This is a single-center, single-arm, open-label, dose-escalation clinical study to evaluate the safety and preliminary efficacy of NKG2D CAR-NK cells followed by NKG2D CAR-T cells in patients with advanced solid tumors (e.g.,colorectal cancer) with liver metastases who have failed standard treatments.

The study primarily focuses on determining the maximum tolerated dose and recommended phase II dose through sequential cohort dose escalation, while secondarily characterizing the pharmacokinetic parameters and collecting initial efficacy data regarding tumor response.

This investigation comprehensively evaluates the pharmacodynamic and pharmacokinetic profile of NKG2D CAR cellular therapy through three primary objectives: (1) systematic monitoring of treatment-emergent adverse events and clinically significant laboratory parameter deviations; (2) assessment of antitumor activity with correlative biomarker analysis; and (3) characterization of cellular kinetics including biodistribution patterns, and mechanistic pathways of therapeutic activity. The protocol clarifies cellular persistence and functional regulation within the tumor microenvironment by longitudinal monitoring of cytokine release and using advanced molecular tracking methods.

This study plans to enroll 9-18 patients with advanced solid tumors (e.g.,colorectal cancer) with liver metastases who have failed standard treatments during a dose-finding phase (Phase IA ). CAR immune cell therapy will be administered primarily via hepatic artery infusion and intravenous infusion. Based on previous experience, hepatic artery infusion will deliver a fixed dose of 3×10^9 NKG2D CAR-NK cells. After catheterization, the catheter will remain indwelling for two days to allow two NKG2D CAR-NK infusions: 1×10^9 cells on Day 0 and 2×10^9 cells on Day1. The preferred source for CAR-NK cells is umbilical cord blood. However, if post-treatment evaluation shows that a patient's results do not meet expected criteria, investigators will adjust the strategy based on clinical circumstances and switch to autologous peripheral blood as the source of CAR-NK cells to ensure optimal therapeutic efficacy. Investigators will determine whether to initiate autologous NKG2D CAR-T cell reinfusion 1-2 weeks post-intervention based on clinical response. The CAR-T cells will be administered via two consecutive intravenous infusions over two days, with dose escalation following the classical "3+3" schema.

Based on prior experience, rapid titration dose escalation will initially proceed with 1 patient per cohort. The starting dose is set at 1.5×10^7 cells/kg. If grade ≥2 CRS or related AEs occur, the study will switch to the 3+3 dose escalation mode. If no such events occur, escalation will follow 3×10^7 cells /kg, 5×10^7 cells /kg, and 7×10^7 cells /kg doses using the 3+3 design, with 3 patients enrolled per dose level. Each CAR-T intravenous infusion involves two consecutive administrations over two days. Approximately two weeks after the first cycle of CAR-T infusions, a second cycle of two consecutive CAR-T infusions will be administered over two days using the same dose as the first cycle.The imaging re-examination will be performed approximately two weeks after completing one cycle of cell reinfusion (i.e., post-second CAR-T infusion), serving as the observation window for safety indicators such as dose-limiting toxicity (DLT) graded per NCI-CTCAE v5.0 criteria in the combined immunotherapy. Based on imaging findings, safety data, and investigator evaluation, subsequent indications for interventional therapy and continued autologous reinfusion will be determined.

The trial progresses through sequential Phase IA (dose-finding) and Phase IB (dose-expansion) stages. Phase IA cessation triggers include either confirmation of Recommended Phase II Dose (RP2D) or investigator-verified favorable therapeutic index at any dose level.

An interim analysis will be conducted upon Phase IA completion, followed by submission of an ethical amendment application prior to initiating Phase IB.

The interim analysis incorporates exploratory subgroup analyses stratified by factors including tumor histology, NKG2D ligand expression levels, prior treatment regimens, baseline tumor burden, tumor markers, peripheral blood granulocyte-to-lymphocyte ratio, and metastatic sites during dose escalation.

Following the completion of Phase IA, investigators may identify tumor types likely responsive to NKG2D CAR immune cell therapy based on preliminary data and existing literature, and select patients with potentially responsive tumor types for expanded cohort enrollment, treating advanced solid tumor patients with liver metastases who have failed standard treatments. In Phase IB, investigators will conduct multiple dose expansion cohorts , divided into colorectal cancer, gastric cancer, esophageal cancer, and other cohorts, with 9-18 cases each.

Inclusion Criteria:

1. Age between 18 and 75 years (inclusive of boundary values), both males and females are eligible.

2. Advanced solid tumors with liver metastasis that have failed prior standard treatment (including disease progression and intolerable adverse reactions). The colorectal cancer cohort will be enrolled first (prior standard treatment includes combination or sequential therapy with fluorouracil, oxaliplatin, and irinotecan, with or without bevacizumab and/or cetuximab and/or regorafenib and/or fruquintinib).

3. Patients with high expression of NKG2D ligands in tumor tissue sections will be prioritized for enrollment.

4. The expected survival of the subjects is ≥12 weeks.

5. Subjects must have at least one target lesion that can be stably assessed according to the RECIST v.1.1 criteria by CT, MRI, or PET-CT. The target lesion should have measurable dimensions (tumor lesion long diameter ≥10 mm on CT scan, lymph node lesion short diameter ≥15 mm on CT scan, and scan slice thickness no more than 5 mm). Alternatively, through laparoscopic exploration, there should be at least one target lesion that can be assessed according to the PCI scoring criteria.

6. Eastern Cooperative Oncology Group (ECOG) performance status (PS) score of 0-1.

7. Subjects must have adequate organ and bone marrow function. Laboratory screening must meet all of the following criteria, with all laboratory test results within the specified stable range and without ongoing supportive therapy.

   1. Hematology: White blood cell count (WBC) ≥1.5×10⁹/L; platelet count (PLT) ≥60×10⁹/L; hemoglobin (Hb) ≥8.0 g/dL; lymphocyte count (LYM) ≥0.4×10⁹/L.
   2. Biochemistry: Serum creatinine ≤1.5×ULN. If serum creatinine >1.5×ULN, creatinine clearance rate must be >50 mL/min (calculated by the Cockcroft-Gault formula); serum total bilirubin ≤1.5×ULN; alanine aminotransferase (ALT) ≤2×ULN; aspartate aminotransferase (AST) ≤2×ULN (for patients with liver metastasis or primary liver cancer, ALT ≤5×ULN and AST ≤5×ULN); amylase and lipase ≤1.5×ULN.
   3. Urinalysis: Urine protein <2+.

8. Echocardiogram within the past month showing left ventricular ejection fraction (LVEF) >45%.

9. Fertility status: Women of childbearing potential or men whose sexual partners are women of childbearing potential must agree to use effective contraception from the time of signing the informed consent form until 6 months after the last cell infusion (women of childbearing potential include premenopausal women and women within 2 years after menopause).

10. Subjects must sign a written informed consent form and date it.

11. Subjects must be willing and able to comply with the prescribed treatment plan, laboratory tests, follow-up visits, and other study requirements.

Exclusion Criteria:

1. Pregnant or breastfeeding women.

2. Known history of human immunodeficiency virus (HIV) infection; acute or chronic active hepatitis B (HBsAg positive); acute or chronic active hepatitis C (HCV antibody positive). Positive syphilis antibody; EB virus DNA quantitative >500 copies; cytomegalovirus (CMV) infection (IgM positive).

3. Active or poorly controlled severe infections.

4. Presence of severe arterial embolism identified by CT angiography or hepatic arterial vascular variations that are unfavorable for HAI treatment.

5. Current presence of cardiac disease requiring treatment or poorly controlled hypertension as judged by the investigator (defined as systolic blood pressure ≥140 mmHg and/or diastolic blood pressure >90 mmHg after standardized antihypertensive drug treatment).

6. Presence of any of the following cardiac clinical symptoms or diseases:

   1. Unstable angina.
   2. Myocardial infarction within the past year.
   3. Resting electrocardiogram (ECG) showing QTc >450 ms (male) or QTc >470 ms (female).
   4. Resting ECG revealing clinically significant abnormalities (such as abnormal heart rate, conduction, or morphological features) or complete left bundle branch block or third-degree heart block or second-degree heart block or PR interval >250 ms.
   5. Presence of factors that increase the risk of QTc prolongation or arrhythmias, such as heart failure, hypokalemia, congenital long QT syndrome, family history of long QT syndrome or sudden unexplained death in first-degree relatives under the age of 40, or use of drugs that prolong the QT interval.

7. Coagulation abnormalities (INR >1.5×ULN), tendency to bleed, or undergoing thrombolytic or routine anticoagulant therapy (e.g., warfarin or heparin). Patients requiring long-term antiplatelet therapy (aspirin, dose >300 mg/day; clopidogrel, dose >75 mg/day).

8. Subjects who require systemic treatment with corticosteroids or other immunosuppressive drugs during the treatment period.

9. Oxygen saturation ≤95% (measured by pulse oximetry) before treatment.

10. Systemic corticosteroid use equivalent to >15 mg/day prednisone within 4 weeks before treatment (excluding inhaled corticosteroids).

11. Development of new arrhythmias in the subject before lymphodepletion conditioning, including but not limited to uncontrolled arrhythmias with medication, hypotension requiring vasopressors, bacterial, fungal, or viral infections requiring intravenous antibiotics. Subjects receiving prophylactic antibiotics for infection will be assessed by the investigator for continued eligibility.

12. Known history of or current need for treatment of hepatic encephalopathy; subjects with current or history of central nervous system disorders, such as seizures, cerebral ischemia/infarction, dementia, cerebellar disease, or any autoimmune diseases involving the central nervous system; subjects with clinically symptomatic central nervous system metastases or leptomeningeal metastases, or other evidence indicating that the central nervous system metastases or leptomeningeal metastases are not controlled, and deemed unsuitable for enrollment by the investigator.

13. Subjects with previous or concurrent other malignancies, with the following exceptions: a) Adequately treated basal cell or squamous cell carcinoma (with sufficient wound healing required before enrollment in the study). b) Cervical carcinoma in situ or ductal carcinoma in situ of the breast, treated with curative intent, with no signs of recurrence for at least 3 years before the study. c) Primary malignancy that has been completely resected and in complete remission for ≥5 years.

14. Subjects with severe psychiatric disorders.

15. Participation in another clinical study within the past month.

16. Subjects assessed by the investigator as unable or unwilling to comply with the requirements of the study protocol.

17. Subjects who have withdrawn from the study for any reason and cannot re-enroll.