Metabolic oncology: clonal selection and the pre-existing origins of cancer resistance

Treatment resistance remains the principal cause of cancer mortality, conventionally modeled as an adaptive response arising under therapeutic pressure. However, a substantial component of resistance is plausibly established before therapeutic exposure, determined by the metabolic environment of the tissue in which the malignancy arises. This article advances the hypothesis that chronic metabolic dysregulation constitutes a selective pressure acting on premalignant tissue. Hyperinsulinemia and increased bioavailable IGF-1 sustain PI3K–AKT–mTOR and RAS–MAPK signaling in tissues where insulin resistance has preserved mitogenic sensitivity, creating an environment permissive to clonal selection. Metabolic intermediates serve as obligate substrates for chromatin-modifying enzymes, coupling metabolic state to the epigenome across the years of subclinical dysregulation that precede diagnosis. Within this landscape, hypoxia stabilizes G9a and promotes locus-specific H3K9 methylation at apoptotic and differentiation loci, whereas one-carbon and S-adenosylmethionine status determine the available H3K9me1 substrate pool. The consequence is an elevated apoptotic threshold and enrichment of epigenetically primed, apoptosis-resistant cells before a tumor is histologically apparent or clinically detected. The argument is fundamentally Darwinian: the metabolic environment selects its survivors before any therapeutic intervention. Clonal dynamics subsequently attributed to therapy may therefore reflect, in part, selection of a pre-existing, resistance-primed population. This is advanced not as an alternative to clonal evolution, of which selection of pre-existing clones is a constituent process, but as a determinant of the pre-treatment resistance heterogeneity upon which therapeutic pressure acts. Existing data on metabolic intervention are discordant, and the discrepancies appear more consistent with differences in intervention potency, timing, and trial design than with failure of the proposed mechanism. Much of the supporting evidence remains indirect and derives from heterogeneous experimental settings. The observations that would distinguish this framework from reverse causation and treatment-induced plasticity are specified, together with its central untested prediction: that pretreatment metabolic state correlates with baseline subclonal architecture, chromatin state, and treatment outcome.