Tumor cell death linked to helpful immune response and immune checkpoint levels in bile duct cancer

Cholangiocarcinoma (CCA) is a highly heterogeneous malignancy that could arise at any level of the biliary tree. CCA is classified into intra- (iCCA) and extra- (eCCA) hepatic cholangiocarcinoma in which eCCA can be further subdivided into perihilar (pCCA) and distal (dCCA)¹. Despite recent advances in the development of early detection and identification of novel therapeutic targets in several human malignancies, most CCA patients are typically diagnosed at advanced clinical stages, and treatment options are enormously limited because overall survival rates have by no means been improved².

The interaction between tumor and inflammatory/immune cells and chronic inflammation in the tumor microenvironment (TME) has been considered to play critical roles in tumorigenesis, progression, recurrence, and cancer patients’ therapeutic responses³. Infiltration of inflammatory and immune cells and their impact on patient outcomes have been continuously reported in CCA. High infiltration of tumor-infiltrating CD8+, CD4+, and FOXp3+ regulatory (Tregs) T cells and low CD163+ M2 macrophages (TAMs) were found to be significantly correlated with a longer overall survival in eCCA patients⁴. In addition, an immune high-risk signature characterized by high tumor-associated neutrophils (TANs), high FOXp3+, and low CD8+ T cells was recently reported to be significantly associated with poor disease-free and overall survival rates (DFS and OS) and resistance to gemcitabine treatment after recurrence in eCCA patients⁵. Therefore, immunotherapy was proposed as a potential strategy for the treatment of CCA patients^2,6. In addition, a particular subgroup of CCA patients represented as an inflamed subtype generally characterized by a massive T cell infiltration, inflammatory and immune checkpoint activation was proposed to potentially benefit from checkpoint blockade immunotherapy⁷. Therefore, systemic analysis toward a better understanding of the interplay between inflammatory/immune cells and their responses, particularly immunogenic dying cells in TME, should lead to the novel development of more efficient cancer immunotherapy.

Necroptosis is a regulated form of cell death in a caspase-independent manner. In contrast to apoptosis, necroptosis is known to elicit marked inflammatory responses and adaptive immunity. Upon necroptosis activation, receptor-interacting protein kinase 1 (RIPK1) interacts with receptor-interacting protein kinase 3 (RIPK3) through their RIP homotypic interaction motif (RHIM) domains, resulting in the formation of a necrosome complex⁸. In this complex, the mixed lineage kinase domain-like protein (MLKL) is phosphorylated by RIPK3, subsequently leading to its oligomerization and translocation to the plasma membrane, resulting in rapid membrane permeabilization^9–11 and the release of intracellular contents, including damage-associated molecular patterns (DAMPs)¹².

Accumulating studies have demonstrated that necroptosis could play important roles in several human diseases, including cancers. Necroptosis in the spectrum of immunogenic cell death (ICD) has also been proposed as a promising novel cancer therapy^13–16. On the one hand, necroptosis evokes strong adaptive immune responses and, therefore, may trigger and enhance anti-tumor immunity and cancer immunotherapy; on the other hand, the associated inflammatory responses could also promote cancer development and progression. Necroptosis of endothelial cells in the tumor microenvironment has been reported to promote tumor cell extravasation and metastasis^17,18, and necroptosis of tumor cells can also promote cancer metastasis¹⁹. RIPK3 depletion could generate an immunosuppressive tumor microenvironment and tumorigenesis in pancreatic adenocarcinoma (PDA) experimental mouse model²⁰. A recent study demonstrated that necroptosis in hepatic microenvironments can direct lineage commitment to switch from hepatocellular carcinoma (HCC) to CCA²¹. In addition, radiation-induced necroptosis was also reported to contribute to tumor cell repopulation and recurrence in colorectal cancer²². However, it is also true that necroptosis was found to function as a tumor suppressor, although its mechanisms have remained virtually unknown^23,24. Most necroptosis in cancer, however, has been reported in vitro and/or animal models, but it is entirely true that these studies are enormously limited by the lack of comparative evaluation in clinical settings.

Therefore, in this study, we systematically analyzed the associations of key necroptotic proteins, necroptosis activation, PD-L1, tumor-infiltrating immune cells and their impact on clinical outcomes and anti-tumor immunity in CCA patients. In addition, in vitro experiments were performed to provide the possible underlying mechanisms of tumor necroptosis-promoted immune cell infiltration and PD-L1 expression. Our results provide valuable information toward a better understanding of necroptosis, MLKL in CCA, and for a potential cancer immunotherapy as well as novel prognostic markers.

Necroptosis-based cancer therapy has been proposed as a potential novel strategy that could enhance cancer immunotherapy, but its functional roles in tumorigenesis have remained unknown because both pro- and anti-tumoral effects have been reported as described above. Our studies provided a list of novel findings on necroptosis in human malignancy and demonstrated the differential roles between MLKL and necroptosis activation (pMLKL) in CCA patients. MLKL, which might function independently of its role in necroptosis, was an unfavorable prognostic factor for patients with CCA. Both pMLKL and RIPK1–RIPK3 interaction were detected in human CCA primary tissues. Of particular interest, pMLKL status was significantly positively correlated with a clinically favorable immune signature (high CD8+/high FOXp3+/low CD163+), whereas MLKL was positively associated with a clinically unfavorable immune signature (low CD8+/low FOXp3+/high CD163+). In addition, this is the first study to demonstrate the relationship between necroptosis activation and PD-L1 expression and revealed that their associations also impact patient survival rates. In vitro experiments demonstrated that tumor necroptosis promoted the expression of proinflammatory cytokines and chemokine genes and induced Jurkat T cells to upregulate PD-L1 expression in CCA cells. Collectively, the results of the present study suggested that in contrast to MLKL, necroptosis activation may have anti-tumorigenic roles, and necroptosis-based treatment in combination with immune checkpoint inhibitors, could provide more efficient treatment for CCA patients.

In our previous study analyzed from the TCGA database, we demonstrated the upregulation of key necroptotic proteins, including RIPK1, RIPK3, and MLKL in CCA primary tissues compared to normal bile ducts²⁵. We confirmed the expression of RIPK3 and MLKL in CCA clinical specimens by IHC staining. Our study was the first to demonstrate that MLKL expression was significantly upregulated in CCA clinical specimens. In agreement with our study, high expression of MLKL was also reported in PDA²⁰. Interestingly, patients with high MLKL were associated with worse survival rates; this result was not consistent with other malignancies, including PDA³³, gastric cancer³⁴, colon cancer³⁵, ovarian cancer³⁶, cervical cancer³⁷, and breast cancer³⁸ in which MLKL expression level was decreased in tumor tissues, and low MLKL was associated with worse survival rates in those cancers. Hence, those studies suggested that MLKL might play an anti-tumorigenic role, probably through necroptosis activation. Further studies are required to explore the functional roles of MLKL in CCA, either through necroptosis activation or independent of its role in necroptosis.

Necroptosis has been implicated in various malignancies. However, growing evidence in most of the studies relied on experimental knockout of key necroptotic proteins, including RIPK3 and MLKL or chemical inhibitors in experimental animal models³⁹. In addition, the clinical relevance of necroptosis activation in human clinical samples has remained largely unknown. To the best of our knowledge, only few studies have reported necroptosis activation by pMLKL detection in human malignancies^19,40,41. We used the same antibody obtained from Abcam (ab187091) for IHC^19,40 and IF⁴¹ staining of pMLKL as in those studies above, but in our present study, pMLKL immunoreactivity was clearly located in the cell membrane in both HT-29 cell block, a positive control, and human CCA primary tissues compared to an elevated cytoplasmic staining or no clear staining pattern reported in the other studies aforementioned. In addition, no research has been reported regarding the application of an in situ PLA assay to determine RIPK1 and RIPK3 interaction in human clinical specimens. In the present study, RIPK1–RIPK3 interaction and pMLKL were also detected in the very same tumor cells, indicating the activation of necroptosis in CCA. It is true that the functional roles of necroptosis in cancer have remained not necessarily established and, therefore, we further explored the potential impact of necroptosis activation on the survival of CCA patients in this study. In contrast to MLKL expression, we revealed that patients with high pMLKL tended to survive for longer. However, one of the limitations in our present study was a rather unsatisfactory number of cases available for examination, thereby resulting in an inadequate statistical power. These results were inconsistent with a study in colon and esophageal patients, in which a relatively high level of pMLKL was reported to be associated with adverse clinical outcome of the patients⁴⁰. The results of a recent study in a breast cancer mouse model were also consistent with those above, in which pMLKL was increased during the progression of mouse breast cancer, indicating that necroptosis might be involved in cancer progression and metastasis¹⁹. Therefore, further investigations are required to clarify the roles of necroptosis in CCA patients.

Necroptosis associated with marked inflammatory response has been reasonably postulated to modulate the tumor microenvironment. However, none has demonstrated an association between the activation of intratumoral necroptosis and inflammatory/immune cell infiltration in human malignancies. Our study was consistent with a previous study in CCA patients, in which their preliminary data demonstrated a direct association between RIPK3 expression and the intratumoral infiltration of CD8+ T cells (only abstract available^42,43). In contrast, the analysis of PDA in an experimental mouse model with RIPK3 deletion demonstrated a significantly increased number of CD4+ and CD8+ T cells but decreased number of FOXp3+, TAMs, and myeloid-derived suppressor cells (MDSC), indicating that deletion of RIPK3 was indeed associated with increased T cell infiltration and reduced immunosuppressive myeloid cells²⁰. Therefore, this study is the first one to demonstrate the correlation of necroptosis activation and tumor-infiltrating immune cells in the actual tumor microenvironment of human malignancies.

MLKL was a clinically unfavorable independent prognostic marker and inversely correlated with a clinically favorable immune cell signature (high CD8+/high FOXp3+/low CD163+). Conversely, necroptosis activation (pMLKL) was positively correlated with a clinically favorable immune cell signature in our study of CCA patients. These results demonstrated the differential roles of MLKL and its role in necroptosis activation in CCA patients. Accumulating evidence indicates the newly identified functions of MLKL beyond necroptosis, including regulating the generation of intraluminal and extracellular vesicles^44,45 and endothelial cell adhesion gene expression⁴⁶; however, the functional roles and a deeper mechanistic understanding of non-necroptotic functions of MLKL in cancers, including CCA, require further investigation⁴⁷. Consistent with other studies in these patients, infiltration of CD8+ T cells and CD163+ M2 macrophages were both significantly associated with longer and shorter patient survival in our CCA cohorts, respectively^4,5,48–50 but the FOXp3+ regulatory T cells reported to suppress immune responses against tumor cells and to be correlated with poor clinical outcome has remained an opposing finding compared to other types of cancers⁵¹. Previous studies in two different cohorts of extrahepatic CCA patients also reported a discrepancy regarding the potential roles of FOXp3+ T cells; the results of our present study were consistent with those of Gosper et al.⁴ and also two recent reports in colorectal cancer patients^52,53. FOXp3+ T cells can be further divided into subsets based on their expression levels of FOXp3+ and CD45RA indicating naïve and effector Treg T cells and non-suppressive T cells⁵⁴. Therefore, the quantification of the percentage of non-suppressive T cells is likely to help clarify the clinical relevance of FOXp3+ in CCA and reduce the limitation in this study, although further investigations are necessary for clarification.

Necroptosis has recently emerged as a potential target of cancer therapy in conjunction with cancer immunotherapy. Necroptotic cancer cells can provide both tumor-specific antigens and DAMPs to dendritic cells, potentially also amplifying the process of antigen cross-priming and the activation of CD8+ cytotoxic T cells^13,55. In this study, we firstly demonstrated a significant positive correlation between necroptosis activation (pMLKL) and infiltration of CD8+ T cells in human cancer patients. Interestingly, CD8+ T cell infiltration was observed in tumor areas with pMLKL-positive staining. Infiltration of CD8+ T cells, a key player in anti-tumor immunity, was significantly associated with a longer DFS, suggesting that necroptosis-based cancer therapy could be a potential therapeutic option for CCA patients, improving the clinical outcomes through activation of the immune system. We conducted in vitro experiments to further explore the association of necroptosis and intratumoral CD8+ T cell infiltration. Tumor necroptosis has been shown to create an inflammatory tumor microenvironment and intratumoral infiltration of CD8+ T cells²⁸, probably mediated through cell-autonomous production of cytokines and release of DAMPs^13,29. In agreement with a previous study²⁹, we showed that necroptosis of CCA cells promoted the upregulation of proinflammatory cytokine and chemokine gene expression. The proinflammatory cytokines and chemokines, including CXCL1, CXCL2, CXCL8, CXCL10, CCL4, CCL20, MCP-1, and ICAM-1 have been shown to be involved in the recruitment of various immune cells including CD8+ T cells⁵⁶. Therefore, the production of these cytokines and chemokines upon necroptosis activation in CCA cells might contribute to the recruitment of immune cells into the CCA tumor microenvironment, supporting the positive correlation of necroptosis activation and the infiltration of CD8+ T cells in CCA patients. Further studies in an immunocompetent animal model of CCA are required to clarify the association of tumor necroptosis and intratumoral CD8+ T cell infiltration in an in vivo setting.

In conclusion, this is the first study in human cancer patients to systematically analyze and compare the associations of MLKL and necroptosis activation with tumor-infiltrating immune cells in the tissue CCA microenvironment and their impact on clinical outcomes of the patients themselves. Our findings demonstrated the differential associations of MLKL and necroptosis activation with a clinically favorable immune signature and survival and should contribute to a better understanding of the clinical significance of necroptosis activation and more likely a non-necroptotic function of MLKL. Together with the significant positive correlation between necroptosis activation and an immune checkpoint, the presence of high PD-L1 and high pMLKL in tumor cells were implicated in the longer OS. Moreover, our in vitro experiments further support the association of tumor necroptosis and intratumoral infiltration of CD8+ T cells and demonstrate the potential role of the relationship between tumor necroptosis and PD-L1 expression that is probably mediated through tumor necroptosis-promoted T cell activation. The results of our study, therefore, highlighted a novel therapeutic possibility for the combination of a necroptosis-based therapeutic approach with immune checkpoint inhibitors toward prolonging the survival time of CCA patients.