The Dark Side of Innate Immunity Fosters Tumor Growth and Offers New Diagnostic

In this issue of Clinical Cancer Research, Duan and colleagues investigate the role of mucosal-associated invariant T (MAIT) cells in the hepatocellular carcinoma (1). The merits of this work are several, and I will first discuss the basic immunologic aspects to later highlight their present and future translational value.

The tumor landscape and its prognostic value named “Immunoscore” by the French school (2) revitalized the field of tumor immunology in the end of the last century, anticipating the anti-immune checkpoint era and the related cancer immunotherapy revolution.

The “Zeitgeist” (spirit of the age) is positively inspiring the work of Duan and colleagues. They start their research making clear that innate immunity plays a crucial role in the tumor progression and that inhibitory receptors such as PD1, CTLA-4 and TIM-3, also known as immune checkpoints, may tilt the balance between immune surveillance and immunotolerance to cancer cells. They start their research focusing on the quantity and quality of the MAIT cells infiltrating the hepatocellular carcinoma lesion. The data reported showed a reduced amount of MAIT cells in the hepatocellular carcinoma lesion when compared with healthy liver tissues. On the other hand, the quality check showed that the phenotype resembled a classical effector memory characterized by the expression of CD45RO⁺CD95⁺ and lack of CCR7. Other phenotype changes were the downregulation of an additional two markers, CD160 and CD127, which are also associated with two other innate immunity cell subsets: natural killer (NK) cells and innate lymphoid cells (ILC). The chemokine receptor expression analysis leads to the appreciation of the mechanism behind the lack of MAIT cell infiltration in hepatocellular carcinoma tumor lesion. Several gut-homing receptors such as CXCR16, CCR6, and CCR9 were found downregulated on MAIT cells' surface indicating an aberrant chemotaxis.

The most intriguing and up-to-date part of the study shows up when the authors focus their attention on the possible role of the MAIT cells in the hepatocellular carcinoma pathogenesis. Considering the immunosuppressive tumor environment, they start to investigate the possible expression of inhibitory receptors on the MAIT cells' membrane. A clear increase in the expression of canonical immune checkpoints was found: higher levels of PD1 were detected on MAIT cells present in the tumor microenvironment compared with those found in the peri-tumor area. Moreover, other two immune checkpoints, CTLA-4 and TIM-3, were found to be increased and coexpressed on MAIT cells. This is an important finding since, so far, the general idea in the field has been that the treatment with anti-immune checkpoint targets mostly T cells compartments (CD4⁺, CD8⁺, or T regulatory cells) unleashing their antitumor effector mechanisms. Thus, the data by Duan and colleagues suggests that in the late stage of the cancer disease, the immune system could become educated by the tumor and foster the disease progression. Furthermore, the study sheds a new light on the possible beneficial effect of the anti-immune checkpoint therapy. The use of anti-PD1, anti–CTLA-4, and anti–TIM-3 antibodies may trigger antibody-dependent cell cytotoxicity that can lead to physical elimination of protumorigenic exhausted components of immune system (Fig. 1).

To understand why and how the hepatocellular carcinoma–associated MAIT memory effector cells express immune checkpoints and inhibitory receptors, the authors wisely model their in vitro findings. In coculture experiments, they demonstrate that the addition of hepatocellular carcinoma to MAIT cells culture induced the expression of PD1, CTLA-4, and TIM-3. The authors speculate correctly that the tumor cells may educate MAIT cells. However, the nature of the signal/s that induce/s the exhausted phenotype and the immune checkpoints expression on MAIT cells remain obscure. Here, a broader scenario should be considered and at least two hypotheses, not mutually exclusive, could be formulated:

• (i) The tumor may produce chemokines ectopically that might recruit the exhausted MAIT cells, as described for the aberrant production of IL8 by melanoma cells. Then, the exhausted MAIT cells can foster the tumor growth as proposed for M2 macrophages (3).

• (ii) The immune cells, following editing of the tumor population accordingly with Schreiber's “Tumor Editing Theory,” become physiologically exhausted or tolerant, allowing the nonimmunogenic tumor cells' growth (4).

The cytokine production demonstrated a shift from the production of IFNγ and Th1/Th17 able to support the immunity against cancer to a tolerogenic profile composed by IL10, IL4, and IL22. Another consideration that should stem from the manuscript of Duan and colleagues is the following: the immune system chronically stimulated by tumor antigens during the long clinical course may activate peripheral tolerance mechanisms to avoid autoimmunity. It has been recently demonstrated in oncological patients that the resistance to anti–CTLA-4 (ipilimumab) therapy depends on the induction on effector cells of alternative immune checkpoints (PD1, TIM-3) due to the chronical stimulation by proinflammatory cytokines present in the patients' sera (5), as it might be the case for MAIT cells in Duan and colleagues' study.

The loss of protective functions by the MAIT cells is observed by the poor amount of perforin and granzyme B with relative loss of cytotoxicity against hepatocellular carcinoma targets.

The translational value of the research comes with the stratification of the absolute MAIT cell number measured in the tumor lesion and the patient overall survival (OS) or relapse-free survival (RFS). Using both clinical scores, an inverse correlation was demonstrated between MAIT cell number and prognosis. It will require further effort to translate these finding in the clinical context and use the exhausted MAIT cells infiltrate as prognostic markers for hepatocellular carcinoma, but this will contribute to increase the precision medicine handling of patients with hepatocellular carcinoma.

Another interesting aspect is the contribution of Duan and colleagues' data in the design of new therapy intervention in patients with hepatocellular carcinoma. Both PD1 and TIM-3 inhibit IFNγ production and cytotoxicity mediated by NK and T cells. Thus, if this turns out to be the case also for MAIT cells, it is conceivable to suggest that anti-immune checkpoint therapy could be well suited also for patients with hepatocellular carcinoma.

Other cells of innate immunity share with MAIT cells an exhausted phenotype inside the tumor infiltrate as NK cells. It will be relevant to extend the analysis on the innate immunity tumor landscape and immunoscore for at least two reasons: C. Janeway Jr demonstrates that innate immunity plays a major role in the activation of the immune system during infectious diseases, and this may be true also for cancer lesion. The innate immunity does not rely on antigen presentation by polymorphic genes as for adaptive immunity. Thus, we can expect to find common prognostic molecular patterns associated with a given tumor shared among the population.

No potential conflicts of interest were disclosed.

This work was supported by Associazione Italiana per la Ricerca sul Cancro (AIRC) grant 15521, Wenner-Gren Stiftelse, NIBIT Foundation, and Fondazione Melanoma ONLUS Naples.