As mentioned above, based on the MMR/MSI classification system, dMMR is mainly characterized by MSI-H, which is defined as instability at two or more sites that results in a large number of DNA replication errors, highlighted by genetic and accidental changes in the MMR gene (51). Through whole exome sequencing of the same number of CRC tumor cells, an average of 1,782 individual cell mutations were found in dMMR/MSI-H tumors, while only 73 individual cell mutations were detected in pMMR/MSS tumors (10). Since most of the dMMR/MSI-H CRC subtypes have a high TMB, ICIs and immunotherapy exert excellent therapeutic effects on patients with a high TMB (52). Therefore, ICI immunotherapy strategies have become the primary clinical treatment for patients with the dMMR/MSI-H subtype, including PD-1 inhibitor (pembrolizumab or nivolumab) monotherapy (19, 53), combination therapy with a PD-1 inhibitor (nivolumab), and CTLA4 inhibitor (ipilimumab) (54) and combination therapy with a PD-L1 inhibitor (atezolizumab) and anti-angiogenic vascular endothelial growth factor (VEGF) antibodies (bevacizumab) (55). For example, Overman enrolled 74 patients with the MSI-H subtype in a clinical study (NCT02060188) and treated them with nivolumab, a PD-1 inhibitor (56, 57). After treatment with nivolumab, approximately 68.9% of patients required longer than 12 weeks for disease control. In addition, eight patients (34.8%) experienced an immune response lasting longer than 12 months, indicating that nivolumab provides long-lasting response and disease control in patients with dMMR/MSI-H mCRC. More interestingly, in patients with the MSI-H subtype, the combination strategy incorporating ICIs appears to exert a greater antitumor effect. In 2018, Overman and colleagues further explored the therapeutic effect of the combination of the PD-1 inhibitor nivolumab and the CTLA4 inhibitor ipilimumab on MSI-H tumors (54). Among 119 patients, the combination inhibitor treatment achieved 80% effective tumor control at 12 weeks, and more than 94% of the immune response was sustained. All these data showed that the combination strategy of a PD-1 inhibitor and CTLA4 inhibitor results in a higher immune response rate and longer overall survival (OS) and progression-free survival (PFS) durations for patients with the MSI-H subtype, prompting the Food and Drug Administration (FDA) to approve nivolumab and ipilimumab as treatments for patients with the dMMR/MSI-H subtype (22). In conclusion, significant therapeutic efficacy has been documented for ICIs for the MSI-H subtype, providing a promising new treatment option for patients.

Recently, a randomized trial was conducted with the PD-L1 and CTLA4 inhibitors durvalumab and tremelimumab, respectively, for supportive pMMR/MSS CRC therapy (58). One hundred eighty patients with CRC were divided into the D+T group (durvalumab (D) and tremelimumab (T)) and BSC group (best supportive care). Although the objective response rate (ORR) and PFS were similar between the two groups, the OS was improved in the D+T group, indicating that the survival time was extended using durvalumab and tremelimumab immunotherapy. This study is the first to combine a PD-L1 inhibitor and CTLA4 inhibitor to prolong the survival of patients with pMMR/MSS advanced refractory CRC, raising new hopes for immunotherapy for pMMR/MSS mCRC. However, the clinical benefit is limited to a small subset of patients with the pMMR/MSS CRC subtype, accounting for approximately 4% of all patients with CRC (59, 60). For pMMR/MSS CRC subtype, ICI treatment cannot achieve the best therapeutic effect. When MSI is present, tumor cells release many tumor-associated antigens (TAAs) that are normally located inside tumor membranes, which are then taken up and presented by antigen-presenting cells (APCs) located in the tumor immune microenvironment, enhancing the antitumor ability of T cells (61, 62). Nevertheless, for the pMMR/MSS subtype of CRC, the DNA structures are too stable to release TAAs, thus blocking immune system activation or inducing a failure of activated immune cells to recognize tumor cells. Therefore, some studies have indicated that PD-1 inhibitors are less effective in patients with pMMR/MSS CRC subtypes (10, 63–66). New therapeutic strategies are urgently needed to enhance tumor immunity in patients with pMMR/MSS CRC. To date, many studies have indicated that chemotherapy, molecular targeted therapy, and radiotherapy cause immunogenic cell death (ICD) in cancer cells (67). After ICD, tumor cells are exposed to a large number of TAAs, and damage-related molecular patterns and proinflammatory cytokines are released and effectively promote immune cell infiltration and activate APCs (such as dendritic cells (DCs) and macrophages). Then, DCs and macrophages gradually mature and cross-present tumor antigens after their uptake, resulting in antigen-specific immune responses to tumors (68). The immunogenic-based treatment strategy covers all associated antigens of tumor cells and minimizes the incidence of immune escape, thereby mediating a systemic immune response through site-specific in situ cell death (69–72). These theories might provide new immunotherapy strategies for patients with pMMR/MSS CRC presenting a low immune response.

In addition, a recent study identified a key mechanism of T-cell failure and resistance to checkpoint blockade in patients with MSS CRC. The depletion of tumor-infiltrating T cells and a simultaneous increase in VEGF-α levels were observed in MSS colorectal tumors, which might explain why human T cells deplete the associated transcriptional programming in a virulence-dependent manner (73–75). After VEGF-α was inhibited, T cells restored antitumor viability. The combination of a PD-1 inhibitor and VEGF-α inhibitor effectively restores the antitumor function of T cells in MSS CRC, achieving a better therapeutic effect (76).

For patients with the pMMR/MSI-L CRC subtype, the microsatellite stability of the cancer cells ranges between MSI-H and MSS subtypes. Therefore, this subtype has few specific tumor characteristics. To date, commonly used immunotherapy strategies for the MSI-L subtype are divided into three categories: CTLA4 inhibitors (ipilimumab), PD-1 inhibitors (pembrolizumab or nivolumab), and PD-L1 inhibitors (atezolizumab or durvalumab), as well as their combinations. Therefore, the combination of pembrolizumab and nivolumab and the combination of nivolumab and ipilimumab have been approved for the treatment of CRC by the FDA (77). ICIs are not absolutely effective against the MSI-L subtype. In a study published in 2015, Le et al. observed that pembrolizumab treatment did not exert favorable immune-mediated antitumor effects on patients with PMMR MSI-L tumors (10). In addition, in the study by Overman, 142 patients with pMMR/MSI-L had a limited response to immunotherapy. Only one of 20 patients exhibited an immune-mediated antitumor response to the combination of PD-1 and CTLA4 antibodies (54). In recent years, with the discovery of new ICI combination strategies, substantial progress has been achieved in improving the efficacy of combination immunotherapy for patients with this type of tumor. According to the study by Liu published in 2015, the combination of a PD-L1 inhibitor and MEK inhibitor showed significant efficacy in patients with pMMR/MSI-L CRC (78). The RAS–MAPK pathway is the intersection or the last common pathway of transmembrane transduction of growth signals by multiple membrane receptors. Activation is related to the reduction of tumor infiltration by T cells and directly promotes the proliferation of tumor cells. MEK is the downstream effector of the RAS–MAPK pathway. Mekinist, a MEK inhibitor, increases the tumor infiltration of CD4⁺ T lymphocytes but does not affect CD8⁺ T cells. More importantly, in murine models of KRAS-mutated CT26 colorectal tumors, when the two drugs were used together, more potent and long-lasting antitumor activity was observed than in mice treated with either of the two drugs alone (18, 78, 79).

Endocellular peptides are processed and expressed on major histocompatibility complex class I (MHC I) molecules on the surface of almost all human cells, including cancer cells. This type of peptide is specifically recognized by T-cell receptors (TCRs) (18). The response of T cells is modulated by a series of coinhibitory or costimulatory signals. Among them, the membrane-binding ligands CD80 and CD86 of the B7 family bind to the costimulatory protein CD28, especially in activated T cells, resulting in an interaction with CTLA4 and the inhibition of T-cell activity (80). Similarly, membrane-bound PD-L1 and programmed cell death 2 ligand (PD-L2) bind to PD-1 to further incapacitate T cells and induce apoptosis (81–83). When these inhibitor signals are antagonized by the corresponding antibodies, T cells are activated and produce an antitumor effect. However, the T-cell content appears to be deficient in the MSI-L subtype, which may be the root cause of the poor response of this type to immunotherapy (56). In addition, the combination of a PD-1 inhibitor and other modulators of immune checkpoints, such as CTLA4, may benefit a small subset of patients with pMMR/MSI-L tumors. This benefit might be associated with the low microsatellite instability of the tumor cells themselves. Due to its poor specificity, this subtype displays low sensitivity to ICIs. Currently, few studies have been conducted on immunotherapy for this type of CRC, and more effective strategies should be designed that mainly focus on approach to increase the infiltration of immune cells into pMMR/MSI-L tumors.

The TME of hot CRC contains many lymphocytes and inflammatory infiltrates, which mainly include the CMS1 and CMS4 subtypes. As the main antitumor cells, T cells are abundant in hot tumors and are more easily activated, and patients with these subtypes are more likely to benefit from immunotherapy (86). Although a large number of immune cells are present around both the CMS1 and CMS4 subtypes, the CMS4 mesenchymal subtype seems to be more prone to an adverse inflammatory immunophenotype characterized by the activation of transforming factor TGF-β and enhanced tumor angiogenesis, tumor growth, and metastasis. The malignant inflammatory environment will block the antitumor effect of immune cells, resulting in immunosuppression. Therefore, different immunotherapy strategies for the two subtypes of hot tumors are expected to obtain the maximum therapeutic benefits from personalized immunotherapy.

CMS2 and CMS3 subtypes are the main representative types of cold CRC tumors. The CMS2 subtype is the most common subtype, accounting for approximately 37% of all CRC cases (85). The subtype is mainly gene-specific, showing activation of WNT and MYC pathways and chromosomal instability. CMS3 mainly presents as an epithelial neoplasm. This subtype is characterized by KRAS gene mutations, a mixed MSI status and abnormal metabolic pathways. Both of these types exhibit gene-specific expression and methylation abnormalities. It has been explored that ICIs drugs alone have poor efficacy, which is related to the immunosuppressive nature of cold CRC. Currently, a variety of immunotherapeutic strategies have been developed, such as oncolytic virus, cytokine therapy, CAR-T therapy, and passive immunotherapy against TAAs (97), which have contributed to the immune-activated antitumor therapy of cold tumors.

The main characteristic of the CMS2 and CMS3 subtypes is the lack of tumor immunogenicity in the TME (98). The immunosuppression mechanism is different from that of hot tumors. Therefore, methods to activate the tumor cells of CMS2 and CMS3 to release TAAs are the starting point of immunotherapy for cold tumors. ICD is mediated by the release of TAAs when tumor cells undergo apoptosis, which stimulates long-lasting antitumor immune effects in the body (99). ICD of tumor cells substantially improves the low immunogenic microenvironment of cold tumors and promotes immune activation. Chemotherapy is one of the main methods to induce ICD in tumor cells. Some anthracyclines and oxaliplatin not only induce tumor cell apoptosis but also induce ICD (67). When TAAs are released, APCs such as DCs and macrophages are activated, along with the activation of T cells (100, 101). In addition, the efficacy of chemotherapy combined with ICIs becomes an important therapeutic strategy for cold CRC and has been well established. In 2017, chemotherapy combined with pembrolizumab (Keytruda) was approved by the FDA as a first-line treatment for metastatic non-small cell lung cancer (NSCLC) (102, 103). The effective rate of Keytruda combined with chemotherapy was 55%, while the effective rate of chemotherapy alone was only 29%. Combination therapy reduced the risk of disease progression. Since then, the FDA approved Keytruda in combination with pemetrexed and platinum-based chemotherapy as a first-line treatment for metastatic non-squamous NSCLC in patients without EGFR and ALK cancer genetic variants in August 2018 and approved Keytruda in combination with standard chemotherapy (carboplatin and paclitaxel/albumin paclitaxel) as a first-line treatment for squamous NSCLC in October 2018.

In addition, lower immune cell infiltration is another major factor contributing to a poor immune response (104), which is also a reason for ineffective ICIs. The low immune cell infiltration of cold tumors results in the insufficient effect of the activated immune system on eradicating the tumor tissue. Therefore, approaches that promote the enrichment of immune cells and activate the strong and lasting immune antitumor response have become the main challenge to overcome in immunotherapy for cold CRC. In a study presented at the ASCO-SITC Symposium on Clinical Immuno-Oncology in 2020, Shen and colleagues found that the methylation and expression levels of ARHGAP9, TBX21 (T-bet), and LAG3 genes were associated with the infiltration of CTLs (105). In CMS2 tumor cells with low T-cell infiltration, the expression of ARHGAP9 and T-bet genes was downregulated, and the methylation level was low. The expression of ARHGAP9, T-bet, FML1, HLA-DPB1, and STX11 genes in the CMS3 subtype was also downregulated, and the methylation level was low (105). However, in the CMS1 subtype with high infiltration of immune cells, the expression level and methylation level of T-bet gene were higher. Therefore, the expression levels of these genes differ between CMS subtypes, and the methylation level is related to the extent of CTL infiltration. T-bet gene was the only gene that was strongly correlated with CTL infiltration, which might directly reflect the degree of T-cell infiltration in patients with CRC. Therefore, the T-bet protein is likely to be a key regulator of T-cell infiltration in CRC (106). An increase in the methylation level of T-bet gene in the CMS2 and CMS3 subtypes and upregulation of the expression level of T-bet gene have become effective strategies for cold tumor immunotherapy.

ICIs has always been the most mature and effective immunotherapy for CRC. However, it appears to be limited for CRC subtypes with low immunogenicity or low immune cell infiltration. Therefore, immunotherapy by using ICIs alone is not effective in some specific CRC subtypes. Chemoimmunotherapy has become an important therapeutic strategy in the treatment of CRC. CAVE-Colon study has been proposed in recent years (NCT04561336). The researcher found that the combination strategy of cetuximab, an EGFR-targeted drug, and avelumab, an ICI drug, would significantly improve survival of CRC patients. Cetuximab improved the T-cell activity by increasing immune cell invasion and inducing APC maturation, while ICIs eliminated the tumor immunosuppressive effect on T cells.

In addition, in order to further improve the survival of CRC patients treated by FOLFOX or FOLFIRI strategies, a new GOLFIG was developed (107, 108). This kind of strategy combined low-dose recombinant interleukin-2 (rIL-2) and granulocyte macrophage colony-stimulating factor (GM-CSF) based on FOLFOX and gemcitabine, a chemotherapy drug. In the GOLFLG strategy, multiple chemotherapy drugs can successfully destroy tumor cells in large quantities, while rIL-2 and GM-CSF would promote the activation of APCs, then activate CTLs, and then accurately destroy the remaining tumor cells (109, 110). Correale et al. firstly explored the effect of FOLFIG trials. They found the remission rate (RR) and disease control rate (DCR) of patients with mCRC were higher than those of FOLFOX strategy, indicating significant immune response and antitumor activities (111). In addition, PFS was significantly lower in patients treated with GOLFIG than with FOLFOX, indicating GOLFIG could effectively reduce the recurrence rate (112, 113). Chemoimmunotherapy has broadened the scope of application of ICIs and maximized the antitumor effect.

Oncolytic virus immunotherapy is a promising approach for the treatment of solid tumors and has the ability to selectively replicate the tumor, achieve ideal immunogenicity, and deliver foreign genes to the tumor in a targeted manner (114, 115). In 1991, Martuza et al. showed that transgenic HSV was an effective treatment for glioblastoma (116). Since then, oncolytic virus therapy based on HSV has been developed. The antitumor strategy of oncolytic viruses is mainly mediated by the genetic modification of some of the less virulent viruses existing in nature. These approaches use the inactivation or defect of tumor suppressor genes in target cells to produce therapeutic oncolytic viruses with a specific recognition ability, which selectively infect tumor cells, then replicate in large numbers, and eventually destroy tumor cells. Meanwhile, damaged tumor cells release a large number of TAAs, and thus the immunogenic death of tumor cells is induced to attract more immune cells and clear the tumor tissue (116). In recent decades, oncolytic virus therapy has attracted increasing attention and achieved great progress in related research. The oncolytic vaccinia virus (VV) treatment strategy has been shown to transform cold tumors of the CRC type into hot tumors by promoting the infiltration of immune cells, resulting in effective antitumor immunity (117, 118). However, the therapeutic potential of oncolytic viruses is also affected by the immunosuppression of immune cells by tumors, such as immune checkpoints. According to the study published by Wei, compared with oncolytic VV therapy alone, the strategy of VV combined with monoclonal antibody against T-cell immunoglobulin and ITIM domain (TIGIT) exerts better effects on reducing the tumor burden and prolonging survival (119). They found that the number of CD8⁺ T cells increased and the microvascular density decreased in the TME of CRC using the VV-ATIGIT strategy for immunohistochemistry, which proved that the combination strategy significantly inhibited tumor growth in mice with colon cancer. In conclusion, oncolytic viruses combined with ICIs may be a new direction for the treatment of CRC ( Figure 2 ) (119).

For CMS2 and CMS3 CRC subtypes, the low level of immune cell infiltration is the main factor contributing to a poor immune response. The emergence of CAR-T cell therapy has become the main and precise treatment strategy to solve this difficulty. The CAR-T strategy is a classic adoptive cellular immunotherapy that kills tumor cells by injecting genetically modified and amplified immune effector T cells (120). CAR-T immunotherapy has many advantages, such as strong specificity, strong targeting, a high fatality rate, few side effects, and no drug resistance (119). For some CRC subtypes with low immune cell infiltration, CAR-T therapy with high efficiency and specific targeting of tumors is undoubtedly an appropriate immunotherapy strategy. Carcinoembryonic antigen (CEA) is a sensitive biomarker for gastrointestinal tumors that is widely expressed in CRC tissues and serum and not expressed in many other normal tissues or organisms (121), making it an effective target for CAR-T therapy. In 2017, Zhang et al. conducted a phase I clinical trial of systemic intravenous CAR-T therapy for CRC with high positive CEA expression (NCT02349724). In the trial, five stepped-up CAR-T cell doses were administered to 10 patients with CRC, and no serious adverse events associated with CAR-T therapy were observed during follow-up. Seven of the 10 patients had progressed at the time of previous treatment but were stable after CAR-T treatment. Imaging analysis showed that tumors shrank after treatment in two patients. After long-term observation, the serum CEA level decreased significantly in most patients. In addition, CAR-T persisted in the peripheral blood of patients who received high doses of CAR-T. More importantly, proliferation of CAR-T was observed in the trial, especially in patients following the second CAR-T treatment. CEA-targeted CAR-T therapy was well tolerated by patients with CEA-positive CRC, even at high doses. Some efficacy was observed in the majority of patients who received the therapy (17).

In 2019, Ying et al. loaded the CD19 truncated (CD19T) gene in an oncolytic virus to increase the expression of the CD19 antigen on the tumor cell membrane surface by taking advantage of the transgenic transmission potential of the oncolytic virus (122). Meanwhile, they combined CAR-T immunotherapy with the chimeric CD19 gene to enhance the specific targeting effect of CAR-T. The combination of the oncolytic virus and truncated CD19 (OV-CD19T) specifically replicated and expressed CD19T antigen in different solid tumor cells. The combination of OV-CD19T and CD19 CAR-T therapy is a safe and effective treatment for multiple solid tumors both in vitro and in vivo, without causing neurotoxicity or severe cytokine release syndrome (CRS), suggesting long-lasting tumor killing effects (122). In the next year, Priceman further explored a novel and effective combination immunotherapy based on the combination method (123). They used a genetically engineered oncolytic virus to overexpress the CD19 antigen on the surface of tumor cell membranes, thereby enhancing the antitumor response of CD19 CAR-T therapy to solid tumors. CD19 has long been an ideal target for CAR-T against hematological malignancies due to its highly restricted expression on B cells (124–126). CD19-mediated CAR-T therapy has been approved by the FDA for B-cell malignancies. However, in solid tumors, the coexpression of most tumor antigens in normal tissues limits the potential of CAR-T in solid tumors. The strategy of introducing specific targets into solid tumors to promote CAR-T immune activation might advance the suitability of CAR-T therapy for solid tumors.

The biological process of ideal immune-activated antitumor therapy is that ICD tumor cells release TAAs, damage-associated molecular patterns, and proinflammatory cytokines, which function as an alarm to effectively promote immune cell infiltration and activation. APCs promote cross-presentation of tumor antigens, triggering a wide range of tumor antigen-specific immune responses (127). T cells, the terminal killer cells of immune antitumor cells, are of decisive significance for immune efficacy. However, cold CRC tumors have low immunogenicity or less T-cell infiltration in the TME, which may be due to the presence of initiation defects or the absence of high-affinity T cells. Therapeutic cancer vaccines with TAAs or tumor-specific antigens (TSAs) might directly stimulate the immune system and induce an immune response by delivering these antigens directly to professional APCs, thus activating CD4⁺ and CD8⁺ T cells to kill tumors (128, 129). Many cancer vaccines have been developed, including cell vaccines, nucleic acid vaccines, protein peptide vaccines, and genetically engineered vaccines (130–134). Adoptive cell immunotherapy (ATT) is a major cellular vaccine strategy and a novel approach to TSA presentation. When the immune cell antigen load is low or the relevant tumor antigen load is not available, ATT might achieve better tumor targeting and killing effects by directly overexpressing tumor cell antigen antibodies in immune cells (135). For example, CD8⁺ T cells targeting k-ras or TP53 mutations have been identified (136, 137). In addition, recent studies have shown that circulating PD-1-positive lymphocytes recognize neoantigens in human digestive tract tumors (32, 138).

Thymidylate synthase poly-epitope-peptide (TSPP) is an anticancer poly-epitope peptide vaccine, which can promote the promote cross-presentation of tumor antigens through APCs and trigger a highly specific immune response with multi-antigen specificity (139). TSPP has shown good antitumor activity in preclinical studies. However, the antitumor effect of TSPP is significantly influenced by the TME such as inflammation and immune response. K-ras and IL17/A have been found to be important factors to predict the resistance of TSPP. Patients with k-ras mutation have no statistically significant inflammatory markers and also have poorer immune response (139). IL17/A, as an inflammatory cytokine independent of k-ras status, could amplify and enhance the cytotoxic effects of CTLs at CRC tumor sites (140). By predicting k-ras mutation and IL-17/A level, the immunoreactivity and antitumor efficacy of immunotherapy like TSPP vaccine against CRC could be evaluated.

In addition, the immunotherapy response is usually associated with the TMB, and patients with dMMR/MSI-H have higher genetic mutation characteristics (52). Therefore, vaccination against the associated antigens of specific genetic mutations has proven effective against this type. In 2017, Maletzki et al. constructed a mouse model of dMMR induced by knockout of MLH1. After the treatment dMMR cancer vaccine was administered to the mice, the OS time was prolonged, and the tumor mutation load was reduced, indicating that the specific vaccine is a viable option for the treatment of the dMMR subtype (141, 142). Similarly, human clinical trials of therapeutic cancer vaccines have shown good results based on the MSI status (143). The main question to be answered in order to promote increased immune efficacy is whether the combined ICI immunotherapy and vaccine strategy is more effective than ICIs alone in dMMR/MSI-H CRC or whether it is capable of causing a response in pMMR/MSI-L CRC that does not respond to ICIs.

The TME is complex and changeable, and the immune response is affected by many factors. Different types of CRC are heterogeneous. Even for the same solid tumor, a large difference in gene or protein expression may be observed. Therefore, extensive clearance of tumor cells is difficult to achieve with a single immunotherapy (144). Immune adjuvant, a non-specific immunoproliferative agent, is an adjuvant that effectively activates the immune response to antigens or changes the type of immune response, which is important for enhancing the efficacy of immunotherapy for CRC. At present, immune adjuvants are mainly divided into biological and abiotic adjuvants, and the latter exerts a better effect on immune sensitization. In a phase I clinical trial, Hou used Mn²⁺ combined with an anti-PD-1 antibody to safely enhance the antitumor efficacy of treatment for advanced mCRC (145). A lack of Mn²⁺ promotes tumor growth. When B16F10 melanoma cells were implanted in Mn²⁺-deficient mice, at least 2.5 * 10³ cells were needed. However, seven times more tumor cells were needed in normal mice to achieve 100% tumor incidence. Moreover, the tumor volume and weight of Mn²⁺-deficient mice were significantly increased. Tumor infiltration of CD4⁺ and CD8⁺ T cells was significantly decreased (146, 147). After the Mn²⁺ injection, tumor growth was inhibited. At the same time, Mn²⁺ injection on one side of the tumor also inhibited tumor growth on the other side, suggesting a systemic antitumor response. Moreover, Mn²⁺ promoted DC maturation and enhanced antitumor effects through the cGAS–STING pathway, which also exerts an antitumor effect on patients with immune-tolerant tumors. From 2018 to 2019, Lv et al. initiated a phase I clinical trial on the effect of Mn²⁺ on advanced mCRC. In this study, a MnCl₂ solution was administered intranasally or inhaled. The authors documented a clinical RR of 45.5% and a tumor control rate of 90.9% after treatment with the MnCl₂ solution (148). More importantly, five patients who had previously failed to respond to anti-PD-1 antibodies or chemotherapy in combination with radiation showed good DCRs during the treatment, suggesting that Mn²⁺ restored immunotherapeutic efficacy in immune treatment-resistant patients. As an inorganic adjuvant, Mn²⁺ has undoubtedly become the most important immune sensitizer for immunotherapy of CRC. It potentially activates the immune response of patients with immune treatment-resistant CRC, providing a new and more effective therapeutic strategy (149).