Understanding of the tumorigenic mechanisms associated with such metabolic dysfunctions has greatly progressed. Our team has been particularly active in these advances and we have published pioneer studies that revealedthe oncometabolite role of succinate, modulating 2-oxoxglutarate dependent dioxygenases such as HIF prolyhydroxylases, DNA and histones demethylases. Inhibition of these enzymes explains the pseudohypoxic and hypermethylator phenotypes of SDH-deficient PPGL, which participate to angiogenesis, dedifferentiation and EMT. Despite these major progresses, many questions remain unsolved. First, the tissue-specificity of the tumor spectra associated with these respective mutations remains a complete mystery. Mitochondrial function is indeed a ubiquitous process. However, SDHx mutations cause PPGL, rare cases of kidney cancers and GIST, while FH mutations are predominantly associated with cutaneous and uterine leiomyomas, papillary kidney cancers, and rarely with PPGL. Finally, somatic mutations in IDH1/2 genes are identified in 80% of secondary glioblastomas, as well as in chronic myeloid leukemia and chondrosarcomas or cholangiocarcinomas. It is still unknown if the recently identified genes (MDH2 and SLC25A11) are to be found in other tumor types than PPGL. Another if not the most crucial question in the field is the heterogeneity in the prognosis associated with these different genes mutations. Indeed, our team has demonstrated that mutations in the SDHB, FH and SLC25A11
genes are specifically associated with malignancy and (at least for SDHB), poor prognosis, while SDHA, C or D mutations are mostly found in benign tumors. Similarly, IDH mutations are considered to be of good prognosis in glioma. The aim 3 will be to elucidate this phenotypic heterogeneity in TCA cycle associated cancer predispositions, both in terms of tissue-specificity and clinical outcome.