Objective To investigate the application value of peripheral blood circulating tumor cell (CTC) classification in the prediction of preoperative microvascular invasion of hepato-cellular carcinoma (HCC).

Methods The retrospective case-control study was conducted. The clinico-pathological data of 102 HCC patients who were admitted to Zhengzhou University People's Hospital from September 2018 to September 2020 were collected. There were 71 males and 31 females, aged from 29 to 80 years, with a median age of 57 years. Observation indicators: (1) surgical situations; (2) results of CTC detection and microvascular invasion in HCC patients; (3) results of CTC classification and the best cut-off value of CTC classification in the prediction of microvascular invasion in HCC; (4) influencing factors for microvascular invasion in HCC; (5) comparison of clinicopathological features in HCC patients with different cell counts in mesenchymal phenotype of CTC (M-CTC). Measurement data with normal distribution were represented as Mean±SD, and comparison between groups was analyzed using the independent sample t test. Measurement data with skewed distribution were represented as M(range) or M(Q₁,Q₃), and comparison between groups was analyzed using the nonparametric rank sum U test. Count data were described as absolute numbers or percentages, and comparison between groups was analyzed using the chi-square test. The receiver operating characteristic (ROC) curve was used to determine the best cut-off value for the risk of microvascular invasion in patients. Univariate and multivariate analysis were performed using the Logistic regression model.

Results (1) Surgical situations. All 102 patients underwent surgery successfully, including 17 cases undergoing local hepatectomy, 43 cases under-going segmentectomy, 22 cases undergoing hepatic lobectomy, 13 cases undergoing hemilectomy and 7 cases undergoing enlarged hemilectomy. The operation time and the volume of intraoperative blood loss were 235(147,293)minutes and 300(110,500)mL of the 102 patients, respectively. (2) Results of CTC detection and microvascular invasion in HCC patients. Of 102 patients, there were 36 casas with epithelial phenotype of CTC (E-CTC), 86 cases with hybrid phenotype of CTC (H-CTC), 30 cases with M-CTC, respectively, and the total CTC (T-CTC) were positive in 89 cases. Results of postoperative pathological examination showed that there were 40 cases with micro-vascular inva-sion and 62 cases without microvascular invasion in the 102 patients. Of the 40 patients with micro-vascular invasion, the count of E-CTC, H-CTC, M-CTC and T-CTC were 0(0,1) per 5 mL, 4(2,5) per 5 mL, 1(0,2) per 5 mL and 5(3,8) per 5mL, respectively. The above indicators of the 62 cases without microvascular invasion were 0(0,1) per 5 mL, 3(1,5) per 5 mL, 0(0,0) per 5 mL and 3(2,6) per 5 mL, respectively. There were significant differences in the count of M-CTC and T-CTC between patients with and without microvascular invasion (Z=-4.83, -2.96, P<0.05). (3) Results of CTC classi-fication and the best cut-off value of CTC classification in the prediction of microvascular invasion in HCC. The ROC curve showed that best cut-off value of M-CTC and T-CTC counts in the prediction of microvascular invasion in HCC were 1 per 5 mL and 4 per 5 mL, respectively, with the area under curve, the corresponding specificity, sensitivity were 0.70 (95% confidence interval as 0.60-0.81, P<0.05), 75.8%, 62.9% and 0.67 (95% confidence interval as 0.57-0.78, P<0.05), 60.0%, 72.5%, respec-tively. (4) Influencing factors for microvascular invasion in HCC. Result of univariate analysis showed that alpha fetoprotein (AFP), aspartate aminotransferase (AST), tumor diameter, tumor number, tumor margin, Barcelona clinic liver cancer staging, M-CTC counts and T-CTC counts were related factors influencing microvascular invasion in HCC (odds ratio=3.13, 0.43, 4.92, 5.65, 2.54, 2.93, 8.25, 4.47, 95% confidence interval as 1.34-7.33, 0.19-0.98, 2.09-11.58, 2.35-13.63, 1.13-5.75, 1.27-6.74, 3.13-21.75, 1.88-10.61, P<0.05). Result of multivariate analysis showed that tumor diameter >5 cm, tumor number as multiple and M-CTC counts ≥1 per 5 mL were independent risk factors influencing microvascular invasion in HCC (odds ratio=2.97, 4.14, 4.36, 95% confidence interval as 1.01-8.70, 1.14-15.02, 1.36-13.97, P<0.05). (5) Comparison of clinicopathological features in HCC patients with different cell counts in M-CTC. The 102 HCC patients were divided into the high M-CTC group of 30 cases with M-CTC counts ≥1 per 5 mL and the low M-CTC group of 72 cases with M-CTC counts <1 per 5 mL, according to the best cut-off value of M-CTC counts. Cases with hepatitis, cases with AFP >400 µg/L, cases with AST >35 U/L, cases with irregular tumor margin, cases with tumor diameter >5 cm, cases with tumor number as multiple and cases with micro-vascular invasion were 22, 17, 13, 21, 18, 16 and 22 in the high M-CTC group of 30 cases. The above indicators were 35, 18, 48, 26, 25, 21 and 18 in the low M-CTC group of 72 cases. There were significant differences in the above indicators between the high M-CTC group and the low M-CTC group (χ²=5.25, 9.42, 4.80, 9.79, 5.55, 5.35, 20.75, P<0.05).

Conclusions The epithelial-mesen-chymal phenotype of peripheral blood CTC can be used to predict the preoperative microvascular invasion in HCC. Tumor diameter >5 cm, tumor number as multiple and M-CTC counts ≥1 per 5 mL are independent risk factors influencing microvascular invasion in HCC patients.