Even though Everolimus has been investigated in a phase II randomized trial as a host-directed therapy (HDT) to treat tuberculosis (TB), an oncological patient treated with Everolimus for a neuroendocrine pancreatic neoplasia developed active TB twice and a non-tuberculous mycobacterial (NTM) infection in a year and a half time span. To investigate this interesting case, we isolated and genotypically characterized the Mycobacterium tuberculosis (Mtb) clinical strain from the patient and tested the effect of Everolimus on its viability in an axenic culture and in a peripheral blood mononuclear cell (PBMCs) infection model. To exclude strain-specific resistance, we tested the activity of Everolimus against Mtb strains of ancient and modern lineages. Furthermore, we investigated the Everolimus effect on ROS production and autophagy modulation during Mtb infection. Everolimus did not have a direct effect on mycobacteria viability and a negligible effect during Mtb infection in host cells, although it stimulated autophagy and ROS production. Despite being a biologically plausible HDT against TB, Everolimus does not exert a direct or indirect activity on Mtb. This case underlines the need for a careful approach to drug repurposing and implementation and the importance of pre-clinical experimental studies.