Diagnostic performance of FDG PET/CT radiomics in predicting microvascular invasion in hepatocellular carcinoma compared to conventional metabolic parameters: a systematic review and meta-analysis

Purpose

The purpose of the current study was to evaluate the diagnostic accuracy of FDG PET/CT radiomics in predicting microvascular invasion (MVI) in hepatocellular carcinoma (HCC), and to compare it with conventional metabolic parameters of FDG PET/CT through a systematic review and meta-analysis.

Methods

The PubMed, EMBASE, and Cochrane databases were searched for studies evaluating the diagnostic performance of FDG PET/CT in predicting MVI in HCC patients. We calculated the pooled area under the curve (AUC) for predicting MVI using FDG PET/CT analyzed with radiomics methods and compared the results with those predicted through visual or semi-quantitative analysis. The study was conducted and registered in PROSPERO (International Prospective Register of Systematic Reviews) with the registration number CRD42023466842.

Results

The pooled AUC for predicting MVI from three studies (274 patients) analyzed using radiomics methods was 0.79 (95% CI; 0.75–0.84), with various model algorithms and selected features. The pooled AUC for six studies (368 patients) using visual analysis was 0.76 (95% CI; 0.73–0.80), and the pooled AUC for nine studies (661 patients) using semi-quantitative analysis was 0.80 (95% CI; 0.76–0.83). The diagnostic performance of the three analysis methods did not show a statistically significant difference.

Conclusion

FDG PET/CT radiomics for predicting MVI in HCC showed diagnostic performance similar to that of conventional visual or semi-quantitative analysis methods. Further large-scale multicenter studies are necessary to substantiate the diagnostic accuracy of FDG PET/CT radiomics for predicting MVI in HCC patients.