Review: Ketogenic diet in the treatment of cancer – Where do we stand? 評論：生酮飲食治療癌症-我們的立場是什麼？

4.2. Ketogenic diet targets mitochondrial metabolism of cancer cells

Some tumor entities are not able to properly respire due to a dysfunctional OXPHOS system. The mode of downregulation of OXPHOS can differ in different types of cancer. Thus, some tumors show a reduction of mitochondrial mass, others a reduction of all OXPHOS complexes and some, such as paragangliomas and oncocytic tumors, have pathogenic mutations in OXPHOS genes [110], [111], [112], [113], [114]. Tumors with dysfunctional mitochondria or decreased mitochondrial activity seem to compensate their energy requirements by aerobic fermentation [115]. Replacing glucose by ketone bodies requires that the tumors have functional mitochondria to be able to use ketone bodies efficiently for growth and survival. Thus, tumors with dysfunctional or low levels of mitochondria might suffer from high metabolic energy stress triggered by a KD [51], [60], [115], [116]. Analysis of the cellular energy sensor AMP-activated protein kinase (AMPK) in neuroblastoma tumors revealed that the KD increased the levels of activated AMPK [51].

On the other hand, tumor mitochondria can possess high activity in terms of respiration and energy production [117], [118], [119]. The question is whether the KD may also target tumors that have functional mitochondria. Rapidly growing tumors develop hypoxic areas in which oxygen supply is sparse [117]. Due to the capability of tumor cells to metabolize ketone bodies solely if enough oxygen is available [120], tumor cells at hypoxic sites would fail to produce energy from ketone bodies even though these cells have functional mitochondria.

The three mitochondrial enzymes SCOT, BDH1, and ACAT1 are key players in ketone body utilization. Thus, therapeutic efficacy might be influenced by the expression of these enzymes. For example, neuroblastoma and pancreatic cell lines and mouse xenografts with very low or no SCOT expression can be targeted by ketone bodies and KD [62], [63], [121]. In a recent clinical trial, differential expression of ketolytic enzymes (including BDH1 and OXCT1) was described in gliomas. The authors hypothesized that patients with low or very low expression of BDH1 and OXCT1 in malignant gliomas may respond better to KD therapy than patients with gliomas that express higher levels of ketolytic enzymes [26], [62]. In contrast, it has been shown that cancer cells of different origin can indeed take up and metabolize ketone bodies [122], [123], [124], [125]. In vitro analyses of several different breast cancer cell lines revealed that physiologic concentrations of ketone bodies did not reduce cell proliferation independent of the expression level of ketolytic enzymes [126]. Moreover, in a rat model of glioma, where the tumor cells were competent in the transport and oxidation of ketone bodies, a KD had no effect on cancer growth [65]. Taken together, it is still unclear whether ketone bodies play a major causal role in the antitumor effect of KDs.
4.3. Ketogenic diet targets amino acid metabolism of cancer cells

Based on the results of several animal model studies, the KD alters amino acid (AA) metabolism and urea cycle metabolites [51], [65], [127], [128], [129], [130]. The most consistent and pronounced changes observed were decreased blood levels of most essential AAs in mouse or rat [51], [128], [130]. In addition, in different studies, alterations of metabolism of other AAs such as glutamate/glutamine, glycine, serine, proline, tryptophan, and aspartate were reported [51], [65], [127], [128], [129], [130]. Douris et al. concluded that the KD led to down-regulation of AA catabolic processes in mice to conserve AA levels [128].

In a preclinical neuroblastoma model, reductions of essential AAs and urea cycle metabolites in plasma and tumors were induced by low-protein KDs, whereas the plasma levels of serine, glycine and glutamine were elevated [51]. Mouse models of glioma administered a KD also showed higher levels of glutamate in the cortex and tumor tissue [65]. In agreement, a clinical study reported increased levels of glutamine and/or glutamate in some patients with brain tumors after administration of a KD [131]. Considering the dependence of a range of tumor cells on glutamine and glutamate metabolism, it is surprising that the observed elevated levels of these AAs did not trigger tumor proliferation.

The impact of the KD on down-regulation of essential AAs most likely contributes to the inhibition of tumor growth, but this needs further investigation. It can be postulated that the reduction of essential AAs might result from relatively low amounts of protein in the KD. In contrast, Aminzadeh-Gohari et al. found neither a reduction of plasma essential AAs nor a reduction in tumor growth in mice fed a control diet containing the same low amount of protein as the KD [51].