OA03-RF01.09. Value of 68Ga-FAPI-04 and 18F-FDG PET-CT in Prediction of Pathological Response to Neoadjuvant Immunochemotherapy in Esophageal Squamous Cell Carcinoma

Topic Esophageal Cancer: New Diagnostic Modalities (Including PET, PET-CT, New Tracers) Background Fibroblast activation protein (FAP)-positive cancer-associated fibroblasts, predominant in the tumor microenvironment, play an important role in tumor progression and therapeutic efficacy. 68Ga-labeled FAP inhibitor (FAPI), an emerging PET tracer, provided localization and quantification for FAP-positive ESCC tumors and was superior to 18F-FDG for the precise staging of ESCC. However, the value of 68Ga-FAPI PET-CT in the assessment of neoadjuvant treatment response is still unclear. This study was to investigate whether dual-tracer 68Ga-FAPI-04 and 18F-FDG PET-CT metabolic parameters could predict pathological response after neoadjuvant immunochemotherapy (nICT) in patients with locally advanced esophageal squamous cell carcinoma (ESCC). Methods This prospective study enrolled 19 patients with locally advanced ESCC who underwent baseline dual-tracer PET-CT before nICT. The participants underwent surgery after 2 cycles of nICT. PET parameters including maximum, mean, and peak SUVs, metabolic tumor volume (MTV), total lesion glycolysis/FAP expression (TLG/TLF), and tumor-to-background ratios (TBRs) were measured for both tracers, and the ratio indices were calculated by dividing FAPI-derived parameters by corresponding FDG-derived values. Major pathological response (MPR) was defined as ≤10% residual viable tumor cells in the primary tumor bed. Patients were classified into good and poor responders groups based on their pathological results. We compared these indices between the good and poor pathological response groups and analyzed their predictive performance for tumor pathological response. Univariate and multivariate logistic regression analyses were performed to identify predictors of MPR. Model performance was evaluated using ROC and decision curve analyses. Results Nine patients were classified as good responders, and 10 were poor responders. The baseline characteristics of these two groups were comparable. A significant difference was observed in the FAPI-SUVmean and FDG-TBR, with the good responders having higher values compared to the poor (P < 0.05). To identify potential predictive factors, we calculated ratios between the FAPI and FDG PET/CT parameters (e.g. MTV ratio = FAPI-MTV/FDG-MTV). The univariate analysis revealed that MTV ratio, TLF/TLG ratio, TBR ratio, and SUVmean ratio were significantly associated with pathological response (all OR > 1, p < 0.05). Multivariate analysis indicated the combined model incorporating MTV ratio and TBR ratio achieved the highest predictive performance (AUC = 0.822), superior to either parameter alone. Higher ratio values, indicating stromal predominance relative to tumor metabolism, predicted poorer response to nICT. Conclusion Dual-tracer FAPI/FDG ratio indices represent novel imaging biomarkers for predicting pathological response to nICT in ESCC, offering a noninvasive tool to identify patients unlikely to benefit from nICT.