CELL INTELLIGENCE · 肿瘤细胞治疗研究

Inclusion Criteria:

* Age ≥18 years.
* Histologically confirmed colorectal liver metastases.
* Administration of neoadjuvant chemotherapy prior to liver resection, with objective tumour response classified as partial response (PR) or stable disease (SD) according to RECIST criteria.
* Availability of a hepatobiliary contrast-enhanced MRI performed within 2 months before surgery.
* Provision of informed consent for prospectively enrolled participants, or eligibility under Article 110-bis of the Italian Privacy Code for retrospectively enrolled participants.

Exclusion Criteria:

* Recurrent metastatic disease.
* Liver resection performed with non-curative intent.
* Current or previous hepatitis B virus (HBV) or hepatitis C virus (HCV) infection.
* Concomitant malignancies or history of another malignancy treated within the previous 5 years.

主要终点：Prediction of Colorectal Liver Metastasis Recurrence · Prediction of colorectal liver metastasis recurrence following curative-intent liver resection using an integrated model based on radiogenomic, immune, and clinical profiling. Recurrence will be assessed through routine clinical follow-up and radiological evaluation. · 24 Months after liver resection
次要终点：Correlation Between Myeloid Immune Cell Frequencies and Clinical Outcome Measures；Genomic Alterations Associated With Immune Landscape Patterns；Radiomic Features Associated With Immune Cell Distribution；Performance of the Machine Learning-Based Recurrence Prediction Model

• Retrospective CRLM Cohort · Patients with histologically confirmed colorectal liver metastases who underwent curative-intent liver resection between January 2009 and December 2023 and for whom tumour tissue and clinical data are available. Retrospective collection and analysis of tumour tissue, radiological, genomic, immune and clinical data.
• Prospective CRLM Cohort · Patients with histologically confirmed colorectal liver metastases undergoing standard clinical management and curative-intent liver resection, enrolled prospectively with collection of tumour tissue and peripheral blood samples. Prospective collection and analysis of tumour tissue, peripheral blood-derived ctDNA, radiological, genomic, immune and clinical data.

The RaP-DMac-LiMe study (Radiogenomic Profiling of Dendritic Cells and Macrophages to Predict Recurrence in Colorectal Liver Metastasis) is a monocentric, non-profit observational study promoted by Fondazione Policlinico Universitario A. Gemelli IRCCS. Its primary aim is to identify immunological, genomic, and radiomic biomarkers associated with recurrence risk in patients with colorectal liver metastases (CRLM) undergoing curative-intent liver resection. The study is based on the need to improve prognostic stratification in CRLM by integrating information from the tumor immune microenvironment, tumor genomics, radiomics, and clinical data. Particular attention is given to myeloid immune cells, especially dendritic cells and tumor-associated macrophages, whose role in metastatic progression and recurrence remains insufficiently understood. The primary objective is to assess the association between myeloid immune profiles and recurrence risk through integrated molecular, spatial, genomic, and radiological analyses. Secondary objectives include characterizing the transcriptomic and genomic features of dendritic cells and macrophages, identifying radiomic and circulating tumor DNA (ctDNA) biomarkers, and evaluating their potential as non-invasive tools for recurrence prediction and patient stratification. The study includes a retrospective cohort of approximately 160 patients treated between 2009 and 2023 and a prospective cohort of approximately 50 patients who will be followed for 24 months. Tumor tissue samples, peripheral blood, imaging data (CT/MRI), and clinical information collected during routine care will be analyzed without introducing any experimental interventions or deviations from standard clinical practice. Analyses will include transcriptomic profiling, multiplex spatial characterization of immune cells, circulating tumor DNA sequencing using next-generation sequencing technologies, radiomic feature extraction, and integration of all data using statistical and machine learning approaches. Predictive models will be trained on retrospective data and independently validated in the prospective cohort. The primary endpoint is the prediction of colorectal liver metastasis recurrence within two years after liver resection. Ultimately, the study aims to develop and validate a multimodal predictive model integrating immune, genomic, radiomic, and clinical variables to improve recurrence risk assessment and support personalized patient management. The overall study duration is 36 months. All procedures will be conducted in accordance with ethical standards and data protection regulations, with samples and clinical data pseudonymized and handled in compliance with the GDPR.