Heterogeneity in Human NSCLC Tumor Glucose Metabolism: On the Origins and Role of Pyruvate Metabolism

In 1956 Otto Warburg, the most prominent scientist in cancer metabolism, stated that "the problem of cancer is...to discover the differences between cancer cells and normal growing cells" (Warburg, 1956a). Over fifty years later, the field still lacks a valid experimental framework to discover such metabolic differences in human tumors. A major limitation is the inability to faithfully recapitulate the microenvironment of primary human tumors in model systems. As a result, fundamental questions about tumor metabolism, including the suppression of pyruvate oxidation upon transformation proposed by Warburg more than 50 years ago, have only rarely been subjected to direct experimental assessment. To provide a direct readout of primary human tumor metabolism in vivo, we have used intra-operative 13C-glucose infusions in non-small cell lung cancer (NSCLC) patients to compare metabolism between tumors and non-cancerous lung. Pre-surgical imaging, including non-invasive assessment of tissue perfusion using dynamic contrast enhanced magnetic resonance imaging (DCE-MRI), allowed us to select areas of microenvironment-based heterogeneity, to guide sample acquisition. Specifically, this microenvironment-based heterogeneity was assessed relative to the oncogenotype, histological parameters, and metabolism of glucose through glycolysis and the TCA cycle. Diverse tumors displayed enhanced glycolysis and glucose oxidation. Furthermore, we discovered that due to the low enrichment in acetyl-CoA and other TCA cycle intermediates, all tumors had evidence for oxidation of multiple nutrients. We identified lactate as a carbon source for tumor oxidative metabolism. Additionally, metabolically heterogeneous regions were identified within and between tumors using DCE-MRI. Regions of lesser contrast enhancement demonstrated higher 13C enrichment, likely reflecting contributions of non-glucose nutrients to central carbon metabolism in well-perfused areas, or the cause or consequence of aberrant proliferation of aggressive clones resulting in inadequate perfusion. The data indicate that the heterogeneous metabolism of these tumors is highly and predictably related to the microenvironment. In summary, we have made novel, significant progress in assaying and analyzing primary human tumor metabolism and its relation to the microenvironment in vivo. I close with a separate project for future directions to begin to dissect the cellular origins of the whole tumor fragment signal that is amenable to direct assays in patients.