Background: Although metabolic disfunction is one of the hallmarks of cancer and metabolism-based therapies are a key class of chemotherapeutics, there remains no systematic way to identify metabolism-based targets for therapeutic intervention. Because of this deficiency, metabolism-based therapies offer a largely untapped area for therapeutic discovery, especially for difficult to target tumors, like basal-like breast cancer (BLBC). Adding to the challenge of discovering metabolism-based therapeutics for BLBC, however, is the heterogeneity of tumor behavior across patients, both in terms of aggressiveness and molecular regulation.
Methods: We addressed these challenges by introducing a systems medicine-based technique integrating metabolic network modeling with single-patient RNA-sequencing (RNA-seq). We then applied this technique to investigate metabolic weaknesses in primary BLBC. We constructed a genome-scale metabolic model of breast cancer (ecBRCA) along with RNA-seq profiles constructed from The Cancer Genome Atlas (TCGA) to predict metabolic targets whose inhibition is lethal to BLBC tumors. We then evaluated the effectiveness of drugging these targets across patients by simulating the effect of a targeted knock-out on primary tumors with published RNA-seq profiles from 534 BLBC patients in TCGA and the Swedish Cancerome Analysis Network for Breast (SCANB).
Results: We found multiple metabolic targets that may prove clinically useful for complementary treatment of primary BLBC tumors. Among these, we found UDP-Glucose 6-Dehydrogenase (UGDH) and Heparan Sulfate 2-O-Sulfotransferase 1 (HS2ST1) to be particularly promising protein targets for chemotherapy in BLBC, with 80-100% of simulated BLBC tumors responding to their inhibition. Currently FDA-approved drugs down-regulating these proteins, such as metformin (for UGDH) and surfen (for HS2ST1), are potentially viable complementary treatments for BLBC patients. Additionally, we found several metabolite targets whose inhibition may inhibit BLBC growth, including leukotriene A4, creatinine, vitamin D3, and the oncometabolites 2- and 3-hydroxy-glutarate.
Conclusions: Taking a systems medicine-based approach to understanding BLBC metabolism allows for a systematic way to identify metabolic targets for therapeutic intervention. Additionally, certain metabolic weaknesses appear to be conserved in BLBC tumors across patients, including a dependence on UGDH and HS2ST1, which could be treated with currently available drugs like metformin and surfen.