Lipocalin-2: A novel diagnostic marker for hepatocellular carcinoma

Abstract

BACKGROUND:

Hepatocellular carcinoma (HCC) is a leading cause of cancer-related mortality worldwide. Viral hepatitis, alcoholism and non-alcoholic steatohepatitis are the most common risk factors. Despite the advances in HCC screening and treatment options, HCC still has a high mortality rate and a high rate of recurrence after treatment.

Lipocalin-2 (LCN-2) is a glycoprotein transporter that is highly expressed in HCC tissues.

OBJECTIVE:

To evaluate serum LCN-2 as a diagnostic marker for HCC.

METHODS:

The study was carried out in Zagazig university hospitals. It included 210 HCC patients (subdivided in three subgroups), 72 liver cirrhosis patients without HCC and 18 normal control persons (the total is 300 subjects). All the study subjects were evaluated by history taking, physical examination, routine laboratory investigations, alpha-fetoprotein (AFP) and LCN-2 in addition radiology.

RESULTS:

In comparison between HCC and control, there was a statistically significant difference in hemoglobin percent (HB%), platelet count, serum ALT, AST, ALP, bilirubin, albumin and creatinine. In comparison to AFP, LCN-2 $>$ 225 ng/ml had a higher diagnostic performance in HCC patients and was more accurate in differentiation between cirrhosis and HCC patients.

CONCLUSION:

LCN-2 is a good candidate for HCC diagnosis and screening.

Keywords

Hepatocellular carcinoma alpha-fetoprotein lipocalin-2 screening cirrhosis hepatitis

1. Introduction

Hepatocellular carcinoma (HCC) is the most common primary liver cancer. It is the 5 ${}^{\text{th}}$ most common cancer and the 3 ${}^{\text{rd}}$ leading cause of cancer-related death. Liver cirrhosis mainly due to hepatitis B virus (HBV) and hepatitis C virus (HCV) infections, chronic alcohol intake and non-alcoholic steatohepatitis (NASH) are the leading causes of HCC. In Egypt, HCV is a major national health problem and is the leading cause of HCC. HCC affects men more than women most probably due to more prevalence of HCV and HBV infections, smoking and alcohol consumption. Additionally, a positive association between serum testosterone level and risk of HCC has been reported [1, 2].

Tri-phasic contrast-enhanced computed tomography (triphasic CT) is the imaging modality of choice for diagnosis of HCC. Majority of HCC lesions are characterized by rapid arterial phase enhancement and washout of the contrast during the portal and late phases. Liver biopsy should only be considered when diagnostic imaging results are doubtful, as well as confirmation of the diagnosis for small lesions ( $<$ 2 cm) prior to surgical resection [3, 5].

Currently, abdominal ultrasonography (US) is the technique of choice for HCC screening in cirrhotic patients. US can also detect vascular invasion (portal vein thrombosis) and hilar lymphadenopathy in advanced cases [4].

Alpha fetoprotein (AFP) is the most widely used tumor biomarker currently available for the early detection of HCC. False-positive AFP results frequently occur with pregnancy, chronic hepatitis, cirrhosis, colon and testicular malignancy. Additionally, AFP usually shows normal values with small HCC lesions [6].

Glypican-3 (GPC3), Des- $\gamma$ -carboxyprothrombin (DCP), Transforming growth factor- $\beta$ 1 (TGF- $\beta$ 1) and MicroRNAs are also possible HCC biomarkers [7, 8, 9, 10].

HCC still has a high mortality rate despite the previous diagnostic and screening options. This is most probably because most patients present with advanced or unresectable disease in addition to the high recurrence rates after successful treatment. Therefore, new approaches targeting molecular characteristics of HCC are needed to improve screening and outcome [11].

Lipocalin-2 (LCN-2), or neutrophil gelatinase- associated lipocalin (NGAL), also known as siderocalin and uterocalin, is a 25-KDa transporter protein that has a significant role in cell differentiation and apoptosis. It has a role in suppressing bacterial growth and has been correlated with hepatic inflammation in NASH [12].

LCN-2 may induce abnormal cellular proliferation, growth and progression in many types of malignant tumors as breast, thyroid and colon cancer as well as leukemia. On the other hand, although LCN-2 is highly expressed in HCC nodules, its role in the growth and progression of HCC cells is still unclear. Moreover, LCN-2 is found in some studies to inhibit tumor growth and invasion in liver, lung and pancreatic cancers patients. The available theories for this protective role of LCN-2 in HCC discuss its interaction and inhibitory effects on the matrix metalloproteinases (MMP), especially MMP-2 and MMP-9, which are extracellular matrix-degrading enzymes secreted by tumor cells to enhance tumor invasiveness and metastasis [13].

The aim of our study was to evaluate serum LCN-2 as a diagnostic marker for HCC.

2. Patients and methods

This study was carried out in the Internal Medicine Department of Zagazig University Hospitals, during the period from May 2016 till November 2018. The hepatology unit of our department receives annually 2000–2500 patients with liver cirrhosis and 450–500 patients with HCC. Inclusion criteria for all patients include age $>$ 18 years, acceptance to be enrolled in the study, no evidence of extrahepatic malignancy and no evidence of renal impairment.

We randomly selected 282 patients and 18 normal healthy persons (total 300 subjects). The study subjects were grouped as following:

•
Group-1: included (210) patients with HCC, diagnosed by typical criteria in Triphasic-CT scan and were initially classified according to Milan criteria (single lesion $\leqslant$ 5 cm or up to 3 lesions, none of them $>$ 3 cm in diameter, without any evidence of vascular invasion or distant metastases) [14].

–
HCC patients were classified into those outside Milan criteria (HCC-I) and those within (HCC-II) Milan criteria. Each group included 70 patients. HCC-II patients were comparable to stages 0, A and B of Barcelona clinic liver cancer (BCLC) criteria and were assigned to receive the recommended treatment, while HCC-I patients were comparable to stages C and D of BCLC criteria and received only conservative treatment [15].
–
A third group of HCC patients (HCC-III) included also 70 patients with HCC, who underwent the recommended treatment and attended our clinic for follow up for at least one year after successful loco-regional treatment of HCC.

•
Group-2 (LC): included (72) patients with liver cirrhosis (without HCC), diagnosed by combination of clinical, laboratory and US criteria. They attended our outpatient clinic for treatment and follow up of features of hepatic decompensation as variceal bleeding, hepatic encephalopathy, ascites etc.
•
Group-3 (control group): included (18) normal healthy persons with no history, laboratory or US evidence of liver diseases.

Informed written consents were obtained from all patients and control subjects. Also, approval of the review board and ethical committee of Zagazig university hospitals was taken.

All patients and control persons were subjected to history taking, thorough physical examination, complete blood count (CBC), liver function tests, serum creatinine level, viral markers for hepatitis C (HCV) and hepatitis B (HBV) viruses (HCV-Ab & HBs-Ag), serum Alpha fetoprotein level and serum Lipocalin-2 level in addition to abdominal US and triphasic CT scan.

Table 1
Demographic data of the study groups

Patient groups HCC-I HCC-II HCC-III LC Control $P$ -value

Patient criteria

Age (years) 51.6 $\pm$ 6.3 53.9 $\pm$ 6 55.1 $\pm$ 7.2 54.6 $\pm$ 6.6 55.1 $\pm$ 4.1 0.404

Male sex 77.8% 77.8% 83.3% 72.2% 77.8% 0.995

Smoking 19 (27.1%) 17 (24.3%) 27 (38.6%) 16 (22.2%) 6 (33.3%) 0.848

HCV infection 64 (91.4%) 67 (95.7%) 68 (97.1%) 68 (94.4%) Zero 0.640

HBV infection 5 (7.1%) 9 (12.9%) 3 (4.3%) 6 (8.3%) Zero 0.640

Child score 9 $\pm$ 2.1 9 $\pm$ 2.4 9.9 $\pm$ 2.9 8.4 $\pm$ 1.9 N/A 0.278

PLT count (/dl) 147.8 $\pm$ 66.9 151.7 $\pm$ 61.5 154.4 $\pm$ 96.1 163.9 $\pm$ 74.8 278.0 $\pm$ 33.4 $<$ 0.001

HB (gm %) 10.1 $\pm$ 1.4 10.3 $\pm$ 1.3 10.6 $\pm$ 1.6 10.9 $\pm$ 1.8 13.9 $\pm$ 0.6 $<$ 0.001

AST (IU/L) 93.5 $\pm$ 49.0 91.8 $\pm$ 48.2 90.0 $\pm$ 40.6 74.3 $\pm$ 40.1 19.7 $\pm$ 7.2 $<$ 0.001

ALT(IU/L) 65.2 $\pm$ 41.8 64.2 $\pm$ 40.0 55.9 $\pm$ 32.0 51.4 $\pm$ 33.0 20.2 $\pm$ 7.1 $<$ 0.001

ALP (IU/L) 175.6 $\pm$ 81.7 176.8 $\pm$ 85.6 171.4 $\pm$ 72.8 157.4 $\pm$ 69.4 56.8 $\pm$ 12.7 $<$ 0.001

Serum bilirubin (mg/dl) 2.8 $\pm$ 2.0 2.6 $\pm$ 2.2 2.4 $\pm$ 1.2 1.9 $\pm$ 1.6 0.4 $\pm$ 0.1 $<$ 0.001

Albumin (gm/dl) 2.7 $\pm$ 0.5 2.8 $\pm$ 0.5 2.7 $\pm$ 0.6 2.8 $\pm$ 0.7 4.3 $\pm$ 0.3 $<$ 0.001

INR 1.48 $\pm$ 0.19 1.48 $\pm$ 0.23 1.54 $\pm$ 0.23 1.37 $\pm$ 0.27 0.99 $\pm$ 0.09 $<$ 0.001

Serum creatinine (mg/dl) 1.17 $\pm$ 0.51 1.08 $\pm$ 0.42 1.39 $\pm$ 0.59 1.24 $\pm$ 0.40 0.60 $\pm$ 0.13 $<$ 0.001

Table 2
Comparison between serum levels of AFP and LCN-2 in the study groups

Patient groups HCC-I (N. $=$ 70) HCC-II (N. $=$ 70) HCC-III (N. $=$ 70) LC (N. $=$ 72) Control (N. $=$ 18) $P$ -value

Serum markers

AFP (ng/ml) 587.4 $\pm$ 813.3 406.4 $\pm$ 655.1 288.3 $\pm$ 481.7 17.3 $\pm$ 20.5 9.5 $\pm$ 6.7 0.003

LCN-2 (ng/ml) 371.5 $\pm$ 34.3 335.7 $\pm$ 40.5 329.8 $\pm$ 67.4 187.8 $\pm$ 24.6 52.4 $\pm$ 7.0 0.001

N. $=$ number of patients.

The collected data were tabulated and analyzed using statistical package for social sciences (IBM SPSS) software version 22.0, IBM Corp., Chicago, USA, 2013. Quantitative data (continuous variables) were expressed as mean $\pm$ standard deviation (mean $\pm$ SD), while qualitative data (categorical variables) were expressed as numbers and percentages. Analysis for quantitative variables was done using ANOVA test. Post hoc Tukey test was done for more than two independent groups with normally distributed data. Qualitative data were analyzed using Chi square test for differences between proportions. The level of significance was taken at $P$ value $<$ 0.05. Pearson correlation was done for numerical normally distributed data. A receiver operating characteristic (ROC) curve was used to evaluate the performance of different tests and to differentiate between certain groups. Cut-off point is the cutoff value that has the maximum Youden’s index (sensitivity $+$ specificity $-$ 1), that is also equivalent to the highest sensitivity and specificity on ROC curve.

Diagnostic characteristics of the estimated cut-off points of both AFP and LCN-2 were calculated and compared for sensitivity, specificity, predictive and negative positive values, likelihood ratio for positive and negative values, diagnostic accuracy and Youden’s index. Kappa coefficient was used to assess inter-rater agreement and reliability between results of AFP $>$ 17 ng/ml and those of LCN-2 $>$ 225 ng/ml in differentiating between liver cirrhosis and HCC.

Table 3
Performance of AFP and LCN-2 in differentiation between HCC and LC groups and differentiation between HCC-I and HCC-II groups

Performance test AUC SE $P$ . value 95% CI

Serum markers

HCC (N. $=$ 210) from LC (N. $=$ 72)

AFP 0.86 0.05 $<$ 0.001 0.76–0.97

LCN-2 0.95 0.03 $<$ 0.001 0.89–1.00

HCC-I (N. $=$ 70) from HCC-II (N. $=$ 70)

AFP 0.58 0.08 0.32 0.42–0.75

LCN-2 0.61 0.08 0.18 0.45–0.78

AUC: Area under curve, SE: Standard error, CI: Confidence interval.

Table 4
Diagnostic characteristics of AFP and LCN-2 cut-off levels in differentiating HCC from LC groups

Serum cut-off level AFP $\geqslant$ 17 ng/ml LCN-2 $\geqslant$ 225 ng/ml

Diagnostic character Value 95% CI Value 95% CI

Sensitivity 87. 5% 75.9 –94.8% 92. 9% 82.7 –98.0 %

Specificity 68. 8% 41.3 –89.0% 93. 8% 69.8 –99.8%

Diagnostic accuracy (DA) 83. 3% 72.7 –91.1% 93. 1% 84.5 –97.7%

Youden’s index 56. 3% 31.9 –80.6% 86. 6% 73.0 –100 %

Positive predictive value (PPV) 90. 7% 79.7 –96.9% 98. 1% 89.9 –100%

Negative predictive value (NPV) 61. 1% 35.7 –82.7% 78. 9% 54.4 –93.9%

Positive likelihood ratio (LR $+$ ) 2. 8 1.3 –5.8 14. 86 2.2 –99.3

Negative likelihood ratio (LR $-$ ) 0. 2 0.1 –0.4 0. 08 0.03 –0.2

Diagnostic odd ratio (LR) 15. 4 4.1 –57.7 195. 00 20.2 –1878.7

Kappa coefficient 0. 5 0.3 –0.8 0. 81 0.7 –1.0

3. Results

Patient groups	HCC-I	HCC-II	HCC-III	LC	Control	$P$ -value
Patient criteria
Age (years)	51.6 $\pm$ 6.3	53.9 $\pm$ 6	55.1 $\pm$ 7.2	54.6 $\pm$ 6.6	55.1 $\pm$ 4.1	0.404
Male sex	77.8%	77.8%	83.3%	72.2%	77.8%	0.995
Smoking	19 (27.1%)	17 (24.3%)	27 (38.6%)	16 (22.2%)	6 (33.3%)	0.848
HCV infection	64 (91.4%)	67 (95.7%)	68 (97.1%)	68 (94.4%)	Zero	0.640
HBV infection	5 (7.1%)	9 (12.9%)	3 (4.3%)	6 (8.3%)	Zero	0.640
Child score	9 $\pm$ 2.1	9 $\pm$ 2.4	9.9 $\pm$ 2.9	8.4 $\pm$ 1.9	N/A	0.278
PLT count (/dl)	147.8 $\pm$ 66.9	151.7 $\pm$ 61.5	154.4 $\pm$ 96.1	163.9 $\pm$ 74.8	278.0 $\pm$ 33.4	$<$ 0.001
HB (gm %)	10.1 $\pm$ 1.4	10.3 $\pm$ 1.3	10.6 $\pm$ 1.6	10.9 $\pm$ 1.8	13.9 $\pm$ 0.6	$<$ 0.001
AST (IU/L)	93.5 $\pm$ 49.0	91.8 $\pm$ 48.2	90.0 $\pm$ 40.6	74.3 $\pm$ 40.1	19.7 $\pm$ 7.2	$<$ 0.001
ALT(IU/L)	65.2 $\pm$ 41.8	64.2 $\pm$ 40.0	55.9 $\pm$ 32.0	51.4 $\pm$ 33.0	20.2 $\pm$ 7.1	$<$ 0.001
ALP (IU/L)	175.6 $\pm$ 81.7	176.8 $\pm$ 85.6	171.4 $\pm$ 72.8	157.4 $\pm$ 69.4	56.8 $\pm$ 12.7	$<$ 0.001
Serum bilirubin (mg/dl)	2.8 $\pm$ 2.0	2.6 $\pm$ 2.2	2.4 $\pm$ 1.2	1.9 $\pm$ 1.6	0.4 $\pm$ 0.1	$<$ 0.001
Albumin (gm/dl)	2.7 $\pm$ 0.5	2.8 $\pm$ 0.5	2.7 $\pm$ 0.6	2.8 $\pm$ 0.7	4.3 $\pm$ 0.3	$<$ 0.001
INR	1.48 $\pm$ 0.19	1.48 $\pm$ 0.23	1.54 $\pm$ 0.23	1.37 $\pm$ 0.27	0.99 $\pm$ 0.09	$<$ 0.001
Serum creatinine (mg/dl)	1.17 $\pm$ 0.51	1.08 $\pm$ 0.42	1.39 $\pm$ 0.59	1.24 $\pm$ 0.40	0.60 $\pm$ 0.13	$<$ 0.001

Patient groups	HCC-I (N. $=$ 70)	HCC-II (N. $=$ 70)	HCC-III (N. $=$ 70)	LC (N. $=$ 72)	Control (N. $=$ 18)	$P$ -value
Serum markers
AFP (ng/ml)	587.4 $\pm$ 813.3	406.4 $\pm$ 655.1	288.3 $\pm$ 481.7	17.3 $\pm$ 20.5	9.5 $\pm$ 6.7	0.003
LCN-2 (ng/ml)	371.5 $\pm$ 34.3	335.7 $\pm$ 40.5	329.8 $\pm$ 67.4	187.8 $\pm$ 24.6	52.4 $\pm$ 7.0	0.001

Performance test	AUC	SE	$P$ . value	95% CI
Serum markers
HCC (N. $=$ 210) from LC (N. $=$ 72)
AFP	0.86	0.05	$<$ 0.001	0.76–0.97
LCN-2	0.95	0.03	$<$ 0.001	0.89–1.00
HCC-I (N. $=$ 70) from HCC-II (N. $=$ 70)
AFP	0.58	0.08	0.32	0.42–0.75
LCN-2	0.61	0.08	0.18	0.45–0.78

Serum cut-off level	AFP $\geqslant$ 17 ng/ml	LCN-2 $\geqslant$ 225 ng/ml
Sensitivity	87.	5%	75.9	–94.8%	92.	9%	82.7	–98.0 %
Specificity	68.	8%	41.3	–89.0%	93.	8%	69.8	–99.8%
Diagnostic accuracy (DA)	83.	3%	72.7	–91.1%	93.	1%	84.5	–97.7%
Youden’s index	56.	3%	31.9	–80.6%	86.	6%	73.0	–100 %
Positive predictive value (PPV)	90.	7%	79.7	–96.9%	98.	1%	89.9	–100%
Negative predictive value (NPV)	61.	1%	35.7	–82.7%	78.	9%	54.4	–93.9%
Positive likelihood ratio (LR $+$ )	2.	8	1.3	–5.8	14.	86	2.2	–99.3
Negative likelihood ratio (LR $-$ )	0.	2	0.1	–0.4	0.	08	0.03	–0.2
Diagnostic odd ratio (LR)	15.	4	4.1	–57.7	195.	00	20.2	–1878.7
Kappa coefficient	0.	5	0.3	–0.8	0.	81	0.7	–1.0

Figure 1.

Comparison between study groups regarding AFP.

Figure 2.

Comparison between study groups regarding LCN-2.

Figure 3.

ROC curve for AFP and lipocalin-2 in differentiating HCC from LC groups.

Figure 4.

ROC curve for AFP and lipocalin-2 in differentiating HCC-I from HCC-II.

We conducted a prospective study in Zagazig University Hospitals to assess LCN-2 as a reliable diagnostic marker for HCC in comparison to the standard AFP marker. The study included 300 persons; 210 HCC patients in 3 subgroups (HCC-I, HCC-II & HCC-III), 72 LC patients and 18 normal control persons. Table 1, showed the demographic, clinical and laboratory data of the study groups. There was statistically significant deterioration of hemoglobin percent (HB%), platelet (PLT) count, liver function tests and kidney function tests in HCC and LC groups in comparison to control group. Table 2 and Figs 1 and 2 showed significantly ( $p<$ 0.003) higher serum levels of AFP in early HCC patients (406 ng/ml) and late HCC patients (587 ng/ml) compared to control group (9.5 ng/ml) and LC group (17.3 ng/ml). LCN-2 serum levels were also significantly ( $p<$ 0.001) higher in HCC-I group (371.5 ng/ml), HCC-II group (335.7 ng/ml) and HCC-III group (329.8 ng/ml) when compared to LC (187.8 ng/ml) and control groups (52.4 ng/ml). The study showed no significant correlation between LCN-2 and other demographic, clinical or laboratory variables of the study groups including AFP. Regarding differentiating cirrhosis from HCC, Table 3 and Fig. 3 revealed that area under receiver operating characteristic curve (AUROC) for LCN-2 at a cut-off $>$ 225 ng/ml was 0.95 ( $p<$ 0.001), while AUROC for AFP at a cut-off $>$ 17 ng/ml was 0.86 ( $p<$ 0.001). While in differentiating between HCC-I and HCC-II groups, both AUROC of AFP and LCN-2 were non-significant as shown by Table 3 and Fig. 4. Table 4 showed better diagnostic performance for LCN-2 in differentiating HCC groups from LC group in comparison to AFP, with sensitivity of 93%, positive predictive value (PPV) of 98%, specificity of 94%, negative predictive value (NPV) of 79% and a diagnostic accuracy of 93%.

4. Discussion

Hepatocellular carcinoma (HCC) is one of the most prevalent human cancers and a leading cause of cancer-related deaths worldwide. The most prominent etiological factors associated with HCC are chronic viral hepatitis B and C infections, non-alcoholic fatty liver disease, toxins and alcohol exposure [1].

A substantial percentage of HCC cases are diagnosed in advanced stages when the treatment options are limited with lower survival rate. Imaging techniques such as US, CT scans or magnetic resonance imaging (MRI) are used for early detection of HCC. However, they are operator-dependent and are too expensive to be used in routine medical checks. As the five-year survival rate after diagnosis at an advanced stage is less than 5%, novel diagnostic techniques allowing for detection at early stage HCC are in high demand [14].

Serum AFP is a widely used marker for the diagnosis of HCC or follow-up after surgery or chemotherapy. High serum level of AFP correlates with poor prognosis of HCC patients. However, two thirds of HCC patients with nodules less than 3 cm have serum AFP levels less than 200 ng/ml and up to 20% of HCC patients do not produce AFP [11].

LCN-2 is a glycoprotein of the lipocalin transporter protein family with roles in immunity and carcinogenesis. It plays an important role in promoting tumorigenesis through enhancing tumor cell survival, proliferation and metastatic potential. Elevated LCN-2 expression has been detected in a variety of human malignancies, including breast cancer, endometrial carcinoma, ovarian carcinoma, pancreatic cancer, gastric cancer and thyroid carcinoma. LCN-2 is over-expressed in HCC tissues and uncertain roles in the proliferation and spread of HCC lesions. Some studies claim it may have a protective effects, others showed a promoting influence on tumor growth and invasion. However, the clinical significance of LCN-2 in HCC detection and early diagnosis has not been previously fully investigated [16, 17].

The aim of this work was to evaluate serum LCN-2 as a diagnostic marker of HCC in cirrhotic patients in comparison to AFP.

This study included 300 persons; 210 HCC patients, 72 patients with liver cirrhosis and 18 normal control persons. There were no significant differences between the study groups regarding age, sex, smoking, HCV or HBV infections. However, the diseased groups (with LC or HCC) had significantly lower hemoglobin, platelet count and serum albumin with significantly higher serum ALT, AST, ALP, bilirubin, INR and serum creatinine in comparison to the control group indicating both vascular and cellular hepatic decompensation in the former groups.

There was no significant difference between the HCC and LC groups regarding Child score and this goes in agreement with Trevisani et al. 1995 who reported no clinical or laboratory differences between cirrhotic and HCC patients [18].

In this study, there was significant higher serum levels ( $p<$ 0.003) of AFP in early HCC patients (406 ng/ml) and late HCC patients (587 ng/ml) when compared with control group (9.5 ng/ml) and when compared with cirrhotic group (17.3 ng/ml).

This goes in agreement with Biondi et al. who reported a significant difference in AFP between HCC and liver cirrhosis in comparison to healthy subjects in contrary to Massironi et al. who reported no significant difference in AFP between HCC and liver cirrhosis [19, 20].

LCN-2 serum levels were significantly higher in HCC-I group (371.5 ng/mL), HCC-II group (335.7 ng/ml) and HCC-III group (329.8 ng/mL) when compared to LC group (187.8 ng/ml) and control groups (52.4 ng/ml) ( $p<$ 0.001). In differentiation between HCC and LC groups, the present study revealed that the AUROC for LCN-2 at a cut-off $>$ 225 ng/ml was 0.945 ( $p<$ 0.001), while AUROC for AFP at a cut-off $>$ 17.0 ng/ml was 0.864 ( $p<$ 0.001), also the diagnostic performance for LCN-2 in HCC groups showed sensitivity of 92.9%, PPV of 98.1%, specificity of 93.8%, NPV of 78.9% and a diagnostic accuracy of 93.1%, indicating that LCN-2 had a significant high diagnostic performance in comparison to the significant moderate diagnostic performance of AFP. This goes in agreement with Asimakopoulou et al. who stated that LCN-2 levels were elevated in HCC tissues and higher LCN-2 expression was correlated with shorter overall survival in HCC patients [21].

Our study also revealed that the AUROC for LCN-2 was 0.614 ( $p=$ 0.175), while AUROC for AFP was 0.317 ( $p=$ 0.317) when comparing group-I and group-II HCC patients, indicating the poor performance of both AFP and lipocalin-2 in differentiating between early and late HCC patients.

In conclusion, serum LCN-2 $>$ 225 ng/ml is a good candidate as a novel diagnostic marker for HCC with higher diagnostic performance and more accurate differentiation between LC and HCC than AFP. Also, LCN-2 may be a potential target for future therapy to suppress the tumor growth and invasion.

Footnotes

Acknowledgments

This work is funded only by its authors.

Conflict of interest

None declared.

References

Yapali

and Tozun

, Epidemiology and viral risk factors for hepatocellular carcinoma in the eastern mediterranean countries, Hepatoma Res 4 (2018), 24–35.

Ming-Whei

and Chien-Jen

, Elevated serum testosterone levels and risk of hepatocellular carcinoma, Cancer Res 53(7) (1993), 794.

Lewis

Sirlin

Almasri

et al., Imaging for the diagnosis of hepatocellular carcinoma: a systematic review and meta-analysis, Hepatology 67(1) (2018), 401–421.

Marrero

Kulik

Sirlin

et al., Diagnosis, staging, and management of hepatocellular carcinoma: practice guidance by the american association for the study of liver diseases, Hepatology 68(2) (2018), 723–750.

Sparchez

and Mocan

, Contemporary role of liver biopsy in hepatocellular carcinoma, World J Hepatol 10(7) (2018), 452–461.

Nakagawa

Beppu

Okabe

et al., Triple positive tumor markers predict recurrence and survival in early stage hepatocellular carcinoma, Hepatol Res 44(9) (2014), 964–974.

Filmus

and Capurro

, Glypican-3: a marker and a therapeutic target in hepatocellular carcinoma, FEBS J 280 (2013), 2471–2476.

Chaur-Shine

Chih-Lin

Hsi-Chang

et al., Usefulness of serum des-γ-carboxy prothrombin in detection of hepatocellular carcinoma, World J Gastroenterol 11(39) (2005), 6115–6119.

Kohla

Attia

Darwesh

et al., Association of serum levels of transforming growth factor β1 with disease severity in patients with hepatocellular carcinoma, Hepatoma Res 3 (2017), 294–301.

10.

Zhong

and Guo-Ping

, MicroRNAs in hepatocellular carcinoma: pathogenesis and therapy, Journal of Medicine & Medical Sciences 2(12) (2011), 1235–1242.

11.

Al-Salloom

, An update of biochemical markers of hepatocellular carcinoma, Int J Health Sci 10(1) (2016), 121–136.

12.

Abd El Moety

El Sharkawy

and Hussein

, Lipocalin: a novel diagnostic marker for hepatocellular carcinoma in chronic liver disease patients in Egypt, International Journal of Clinical Medicine 4 (2013), 440–450.

13.

Eun-Kyoung

Hyun-Jun

Kong-Ju

et al., Inhibition of the proliferation and invasion of hepatocellular carcinoma cells by lipocalin 2 through blockade of JNK and PI3K/Akt signaling, Int J Oncol 38 (2011), 325–333.

14.

Rostambeigi

Taylor

et al., Effect of MRI versus MDCT on milan criteria scores and liver transplantation eligibility, Am J Roentgenol 206(4) (2016), 726–733.

15.

Kinoshita

Onoda

Fushiya

et al., Staging systems for hepatocellular carcinoma: current status and future perspectives, World J Hepatol 7(3) (2015), 406–424.

16.

Ata

Abdel Sameea

El Sabawy

et al., Urinary lipocalin-2 as a potential marker for diagnosis of early hepatocellular carcinoma, Journal of Gastroenterology and Digestive Diseases 2(3) (2017), 17.

17.

Bauvois

and Susin

, Revisiting neutrophil gelatinase-associated lipocalin (NGAL) in cancer: saint or sinner? Cancers (Basel) 10 (2018), 336.

18.

Trevisani

D’Intino

P.E.

Caraceni

et al., Etiologic factors and clinical presentation of hepatocellular carcinoma. Differences between cirrhotic and non-cirrhotic Italian patients, Cancer 75(9) (1995), 2220–2232.

19.

Biondi

Malaguarnera

Vacante

et al., Elevated serum levels of chromogranin a in hepatocellular carcinoma, BMC Surg 12(1) (2012), 7.

20.

Massironi

Fraquelli

Paggi

et al., Chromogranin a levels in chronic liver disease and hepatocellular carcinoma, Dig Liver Dis 41(1) (2009), 31–35.

21.

Asimakopoulou

Weiskirchen

and Weiskirchen

, Lipocalin 2 (LCN2) expression in hepatic malfunction and therapy, Front Physiol 7 (2016), 430–448.

Serum cut-off level	AFP $\geqslant$ 17 ng/ml				LCN-2 $\geqslant$ 225 ng/ml
Diagnostic character	Value		95% CI		Value		95% CI
Sensitivity	87.	5%	75.9	–94.8%	92.	9%	82.7	–98.0 %
Specificity	68.	8%	41.3	–89.0%	93.	8%	69.8	–99.8%
Diagnostic accuracy (DA)	83.	3%	72.7	–91.1%	93.	1%	84.5	–97.7%
Youden’s index	56.	3%	31.9	–80.6%	86.	6%	73.0	–100 %
Positive predictive value (PPV)	90.	7%	79.7	–96.9%	98.	1%	89.9	–100%
Negative predictive value (NPV)	61.	1%	35.7	–82.7%	78.	9%	54.4	–93.9%
Positive likelihood ratio (LR $+$ )	2.	8	1.3	–5.8	14.	86	2.2	–99.3
Negative likelihood ratio (LR $-$ )	0.	2	0.1	–0.4	0.	08	0.03	–0.2
Diagnostic odd ratio (LR)	15.	4	4.1	–57.7	195.	00	20.2	–1878.7
Kappa coefficient	0.	5	0.3	–0.8	0.	81	0.7	–1.0