In this single-center,open-label, randomized, phase II study, the efficacy and feasibility of GSL synthetase inhibitor in combination with immune checkpoint inhibitor and/or regorafenib therapeutic regimen will be evaluated in patients with advanced/metastatic proficient mismatch repair/microsatellite stable (pMMR/MSS) colorectal cancer (CRC).In this clinical trial, a total of 120 eligible patients were stratified randomly (with/without liver metastases) assigned to the 3 arms in a 1:1:1 ratio: comparator group-arm A (Regorafenib+Immune checkpoint inhibitor) ,experimental group-arm B (Eliglustat+Immune checkpoint inhibitor) and experimental group-arm C (Eliglustat+Immune checkpoint inhibitor+Regorafenib).It aims to: 1).assess the antitumor effects of GSL synthetase inhibitor in combination with immune checkpoint inhibitor and/or regorafenib;2).evaluate the immunological or clinical predictive biomarkers for efficacy and toxicity; 3).detect the transformation of tumor microenvironment (TME) and dynamic changes of immune cells in peripheral blood after the treatment with GSL synthetase inhibitor in combination with immune checkpoint inhibitor and/or regorafenib.

Immunotherapy has achieved significant therapeutic effect in DNA mismatch repair-deficient or microsatellite instability-high (dMMR/MSI-H) metastatic colorectal cancer(mCRC). Distinct from those with dMMR/MSI-H mCRC, isolated immunotherapy has proven to be almost ineffective for patients with pMMR/MSS type mCRC,which indicating a worse prognosis.

Previous work has established that the TME is distinct between MSI-H and MSS CRC. Therapeutic combinations of targeted therapy and immunotherapy,such as regorafenib combined with programmed death 1(PD-1) monoclonal antibody ,which can alter the TME and successfully promote favorable immune modulation has attracted extensive attention.Based on the small sample clinical trial results of other regorafenib combined with anti-PD-1 monoclonal antibody , the overall ORR is between 0% and 33.3%, the ORR for non-liver metastases is between 20% and 50%, and the ORR for liver metastases is between 0% and 15%,demonstrating limited clinical benefit.

Besides TME,another important reason is that tumor cells often escape from immune surveillance by downregulating one or multiple molecules critical in human leukocyte antigen (HLA ) antigen presentation. As a consequence, options that could restore HLA antigen presentation may augment immune checkpoint inhibitor-mediated immune responses.

Abnormal expression of glycosphingolipid (GSL) synthetase is a basic and specific characteristic of most tumors and tumor microenvironment, such as Globo H Ceramide, which is overexpressed in multiple epithelial-derived tumors. Several studies also reported that GSL synthetase was overexpressed in chemotherapy-resistant tumors. Eliglustat is an orally GlcCer synthase inhibitor, which is approved for treating Type-1 Gaucher disease. However, one most recent study reveals that it could inhibit glycosphingolipids synthesis and restore HLA antigen presentation, and transforming the immunogenicity of tumor cells.The investigators has demonstrated the excellent safety and efficacy of the combination of Eliglustat and immune checkpoint inhibitor in advanced/ metastatic solid tumors and r/r hematological malignancies,especially in pMMR/MSS mCRC (even with liver metastases).

Based on the above reasons, we designed this open-label, randomized,phase II study to observe the efficacy and feasibility of the GSL synthetase inhibitor in combination with immune checkpoint Inhibitor and/or regorafenib for patients with advanced/metastatic pMMR/MSS CRC and strive to provide a high-level evidence-based basis for combination therapy regimen for these patients. A total of 120 advanced/metastatic pMMR/MSS CRC patients were stratified randomly (with/without liver metastases) assigned to the 3 arms in a 1:1:1 ratio: comparator group-Arm A (Regorafenib+Immune checkpoint inhibitor),experimental group-Arm B(Eliglustat+Immune checkpoint inhibitor) and experimental group-Arm C (Eliglustat+Immune checkpoint inhibitor+Regorafenib).The primary objective of this study is to assess the efficacy and feasibility of the above two experimental groups.The exploratory objectives are to evaluate the immunological or clinical predictive biomarkers for efficacy and toxicity, transformation of tumor microenvironment and dynamic changes of immune cells in peripheral blood.

Inclusion Criteria:

1. Age ≥18 years old and ≤75 years old.
2. Histologically confirmed diagnosis of unresectable locally advanced, recurrent or metastatic colorectal cance have failed at least two lines of prior treatment.
3. Tumor tissues were identified as pMMR by immunohistochemistry (IHC) method or MSS by polymerase chain reaction (PCR).
4. CYP2D6 extensive metabolizers (EMs), intermediate metabolizers (IMs), or poor metabolizers (PMs).
5. Eastern Cooperative Oncology Group (ECOG) performance status score≤2 and Estimated life expectancy of more than 3 months.
6. At least one measurable lesion at baseline according to RECIST version 1.1.
7. Fresh solid tumor samples or formalin-fixed paraffin embedded tumor archival samples within 6 months are necessary; Fresh tumor samples are preferred. Subjects are willing to accept tumor rebiopsy in the process of this study.
8. Have adequate organ function as assessed by the laboratory required by protocol, which should be confirmed within 2 weeks prior to the first dose of study drugs.
9. Previous treatment must be completed for more than 4 weeks prior to the enrollment of this study, and subjects have recovered to <= grade 1 toxicity.
10. Previous treatment with anti-PD-1/PD-L1 antibodies or cytotoxic T lymphocyte associated antigen 4 (CTLA-4) inhibitors are allowed.
11. Pregnancy tests for women of childbearing age shall be negative; Both men and women agreed to use effective contraception during treatment and during the subsequent 1 year.
12. Ability to understand and sign a written informed consent document.

Exclusion Criteria:

1. Participants with dMMR /MSI-H colorectal cancer.
2. CYP2D6 ultra-rapid metabolizers (URMs).
3. Active, known or suspected autoimmune diseases.
4. The patients is taking a CYP2D6 inhibitor and/or concomitantly with a strong or moderate CYP3A inhibitor.
5. Subjects are being treated with either corticosteroids (>10 mg daily prednisone equivalent) or other immunosuppressive medications within 14 days of enrollment.
6. History of severe hypersensitive reactions to other monoclonal antibodies.
7. History of allergy or intolerance to study drug components.
8. Known brain metastases or active central nervous system (CNS). Subjects with CNS metastases who were treated with radiotherapy for at least 3 months prior to enrollment, have no central nervous symptoms and are off corticosteroids, are eligible for enrollment, but require a brain MRI screening.
9. Uncontrolled intercurrent illness, including ongoing or active systemic infection, symptomatic congestive heart failure, unstable angina pectoris, cardiac arrhythmia (excluding insignificant sinus bradycardia and sinus tachycardia) or psychiatric illness/social situations and any other illness that would limit compliance with study requirements and jeopardize the safety of the patient.
10. Known positive test result for human immunodeficiency virus (HIV) or acquired immune deficiency syndrome (AIDS).
11. Previous or concurrent cancer within 3 years prior to treatment start EXCEPT for curatively treated cervical cancer in situ, non-melanoma skin cancer, superficial bladder tumors [Ta (non-invasive tumor), Tis (carcinoma in situ) and T1 (tumor invades lamina propria)].
12. Major surgery or trauma occurred within 28 days prior to enrollment, or major side effects have not been recovered.
13. Vaccination within 30 days of study enrollment.
14. Active bleeding or known hemorrhagic tendency.Any life Threatening bleeding within 3 months prior to the enrollment.
15. Uncontrolled hypertension (systolic pressure >150 mm Hg or diastolic pressure > 100 mm Hg on repeated measurement) despite optimal medical management.
16. Pregnant or breast-feeding. Women of childbearing potential must have a pregnancy test performed within 7 days before the enrollment, and a negative result must be documented
17. Being participating any other trials or withdraw within 4 weeks.
18. Researchers believe that other reasons are not suitable for clinical trials.