Early prediction of microvascular invasion (MVI) occurrence in hepatocellular carcinoma (HCC) by 18F-FDG PET/CT and laboratory data

Abstract

Background: Hepatocellular carcinoma (HCC) is one of the deadliest malignant tumors in China. Microvascular invasion (MVI) often indicates poor prognosis and metastasis in HCC patients. ¹⁸F-FDG PET-CT is a new imaging method commonly used to screen for tumor occurrence and evaluate tumor stage.

Purpose: This study attempted to predict the occurrence of MVI in early-stage HCC through ¹⁸F-FDG positron emission tomography (PET)/computed tomography (CT) imaging findings and laboratory data.

Patients and methods: A total of 113 patients who met the inclusion criteria were divided into two groups based on postoperative pathology: the MVI-positive group and MVI-negative group. We retrospectively analyzed the imaging findings and laboratory data of 113 patients. Imaging findings included tumor size, tumor _maximum standard uptake value (SUV_maxT), and normal liver _maximum standard uptake value (SUV_maxL). The ratios of SUV_maxT to SUV_maxL (SUV_maxT/L) and an SUV_maxT/L > 2 were defined as active tumor metabolism. The tumor size was indicated by the _maximum diameter of the tumor, and a diameter greater than 5 cm was defined as a mass lesion. The laboratory data included the alpha-fetoprotein (AFP) level and the HBeAg level. An AFP concentration > 20 ng/mL was defined as a high AFP level. A HBeAg concentration > 0.03 NCU/mL was defined as HB-positive.

Results: The SUV_maxT/L (p = 0.003), AFP level (p = 0.008) and tumor size (p = 0.015) were significantly different between the two groups. Patients with active tumor metabolism, mass lesions and high AFP levels tended to be MVI positive. Binary logistic regression analysis verified that active tumor metabolism (OR = 4.124, 95% CI, 1.566-10.861; p = 0.004) and high AFP levels (OR = 2.702, 95% CI, 1.214-6.021; p = 0.015) were independent risk factors for MVI. The sensitivity of the combination of these two independent risk factors predicting HCC with MVI was 56.9% (29/51), the specificity was 83.9% (52/62) and the accuracy was 71.7% (81/113).

Conclusion: Active tumor metabolism and high AFP levels can predict the occurrence of MVI in HCC patients.

Keywords: 18F-FDG PET/CT; Hepatocellular carcinoma; Microvascular invasion.

Fig. 1

Flowchart of the study population. HCC hepatocellular carcinoma, MVI microvascular invasion

Fig. 2

A 63-year-old male with MVI of HCC. a PET images showed an uneven increase in FDG uptake, suggesting tumor metabolic activity. b CT showed that the density of the liver was unevenly decreased, lobed, and the boundary was not clear. c PET/CT fusion images showed that the degree of metabolism in the tumor area was significantly higher than that in the surrounding liver parenchyma. d PET/CT MIP showed concentration of radioactivity in the liver area with unclear boundary with the right kidney, and physiological concentration of radioactivity in the heart, pelvis, ureter, and bladder

Fig. 3

Two patients with HCC. a, b MVI-negative patient ¹⁸F-FDG PET/CT imaging showed that a low-density mass lesion with FDG moderate metabolism. c, d MVI-positive patient ¹⁸F-FDG PET/CT imaging showed that a slightly low-density small lesion with FDG active metabolism

Fig. 4

A nomogram predicting the occurrence of MVI in HCC. SUV_maxT/L and AFP level were independent risk factors for MVI

Fig. 5

ROC curve was constructed to verify the accuracy of different indicators in predicting the occurrence of MVI in HCC. a The ROC curve constructed by tumor size. b The ROC curve constructed by SUV_maxT/L value. c The ROC curve constructed by AFP value. d The ROC curve constructed by SUVmaxT/L and AFP level