Establishment and Validation of a Blood Test-based Nomogram to Diagnose Patients with AFP-negative HCC.

BACKGROUND

Hepatocellular carcinoma (HCC) is the fourth leading cause of cancer death worldwide. Alpha-protein (AFP) is the most widely used blood biomarker for HCC. However, elevated serum AFP is only observed in part of HCC.

AIMS

This study aimed to develop an efficient nomogram model to distinguish patients with alpha- protein-negative HCC and liver cirrhosis.

OBJECTIVES

A total of 1130 patients (508 HCC patients + 622 cirrhosis patients) were enrolled in the training cohort. A total of 244 HCC patients and 246 cirrhosis patients were enrolled in the validation cohort.

METHODS

A total of 41 parameters about blood tests were analyzed with logistic regression. The nomogram was based on independent factors and validated both internally and externally.

RESULTS

Independent factors were eosinophils %, hemoglobin concentration distribution width, fibrinogen, platelet counts, total bile acid, and mitochondria aspartate aminotransferase. The calibration curve for the probability of HCC showed good agreement between prediction by nomogram and actual observation. The concordance index was 0.851. In the validation cohort, the nomogram distinguished HCC from liver cirrhosis with an area under the curve of receiver operating characteristic of 0.754.

CONCLUSION

This proposed nomogram was an accurate and useful method to distinguish patients with AFP-negative HCC from liver cirrhosis.