F]FDG), catalyzed by hexokinase II, and 13C magnetic resonance spectroscopy (MRS) measurements of hyperpolarized 13C label exchange between [1-13C]pyruvate and the endogenous tumor lactate pool, catalyzed by lactate dehydrogenase A.

Results: Lactate labelling was higher in Int-Clusters dominated by ER- basal-like PDXs from patients with a poor short-term prognosis. Lower levels of lactate labelling were observed in Int-Clusters dominated by estrogen receptor positive (ER+) PDXs and the Pam50 subtypes, Luminal A and B, from patients which have an intermediate to good prognosis. More importantly, tumors that had a similar genomic profile nevertheless showed differences in lactate labeling, that was not observed in the [18F]FDGPET measurements. Although lactate labelling was low in the ER+ tumor models, decreases in lactate labeling could be used to detect treatment response, for example to drugs targeted at the most frequently mutated oncogene in this tumor subtype, PI3KCA.

Discussion: These results illustrate the potential importance of metabolism-based stratification of breast cancer and demonstrate that metabolic imaging can provide diagnostic information that could be important for non-invasive diagnosis and for designing and monitoring therapeutic strategies.

Citation Format: Susana Ros, Alan J. Wright, Flaviu Bulat, Wendy Greenwood, De-en Hu, Maurizio Callari, Ming Li Chia, Violeta Serra, Oscar Rueda, Alejandra Bruna, Carlos Caldas, Kevin M. Brindle. Metabolic heterogeneity in breast cancer integrative clusters to screen metabolic vulnerabilities for diagnosis and treatment response [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2021; 2021 Apr 10-15 and May 17-21. Philadelphia (PA): AACR; Cancer Res 2021;81(13_Suppl):Abstract nr 2817.