Contribución de la metabolómica en el diagnóstico histopatológico de la hepatitis C y de las hepatopatías autoinmunes

Resumen

The histopathological diagnosis of hepatitis C (HCV) and autoimmune hepatopathies such as primary biliary cholangitis (PBC) and autoimmune hepatitis (AIH) requires the integration of biochemical, serological, immunological, and molecular data, because there are no pathognomonic morphological aspects in the biopsy that characterize these diagnostic entities. In the absence of data favoring an entity, biopsy is an essential element to assist in the diagnosis, which in these cases is presumptive. Sixty-two patients were selected and divided into four groups: control (n=13); PBC (n=11); HCV (n=21); HAI (n=17). The metabolite extraction product from paraffin-embedded liver biopsy tissue, analyzed by HPLC-MS/MS, revealed 95 metabolites of which 53 met the imposed precision requirement (35%). The predictive models generated, with these 53 metabolites for the different diagnostic groups, showed good discriminatory ability for Control vs HCV (AUC= 0.88), Control vs HAI (AUC=0.81), HCV vs HAI (AUC=0.88) and excellent for Control vs CBP (AUC=0.96), HCV vs CBP (AUC=0.97), HAI vs CBP (AUC=0.92). The study of metabolic pathways from the modules generated in the network of perturbed correlations between the different diagnostic groups suggests a perturbation of carbohydrate metabolism in the Control vs HCV and HCV vs CBP groups, of amino acid metabolism in the HCV vs HAI and HAI vs CBP groups, of nucleotide metabolism in the HCV vs HAI, HAI vs CBP and control vs CBP groups, of energy metabolism in the control vs CBP group. The low discriminative capacity of the morphological variables does not contribute to improve the discriminative power of the models generated with the metabolites. We did not find a set of metabolites sufficiently discriminative to build a model to predict the degree of fibrosis and necroinflammatory activity. In summary, we demonstrated in this thesis that metabolomics might constitute an extremely relevant basis to perform adjunct technique to the histopathological diagnosis of hepatitis C and autoimmune hepatopathies. The body of knowledge reported in this thesis should be of general interest for researchers using cutting edge technologies, including metabolomics, for the refinement of histopathological diagnosis in the 21st century.