Development and clinical validation of a novel platelet count-based nomogram for predicting microvascular invasion in HCC

Abstract

We aimed to develop a convenient nomogram to predict preoperative MVI in patients with hepatocellular carcinoma (HCC). Patients who underwent surgical resection due to HCC from June 2018 to June 2023 at the Third Affiliated Hospital of Sun Yat-sen University were retrospectively reviewed. Univariate and multivariable logistic linear regression analyses were used to investigate potential risk factors for MVI. A nomogram was plotted based on these risk factors. The tumor diameter (≥ 5 cm), BCLC stage, PLT (>127.50 × 10⁹/L), AST (>29.50 U/L) and AFP (>10.07 ng/ml) were identified as independent preoperative risk factors for MVI by univariate and multivariable logistic analysis. The nomogram demonstrated decent accuracy in estimating the presence of MVI, with an AUC of 0.69 (95%CI: 0.64–0.73). The calibration curves exhibited a close match between the predicted probabilities and the actual estimates of MVI in the nomogram (p = 0.947). Decision curve analysis (DCA) revealed that the prediction model had a high net benefit if the threshold probability>20%. High platelet counts were strongly associated with the presence of MVI in HCC patients. Our convenient nomogram demonstrated decent accuracy in estimating the presence of MVI and had notable clinical application.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-025-88343-3.

Hepatocellular carcinoma (HCC) is one of the most common malignancies in the world and a leading cause of cancer-related mortality worldwide^1,2. Surgical resection remains the mainstay treatment for HCC³, however, the incidence of tumor recurrence can be as high as 70% at 5 years after surgery⁴. Microvascular invasion (MVI), which is defined under microscopy as a cluster of tumor cells in the vessels of the surrounding hepatic tissue lined by endothelium⁵, has already been demonstrated to be associated with early recurrence^6–9 and a poor overall survival rate after resection¹⁰. However, the diagnosis of MVI can only be confirmed by microscopic examination of surgical samples, and even preoperative biopsies could not confirm the diagnosis.

In the treatment of very early-stage HCC (≤ 2 cm) with BCLC (stage 0), although surgical resection and ablation are both recommended, it is argued that some patients may not receive appropriate treatment because there is no pathological specimen to identify the MVI situation after ablation¹¹. In addition, some scholars believe that the presence or absence of MVI may affect the indications for liver transplantation, as the presence of MVI may discourage transplantation and the absence of MVI may favor transplantation for both tumors within the extended criteria for transplantation¹².

Thus, it is vital to explore efficacy methods for precisely diagnosing MVI preoperatively, which is of great significance for surgeons to select more appropriate therapies and improve patient prognosis. Recent years, several studies have focused on predicting MVI, ranging from patients` laboratory results^13–20 to imaging data^21–24 to genomic data²⁵. However, studies rarely research the relationship between platelet counts and MVI presence and its efficiency to predict MVI. Furthermore, the precise relationship between PLT counts and the prognosis of HCC remains to be conclusively determined^26–28. And some research that focuses on genomic data or imaging data might be costly and complicated. Our study aimed to identify preoperative predictors of MVI and plot a more convenient nomogram.

Materials and methods

Patients selection and data acquisition

Patients who underwent surgical resection due to HCC from June 2018 to June 2023 at the Third Affiliated Hospital of Sun Yat-sen University were retrospectively reviewed. The inclusion criteria were as follows: (a) Age between 18 and 80 years old; (b) Without history of prior intervention therapy, immunotherapies, targeted therapies or partial hepatectomy; (c) Postoperative pathology confirmed HCC. The exclude criteria included: (a) Patients with hypersplenism/ portal hypertension; (b) Patients with other organ metastases or other malignant tumors; (c) Child-Pugh class C patients. Supplementary Fig. 1 showed the flow chart diagram that outlines the patient selection criteria and process used in our study. Basic characteristics including age, gender, Barcelona clinic liver cancer stage (BCLC), and history of smoking, drinking, and illness were collected from medical records. Routine preoperative laboratory examinations including blood examination, liver function, coagulation test and alpha-fetoprotein (AFP) level were also collected.

The need for informed consent was waived by approving ethics committee because of the retrospective nature of the study, and the study design was approved by the Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University.

Histology

All specimens were sampled at the junction of the tumor and adjacent liver tissues in a 1:1 ratio at the 12, 3, 6, and 9 o’clock reference positions²⁹and examined by two experienced abdominal pathologists. The diagnostic criteria for MVI included the presence of cancer cell clusters in blood vessels lined with endothelial cells, which are most pronounced in the branch of the portal vein (including the intracapsular blood vessels) under microscopic observation³⁰.

Immunofluorescent staining

Immunofluorescent staining was performed according to previously described procedures³¹. Briefly, the sections were routinely deparaffinized and rehydrated, blocked with 5% BSA for 30 min after the antigen retrieval procedure. The specimens were then incubated with rabbit polyclonal CD41 antibody (ab134131) overnight at 4 °C. The sections were then incubated with fluorescence-conjugated secondary antibody (CY5 labeled anti-rabbit IgG, Abcam, ab6563). The results were analyzed by Image J to define the mean fluorescence intensity.

Statistical analyses

The data were processed by SPSS 26.0 and R 3.6.3. Variables were analyzed by the Student t test and rank-sum test according to whether they conformed to the normal distribution and homogeneity of variance, and were expressed by mean ± standard deviation or quartile values. The Chi-square test or Fisher’s exact test was used for categorical variables. The univariate logistic linear regression analysis was used to assess the single factor for discriminating MVI presence. Risk factors with p < 0.05 at univariate analysis and those factors considered clinically relevant were entered into multivariable logistic regression analysis to determine potential risk factors of MVI. A nomogram was constructed based on the results, and the calibration curve and ROC curve were used to detect the accuracy of the nomogram. Decision curve analysis (DCA) was used to evaluate the clinical value and net benefits of the nomogram.

Results

Patient characteristics

A total of 512 HCC patients who underwent curative hepatectomy from June 2018 to June 2023 at the Third Affiliated Hospital of Sun Yat-sen University were enrolled in this study. Patients were divided into two groups based on whether they were MVI-positive or not. There were 203 (39.65%) patients with positive MVI and 309 (60.35%) patients with negative MVI. The clinicopathological characteristics of the two groups are shown in Table 1. Patients in the MVI-positive group showed a significantly greater proportion of tumors with a diameter ≥ 5 cm and a more advanced BCLC stage. The positive MVI group had significantly higher median levels of AST, GGT, ALP, PCT and AFP, and counts of PLT and NEU.

Table 1.

Clinical characteristic of patients in the two group.

Characteristics	MVI-(n = 309)	MVI+(n = 203)	p value
Gender			0.340
Male	260(84.1%)	177(84.1%)
Female	49(15.9%)	26(84.1%)
Age	56(25,87)	53(22,76)	0.520
BMI (kg/m2)	23.03(15.57,42.00)	23.59(17.09,34.11)	0.057
Hypertension			0.138
Yes	98(31.7%)	52(25.6%)
No	211(68.3%)	151(74.4%)
Diabetes			0.115
Yes	36(11.7%)	15(7.4%)
No	273(92.6%)	188(88.3%)
Smoke			0.909
Yes	102(33.0%)	68(33.5%)
No	207(67.0%)	135(66.5%)
Alcohol			0.568
Yes	33(10.7%)	25(12.3%)
No	276(89.3%)	178(87.7%)
Cirrhosis			0.708
Yes	175(56.6%)	111(55.0%)
No	134(43.4%)	91(45.0%)
Ascites			0.328
No	254(82.2%)	159(78.7%)
Mild	55(17.8%)	43(21.3%)
Tumor diameter(cm)			<0.001
≥ 5	73(23.6%)	88(43.3%)
<5	236(76.4%)	115(56.7%)
Tumor number			0.907
Single	290(93.9%)	190(93.6%)
Multiple	19(6.1%)	13(6.4%)
BCLC stage			<0.001
0	67(21.7%)	15(7.4%)
A	237(76.7%)	184(90.6%)
B	5(1.6%)	4(2.0%)
HBV			0.631
Negative	41(13.3%)	24(11.8%)
Positive	268(86.7%)	179(88.2%)
Child
A	289(93.5%)	190(93.6%)	0.975
B	20(6.5%)	13(6.4%)
AST (U/L)	28(13,253)	32(14,536)	0.002
ALT (U/L)	29(6,474)	34(10,235)	0.065
TBIL (umol/L)	11.05(3.20,42.40)	11.4(2.3,61.70)	0.414
IBIL (umol/L)	4.30(0.60,20.00)	4.45(1.19,25.70)	0.717
BUN (mmol/L)	5.16(2.01,523.00)	5.06(2.41,12.00)	0.350
CR (umol/L)	73(34,147)	75(3.2,226)	0.303
GGT (U/L)	42(11,656)	54.5(10.1184)	<0.001
TBA (umol/L)	7.00(0.7,123)	7.2(0.5,151.6)	0.336
ALP (U/L)	76(3.10,411.00)	80(37,478)	0.049
ALB (g/L)	40.8(27.9,51.3)	40.85(25.6,51.7)	0.883
WBC (×109/L)	5.49(2.63,13.7)	5.92(1.93,13.3)	0.064
RBC (×1012/L)	4.66(2.65,6.97)	4.71(2.15,7.15)	0.553
MONOCYTE (×109/L)	0.47(0.08,3.24)	0.50(0.18,1.34)	0.071
NEU (×109/L)	3.04(0.79,12.48)	3.31(0.94,10.01)	0.031
PLT (×109/L)	171(56,428)	181(69,413)	0.016
PDW (%)	12.6(7.9,21.8)	12.65(8.9,20.2)	0.277
MPV (fL)	10.6(8.5,15.9)	10.6(8.5,13.7)	0.547
P-LCR (%)	0.29(0.12,10.4)	0.28(0.14,10.20)	0.294
PCT (%)	0.19(0.06,0.44)	0.19(0.06,0.43)	0.007
LYMP (×109/L)	1.74(0.38,223)	1.7(0.44,106.00)	0.383
APTT (s)	37.7(29.6,49.6)	37.2(24.7,46.6)	0.091
PT (s)	13.6(11.8,17.8)	13.5(10.3,18.4)	0.375
INR	1.03(0.86,104.00)	1.03(0.87,1.51)	0.580
AFP (ng/ml)	25.41(0,78833.74)	48.28(1.53,80000.00)	0.006

MVI: microvascular invasion; BMI: body mass index; BCLC: Barcelona clinic liver cancer; HBV: hepatitis B virus; AST: aspartate aminotransferase; ALT: alanine aminotransferase; TBIL: total bilirubin; IBIL: indirect bilirubin; BUN: blood urea nitrogen; CR: creatinine; GGT: glutamine transpeptidase; TBA: total bile acids; ALP: alkaline phosphatase; ALB: albumin; WBC: white blood cell; RBC: red blood cell; NEU: neutrophil; PLT: platelet; PDW: Platelet distribution width; MPV: mean platelet volume; P-LCR: platelet -larger cell ratio; PCT: platelet hematocrit; LYMP: lymphocyte count; APTT: activated partial thromboplastin time; PT: prothrombin time; INR: international normalized ratio; AFP: alpha fetal protein.

The plot of ROC curve and the determination of cut-off value

According to significantly different laboratory indexes between the two groups, the ROC curves are plotted. The cut-off values were identified when the Youden index was at its maximum. The results revealed that the cut-off values of AST, GGT, ALP, NEU, PLT, PCT, and AFP were 29.50, 50.50, 91.50, 3.35, 127.50, 0.125, and 10.07 respectively. The areas under the curve (AUCs) and 95% confidence interval (CI) were 0.58 (0.53, 0.63), 0.59 (0.54, 0.64), 0.552 (0.50, 0.60), 0.56(0.51, 0.61), 0.564 (0.51, 0.61), 0.57 (0.52, 0.62) and 0.57 (0.52, 0.62), respectively (Fig. 1; Table 2).

Fig. 1.

The ROC curve for AST, GGT, ALP, NEU, PLT, PCT and AFP. ROC: receiver operating characteristic; AST: aspartate aminotransferase; GGT: glutamine transpeptidase; ALP: alkaline phosphatase; NEU: neutrophil; PLT: platelet; PCT: platelet hematocrit; AFP: alpha fetal protein.

Table 2.

Value of Preoperative AST, GGT, ALP, NEU, LPT, PCT and AFP in the diagnosis of MVI.

Variables	AUC	p value	95%CI	Cut-off value
AST (U/L)	0.58	0.002	0.529–0.630	29.5
GGT (U/L)	0.594	<0.001	0.544–0.644	50.5
ALP (U/L)	0.552	0.049	0.500-0.603	91.5
NEU (×109/L)	0.559	0.025	0.507–0.610	3.35
PLT (×109/L)	0.564	0.014	0.514–0.614	127.5
PCT (%)	0.572	0.006	0.522–0.622	0.125
AFP (ng/ml)	0.57	0.007	0.519–0.621	10.07

AUC: area under curve; CI: confidence interval; AST: aspartate aminotransferase; GGT: glutamine transpeptidase; ALP: alkaline phosphatase; NEU: neutrophil; PLT: platelet; PCT: platelet hematocrit; AFP: alpha fetal protein.

Independent risk factors of MVI

Univariate logistic regression analyses were performed to find factors related to significant regression according to basic characteristics and cut-off values. Multivariable logistic analyses were conducted based on the results with a p value <0.05 in univariate logistic regression analyses. The tumor diameter (≥ 5 cm) (OR = 1.63, 95% CI: 1.08–2.48, p = 0.02), BCLC stage (A vs. 0) (OR = 2.53, 95%CI: 1.35–4.73, p = 0.004), AST(>29.50 U/L) ( OR = 1.67, 95%CI: 1.14–2.45, p = 0.009) and AFP (>10.07 ng/ml) ( OR = 1.59, 95%CI: 1.08–2.36, p = 0.02) were identified as independent preoperative risk factors for MVI. Interestingly, the assessment of platelet counts(>127.5 × 10⁹/L) ( OR = 2.28, 95%CI: 1.31–3.95, p = 0.003) was found to be a significant independent preoperative risk factor for MVI(Table 3).

Table 3.

Univariate and Multivariable logistic analysis of predictors of MVI.

Variables	Univariate OR (95%CI)	p value	Multivariable OR (95%CI)	p value
Gender (Male)
Female	0.779(0.467,1.301)	0.341
Age	0.608(0.982,1.011)	0.996
BMI (kg/m2)	1.051(0.986,1.120)	0.124
Hypertension (No)
Yes	0.741(0.499,1.102)	0.139
Diabetes (No)
Yes	0.605(0.322,1.136)	0.118
Smoke (No)
Yes	1.022(0.702,1.488)	0.909
Alcohol (No)
Yes	1.175(0.676,2.042)	0.568
Cirrhosis (No)
Yes	0.934(0.654,1.335)	0.708
Ascites (No)
Mild	1.249(0.800,1.950)	0.328
Tumor diameter(cm) <5
≥ 5	2.474(1.688,3.625)	<0.001	1.634(1.078,2.477)	0.021
Tumor number (Single)
Multiple	1.044(0.504,2.164)	0.907
BCLC stage (0)
A	3.468(1.918,6.269)	<0.001	2.526(1.351,4.725)	0.004
B	3.573(0.856,14.917)	0.084	2.433(0.546,10.843)	0.243
B vs. A	0.970(0.257,3.665)	0.965	1.038(0.259,4.157)	0.958
HBV (-)
(+)	1.141(0.666,1.954)	0.631
Child(A)
B	0.989(0.480,2.035)	0.975
AST (U/L) ≤ 29.5
>29.5	1.842(1.286,2.639)	0.001	1.667(1.137,2.446)	0.009
ALT (U/L)	1.001(0.997,1.006)	0.534
TBIL (umol/L)	1.013(0.984,1.042)	0.391
IBIL (umol/L)	1.019(0.974,1.067)	0.410
BUN (mmol/L)	0.969(0.887,1.059)	0.491
CR (umol/L)	1.006(0.996,1.016)	0.231
GGT (U/L) <50.5
>50.5	1.849(1.292,2.646)	0.001		0.190
TBA (umol/L)	0.996(0.986,1.006)	0.446
ALP (U/L) <91.5
>91.5	1.586(1.076,2.336)	0.020		0.243
ALB (g/L)	0.995(0.955,1.036)	0.802
WBC (×109/L)	1.085(0.986,1.194)	0.093
RBC (×1012/L)	1.035(0.786,1.363)	0.807
MONOCYTE (×109/L)	1.513(0.719,3.199)	0.279
NEU (×109/L) <3.35
>3.35	1.657(1.158,2.372)	0.006		0.193
PLT (×109/L) <127.5
>127.5	1.492(1.041,2.140)	0.029	2.277(1.314,3.948)	0.003
PDW (%)	1.036(0.950,1.129)	0.429
MPV (fL)	1.028(0.872,1.212)	0.742
P-LCR (%)	0.978(0.824,1.16)	0.978
PCT (%) <0.125
>0.125	59.725(3.234,1103,055)	0.006		0.149
LYMP (×109/L)	0.999(0.982,1.016)	0.902
APTT(s)	0.952(0.904,1.003)	0.066
PT(s)	0.933(0.768,1.133)	0.484
INR	0.683(0.092,5.047)	0.709
AFP (ng/ml) <10.07
>10.07	1.630(1.121,2.370)	0.011	1.592(1.076,2.358)	0.020

MVI: microvascular invasion; BMI: body mass index; BCLC: Barcelona clinic liver cancer; HBV: hepatitis B virus; AST: aspartate aminotransferase; ALT: alanine aminotransferase; TBIL: total bilirubin; IBIL: indirect bilirubin; BUN: blood urea nitrogen; CR: creatinine; GGT: glutamine transpeptidase; TBA: total bile acids; ALP: alkaline phosphatase; ALB: albumin; WBC: white blood cell; RBC: red blood cell; NEU: neutrophil; PLT: platelet; PDW: Platelet distribution width; MPV: mean platelet volume; P-LCR: platelet -larger cell ratio; PCT: platelet hematocrit; LYMP: lymphocyte count; APTT: activated partial thromboplastin time; PT: prothrombin time; INR: international normalized ratio; AFP: alpha fetal protein.

Immunofluorescent staining of platelets in paraffin sections of tumors in HCC patients

To further confirm the relationship between the counts of PLT and the presence of MVI, immunofluorescent staining was conducted for patients in the two groups. CD41 was used to identify platelets. Patients with positive MVI showed significantly higher PLT counts in paraffin sections of tumors compared to MVI-negative patients (p<0.001) (Fig. 2A, B). Therefore, it can be inferred that there is a significant correlation between increased platelet counts and the occurrence of MVI in patients with hepatocellular carcinoma.

Fig. 2.

Platelet staining between the two groups. (A) Histology of HCC patients with or without MVI. The part circled by the red line showed MVI. Immunostaining images showed the expression of CD41 in the two groups. (B) Analysis of CD41 expression of the two groups. (***, p<0.001).

Development and validation of the Nomogram for prediction of MVI preoperatively

Based on the results of the multivariable logistic regression analysis, tumor diameter, BCLC stage, AST, PLT and AFP were selected to construct a nomogram for the prediction of preoperative MVI in HCC patients (Fig. 3A). The nomogram demonstrated decent accuracy in estimating the presence of MVI, with an AUC of 0.69 (95% CI: 0.64–0.73) (Fig. 3B). The calibration curves exhibited a close match between the predicted probabilities and the actual estimates of MVI in the nomogram (p = 0.947) (Fig. 3C). Decision curve analysis (DCA) revealed that the prediction model had a high net benefit if the threshold probability was >20%, indicating that the novel nomogram has notable clinical application(Fig. 3D).

Fig. 3.

(A) Nomogram based on risk factors for preoperative prediction of MVI in HCC patients. (B) The performance of the nomogram model for predicting MVI by receiver operating characteristic (ROC) curve. (C) The calibration of the nomogram model. (D) Decision curve analysis (DCA) for the nomogram model. BCLC: Barcelona clinic liver cancer; AST: aspartate aminotransferase; PLT: platelet; AFP: alpha fetal protein.

Discussion

The aim of our study was to develop and validate a novel and convenient nomogram for preoperative prediction of the presence of MVI. We found that patients in the MVI-positive group showed a significantly greater proportion of tumors with a diameter ≥ 5 cm and a more advanced BCLC stage. And HCC patients with MVI had significantly higher median levels of AST, GGT, ALP, PCT and AFP, and counts of PLT and NEU. ROC curves were drawn to identify the cut-off value to distinguish a positive MVI. The tumor diameter(≥ 5 cm), BCLC stage, PLT(>127.5 × 10⁹/L), AST(>29.50 U/L) and AFP(>10.07 ng/ml) were identified as independent preoperative risk factors for MVI by univariate and multivariable logistic analysis. The relationship between high platelet counts and the presence of MVI in HCC patients was further identified by immunofluorescent staining. We then developed and validated a new nomogram that included tumor diameter, BCLC stage, AST, PLT and AFP for the prediction of preoperative MVI in HCC patients. The nomogram has demonstrated decent accuracy in estimating the presence of MVI and might have notable clinical application.

Previous studies developed diverse predictive models using preoperative serum biomarkers to predict MVI preoperatively. Gu et al.³² developed and validated a novel nomogram that involved gamma-glutamyl transpeptidase to lymphocyte ratio, GGT to platelet ratio, AST to platelet ratio, and GGT to albumin ratio for the preoperative prediction of MVI in patients with solitary primary HCC. Rungsakulkij et al.³³ identified large tumor size (≥ 5 cm) and high platelet-to-lymphocyte ratio (≥ 102) as independent predictive factors for MVI in HCC. In our study, we confirmed that tumor diameter (≥ 5 cm), BCLC stage, PLT (>127.5 × 10⁹/L), AST (>29.50 U/L) and AFP(>10.07 ng/ml) were independent preoperative risk factors for MVI, which is consistent with the research of Li who also suggested that AFP, tumor size and inflammatory score were independent factors for MVI³⁴. Moreover, from an etiological perspective, our study primarily focuses on the presence of HBV infection in patients. However, to comprehensively address other etiologies, such as HCV and metabolic factors associated with HCC, which are significant in the patient population with hepatoma³⁵, it is imperative to incorporate a larger and more diverse patient population to facilitate subgroup analyses.

Currently, a consensus on the impact of platelet status on HCC prognosis has not yet been established. Platelet count plays a pivotal role in the progression of HCC, influencing tumor growth, invasiveness, and angiogenesis, which are key determinants of patient survival³⁶. Studies indicate that preoperative thrombocytopenia is correlated with poor postoperative survival in HCC patients³⁷, while in those with thrombocytosis, survival is significantly reduced³⁸. Furthermore, activated platelets are considered an independent risk factor for poor prognosis in HCC³⁹. Particularly in the context of circulating tumor cells, HCC patients with high platelet counts exhibit inferior relapse-free survival⁴⁰. Our study’s findings align with this, revealing a strong correlation between high platelet counts and MVI in HCC patients, which portended a poor prognosis. However, contrary to our study’s conclusions, some research suggests that thrombocytopenia is associated with poor prognosis, potentially including patient populations with more severe liver pathological backgrounds^28,41. Clinically, plenomegaly may affect the prognostic value of platelet counts²⁸. Therefore, our study excluded HCC patients with hypersplenism/portal hypertension, which may have influenced the results.

From a mechanistic perspective, the formation of new blood vessels is essential for tumor growth, and circulating platelets are postulated to regulate tumor angiogenesis by interacting with endothelial cells and releasing angiogenic regulators from specialized α granules^42,43. Several studies confirmed the contribution of platelets to tumor angiogenesis^44,45. Furthermore, platelets were proved to be able to affect the inflammatory response in cancer by modulating the immune system, influencing the activation state of the endothelium, and recruiting leukocytes to primary and metastatic tumor sites, as well as to distant organs unaffected by tumor growth⁴⁶. Moreover, recent clinical studies showed that the platelet inhibitor aspirin reduces the risk of HCC in patients with viral hepatitis⁴⁷.

The present study acknowledged several inherent limitations. Firstly, the retrospective design and single-center data sourced predispose our analysis to selection bias, lacking external validation in other HCC cohorts. Secondly, the focus on laboratory outcomes, while robust, may limit the generalizability of our findings; despite affirmative evidence supporting the role of gene expression and radiomic analysis in preoperative MVI prediction. Thirdly, the study did not delve into the mechanistic correlation between platelet counts and MVI presence, a gap that warrants future investigation. Furthermore, the practical application of our nomogram within the dynamic clinical setting may face challenges due to time limitations, workflow disruptions, and the exigency of swift clinical decisions. In future, we plan to make this nomogram available as free software or handheld devices, in order to facilitate the clinical usage of our model.

Conclusions

Our study revealed that high platelet counts are strongly associated with the presence of MVI in HCC patients. Compared to previous costly and complicated nomograms, our convenient nomogram demonstrated decent accuracy in estimating the presence of MVI and had notable clinical application.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

MVI

Microvascular invasion

PLT

Platelet

AFP

Alpha fetal protein

HCC

Hepatocellular carcinoma

ROC

Receiver operating characteristic curve

DCA

Decision curve analysis

CI

Confidence interval

BCLC

Barcelona clinic liver cancer

ALP

Alkaline phosphatase

PT

Prothrombin time

AST

Aspartate aminotransferase

ALT

Alanine aminotransferase

ALP

Alkaline phosphatase

PCT

Platelet hematocrit

WBC

White blood cell

RBC

Red blood cell

NEU

Neutrophil

ALB

Albumin

GGT

Glutamine transpeptidase

BMI

Body mass index

TBIL

Total bilirubin

IBIL

Indirect bilirubin

BUN

Blood urea nitrogen

CR

Creatinine

TBA

Total bile acids

ALB

Albumin

WBC

White blood cell

RBC

Red blood cell

PDW

Platelet distribution width

MPV

Mean platelet volume

P-LCR

Platelet -larger cell ratio

LYMP

Lymphocyte count

APTT

Activated partial thromboplastin time

PT

Prothrombin time

INR

International normalized ratio

NAFLD

Non-alcoholic fatty liver disease

HBV

Hepatitis B virus

Author contributions

Study conception and design: Hua Li, Shuguang Zhu, Yang Yang. Acquisition of data: Haoqi Chen, Xiaowen Wang, Xijing Yan, Zexin Lin. Analysis and interpretation of data: Wenfeng Zhu, Wenjie Zheng, Xiaolong Chen. Drafting of manuscript: Wenfeng Zhu, Wenjie Zheng, Haoqi Chen, Kaiming He. Critical revision of manuscript: Hua Li, Jianfeng Zhang, Shuguang Zhu, Yang Yang.

Funding information

This work was supported by National Natural Science Foundation of China (82270688, 82100691), Natural Science Foundation of Guangdong Province(2021A1515010726).

Data availability

The data that support the findings of this study are available from the author (18369609271@163.com) upon reasonable request.

Declarations

Consent for publication

All authors approved the submitted final version and this manuscript is not under consideration for other journal.

Competing interests

The authors declare no competing interests.

Ethics Statement

This article is a retrospective review of patient data, and the study design was approved by the Medical Ethics Committee of the Third Affiliated Hospital of Sun Yat-sen University. This article does not contain any studies with animals or human participants performed by any of the authors. All methods were carried out in accordance with relevant guidelines and regulations.