Combination of notch4 inhibitor and ctla-4 inhibitor for the preparation of drugs for treating hepatocellular carcinoma associated with hepatitis b cirrhosis

Abstract

The present application relates to the combined application of NOTCH4 inhibitor and CTLA-4 inhibitor in the preparation of drugs for treating hepatitis B cirrhosis related hepatocellular carcinoma, and belongs to the technical field of biological medicine. The present application combines NOTCH4 inhibitor and CTLA-4 inhibitor to prepare drugs for treating hepatitis B cirrhosis related hepatocellular carcinoma. The present application finds that the drug resistance of CHCC to CTLA-4 inhibitor is related to the specific expansion of a new endothelial cell subpopulation in the tumor microenvironment, NOTCH4 + Capillary endothelial cell dysfunction is closely related. NOTCH4 inhibitor can target this subpopulation, correct vascular dysfunction and reverse the immunosuppressive microenvironment, thereby producing a synergistic antitumor effect with CTLA-4 inhibitor, significantly inhibiting the growth, metastasis of hepatitis B cirrhosis related hepatocellular carcinoma and improving liver fibrosis. The combined scheme provides a new effective strategy for overcoming the immunotherapy resistance of hepatitis B cirrhosis related hepatocellular carcinoma.

Description

Technical Field

[0001] This invention relates to the combined use of NOTCH4 inhibitors and CTLA-4 inhibitors in the preparation of drugs for treating hepatocellular carcinoma associated with hepatitis B cirrhosis, and belongs to the field of biomedical technology. Background Technology

[0002] Hepatocellular carcinoma (HCC) is one of the leading causes of cancer-related deaths worldwide. Cirrhosis-related hepatocellular carcinoma (CHCC), caused by chronic hepatitis B virus (HBV) infection leading to cirrhosis, accounts for a significant proportion and has an extremely poor prognosis. In recent years, immune checkpoint inhibitors, especially NOTCH4 inhibitors, have become an important option for systemic treatment of advanced HCC. However, a serious clinical challenge lies in the fact that CHCC patients generally exhibit primary or secondary resistance to immunotherapies such as cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4) inhibitors, with significantly lower objective response rates compared to non-cirrhotic HCC patients. The mechanism of immunotherapy resistance driven by a cirrhotic background remains unclear, resulting in a lack of effective treatment strategies for the majority of HCC patients (CHCC), becoming a key bottleneck in current clinical practice.

[0003] To overcome immune resistance, the academic community has explored various combination strategies. One important approach is the combination of drugs targeting tumor angiogenesis (such as VEGFR inhibitors) with immune checkpoint inhibitors. However, in chronic liver cancer (CHCC), the efficacy improvement of such combination regimens is limited and often accompanied by increased adverse reactions. This suggests that the underlying issue in CHCC may not be simply excessive angiogenesis, but rather a deeper vascular dysfunction related to the pathological microenvironment of liver cirrhosis, a fundamental deficiency that current anti-angiogenic drugs have failed to correct.

[0004] Therefore, a key, yet largely unrecognized question remains in this field: In the specific pathological context of CHCC, is there a core mechanism connecting the cirrhotic microenvironment, vascular dysfunction, and immunotherapy resistance? Can an interventional target be found that, while correcting vascular function, reverses the immunosuppressive microenvironment, thereby creating conditions for a response to immunotherapies such as CTLA-4 inhibitors? Answering this question is a prerequisite for developing new therapies that can effectively overcome immunotherapy resistance in CHCC.

[0005] The NOTCH signaling pathway, especially the NOTCH4 receptor, has been reported to be abnormally activated in various tumors, including HCC, and to participate in regulating tumor cell proliferation. However, its mechanism of action is generally understood to be the direct killing of tumor cells.

[0006] This invention is the first to discover the role of NOTCH4 in tumor microenvironment remodeling, particularly in regulating tumor vascular endothelial cell function and immunosuppression. Furthermore, it was found that the combination of NOTCH4 inhibitors and CTLA-4 inhibitors can overcome immunosuppression resistance in CHCC. Summary of the Invention

[0007] The technical problem to be solved by the present invention is to provide a combination drug regimen that can effectively overcome the resistance of hepatocellular carcinoma to CTLA-4 inhibitors. This regimen is based on a new mechanism and achieves synergistic effect by targeting the tumor microenvironment.

[0008] The first objective of this invention is to provide the combined use of NOTCH4 inhibitors and CTLA-4 inhibitors in the preparation of medicaments for the treatment of hepatocellular carcinoma associated with hepatitis B cirrhosis.

[0009] To achieve the first objective of this invention, this invention combines a NOTCH4 inhibitor with a CTLA-4 inhibitor to prepare a drug for treating hepatocellular carcinoma associated with hepatitis B cirrhosis.

[0010] In one specific embodiment, the combined application is the preparation of a drug for treating hepatocellular carcinoma associated with hepatitis B cirrhosis that is resistant to or poorly responsive to CTLA-4 inhibitor monotherapy.

[0011] In one specific embodiment, the mass ratio of the NOTCH4 inhibitor to the CTLA-4 inhibitor is 1.8 to 2.2:1.

[0012] In one specific implementation, the combined application uses a NOTCH4 inhibitor to reverse the immunosuppressive state mediated by dysfunctional capillary endothelial cells in the tumor microenvironment, thereby restoring the efficacy of the CTLA-4 inhibitor.

[0013] In one specific embodiment, the NOTCH4 inhibitor and the CTLA-4 inhibitor produce a synergistic anti-tumor effect when used in combination, with a combination index (CI) of less than 1.

[0014] In one specific embodiment, the NOTCH4 inhibitor is an anti-NOTCH4 monoclonal antibody.

[0015] Preferably, the CTLA-4 inhibitor is an anti-CTLA-4 monoclonal antibody.

[0016] A second object of the present invention is to provide a pharmaceutical composition for treating hepatocellular carcinoma associated with hepatitis B cirrhosis.

[0017] To achieve a second objective of the present invention, the pharmaceutical composition comprises a therapeutically effective amount of a NOTCH4 inhibitor and a CTLA-4 inhibitor, as well as a pharmaceutically acceptable carrier.

[0018] Preferably, the mass ratio of the NOTCH4 inhibitor to the CTLA-4 inhibitor is 1.8 to 2.2:1.

[0019] Beneficial effects

[0020] 1. Synergistic Effect and Reversal of Drug Resistance: This invention is the first to propose a combined approach targeting NOTCH4 and CTLA-4 for the treatment of CHCC. Experiments have demonstrated that this combined regimen is significantly superior to either single-agent therapy in inhibiting tumor growth, reducing metastasis, and alleviating liver fibrosis, exhibiting a clear synergistic effect (CI < 1). More importantly, this regimen can effectively reverse the inherent drug resistance of CHCC to CTLA-4 inhibitors.

[0021] 2. Unique Mechanism: The core mechanism of this invention lies in the fact that NOTCH4 inhibitors do not directly kill tumor cells, but rather target NOTCH4, a novel endothelial cell subset that specifically proliferates and functions abnormally in the CHCC tumor microenvironment. + Capillary ECs (notch receptor 4-positive capillary ECs, NOTCH4) + CapECs. By correcting the "NOTCH4 signaling uncoupling" state of this subset, it promotes the normalization of vascular structure and function, and upregulates the antigen-presenting function of endothelial cells, thereby systematically reshaping the immunosuppressive microenvironment and laying the foundation for the reactivation of T cell function by CTLA-4 inhibitors. Combined therapy can significantly reduce the exhaustion level of tumor-infiltrating CD8⁺ T cells (e.g., a decrease in the proportion of PD-1⁺TOX⁺ cells) while enhancing their effector function.

[0022] 3. Addressing clinical challenges: This invention directly addresses the significant clinical challenge of low immunotherapy response rates in CHCC patients, providing a combination therapy strategy with strong synergistic effects based on novel biological insights, and has extremely high clinical translational value. Attached Figure Description

[0023] Figure 1 Example 1: Quantitative graph of the number of intrahepatic tumor nodules at the endpoint.

[0024] Figure 2 Example 1: Endpoint liver weight quantification graph.

[0025] Figure 3 Quantitative graph of liver / body weight ratio in Example 1.

[0026] Figure 4 Example 1: Quantitative graph of photon flux change factor.

[0027] Figure 5 : Bar chart of mesenteric metastasis incidence in Example 1.

[0028] Figure 6 Bar chart of lung metastasis incidence in Example 1.

[0029] Figure 7 Quantitative graph of collagen area percentage in Example 1.

[0030] Figure 8 Quantitative graph of Sirius red positive area percentage in Example 1.

[0031] Figure 9 Example 1: Quantitative graph of tumor burden percentage.

[0032] Figure 10 Volcano plot of differentially expressed genes G6 vs G5 in Example 1.

[0033] Figure 11 Example 1: Vascular function-related feature scoring chart.

[0034] Figure 12 : Comparison of CD8⁺ T cell exhaustion score, effector / memory score and TCR clonality in Example 1.

[0035] Figure 13 Representative flow cytometry plots from Example 1 (G5 vs G6, co-expression of PD-1 and TOX).

[0036] Figure 14 Flow cytometry quantitative images of Examples 1 and 2: proportion of PD-1⁺TOX⁺ exhausted cells, proportion of GZMB⁺ effector cells, and proportion of CD44⁺ memory cells.

[0037] in, Figure 1-14 In the study, G1 was the isotype control antibody group (Control); G4 was the high-dose anti-NOTCH4 group (0.2 mg / kg); G5 was the anti-CTLA-4 monotherapy group (0.1 mg / kg); and G6 was the combination therapy group of high-dose anti-NOTCH4 (0.2 mg / kg) + anti-CTLA-4 (0.1 mg / kg). Figures 15-20In the study, IgG was the IgG control group, anti-NOTCH4 was the anti-NOTCH4 monotherapy group, anti-CTLA-4 was the anti-CTLA-4 monotherapy group, Combo was the anti-NOTCH4 + anti-CTLA-4 combination therapy group, and SoC was the anti-PD-L1 + anti-VEGFR2 positive control group.

[0038] Figure 15 Example 3: Longitudinal tumor volume curve of humanized PDX tumor-bearing mice (treatment group: IgG control, anti-NOTCH4, anti-CTLA-4, combination therapy, anti-PD-L1 + anti-VEGFR2 positive control).

[0039] Figure 16 Example 3: Quantitative graph of tumor weight / body weight ratio as the endpoint.

[0040] Figure 17 Example 3: Integrated vascular-immune quadrant analysis diagram.

[0041] Figure 18 Example 3: Immunocharacteristics scores of human transcriptomes for each treatment group (CD8⁺ T cells, cytotoxicity, exhaustion / cytotoxicity ratio, immune activation, IFN-γ response, MHC-I antigen presentation, MHC-II antigen presentation, T cell exhaustion).

[0042] Figure 19 Example 3: Transcriptome vascular / matrix feature score maps of mice in each treatment group (barrier integrity, CapEC_NOTCH4, fibrosis / ECM, hypoxia, nonfunctional vessels, stalk / mature, tip-like, vascular leakage, vascular normalization).

[0043] Figure 20 Scatter plot of cross-species association analysis in Example 3. Detailed Implementation

[0044] The specific embodiments of the present invention will be further described below with reference to examples, but the present invention is not limited to the scope of the embodiments described herein.

[0045] Example 1

[0046] A study on the efficacy and synergistic effect of combined use of NOTCH4 inhibitors (anti-NOTCH4 antibodies) and CTLA-4 inhibitors (anti-CTLA-4 antibodies) in the treatment of CHCC.

[0047] 1. Animal model establishment: A CHCC model was established in 8-week-old male BALB / c mice.

[0048] 2. Grouping and Treatment: CHCC mice were randomly divided into six groups: isotype control antibody group (G1, Control), low-dose anti-NOTCH4 antibody group (G2, 0.1 mg / kg), medium-dose anti-NOTCH4 antibody group (G3, 0.15 mg / kg), anti-NOTCH4 antibody monotherapy group (G4, NOTCH4-i-High), anti-CTLA-4 antibody monotherapy group (G5, Anti-CTLA4), and anti-NOTCH4 + anti-CTLA-4 combined treatment group (G6, Anti-CTLA4 + NOTCH4-i-High). The anti-NOTCH4 antibody was administered at a dose of 0.2 mg / kg, and the anti-CTLA-4 antibody was administered at a dose of 0.1 mg / kg. Both were administered via intraperitoneal injection once every 3 days for a total of 3 times.

[0049] 3. Observation indicators:

[0050] Tumor burden (number of liver tumor nodules, liver weight, liver / body weight ratio, fold change in photon flux during in vivo imaging);

[0051] Incidence of intrahepatic / distant metastases (lung, mesentery, adrenal glands, stomach, etc.);

[0052] The degree of liver fibrosis (percentage of collagen area stained by Masson staining, percentage of Sirius red positive area);

[0053] Vascular function-related transcriptomic features (tip / stalk index, barrier / leakage function, hypoxia level, and NOTCH upstream and downstream imbalance index UDI).

[0054] Immune microenvironment (MHC-I / II and antigen-presenting molecule expression, CD8) + T cell exhaustion and effector function, TCR clonality).

[0055] 4. Experimental Results

[0056] (1) Anti-NOTCH4 monotherapy showed dose-dependent inhibition of tumor growth.

[0057] Compared with the control group, anti-NOTCH4 antibody treatment produced a clear dose-dependent inhibition of tumor growth. Gross examination showed that with increasing antibody dose, liver enlargement and tumor burden gradually decreased. Quantitative analysis confirmed that, compared with the control group, the high-dose group (G4) had significantly reduced intrahepatic tumor nodules, liver weight, and liver / body weight ratio (all P < 0.001), and no systemic toxicity was detected by weight assessment.

[0058] Longitudinal bioluminescence imaging further confirmed robust tumor growth inhibition. Baseline tumor burden was comparable across groups, but tumor signal expansion was significantly suppressed after treatment in anti-NOTCH4-treated mice. The tumor fold increase in the high-dose group (G4) was significantly lower than that in the control and low-dose groups (P < 0.001). Anti-NOTCH4 treatment also reduced distant metastasis, particularly to high-risk organs such as the lungs, mesentery, adrenal glands, and stomach, with the highest efficacy observed in the high-dose group (comparisons of categorical variables were performed using chi-square or Fisher's exact test; P < 0.05).

[0059] Histopathological analysis revealed that NOTCH4 inhibition remodeled the liver microenvironment bearing tumors. Control group livers showed extensive tumor invasion, structural disruption, and collagen deposition, while anti-NOTCH4 treatment reduced tumor invasion and collagen-rich matrix deposition in this early in situ model, as confirmed by H&E, Masson trichrome staining, and Sirius red staining. Quantitative analysis confirmed that the high-dose group showed significantly reduced tumor burden percentage, collagen area percentage, and Sirius red positive area percentage (all P < 0.01).

[0060] To elucidate the molecular basis of these effects, we performed RNA sequencing on tumor tissues. Anti-NOTCH4 antibody treatment did not significantly alter Notch receptor mRNA expression, consistent with antibody-mediated pathway regulation rather than Notch receptor transcriptional repression. Endothelial NOTCH downstream activity decreased in a dose-dependent manner, accompanied by extensive transcriptional remodeling. Functionally, anti-NOTCH4 treatment shifted vascular programming toward improved maturation-related characteristics, increased barrier-related characteristics, and dose-dependent reductions in leakage, hypoxia, and the Notch upstream / downstream imbalance (UDI) index (all P < 0.01).

[0061] In addition to vascular normalization, NOTCH4 inhibition profoundly reprogrammed the immune microenvironment. Gene ontology analysis revealed significant upregulation of immune-related pathways, including lymphocyte-mediated immunity, T cell activation, and interferon responses, particularly in the high-dose group. TCR clonal analysis showed enhanced clonal expansion of T cells in the treatment group, indicating a recovery of antigen-specific immunity. Consistently, transcription of MHC-I / II molecules and antigen-presenting components significantly increased after treatment. GSEA further confirmed the activation of cytotoxic, immune effector, and cytokine production pathways. At the cellular level, NOTCH4 blockade shifted CD8⁺ T cells from a state of exhaustion to an effector memory state. Transcriptome characterization showed a dose-dependent decrease in exhaustion scores, while effector memory characteristics increased (both P < 0.01).

[0062] Flow cytometry protein level validation confirmed the above findings. Anti-NOTCH4 antibody treatment significantly reduced the proportion of PD-1⁺TOX⁺ exhausted CD8⁺ T cells, while dose-dependently increasing the expression of the effector molecule GZMB and the activation / memory marker CD44 (both P < 0.001).

[0063] (2) Combined therapy further enhances the anti-tumor effect

[0064] Given that NOTCH4⁺ endothelial cells spatially colocalize with CTLA-4⁺ T cells and preferentially participate in the inhibitory CTLA-4–CD80 / CD86 interaction rather than the PD-1 / PD-L1 axis, we hypothesize that simultaneous targeting of NOTCH4 and CTLA-4 can enhance antitumor activity by combining vascular remodeling with CTLA-4-guided T cell activation. To validate this, we compared control group (G1), high-dose anti-NOTCH4 (G4), anti-CTLA-4 (G5), and combination therapy (G6) in an H22 orthotopic hepatocellular carcinoma model.

[0065] At the endpoint, the combination therapy group (G6) showed a significant reduction in tumor burden. Quantitative analysis confirmed that, compared with all single-agent therapy groups, the G6 group had a significantly lower number of intrahepatic tumor nodules. Figure 1 ), liver weight ( Figure 2 ) and liver / body weight ratio ( Figure 3 All values ​​were the lowest (P < 0.001), and there was no significant change in body weight.

[0066] Longitudinal bioluminescence imaging showed balanced baseline tumor burden across groups. After treatment, the combined treatment group exhibited the strongest tumor growth inhibition, with the lowest fold change in photon flux among all groups. Figure 4 P < 0.001. Combination therapy was also associated with reduced involvement of metastases in multiple organs, including the lungs, mesentery, adrenal glands, and spleen. Figure 5 – Figure 6 Comparisons of categorical variables were performed using the chi-square test or Fisher's exact test, with P < 0.05.

[0067] Histopathological analysis showed that the combined treatment reduced tumor invasion and collagen-rich matrix deposition in the orthotopic model. Quantitative analysis confirmed that the tumor burden percentage in the G6 group ( Figure 9 The percentage of collagen area and Sirius red positive area were significantly lower than those of all other groups (all P < 0.01). Fibrosis indicators (percentage of collagen area and percentage of Sirius red positive area) were significantly lower than those of the control group and anti-CTLA-4 monotherapy, and showed a further decreasing trend compared to anti-NOTCH4 monotherapy. Figure 7 – Figure 8 (All P < 0.01).

[0068] Transcriptome analysis revealed unique molecular characteristics in the combination therapy group. Compared to anti-CTLA-4 monotherapy, the combination therapy significantly downregulated the hemoglobin gene (…). Figure 10 The combination therapy also most effectively suppressed immunosuppressive transcripts and potently induced MHC-I / II molecules and antigen presentation mechanisms (B2m, Cd74).

[0069] Functionally, the combination therapy produced the most favorable vascular remodeling characteristics across all study groups. The G6 group had the lowest tip-like index and the highest stalk-like index, the highest barrier integrity score, and the lowest leakage, hypoxia, and UDI indices. Figure 11 (All P < 0.001). At the immune level, the combination therapy produced the strongest shift to CD8⁺ T cell effector and memory-related programs in all subjects. RNA-seq scores showed that the G6 group had the lowest exhaustion characteristics and the highest effector memory characteristics and TCR clonality among all groups. Figure 12 All P < 0.001. Flow cytometry confirmed that the proportion of PD-1⁺TOX⁺ depleted CD8⁺ T cells ( Figure 13 In the combination therapy group, the proportions of GZMB⁺ effector cells and CD44⁺ memory T cells were significantly reduced. Figure 14 Significantly increased (all P < 0.001).

[0070] Example 2

[0071] Based on the data from Example 1, synergistic effects were calculated: To quantitatively demonstrate the synergistic effect, the Chou-Talalay combination index method was used to reanalyze the tumor volume inhibition data. Specifically, the liver tumor burden (number of tumor nodules) at the endpoint of day 20 was used as the effect index. The actual inhibition rates of anti-NOTCH4 monotherapy (G4, 0.2 mg / kg), anti-CTLA-4 monotherapy (G5, 0.1 mg / kg), and combination therapy (G6) at this fixed dose combination were calculated and compared with the theoretical additive effect when the two drugs were used alone. The combination index was calculated using CompuSyn software based on the intermediate-efficacy principle.

[0072] The results showed that the CI value of the combined treatment group was 0.88, indicating that anti-NOTCH4 antibody and anti-CTLA-4 antibody had a clear synergistic anti-tumor effect in the CHCC model. This synergistic effect was verified by changes in tumor nodule number, liver weight, photon flux fold change, and CD8⁺ T cell function indicators detected by flow cytometry (decreased proportion of PD-1⁺TOX⁺ exhausted cells, increased proportion of GZMB⁺ and CD44⁺ cells). Figure 1 – Figure 14 ).

[0073] Example 3

[0074] Validation of combination therapy in humanized CHCC-derived PDX models

[0075] Fresh tumor tissue was obtained from CHCC patients who had given written informed consent and received institutional ethical approval. Tumor specimens were placed in sterile RPMI-1640 medium containing antibiotics, transported on ice, and processed under aseptic conditions. After removing necrotic tissue, fat, and blood clots, the viable tumor tissue was cut into small pieces approximately 2×2×2 mm. Under isoflurane anesthesia, the tumor fragments were subcutaneously implanted into 4-week-old male M-NSG mice. Tumor growth was monitored using calipers, and tumor volume was calculated using the formula V = L × W. 2 The formula is calculated as / 2, where L is the longest diameter and W is the shortest diameter. After tumor formation, mice are continuously passaged, and tumor-bearing mice with suitable tumor size are used for treatment experiments.

[0076] Human peripheral blood mononuclear cells (PBMCs) were prepared under aseptic conditions, resuspended in PBS before injection, and PBMC viability was confirmed. Tumor-bearing M-NSG mice were injected with 1 × 10⁶ PBMCs via tail vein. 7 To establish a humanized immune system, peripheral blood was collected after humanization, and the reconstitution of human immune cells was assessed by flow cytometry. Mice with a peripheral blood CD45⁺ cell ratio ≥25% were considered successfully humanized and eligible for enrollment in the treatment trial. Mice with insufficient humanized immune reconstitution, unsuitable tumor size, excessive weight loss, obvious graft-versus-host disease, or poor general condition before treatment were excluded before randomization.

[0077] A total of 82 male M-NSG mice were used for PDX establishment and screening. After tumor size qualification and human immune reconstitution assessment, 35 humanized PDX tumor-bearing mice were enrolled for the treatment study, with 7 mice in each group. Qualified mice were stratified according to baseline tumor volume and randomly assigned to the following 5 treatment groups: IgG control group, anti-NOTCH4 monotherapy group, anti-CTLA-4 monotherapy group, anti-NOTCH4 + anti-CTLA-4 combination therapy group, and anti-PD-L1 + anti-VEGFR2 positive control group. The anti-PD-L1 + anti-VEGFR2 group served as a preclinical VEGF / VEGFR pathway blocking control for the clinically established anti-PD-L1 + anti-VEGF treatment axis.

[0078] Dosage regimen: All antibodies were administered intraperitoneally. The IgG control group received isotype control IgG administered according to the regimen matched to the active treatment group. The anti-NOTCH4 dose was 0.2 mg / kg twice weekly; the anti-CTLA-4 dose was 0.1 mg / kg once weekly. The combination therapy group received both anti-NOTCH4 and anti-CTLA-4 simultaneously, following the corresponding monotherapy regimens: anti-NOTCH4 0.2 mg / kg twice weekly; anti-CTLA-4 0.1 mg / kg once weekly. The positive control group received anti-PD-L1 0.1 mg / kg and anti-VEGFR2 0.1 mg / kg, both administered intraperitoneally twice weekly. Tumor volume and body weight were measured twice weekly during treatment. Figure 7 In the treatment cohort shown in Q–S, tumor volume was monitored on days 41, 45, 48, 52, and 55 post-implantation, with endpoint tissue sampling performed on day 55.

[0079] At the endpoint, mice were euthanized according to the institution's animal care guidelines. Tumors were removed, photographed, weighed, and segmented for subsequent analysis. The tumor weight / body weight ratio was calculated for each mouse. Endpoint criteria included excessive tumor burden, ulceration, weight loss ≥15%, signs of graft-versus-host disease, or other distress requiring euthanasia. Tumor tissue was rapidly frozen for RNA sequencing or fixed for histological and immunological analysis. In the enrolled treatment cohort, each group initially consisted of 7 mice. One mouse in each of the IgG control group and the anti-CTLA-4 monotherapy group died the day before the planned endpoint tissue collection. The longitudinal tumor volume data of these mice were still included in the analysis up to the last valid measurement, but their endpoint tumor weight, tumor weight / body weight ratio, and transcriptome data of the tumor tissue required for endpoint analysis were excluded.

[0080] RNA-seq library construction and species-differentiated transcriptome analysis

[0081] Total RNA was extracted from the endpoint PDX tumor tissue using a standard RNA extraction kit, with RNA concentration and integrity assessed prior to extraction. Libraries were constructed using a strand-specific RNA-seq library construction kit according to standard procedures. After RNA fragmentation, first-strand cDNA was synthesized using random primers and reverse transcriptase, followed by second-strand cDNA synthesis. cDNA fragments underwent end repair, A-tailing, adapter ligation, PCR enrichment, and library purification. The purified libraries were quantified using a standard quantitative PCR instrument and quality verified using a bioanalyzer. Sequencing was performed on a standard high-throughput sequencing platform. 33–95 million valid RNA-seq reads were obtained from each sample.

[0082] Because humanized PDX tumors contain both human tumor / immune components and mouse matrix / vascular components, a species-resolved analysis workflow was used for sequencing reads. Effective reads were aligned to a joint human and mouse reference genome. Reads uniquely aligned to either the human or mouse genome were retained for species-specific analysis, while ambiguous reads aligned to both species were discarded. Subsequently, human and mouse aligned reads were analyzed separately.

[0083] Raw gene counts were generated using StringTie, and differential expression analysis was performed using the R package edgeR. TMM normalization was used for inter-sample normalization, and FPKM values ​​were calculated for visualization and transcriptome feature scoring. P-values ​​were corrected using the Benjamini-Hochberg false discovery rate method with multiple tests, and the corrected P-values ​​are expressed as q-values. Differentially expressed genes were defined as those with a fold change >2 and a q-value <0.05. Gene ontology enrichment analysis was performed using the R package clusterProfiler, with a q-value <0.05 considered significant. Heatmaps were generated using the R package pheatmap, and principal component analysis results were visualized using the R package scatterplot3d.

[0084] Human comparative transcriptomics was used to assess the characteristics of human immunization programs, including CD8. + T cell, cytotoxicity, IFN response, immune activation, MHC-I antigen presentation, MHC-II antigen presentation, and T cell exhaustion characteristics. Mouse comparative transcriptomes were used to assess host matrix and vascular programming characteristics, including CapEC_NOTCH4, barrier integrity, stalk / mature, tip-like features, vascular leakage, hypoxia, non-functional vessels, fibrosis / ECM, and vascular normalization characteristics. Characteristic scores were calculated based on normalized expression values ​​using a given gene set and were z-score normalized between samples for visualization and comparison.

[0085] Experimental results:

[0086] (1) Anti-tumor efficacy

[0087] Under conditions combined with a humanized immune environment, the combination therapy group exhibited the strongest tumor control among all test groups. Longitudinal tumor volume curves in humanized PDX-bearing mice showed that the combination therapy group had the slowest tumor growth, even compared to the anti-PD-L1 + anti-VEGFR2 positive control group. Figure 15 Overall comparisons between groups: P < 0.001. Quantitative analysis of the endpoint tumor weight / body weight ratio showed that the combination therapy group had the lowest ratio. Figure 16 (P < 0.001 for overall comparisons between groups).

[0088] (2) Integrated vascular-immune quadrant analysis

[0089] Integrated vascular-immune quadrant analysis based on species-discriminate transcriptome data showed that the combination therapy group was located in the most favorable quadrant of "high mouse vascular normalization + high human immune activation" ( Figure 17 ).

[0090] (3) Species-distinguished transcriptome features

[0091] In the human transcriptome, the immune characteristic scores of the human transcriptome in each treatment group showed that the combination treatment group had the strongest immune activation, including significantly increased CD8⁺ T cells, cytotoxicity, immune activation, IFN-γ response, MHC-I antigen presentation, and MHC-II antigen presentation, while the exhaustion / cytotoxicity ratio was significantly reduced. Figure 18 (P < 0.001 for overall comparisons between groups).

[0092] In the mouse transcriptome, the vascular / matrix characteristic scores of the transcriptomes of mice in each treatment group showed that the combined treatment group had the most significant vascular normalization, as evidenced by significant reductions in CapEC_NOTCH4, hypoxia, nonfunctional vessels, vascular leakage, and fibrosis / ECM characteristic scores, while barrier integrity, stalk / mature, and vascular normalization characteristic scores were significantly increased. Figure 19 (P < 0.001 for overall comparisons between groups).

[0093] Scatter plots from cross-species association analysis showed that mouse vascular normalization scores were positively correlated with human immune activation scores, while CapEC_NOTCH4, hypoxia, and nonfunctional vascular scores were negatively correlated with human immune activation. Figure 20 ***P <0.001).

[0094] The above results indicate that NOTCH4 inhibitors promote vascular normalization by targeting and correcting abnormal NOTCH4⁺ capillary endothelial cells in the CHCC microenvironment, thereby relieving immunosuppression and creating conditions for CTLA-4 inhibitors to exert their anti-tumor immune effects. The combination of the two produces a synergistic anti-tumor effect.

Claims

1. The combined use of NOTCH4 inhibitors and CTLA-4 inhibitors in the preparation of drugs for treating hepatocellular carcinoma associated with hepatitis B cirrhosis.

2. The combined application according to claim 1, characterized in that, The combined application is used to treat hepatocellular carcinoma associated with hepatitis B cirrhosis that is resistant to CTLA-4 inhibitor monotherapy.

3. The combined application according to claim 1, characterized in that, The mass ratio of the NOTCH4 inhibitor to the CTLA-4 inhibitor is 1.8 to 2.2:

1.

4. The combined application according to any one of claims 1 to 3, characterized in that, The NOTCH4 inhibitor and the CTLA-4 inhibitor exhibit a synergistic anti-tumor effect when used in combination, with a synergistic index (CI) < 1.

5. The combined application according to claim 1, characterized in that, The NOTCH4 inhibitor is an anti-NOTCH4 monoclonal antibody.

6. The combined application according to claim 1, characterized in that, The CTLA-4 inhibitor is an anti-CTLA-4 monoclonal antibody.

7. A pharmaceutical composition for treating hepatocellular carcinoma associated with hepatitis B cirrhosis, characterized in that, The composition comprises a therapeutically effective amount of a NOTCH4 inhibitor and a CTLA-4 inhibitor, as well as a pharmaceutically acceptable carrier.

8. The pharmaceutical composition for treating hepatocellular carcinoma associated with hepatitis B cirrhosis according to claim 7, characterized in that, The mass ratio of the NOTCH4 inhibitor to the CTLA-4 inhibitor is 1.8 to 2.2:1.

Images