Parallel Evaluation of TruAI-assisted imaging and flow cytometry for circulating tumor cell detection in hepatocellular carcinoma

Abstract Background Circulating tumor cells (CTCs) and CTC clusters are promising liquid biopsy biomarkers for hepatocellular carcinoma (HCC). Evaluation of both enumeration and morphology remains an important challenge. Flow cytometry (FACS) enables high-throughput quantification and multiparametric analysis, whereas artificial intelligence–based imaging platforms such as TruAI allow structural assessment of CTCs. Methods We performed a parallel evaluation of these approaches in patients with HCC. Peripheral blood samples from patients with HCC and spike-in experiments using tumor cell lines were analyzed following CD45 depletion–based enrichment. TruAI identified CTCs based on a DAPI + /CD45 − /PanCK + phenotype with preserved nuclear morphology. Results In spike-in experiments, both platforms demonstrated strong linearity (R 2 > 0.94), with higher absolute counts observed using FACS. In 17 patients with HCC, CTC counts measured by TruAI and FACS were significantly correlated ( r = 0.8216, R 2 = 0.675, p < 0.0001), and both methods captured longitudinal decreases during treatment. TruAI additionally enabled morphological analysis, revealing reductions in the proportion of CTC clusters in patients achieving disease control. Decreases in cluster proportions paralleled radiologic tumor shrinkage and declining serum tumor markers. Conclusions TruAI-assisted imaging and flow cytometry showed quantitative concordance in CTC assessment in HCC, while imaging provided additional morphology-based information. Their combined use may enhance longitudinal CTC monitoring in HCC by capturing treatment-related changes in both total CTC burden and cluster dynamics.