Metabolic Influences on Complex Tumor Neighborhoods

Project Summary Renal cell carcinoma is among the ten most prevalent malignances in the United States, exhibiting an increased incidence in both men and women since 2001. The most common kidney cancer subtype is “clear cell” renal cell carcinoma (ccRCC), which accounts for ~75% of all cases. For early-stage disease, surgical resection of ccRCCs can be curative, although survival drops significantly for advanced, metastatic cancers. Multiple therapies are now available to ccRCC patients, including anti-angiogenic VEGF/receptor tyrosine kinase inhibitors, immune checkpoint blockade, mTORC1-based drugs, and a novel HIF-2a inhibitor. However, not all patients respond to these treatments and five-year relapse rates now approach 40%, and the majority of these cases develop metastases. Importantly, ccRCCs lack common oncogenic mutations observed in other human cancers, including PI3K, PTEN, TP53, and KRAS, which hinders successful treatment using corresponding targeted therapies. Instead, we have generated copy number variation, transcriptomic, and metabolomic data to identify multiple metabolic pathways that are universally altered in ccRCC tumors. These include loss of the gluconeogenic enzyme fructose-1,6-bisphosphate 1 (FBP1) and urea cycle enzymes, including argininosuccinate synthetase 1 (ASS1), argininosuccinate lyase (ASL), and arginase 2 (ARG2). Finally, ccRCCs exhibit unusually high numbers of lipid droplets, organelles which store triglycerides and cholesterol esters and a hallmark of this disease. By delineating the molecular consequences of these universal metabolic changes, we have developed new therapeutic strategies to target most patients diagnosed with this kidney cancer subtype. Moreover, our findings have been extended to other cancers such as hepatocellular carcinoma (HCC) and soft tissue sarcoma (STS) which appear to engage in highly similar metabolic reprogramming. Our data demonstrate that “senolytics” like ABT-263 could be deployed for the treatment of HCC, whereas ITX-5061, an inhibitor of the HDL cholesterol transporter SCARB1, may be effective for treating ccRCC. The results are paradigm-shifting in that understandable skepticism remains regarding the utility of “drugging” cancer metabolism, considering the metabolic heterogeneity, plasticity, and redundancy observed in various cancers. However, our results using autochthonous in vivo tumor models provide a rationale for deeper exploration. Ongoing and future work will investigate how consistent metabolic adaptations within the tumor parenchyma impact stromal components, such as fibroblasts and immune cells, based on an arsenal of complementary in vitro and in vivo models, that include novel autochthonous HCC and STS mouse models and ccRCC and HCC patient derived xenografts and organoids. A principal conceptual innovation of our recent work is the demonstration that multiple metabolic networks are consistently altered (~100%) in genetically diverse cancers like ccRCC, HCC, and STS, and the identification of novel, highly feasible therapeutic strategies.