The efficacy of some anti-cancer agents, including anthracyclines and oxaliplatin is based on their capacity to induce immunogenic cell death (ICD) in tumor cells. This peculiar type of apoptosis is defined by a sequential emission of specific immunogenic signals from the dying tumor, which in their correct spatio-temporal appearance ignite a specific immune response against therapy resistant and dormant tumor (stem) cells. Thus the early membrane exposure of the ER-resident molecular chaperone calreticulin (CRT) constitutes a critical uptake signal for the engulfment of dying tumor cells by dendritic cells (DCs). Then, at later stages, the autophagy-dependent secretion of ATP and its binding to purinergic receptors on DCs activates the NLRP3 inflammasome. Subsequent release of IL-1 by DC triggers the polarization of IFN-γ producing CD8+ T cells. Finally, during secondary necrosis, the release of the pro-immunogenic high-mobility group box 1 (HMGB1) protein and its interaction with Toll like receptor 4 (TLR4) on DCs facilitates an optimal antigen presentation to T cells. Thus ICD contributes to the tumoricidal activity of chemotherapy and protecting the host from relapse. In order to identify thus far unknown inducers of ICD, a high content screening of compound libraries approved by the Food and Drug Administration (FDA) was conducted by means of robotized automated bioimaging combined with ICD biosensors allowing for the detection of CRT relocation, ATP secretion and HMGB1 release. This multiparametric approach led to the identification of cardiac glycosides (CGs), already well known for their preferential cytotoxic activity on cancer cells, as effective inducers of ICD. The hit compounds were validated by alternative methods in vitro, followed by a mechanistic study of GC induced ICD. Results indicated an on-target inhibition of the Na+/K+ ATPase subunit 1, which in turn interfered with the Ca2+-homeostasis of the target cell, an effect that could be mimicked by Ca2+ ionophores. We then showed in different mouse models that tumor cells killed with a combination of GC and non-immunogenic chemotherapy (cisplatin or mitomycin C) have the ability to immunize syngeneic mice against rechallenge with living cells. In addition the antineoplastic effects of these DNA damaging agents in vivo were increased by GCs in immunocompetent but not in immunodeficient mice. Finally, retrospective clinical analyses revealed that the administration of the GC digoxin during chemotherapy had a significant positive impact on overall survival in cohorts of breast, colorectal, head and neck, and hepatocellular carcinoma patients, especially when they were treated with agents other than anthracyclines and oxaliplatin.
