A Prospective Real-World Observational Study of Intratumoral Oncolytic Virus Injection Combined With Immune Checkpoint Inhibitors for Metastatic Tumors (Colorectal Cancer, Gastric Cancer, and Hepatocellular Carcinoma)

Background and Rationale Advanced metastatic colorectal cancer, gastric cancer, and hepatocellular carcinoma remain major causes of cancer-related mortality worldwide, particularly after failure of multiple lines of standard systemic therapies. Immune checkpoint inhibitors (ICIs) have shown limited efficacy as monotherapy in these tumors, partly due to an immunosuppressive tumor microenvironment (TME) characterized by sparse T-cell infiltration, high myeloid-derived suppressor cell (MDSC) accumulation, and dysfunctional antigen presentation. Oncolytic viruses, such as H101 (recombinant human type 5 adenovirus), can selectively replicate in tumor cells, induce immunogenic cell death, and release tumor-associated antigens, thereby converting a "cold" TME into a "hot" one. Preclinical and early clinical data suggest that intratumoral administration of oncolytic viruses may synergize with systemic ICIs by enhancing T-cell priming, increasing Programmed Death-Ligand 1 (PD-L1) expression, and remodeling the immune cell landscape. However, real-world evidence in the digestive cancer setting, especially for heavily pretreated patients, is scarce. This investigator-initiated study is therefore designed to generate preliminary data on the combination's safety, feasibility, and biological activity in a routine clinical practice context. Study Objectives Primary objective: To assess the safety \[Treatment-Related Adverse Events / Serious Adverse Events(TRAEs/SAEs)\], feasibility (treatment completion and modification rates), and preliminary antitumor activity (ORR and LCR by RECIST 1.1) of intratumoral H101 plus ICIs. Secondary objective: To evaluate disease control rate (DCR), progression-free survival (PFS), and longitudinal changes in the TME, including immune cell infiltration (T cells, NK cells, dendritic cells, macrophages), subset proportions, functional markers, and immune-related signaling pathways. Exploratory objective: To identify potential predictive biomarkers of response by integrating single-cell RNA sequencing (scRNA-seq) data from paired tumor biopsies and peripheral blood, and to characterize the molecular and cellular mechanisms underlying the combination therapy. Study Design This is a single-center, prospective, open-label, single-arm observational study conducted in a real-world clinical setting. Although the treatment is protocol-specified, all clinical decisions are made per routine multidisciplinary team (MDT) evaluation; the study does not add extra interventions beyond planned therapy but standardizes data collection, efficacy assessment, safety monitoring, and biospecimen banking. A total of 30-45 patients (10-15 per cancer type) will be enrolled. Participant Selection Eligible patients are adults (≥18 years) with histologically or cytologically confirmed advanced metastatic colorectal cancer, gastric cancer, or hepatocellular carcinoma, who have progressed on or were intolerant to at least two prior lines of systemic therapy (including chemotherapy and/or targeted agents). Patients must have at least one measurable lesion suitable for intratumoral injection per RECIST 1.1, adequate organ function, and an Eastern Cooperative Oncology Group performance status of 0-2. Key exclusion criteria include prior treatment with oncolytic viruses or ICIs (unless washout \>6 months for ICIs), active central nervous system metastases, uncontrolled intercurrent illnesses, or known allergy to adenoviral vectors. All patients are assessed by the institutional MDT and deemed appropriate for the combination regimen. Treatment Protocol The treatment schedule consists of 4 cycles, each lasting 2-3 weeks, as follows: Cycle 1: On Day 1, H101 is injected intratumorally at a dose stratified by the largest lesion diameter: ≤5 cm → 5.0×10¹¹ viral particles (vp); \>5 cm → 1.0×10¹² vp, divided into multiple injection points if necessary. On Day 5-7 (window), the first dose of ICI is administered intravenously: either anti-PD-1 monotherapy (e.g., sintilimab, toripalimab, or pembrolizumab at standard approved doses) or a combination of anti-PD-1 plus anti-CTLA-4 (e.g., ipilimumab at 1 mg/kg) per the treating physician's discretion and local label. Subsequent cycles (2-4): H101 injection is repeated on Day 1 of each cycle, followed by ICI on Day 5-7. Dose modifications for toxicity follow standard guidelines; treatment delays of up to 2 weeks are permitted, but discontinuation is required if toxicity does not resolve to ≤ Grade 1 within that period. Assessments and Data Collection Safety: All adverse events are recorded and graded using CTCAE v5.0 from the first dose until 30 days after the last dose. SAEs are reported within 24 hours. Efficacy: Tumor imaging (CT/MRI) is performed at baseline and after cycle 4 (or earlier if clinically indicated). Responses are independently reviewed by two radiologists according to RECIST 1.1. PFS is calculated from the first treatment date to disease progression or death from any cause. Biomarker sampling: Paired tumor biopsies (core needle or fine-needle aspiration from the injected lesion) and peripheral blood (10 mL) are collected at baseline (within 7 days before first H101) and at the end of cycle 4 (within 7 days after the last ICI). Samples are processed for scRNA-seq, flow cytometry, and multiplex cytokine assays. Data management: All data are entered into a secured electronic case report form (eCRF) with unique subject codes; personal identifiers are removed to protect privacy. Statistical Methods Given the exploratory nature, the sample size is not based on formal power calculation but is considered adequate to provide descriptive estimates and generate hypotheses. Continuous variables will be summarized as mean (SD) or median (IQR); categorical variables as frequencies and percentages. For primary endpoints: Safety and feasibility will be reported with descriptive statistics. ORR and LCR will be calculated with exact 95% confidence intervals (Clopper-Pearson method). DCR will be similarly reported. PFS will be estimated using the Kaplan-Meier method, with median time and 95% CI. For paired immune parameters (e.g., cell subset frequencies, cytokine levels), changes from baseline to post-treatment will be assessed using paired t-tests for normally distributed data or Wilcoxon signed-rank tests for non-normal distributions. All statistical tests are two-sided, with a significance level of α=0.05 (exploratory, no multiplicity adjustment). For scRNA-seq data, after quality control and normalization, unsupervised clustering (Seurat) will be performed to identify cell populations; differential expression analysis between responder and non-responder groups, or between pre- and post-treatment, will employ the MAST or Wilcoxon rank-sum test. Gene set enrichment analysis (GSEA) and pathway analysis (GO/KEGG) will be applied to characterize functional changes. Correlations between baseline immune features and clinical outcomes will be explored using Spearman's rank correlation. Ethical and Regulatory Considerations The study protocol has been approved by the institutional review board/ethics committee of the participating center. Written informed consent will be obtained from all patients prior to any study-specific procedures. The study will be conducted in full compliance with the Declaration of Helsinki (Fortaleza revision), Good Clinical Practice guidelines, and all applicable local regulations. Any protocol amendments will be submitted for ethical review and updated on this registry. Expected Significance This study is among the first to prospectively evaluate the H101-ICI combination in a real-world, heavily pretreated population of three major digestive cancers. The integrated clinical and multi-omics data are expected to provide crucial evidence on the safety profile and clinical benefit of this approach, while the mechanistic explorations may uncover immune biomarkers that guide patient selection and combination strategy refinement. The findings will inform the design of larger, randomized controlled trials and ultimately contribute to the development of personalized immunotherapeutic regimens for patients with limited treatment options.