CRC is one of the leading causes of cancer death in the United States, accounting for approximately 9% of all cancer deaths [1]. The major therapies for CRC include surgery resection, radiation, chemotherapies, and targeted therapies. The two major targeted therapies are Bevacizumab (Avastin) and Cetuximab (Erbitux), which targets VEGF and blood vessel formation, and epidermal growth factor receptor), respectively [1]. However, these anti-EGFR drugs do not work in CRC with mutations (defects) in the KRAS, NRAS, or BRAF gene [1]. Anti-VEGF agents have been widely used in the treatment of many solid cancers, including CRC [2,3]. However, the systemic administration of anti-VEGF monoclonal antibodies has been reported to cause tremendous adverse events including hypertension, impaired wound healing, hemorrhage and thrombosis, cardiac impairment, stroke, gastrointestinal perforations, and kidney disease [2,3]. These side effects reduce patient compliance and limit the application of anti-VEGF, particularly at the most effective dosage, on tumor therapy [2]. In colorectal cancer patients who were treated with bevacizumab combined with chemotherapy regimens, the median progression-free survival of metastatic colorectal cancer patients remained less than one year [2].

In recent years, immunotherapies have been devel¬oped to stimulate the capacity of immune effectors such as T cells to attack and eliminate tumor cells. Many immune therapies target inhibitory receptors or pathways that prevent or limit antitumor T cell responses and reactivate immune cells. Immune checkpoint blockade (ICB), including anti-PD1 and anti-CTLA4, is a strategy that directly tar¬gets T cells and re-stimulates them [4]. These immune-checkpoint inhibitors have shown clinical efficacy for some tumors, but not for many other tumors, which remain refractory or less susceptible to immune therapy [5,6]. This is the case in colorectal cancer. Most CRC belongs to the microsatellite stable (MSI-S) group, are refractory to the ICB treatment [7], despite occasionally exhibiting exceptionally good infiltration by immune cells. A major challenge in the field of immune therapy has thus become the understanding of primary and acquired resistance to ICB. Resistance to ICB may partly contribute to the redundancy of inhibitory receptors and pathways. Although mechanisms for resistance and/or insensitivity to current checkpoint inhibitors have been described [8], there are more mechanisms for tumor immune modulation yet to be discovered. Thus, we aimed to discover the new immunomodulators and investigate novel pathways in CRC.

Wnt signaling regulates many aspects of cellular processes. Its dysregulation causes developmental defects and diseases including cancer [9-11]. The Wnt/β-catenin signaling pathway is among the most studied. This pathway is deregulated in many tumors, particularly those lacking the gene adenomatous polypo¬sis coli (APC), a critical negative regulator of this pathway [12]. Wnt proteins bind to two types of cell-surface receptors: low-density lipoprotein receptor-related proteins 5 and 6 (LRP5 and LRP6, respectively), and Frizzled proteins, to initiate downstream pathways [9,13]. DKK1 and DKK2 are negative regulators of the Wnt pathway that interact with LRP5 and LRP6, preventing Wnt bind¬ing [9,13]. DKK1 was previously known to be secreted by latent metastatic cancer cells to downregulate ULBP ligands for NK cells and responsible for the evasiveness of NK-cell-mediated clearance [14]. Another study demonstrated that DKK1 regulates the accumulation and function of myeloid-derived suppressor cells in cancer [15]. 
The optimal candidates should be expressed at a lower level in various normal human tissues, especially in immune tissues. By this means, antibody blocking will not become a strong risk factor for autoimmunity. Likewise, the lack of these genes should not impede tissue development and cell function. DKK2 is one of these targets. 
In our previous study published on Nature Medicine, we identified a novel inhibitory mechanism hampering natural killer (NK) and T cell function, involving the DKK2 protein, which confers resistance to ICB in colon cancer [16]. DKK2 was highly expressed in human colorectal cancer and, importantly, was correlated with individuals with poor survival [16]. The expression of DKK2 is regulated by the loss of APC [16]. To test the role of DKK2 in colorectal cancer, a monoclonal antibody against DKK2 was generated [17]. DKK2 blockade with the anti-DKK2 antibody was effective in suppressing tumor progression using a syngeneic CRC model grafted with mouse colon cancer cell MC38 and a genetic benign CRC model using the ApcMin/+ mice [16]. DKK2 secreted by tumor cells acts on cytotoxic lymphocytes, inhibiting STAT5 signaling by impeding STAT5 nuclear localization via LRP5, but independently of LRP6 and the Wnt–b-catenin pathway [16].

In our most recent studies, we generated a mouse genetic CRC model with mutations in APC and KRAS, which more closely mimics advanced human CRC, we confirmed the tumor inhibitory effect of DKK2 blockade, which significantly retarded tumor progression and extended survival, with increased immune effector cell activation and reduced angiogenesis [18]. We also established an ex vivo culture system to study the effect of DKK2 blockade on infiltrated cytotoxic immune cells from human CRC tumor samples. We demonstrated that the human colon tumors expressed high levels of DKK2, and DKK2 blockade caused stronger activation of tumor infiltrating CD8+ T cells. Intriguingly, we observed a correlation of high DKK2 expression with increased lymph node metastasis prevalence in these CRC patients as well [18]. Regarding to the clinical relevance, we found that the anti-DKK2 antibody synergizes with an anti-PD-1 antibody to promote the immune antitumor response in the ApcMin/+ mouse model, in part through the induction of GZMB on cytotoxic lympho¬cytes [16]. We have also expanded DKK2 blockers to treat lung cancer with mutations in APC and KRAS [19]. Previous studies demonstrated that DKK2 induction promoted angiogenesis in cultured human endothelial cells and in in vivo assays using mice [20,21]. In our study, we also reproduced the pro-angiogenic effects of recombinant DKK2 proteins [18]. On the other hand, Braumüller et al. demonstrated that IFNγ not only impedes tumor growth by acting directly on cancer cells [22], but also acted on the tumor stroma and tumor angiogenesis for effective rejection of large, established tumors [23-25]. Given that VEGF/VEGFR blockage had been approved to treat human CRC, we performed a combined administration of DKK2 blockade with sub-optimal anti-VEGFR treatment and observed a synergetic effect on suppressing tumor angiogenesis and progression, as well as extending survival, which yielded better results than those of every single therapy [18] (Figure 1).

Our research has identified previously unknown tumor immune-suppressive mechanism and immunotherapeutic targets specifically related to CRCs. The beneficial effect of the anti-DKK2 monoclonal antibody was not only in colorectal cancer models but also in mouse models of APC-deficient lung cancer [19], suggesting that DKK2 blockade may be used as a therapeutic approach alone or in combination with other ICB strategies in a wide range of tumors. Thus, our study provides further evidence for the therapeutic application of the DKK2 blockade in the clinical treatment of human CRC when combined with other targeted therapies.