Thomas Guiot

PURPOSE: Managing chemorefractory metastatic colorectal cancer (mCRC) requires a meticulous equilibrium between the efficacy and toxicity of interventions, a task compounded by the constrained life expectancy of the patient. While existing prognostic tools, such as the Colon Life nomogram, primarily focus on general patient conditions or a single diagnostic modality, they do not fully integrate the potential predictive value of multimodal data. This study aims to develop and validate an Imaging Score, integrating clinical and imaging features derived from whole-body ¹⁸ F-fluorodeoxyglucose ( ¹⁸ F-FDG) positron emission tomography-computed tomography (PET-CT), predicting death probability within 12 weeks from treatment initiation for refractory disease. MATERIALS AND METHODS: The development cohort comprises 254 patients from three clinical trials. Nine clinical variables and six imaging variables were assessed. After optimal subset selection through recursive Feature Elimination with cross-validation, a support vector classifier-trained machine learning model generated the Imaging Score. Validation was performed on a real-life patient cohort (n = 74). Model performance was assessed on discrimination (Harrell C-index) and calibration. RESULTS: Final prognostic features included whole-body metabolically active tumor volume, Eastern Cooperative Oncology Group performance status, visceral fat density, number of metastatic sites, body mass index, maximum standardized distance, and months since diagnosis. The Imaging Score demonstrated robust discriminative ability in both the development (C-index, 0.797) and validation (C-index, 0.714) sets, outperforming the Colon Life nomogram that tended to overestimate 12-week mortality. CONCLUSION: The Imaging Score, integrating ¹⁸ F-FDG PET-CT imaging with clinical parameters, is an effective prognostic tool for patients with chemorefractory mCRC. This combination of imaging biomarkers with clinical factors improves discrimination, enhancing its potential for clinical decision making, patient stratification for chemorefractory treatments, and trial eligibility.

Gene set enrichment analysis on Reactome gene sets, comparing metabolic response (positive normalized enrichment score – NES) and lack of metabolic response (negative NES).

Gene set enrichment analysis on Reactome gene sets, comparing 89Zr-trastuzumab classes “high” (positive normalized enrichment score – NES) and “low” (negative NES).

Gene set enrichment analysis on Hallmark gene sets, comparing metabolic response (positive normalized enrichment score – NES) and lack of metabolic response (negative NES).

Gene set enrichment analysis on Reactome gene sets, comparing, within the 89Zr-trastuzumab class “high”, metabolic response (positive normalized enrichment score – NES) and lack of metabolic response (negative NES).

Purpose:The ZEPHIR clinical trial evaluated the role of [89Zr]trastuzumab-PET/CT (HER2-PET/CT) and 2-[18F]fluoro-2-deoxy-D-glucose PET/CT ([18F]FDG-PET/CT) in predicting outcomes in patients with advanced HER2-positive breast cancer treated with trastuzumab emtansine (T-DM1). In this study, we combined molecular/metabolic imaging and transcriptomic data to investigate the biological processes associated with [89Zr]trastuzumab and [18F]FDG uptake and to dissect the mechanisms involved in T-DM1 resistance.Experimental Design:RNA was extracted from metastasis biopsies obtained in the ZEPHIR trial. HER2-PET/CT and [18F]FDG-PET/CT imaging data of biopsied lesions were integrated with transcriptomic data. Lesions were compared based on the level of [89Zr]trastuzumab uptake as well as on the presence/absence of metabolic response, defined comparing baseline and on-treatment [18F]FDG-PET/CT.Results:We analyzed matched transcriptomic and molecular/metabolic imaging data for 24 metastases. Genes and pathways involved in extracellular matrix (ECM) organization and glycosylphosphatidylinositol synthesis were enriched in lesions presenting low [89Zr]trastuzumab uptake. [18F]FDG uptake at baseline correlated with proliferation and immune-related processes. Hypoxia and ECM-related processes were enriched in lesions showing no metabolic response to T-DM1, whereas immune-related processes were associated with high [89Zr]trastuzumab uptake and metabolic response. Gene signatures including differentially expressed genes according to [89Zr]trastuzumab uptake and metabolic response showed predictive value in an external cohort.Conclusions:To the best of our knowledge, this study represents the first correlative analysis between [89Zr]trastuzumab tumor uptake and gene expression profiling in humans. Our findings suggest a role for ECM in impairing [89Zr]trastuzumab tumor uptake and T-DM1 metabolic response in advanced HER2-positive breast cancer, highlighting the potential of molecular imaging to depict tumor microenvironment features.

CONSORT diagram showing the patient selection for the present analysis in the ZEPHIR trial

Comparison of abundance scores of immune populations and stromal cell types with MCP-counter according to A, ZrT class (dichotomized into “high” and “low”), and B, metabolic response.

Levels of stroma (%) according to ZrT class (dichotomized into “high” and “low”), and metabolic response

Examples of the four classes of HER2-PET/CT uptake.