Liver metastasis is the main reason that affects the survival rates of patients with colorectal cancer (CRLM), and is also the main cause of death of those patients. Especially after the failure of first-line or second-line system treatment, the prognosis of those patients is extremely poor, with the median OS of only 3.5 months. Even in combination with molecular targeted drugs such as cetuximab or bevacizumab, the median tumor-free survival period is only 4.8-6.8 months, and OS is only 11-15 months. When they have disease progression, treatment is currently a difficult clinical problem. Regofinib is a new targeted drug for the third-line treatment of advanced colorectal cancer in recent years. However, in the prospective multicenter clinical study, compared with the placebo group, the extended OS is only 1.4 months, which is not so satisfactory. How to improve the survival of these advanced patients with drug resistance is an important clinical problem to be solved urgently. Minimally invasive local treatment may be a promising way to solve this problem. Transcatheter arterial chemoembolization (TACE) and hepatic artery infusion chemotherapy (HAIC) are currently the most widely used methods in clinical practice. In theory, TACE combined with HAIC can control small metastasis and embolic residual lesions. The combination of TACE and HAIC can improve the curative effect. Whether the combination of TACE, HAIC and Regofinib can be expected to achieve the effect of 1+1+1>3 in CRLM patients who have failed the previous second-line chemotherapy remains unknown. Therefore, the purpose of this study is to explore the safety and clinical efficacy of irinotecan-loaded drug-eluting beads-TACE (DEBIRI-TACE) combined with HAIC and Regofinib in the treatment of patients with CRLM who failed standard treatment regimens.

Colorectal cancer is one of the most common tumors worldwide, with an annual incidence rate ranking the first in gastrointestinal tract tumors and the third in all malignant tumors in Europe and the United States, the second in gastrointestinal tract tumors and the fourth in all malignant tumors in China, featuring an continuously increasing incidence rate year by year. The liver is the most common site of hematogenous metastasis in colorectal cancer, and about 50% of patients develop liver metastasis during the disease process, at least over 60% of which will die of liver metastasis. Related epidemiological studies have shown that about 25% of colorectal cancer patients are diagnosed with liver metastasis, while another 15% to 25% of patients develop liver metastasis after radical primary lesions, and the vast majority (80%-90%) of liver metastases can't obtain curative resection. Liver metastasis is the main influencing factor of survival and the leading cause of death in CRC patients. The median survival of the untreated patients with focal liver metastases was only 6.9 months, and the 5-year survival rate of the unresectable patients reached nearly 0.

Surgical resection of primary lesions and liver metastases is the preferred method for curing colorectal liver metastases (CLRM), but only 10%-20% of patients are considered suitable for surgical resection of metastasis at initial diagnosis. The standard first-line chemotherapy for FOLFOX or XELOX was used for unresectable CLRM, and the median PFS was about 7.0-12.3 months, while the OS was about 15.0-29.8 months. However, after the failure of the first-line or second-line system treatment, the patient prognosis was extremely poor, with a median natural survival of only 3.5 months. Even with molecular targeted drugs such as cetusimab or bevacizumab, the median PFS was only 4.8-6.8 months, while the OS was only 11-15 months. Treatment for these patients remains the existing clinical difficulty. Regefenib is a new targeted drug for third-line treatment of advanced colorectum, which, as a small-molecule multi-target kinase inhibitor, affects angiogenesis, proliferation, and microenvironment by inhibiting tumor angiogenesis, proliferation, and other targets such as VEGFR1, VEGFR2, VEGFR3, TIE2, Kit, RET, PDGFR, FGFRl, etc. It is now the treatment method recommended for the treatment of the third-line and above CLRM in the major domestic and foreign guidelines such as the National Comprehensive Cancer Network (NCCN), the European Society of Medical Oncology (ESMO), and the Chinese Clinical Cancer Congress (CSCO), etc. The prospective multicenter clinical study CORRECT trial compared the efficacy of regorofenib and placebo in patients with metastatic colorectal cancer above the second line, with the median OS in the regorofenib group significantly better than that in the placebo group (6.4 months vs. 5.0 months, HR=0.77), which established the recommended position of regorofenib in the third-line CLRM. However, the extended OS was also only 1.4 month, which was not satisfactory. To this end, improvement of the survival rate of such advanced drug-resistant CLRM patients becomes an important clinical issue that requires immediate attention.

Timely addition of minimally invasive local treatment is considered a promising way to solve this problem, and the interventional route of chemotherapy and embolization is one of the most used clinical methods. Given that liver metastases of colorectal cancer are mainly supplied by the hepatic artery (90% to 95%), while the normal liver parenchyma is mainly provided by the portal vein, an anatomical basis and clinical basis are provided for the local intervention of the transhepatic artery, including transcatheter arterial chemoembolization (TACE), and hepatic artery infusion chemotherapy (HAIC), etc. These attempts can effectively control the intrahepatic tumor load, and achieve the goal of prolonging patient survival, which have been widely used in clinical practice. The 2020 edition of the Chinese Guidelines for the Diagnosis and Comprehensive Treatment of Liver Metastasis in Colorectal Cancer clearly states that HAIC or TACE can be used on refractory patients with liver metastases mainly, large tumor load and insignificant drug treatment effect, or those who cannot tolerate systemic treatment [4]. Both the 2016 ESMO and 2021 NCCN colorectal guidelines also stipulate that TACE or HAIC can be performed on CRLM patients with poor chemotherapy efficacy in conditional hospitals, or used for clinical research. Among them, irinotecan-loaded drug-eluting beads-TACE (DEBIRI-TACE) is a classic interventional treatment for CLRM, which can embolize the tumor blood vessels, block the tumor blood supply, and make the slow and continuous release of chemotherapeutic drugs in the tumor microcirculation, thereby improving the curative effect. Several prospective studies have demonstrated the superiority of DEBIRI-TACE efficacy to systemic chemotherapy or conventional lipiodol TACE on CLRM patients without increasing the toxic side phase. For patients with bowel cancer liver metastases who have failed standard chemotherapy, the implementation of DEBIRI-TACE achieves a median PFS up to 5 months and a median OS up to 8 months. DEBIRI-TACE is even effective and safe for colorectal cancer liver metastases subject to the failure of irinotecan systemic chemotherapy, which provides basis for CLRM above the second line.

However, it is difficult for TACE to achieve complete tumor necrosis as a non-radical treatment mode, and its induced ischemic damage will lead to the upregulation of angiogenesis-related molecules, thus subsequently promoting tumor growth, invasion, and metastasis. In this case, the combination of regorofenib with TACE can both enhance the tumor reduction effect and inhibit the TACE-induced angiogenic effect as well as tumor growth by complementing each other. In addition, it was reported that the adverse reactions caused by regofinil were mainly fatigue, hand and foot syndrome, and diarrhea, etc., while the abdominal pain, nausea, vomiting and fever above the adverse reactions caused by DEBIRI-TACE were the most common. Adverse effects of these two approaches had no obvious overlapping effects, indicating the safety and feasibility of the combination of TACE with Regofenib, which was confirmed in a recently-published retrospective study. An analysis over 76 CLRM patients who had failed previous chemotherapy found that compared with single regorofenib monotherapy, DEBIRI-TACE combined with single achieved a higher local response rate (ORR 35.3% vs 7.1%, P = 0.002; DCR 76.5% vs 47.6%, P = 0.011), and a longer median PFS (7.6 vs 4.1 months, P <0.001) and OS (15.7 vs 9.2 months, P <0.001), and the side reaction was equivalent (P> 0.05). The above results further validate the important potential of local interventional therapy combined with regofenib in the posterior treatment of CRLM. It is an important clinical problem worth further exploration to further improve the curative effect on this basis for the benefit of more patients.

Different from the TACE vascular embolization effect, HAIC has the advantage to directly deliver high doses of chemotherapeutic drugs to the blood supply branch of the hepatic artery tumor, which can increase the local drug concentration and reduce the systemic side reactions, gradually making it an effective treatment for unresectable CRLM. Besides, it has been shown that even after the failure of systemic chemotherapy regimen, the same chemotherapy regimen is still effective, and the overall survival can be extended to 7.7-19 months. In a phase II clinical study from Sloan Catherine Cancer Memorial Hospital, 65% of the enrolled CRLM patients were on second-or third-line therapy, with an ORR of 76% with 5-Fu and systemic chemotherapy with or without bevacizumab, and about 47% of initially-unresectable patients were converted to resectable ones, with a median OS of 38 months. In a European multicenter study (OPTILIV) published in 2016, all enrolled patients received overline systemic therapy and combined systemic cetuximab with arterial perfusion (oxaliplatin, irinotecan and 5-Fu), and 29.7% of these patients were converted to resectable ones, with a median OS of 35.2 months after resection. The above studies suggest that HAIC can still obtain a high response rate and resection rate as a powerful complementary method after systemic treatment failure.

Theoretically, HAIC combined with TACE can control small metastases and embolic residual lesions, and their combination can complement each other and improve the efficacy. Indeed, as early as 2014, Japanese scholars had reported several cases of TACE combined with HAIC for successfully treating liver metastases of gastric cancer. Recently, this interventional combination model has also been successively reported in primary liver cancer. A study involving 83 patients with middle and advanced HCC reported that the TACE and local response rate, conversion resection rate, PFS and OS in the HAIC group were better than the TACE group alone, and had a comparable grade 3 / 4 adverse reaction rate; another study also demonstrated the efficiency and feasibility of TACE combined with HACI and sorafenib in advanced liver cancer, with more than 50% of the 66 patients ith HCC achieving a PFS for 6 months, and the side reactions could be tolerated. However, in CLRM, only one retrospective study explored the efficacy of 162 patients receiving TACE and HAIC after chemotherapy failure, and found that the total median OS in the combined treatment group could be up to 29.5 months (calculated from the definite diagnosis) and 15.6 months (calculated from the first TACE and HAIC), with a DCR reaching 75%. For severe adverse reactions, 15 patients (9.26%) had grade 3 / 4 myelosuppression, and one (0.62%) was subject to liver abscess, having recovered after drainage. The above evidence suggests the important potential of the TACE combined with HAIC in the posterior treatment of CRLM. Further explorations should be made upon the efficiency of the combination of the TACE, HAIC and regofenib treatment in CRLM patients subject to previous second-line chemotherapy failure, as well as the achievement of the effect of 1 + 1 + 1> 3.

CalliSpheres drug-eluting beads is a new drug-carrying microsphere independently developed with independent intellectual property rights in China, and has obtained the national invention patent. The product is produced with water-soluble and non-toxic polymer biomaterials with no side effect after chemical modification, featuring a microsphere shape uniform in size, smooth surface, good blood suspension, variable elasticity and compressibility, high drug loading speed, large load, long slow release time, the capability of absorbing a variety of chemotherapy drugs, and many other characteristics. The instructions for using CalliSpheres include loading irinotecan methods and indications, and the treatment of CLRM has been reported in some clinical studies.

This study aims to explore the clinical efficacy and safety of CalliSpheres drug-eluting beads loaded with irinotecan combined with HAIC and Regorafenib in the treatment of liver metastasis in colorectal cancer after previous chemotherapy failure, so as to find a more effective posterior treatment plan for liver metastasis in intestinal cancer.

Inclusion Criteria:

1. Patients aged 18-75 years old, gender unlimited;
2. Patients with liver metastasis of colorectal cancer by pathological histology or clinical diagnosis (refer to the Guidelines for the Diagnosis and Comprehensive Treatment of Liver metastasis of Colorectal Cancer, 2016 Edition), those who were mainly exposed to liver metastasis with large tumor load and suffered from no clear extrahepatic metastasis;
3. Patients with a liver area tumor no more than 5 tumor nodules, and a nodule size no more than10cm;
4. Patients subject to the failure of previous chemotherapy regimen containing irinotecan or oxaliplatin after at least six cycles of systemic chemotherapy; they cannot undergo surgery or refuse surgery;
5. Patients with liver tumor who did not receive interventional treatment (TACE, ablation, iodine particle treatment, etc.) within one year;
6. Patients with an expected survival period longer than 3 months;
7. Patients with favorable liver function (7 points for Child-Pugh A or B);
8. Patients with a physical fitness ECOG no more than one point;
9. Patients who understood and signed the Informed Consent Form.

Exclusion Criteria:

1. Patients with distant metastases except the liver;
2. Patients subject to the treatment of TACE, ablation or iodine particles within one year;
3. Patients with obvious artero / venous fistula and cancer plugs in the main portal vein;
4. Patients who had suffered from or were currently developing other malignant tumors (except for the cured basal or squamous cell skin carcinoma or cervical carcinoma in situ);
5. Patients whose white blood cells were less than 3,000 cell / mm3, or platelet count less than 50,000 / mm3;
6. Patients subject to renal insufficiency (creatinine> 2 mg/L);
7. Patients whose AST and / or ALT was / were more than 5 times the upper limit of normal;
8. Patients with poor coagulation function, an INR larger than 1.5, or currently subject to anticoagulant therapy or known hemorrhagic disease;
9. Patients with a history of major disease with heart, kidney, bone marrow, or lung, and central nervous system involvement;
10. Patients requiring antibiotic treatment due to recent infections;
11. Patients subject to comorbidities or social environment that could prevent them from following the study plan or even endanger their safety.

Not applicable. We do not plan to share individual participant data with other researchers so that we can maintain the privacy of our participants.