Tumor Biol. DOI 10.1007/s13277-015-3501-4

RESEARCH ARTICLE

HCC-DETECT: a combination of nuclear, cytoplasmic, and oncofetal proteins as biomarkers for hepatocellular carcinoma Abdelfattah M. Attallah 1 & Mohamed El-Far 2 & Camelia A. Abdel Malak 3 & Mohamed M. Omran 4 & Gamal E. Shiha 5 & Khaled Farid 5 & Lamiaa A. Barakat 6 & Mohamed S. Albannan 1 & Ahmed A. Attallah 1 & Mohamed A. Abdelrazek 1 & Mohamed S. Elbendary 1 & Refaat Sabry 3 & Gehan A. Hamoda 1 & Mohamed M. Elshemy 1 & Abdallah A. Ragab 1 & Basma M. Foda 1 & Sanaa O. Abdallah 7

Received: 17 March 2015 / Accepted: 23 April 2015 # International Society of Oncology and BioMarkers (ISOBM) 2015

Abstract Currently, the search for suitable hepatocellular carcinoma (HCC) biomarkers is very intensive. Besides, efficacy and cost/effectiveness of screening and surveillance of cirrhotics for the diagnosis of HCC is still debated. So, the present study is concerned with the evaluation of cytokeratin-1 (CK-1) and nuclear matrix protein-52 (NMP-52) for identifying HCC. Two-hundred and eighty individuals categorized into three groups [liver fibrosis (F1–F3), cirrhosis (F4), and HCC] constituted this study. Western blot was used for identifying CK-1 and NMP-52 in serum samples. As a result, a single immunoreactive band was shown at 67 and 52 kDa corresponding to CK-1 and NMP-52, respectively. Both CK1 and NMP-52 bands were cut and electroeluted separately. These markers were quantified in sera using ELISA. Patients with HCC were associated with higher concentrations of CK1 and NMP-52 than those without HCC with a significant difference (P < 0.0001). CK-1 showed an area under receiver-operating characteristic curve (AUC) of 0.83 with 75 % sensitivity and 82 % specificity while NMP-52 yielded * Abdelfattah M. Attallah [email protected] 1

Biotechnology Research Center, P.O. Box (14), 23 July St., Industrial Zone, New Damietta 34517, Egypt

2

Faculty of Science, Mansoura University, Mansoura, Egypt

3

Faculty of Science, Damietta University, Mansoura, Egypt

4

Faculty of Science, Helwan University, Cairo, Egypt

5

Faculty of Medicine, Mansoura University, Mansoura, Egypt

6

Faculty of Science, Port-Said University, Port-Said, Egypt

7

Faculty of Science, Cairo University, Giza, Egypt

0.72 AUC with 62 % sensitivity and 70 % specificity for identifying HCC. HCC-DETECT comprising CK-1 and NMP-52 together with AFP was then constructed yielding 0.90 AUC for identifying HCC with 80 % sensitivity and 92 % specificity. HCC-DETECT was then tested for separating HCC from F1–F3 showing 0.94 AUC with 80 % sensitivity and 93 % specificity. In conclusion, CK-1 in conjunction with NMP-52 and AFP could have a potential role for improving the detection of HCC with a high degree of accuracy. Keywords Liver diseases . Serum biomarkers . HCC-DETECT . Western blot

Introduction Liver fibrosis is a wound-healing process characterized by the excessive accumulation of extracellular matrix (ECM) following liver injury [1]. Progression of disease with sustained fibrogenesis leads to cirrhosis which can have a poor outcome and high mortality [2]. It was reported that hepatocellular carcinoma (HCC) develops in a cirrhotic liver in 80 % of cases [3] and is considered the most common primary malignancy of the liver, being the fifth most frequent cancer worldwide [4]. The current diagnosis of HCC is a multistage process that includes clinical, laboratory, imaging, and pathological examinations [5]. Alpha-fetoprotein (AFP) is one of the oncofetal proteins which is considered to be the most common marker that is used in clinical practice in combination with ultrasonography for the detection of HCC in cirrhotic patients [6]. Nevertheless, it was concluded that the use of AFP is not a useful diagnostic test [7–9] and the abdominal ultrasound is

Tumor Biol.

highly dependent on the operator’s experience [10]. Hence, identification of novel serum biomarkers is an important goal in the diagnosis of cancer. Nuclear matrix proteins (NMPs) are a part of the structural frame of the nucleus and released from dead cell nuclei. These nuclear proteins were investigated as a marker of organ damage in liver disease [11, 12]. That is because alterations in the nuclear shape or structures that occur with neoplastic transformation are accompanied by changes in nuclear matrix composition and architecture [13]. On the other hand, cytokeratins (CKs) constitute a diverse group of intermediate filament proteins, which are cytoplasmic cytoskeletal structures, forming the cytoskeleton of epithelial cells, and their main function is to maintain the epithelial cell integrity [14, 15]. It was reported that CKs are the major intermediate filament protein in the liver, and any cellular damage that alters the hepatocyte membrane integrity may cause the release of CKs into the blood circulation [16]. Several reports showed that when hepatocytes undergo necrosis as a result of liver inflammation, both uncleaved and partially degraded CK fragments are released in the bloodstream and can be measured using serological assays [17, 18, 16]. Therefore, NMPs in conjunction with CKs may have a potential role in the diagnosis of liver disease. Previously, we have supported that CK-1 may be clinically valuable as a marker for predicting HCC [19]. The present work is concerned with enhancing the diagnostic ability of CK-1 “cytoplasmic protein” by its combination with NMP-52 “nuclear protein” together with AFP “oncofetal protein” for discriminating patients with HCC from those with non-malignant chronic liver diseases.

Patients and methods Samples Two hundred and eighty consecutive Egyptian individuals constituted the present study. All tissue and serum samples were obtained with informed consent. Patients were categorized into three different groups. The first one included serum samples from 100 to 60 patients who have liver fibrosis (F1– F3). This cohort comprised 100 and 15 males and 45 females with a mean (±SD) age of 43.2(±7.6)years. The second group included serum samples from 60 patients who have liver cirrhosis (F4). This cohort comprised 44 males and 16 females with a mean (±SD) age of 54.5(±9.1)years. Histopathological classification for liver fibrosis and cirrhosis was performed according to the METAVIR score [20]. Needle liver biopsy specimens were obtained with an 18-gauge or larger needle. To be considered as adequate for scoring, the liver biopsies have to measure at least 15 mm and/or contain at least five portal tracts, except for cirrhosis for which no limitation was required. Liver fibrosis was scored on a 5-point scale: F0, no fibrosis; F1, portal fibrosis alone; F2, portal fibrosis with rare

septae; F3, portal fibrosis with many septae; and F4, cirrhosis. The last group included serum samples from 60 patients who developed HCC. This cohort comprised 43 males and 17 females with a mean (±SD) age of 56.3(±8.6)years. The diagnosis of HCC was carried out according to the American Association for the Study of Liver Diseases Practice Guidelines [21]. The diagnosis of HCC was based on AFP levels above 400 U/L, presence of hepatic focal lesion(s) detected by liver ultrasound (US), and confirmed by computed tomography and/or magnetic resonance imaging techniques. The final diagnosis was confirmed by histopathologic analysis on USassisted fine-needle biopsy, when indicated. Patients with other causes of liver diseases, or other suspected malignancies were excluded from the present study. None of the HCC patients had received transarterial embolisation or chemotherapy or underwent radiofrequency ablation or surgical interference. Informed consents were obtained from all participants and they were fully informed concerning the diagnostic procedures involved and disease nature. The study protocol conformed to the ethical guidelines of the 1975 Helsinki Declaration. Laboratory tests Liver function tests [aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), total bilirubin, and albumin] were all measured on fresh serum on an automated biochemistry analyzer (Roche/Hitachi 917, Mannheim, Germany). Complete blood count including platelets counting was determined on blood treated with EDTA-K3 using KX-21 Sysmex automated hematology analyzer (Sysmex Corporation, Japan). Another portion was treated with a citrate solution for prothrombin time-international normalized ratio (PT-INR). AFP level was estimated by chemiluminescence, with Immulite (1000) AFP kit (Diagnostic Products Corporation, Los Angeles, CA, USA). All patients were tested negative for HBsAg (Dia.Pro, Milan, Italy) and were tested positive for anti-HCV antibodies (Biomedica, Sorin, Italy). Patients were then confirmed for the presence of HCV-RNA using quantitative polymerase chain reaction assay (COBAS Ampliprep/COBAS TaqMan, Roche Diagnostics, Pleasanton, USA). Western blot and gel electroelution Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was carried out in 0.75-mm thick, 12 % vertical slab gels according to the method of Laemmli [22]. This technique relies on the migration of charged molecules in a gel matrix in response to an electrical field and facilitates the separation of a mixture of proteins according to molecular weight. Following electrophoretic separation, serum samples were electrotransferred onto nitrocellulose membrane (0.45-

Tumor Biol.

Receiver operating-characteristic (ROC) curves were constructed for determining the diagnostic ability of different variables. Stepwise linear regression analysis was performed to develop an index for identifying HCC. Based on the ROC analysis, the best cutoff points were selected and then sensitivity, specificity, and accuracy in addition to positive and negative predictive values were derived from such a 2×2 contingency table.

mm pore size, Sigma) in a protein transfer unit [23]. They were then immunostained using respective antibodies corresponding to cytokeratin-1 (CK-1) and nuclear matrix protein52 (NMP-52) separately. The dilutions of the antibodies and conjugates were adjusted to eliminate the background, i.e., the presence of the target antigens in the low concentration. Finally, both CK-1 and NMP-52 bands were cut and then electroeluted separately from preparative polyacrylamide gels at 200 V for 3 h in a dialysis bag (Sigma). The protein content of the purified bands was determined [24] and the remainder was stored at −20 °C.

Results

Quantitation of CK-1 and NMP-52 using ELISA

Patients’ characteristics

Diluted serum samples (1:400) in coating buffer (50 mM carbonate/bicarbonate buffer, pH 9.6) were tested (50 μL/ well) for CK-1 bound on a 96-well microtiter plate at 4 °C overnight. After blocking with phosphate buffer saline (PBS)+0.5 % Tween 20 (200 μL/well), mouse monoclonal antibody corresponding to CK-1 at dilution 1:200 in PBS was added separately (50 μL/well) then incubated at 37 °C for 1 h. Goat anti-mouse antibody conjugated with ALP (Sigma) 1:500 in 0.2 % bovine serum albumin (BSA) was incubated at 37 °C for 1 h. The plate was washed with PBS+0.5 % Tween 20 after every step. The substrate was 1 mg/mL pnitrophenyl phosphate and the intensity of the signal was determined by measuring the absorbance at 450 nm after 10 min using a microtiter plate reader (Σ960, Metretech Inc., Germany). Likewise, the aforementioned steps was performed in respect of NMP-52 using the same quantities and intervals but in different concentrations as the following: sera dilution (1:100) in coating buffer, blocking 5 % BSA in coating buffer, mono-specific antibody at dilution 1:1000 in PBS, and ALPconjugated goat anti-rabbit IgG (Sigma) was diluted 1:3000 in 0.2 % BSA. Serial concentrations of the purified CK-1 and NMP-52 were tested in parallel to establish a dose-response curve as a function of the concentration in serum samples.

Laboratory characteristics of all patients are summarized in Table 1. The main endpoint of the current study was concerned with the identification of patients with HCC from cirrhotic patients. The laboratory data of these two groups were analyzed using univariate analysis. As a result, the data showed that only levels of total bilirubin, albumin, and AFP could discriminate significantly patients who have HCC from those who developed F4 (P<0.05). Patients with HCC produced a range of AFP values from normal to more than 29370 U/L.

Statistical analysis All statistical analyses were performed by SPSS software version 15.0 (SPSS Inc., Chicago, IL) and GraphPad Prism package, version 5.0 (GraphPad Software, San Diego, CA). Patient characteristics have been represented as the mean±standard deviation (SD). A value of P>0.05 is considered not significant and P<0.05 is considered statistically significant. The correlation was evaluated by Spearman’s rank correlation coefficient. The deviation of AFP was successfully corrected by log transformation of the data. The main endpoint of this work was concerned with the early detection of patients with clinically HCC. Univariate analysis based on Student’s t test was performed to identify variables that were significantly different between patients with HCC versus those with cirrhosis.

Table 1

Patients characteristics

Variables

Fibrosis, F1–F3 (n=160)

Cirrhosis, F4 (n=60)

HCC (n=60) P*

Male/female HBsAg (−) and HCV Ab (+) Age (years) AST (U/L)a ALT (U/L)a AST/ALT (AAR) ALP (U/L)a T.bilirubin (mg/dL)a Albumin (g/dL)a PT-INRa PLT (109/L)a AFP (U/L)a Log (AFP (U/L)a

115/45 160

44/16 60

43/17 60

0.973

43.2±7.6 56.0±32.3 67.4±34.9 0.9±0.4 96.2±49.4 0.8±0.4 4.3±0.4 1.14±0.1 190.0±55.9 10.2±17.29 0.72±0.43

54.5±9.1 73.6±41.8 47.1±27.1 1.6±0.6 118.4±44.1 1.3±0.9 3.6±0.6 1.38±0.2 139.7±80.0 12.6±16.4 0.9±0.4

56.3±8.6 91.1±57.0 54.8±42.5 1.8±0.9 167.3±100.7 2.4±1.5 3.3±0.7 1.44±0.2 128.9±81.4 838.4±3379 1.82±1.0

0.293 0.089 0.292 0.289 0.069 <0.0001 0.005 0.427 0.538 <0.0001 <0.0001

Variables were expressed as mean±SD HCV Ab hepatitis C virus antibody, HBsAg hepatitis B surface antigen *P value for F4 versus HCC (P>0.05 is considered non-significant and P<0.05 is considered significant) a

Reference values: aspartate aminotransferase (AST) (male up to 37 U/L, female up to 31 U/L), alanine aminotransferase (ALT) (male up to 41 U/ L, female up to 31 U/L), alkaline phosphatase (ALP) 22–92 U/L, total bilirubin (T.bilirubin) up to 1 mg/dL, albumin 3.8–5.4 g/dL, prothrombin time-international normalized ratio (INR) 1, platelet count (PLT) 150– 400×109 /L, alpha-fetoprotein (AFP) up to 10 U/L

Tumor Biol.

Fig. 1 Identification of both cytokeratin-1 (CK-1) and nuclear matrix protein-52 (NMP-52) using Western blot analysis after SDS-PAGE separation and electroelution. a Western blot analysis using respective antibody corresponding to CK-1 at 67 kDa and b Western blot analysis

using respective antibody corresponding to NMP-52 at 52 kDa. Lanes 1– 2: serum of healthy individuals, lanes 3–4: serum of patients with liver fibrosis, lanes 5–6: serum of patients with cirrhosis, and lanes 7–8: serum of patients with HCC

Identification and quantitation of CK-1 and NMP-52

F4. The median values for CK-1 and NMP-52 in patients who have HCC were higher (12.0 μg/mL for CK-1 and 11.0 μg/ mL for NMP-52) as compared to those who developed F4 (1.6 μg/mL for CK-1 and 6.0 μg/mL for NMP-52) with an extremely significant difference (P<0.0001). The results showed that the use of CK-1 per se gave an AUC equal to 0.83 while NMP-52 per se provided an AUC of 0.72 for separating patients with HCC from patients who developed F4. The next step was aimed to enhance the diagnostic accuracies of these biomarkers for diagnosing HCC. Conceptually, if both CK-1 and NMP-52 were allowed to be combined with other routine markers in a single predictive function, the overlap in their values among patients with F4 and HCC would be diminished and their diagnostic capabilities could be enhanced. As mentioned previously, only three out of ten evaluated variables included in the present study showed a significant association with HCC (P<0.05). They were total bilirubin, albumin, and AFP. The best overall formula that could best predict HCC was then constructed by entering these significant variables together with CK-1 and

Western blot analysis was used for identifying CK-1 and NMP-52 antigens in serum samples. The results showed that treatment of resolved antigens with CK-1 and NMP-52 antibodies, separately, resulted in visualization of a single immunoreactive band at 67 and 52 kDa, respectively, as depicted in Fig. 1. These bands were undetectable in serum samples from the healthy controls. Next, they were purified from sera using an electroelution technique and quantified using doseresponse curves of their serial concentrations. Distributions of CK-1 and NMP-52 As the identification of patients with HCC is of clinical importance, the distribution of CK-1 and NMP-52 levels in patients who developed HCC were compared with patients who have F4 and presented as box plots as showed in Fig. 2. In general, patients with HCC were associated with higher concentration (μg/mL) of CK-1 and NMP-52 than patients with Fig. 2 Distributions of cytokeratin-1 (CK-1) and nuclear matrix protein-52 (NMP-52) between patients who developed HCC versus those with liver fibrosis (F1–F3) or cirrhosis (F4). a CK-1 in F1–F3 versus HCC, b CK-1 in F4 versus HCC, c NMP52 in F1–F3 versus HCC, and d NMP-52 in F4 versus HCC

Tumor Biol.

NMP-52 into the stepwise linear regression analysis. As a result, only CK-1, NMP-52, and AFP retained significance when combined with each other and the following index was generated. HCC-DETECT = 0.004 × CK-1 (μg/mL) + 0.007×NMP-52 (μg/mL)+0.248×Log AFP (U/L)+0.951. Thus, it could be said that CK-1, NMP-52, and AFP were selected as the best combination for detecting HCC. Based on ROC analysis, the optimal cutoff points for the three candidate markers constituting HCC-DETECT were determined and their diagnostic performances for identifying HCC were measured and presented in Table 2. The AUCs for CK-1, NMP-52, and AFP for separating patients with HCC from patients who have F4 were 0.83, 0.72, and 0.78, respectively (Fig. 3). The results showed that AFP value at a cutoff level of 400 U/L presents a superior specificity of 100 % but the sensitivity dropped to 33 %. This cutoff value was chosen because this value is frequently reported to be specific for the diagnosis of HCC. Bivariate Spearman’s rank correlation coefficient was then calculated to measure the relationship between these candidate markers to the histological disease progression (Fig. 4a). There was no correlation between theses candidate markers (Fig. 4b). They are not related, which means that there is no redundancy.

Diagnostic performances for HCC-DETECT The areas under the ROC curve estimated the diagnostic value for HCC-DETECT derived from our data set for identifying HCC. It was found that HCC-DETECT yielded an AUC equal to 0.90 and 0.94 for separating patients with HCC from those who developed cirrhosis (F4) and liver fibrosis (F1–F3), respectively (Fig. 4c, d) which was evident to be more efficient than that produced by each marker separately. Finally, a cutoff Table 2 Diagnostic performances for HCC-DETECT and its candidate markers for identifying HCC F4 vs. HCC

Cutoff Sensitivity (%) Specificity (%) Accuracy (%) PPV (%) NPV (%)

CK-1 (μg/mL)

NMP-52 (μg/mL)

AFP (U/L)

HCCDETECT

F1–F3 vs. HCC HCCDETECT

>5.2 75 82 78 79 78

>8.5 62 70 66 65 67

≥400 33 100 67 100 62

>1.4 80 92 86 90 83

>1.4 80 93 90 80 93

CK-1 cytokeratin-1, NMP-52 nuclear matrix protein-52, AFP alpha-fetoprotein, HCC-DETECT: 0.004×cytokeratin-1 (μg/mL)+0.007×NMP52 (μg/mL)+0.248×Log AFP (U/L)+0.951, PPV positive predictive value, NPV negative predictive value

Fig. 3 Area under receiver-operating characteristic curve for different biomarkers constituting HCC-DETECT for separating patients with HCC from cirrhotic patients. a Cytokeratin-1 (CK-1), b nuclear matrix protein-52 (NMP-52), and c alpha-fetoprotein (AFP). An AUC of 1.0 is characteristic of an ideal test, whereas an AUC of 0.5 or less indicates a test of no diagnostic value

point of 1.4 was then chosen for detecting HCC and the calculated sensitivities, specificities, and accuracies in addition to positive and negative predictive values were clarified in Table 2.

Discussion Hepatocellular carcinoma is considered one of the most common cancers worldwide which usually develops in a liver already suffering from chronic damages, often cirrhosis [25]. For this reason, patients with cirrhosis are recommended to

Tumor Biol.

Fig. 4 Development of HCC-DETECT in the present study including 280 patients. a Spearman’s rank correlation coefficient of individual serum markers constituting HCC-DETECT in relation to the progression of liver pathology. b Spearman’s rank correlation coefficient of candidate markers in relation to each other. c area under curve for HCC-DETECT to discriminate patients with HCC from those

with liver cirrhosis (F4) and d area under curve for HCC-DETECT to discriminate patients with HCC from those with liver fibrosis (F1–F3). CK-1 cytokeratin-1, NMP-52 nuclear matrix protein-52, AFP alpha-fetoprotein, AUC area under the ROC curve, HCC-DETECT: 0.004 × cytokeratin-1 (μg/mL)+0.007×NMP-52 (μg/mL) +0.248×Log AFP (U/L)+0.951

undergo regular examinations for early detection of possible HCC [26]. However, the efficacy and cost-effectiveness of screening and surveillance of cirrhotics for the diagnosis of HCC is still debated [27]. No doubt that biomarkers which are detectable in blood can be useful for the clinical management of various disease states, especially various types of cancer. HCC is one such cancer that can benefit from tumor biomarkers [28]. Currently, the search for suitable biomarkers for the detection of HCC is very intensive. The use of AFP as a tumor marker for detecting HCC has been widely debated [7–9]. It was reported that AFP had a sensitivity and specificity of 39–65 and 76–94 %, respectively, for the presence of HCC in previously published studies [29]. It was concluded that AFP is not a useful diagnostic test as there are falsenegative results (10–50 %) depending on ethnic and geographical variations and the techniques employed [26]. There are two major questions during searching for biomarkers to detect HCC: the first, what is the efficacy of its using as a screening test for HCC to improve clinical outcomes in patients who developed CHC. The second, what is its ability to detect HCC in those who have HCC “sensitivity” and what is its ability to rule out HCC when HCC is not present “specificity.” Therefore, this work aimed to develop a sensitive simple diagnostic discriminant equation incorporating one of the CKs (CK-1) together with one of the NMPs (NMP-52) for the accurate detection of HCC. Cytokeratins are present in many malignant epithelial original cells and several subtypes of

them are expressed in HCC [30–33]. CK-1 has the highest molecular weight and the highest isoelectric pH [34, 35]. Although, CK-1 is easily measurable and requires only a microplate colorimetric reader, little information exists regarding its use in patients with HCC. In the present study, CK-1 was identified in serum samples using Western blot analysis at 67 kDa and its concentration (μg/mL) was also found to increase with the progression of liver disease being higher in patients who have HCC than those who developed F4 with an extremely significant difference (P<0.0001). Additionally, CK-1 enabled the correct identification of patients who have HCC with 0.83 AUC. On the other hand, the nuclear matrix is the non-chromatin scaffolding of the cell nucleus which confers the shape of the nucleus, organizes the nuclear chromatin, and regulates many important biochemical events that take place in the nucleus [36]. It was reported that that the changes in nuclear structure which is determined by the nuclear matrix are so prevalent in cancer cells. As well, alterations in nuclear shape that take place with neoplastic transformation are accompanied by changes in nuclear matrix architecture [13]. In this work, NMP-52 was identified in serum samples using Western blot analysis at 52 kDa. The values of NMP-52 concentration (μg/mL) were found to increase with liver disease progression being higher in patients who have HCC than those who developed F4 with an extremely significant difference (P<0.0001). As well, NMP-52 enabled the correct identification of HCC patients with 0.72 AUC. These findings

Tumor Biol.

suggest that these components “CK-1 and NMP-52” could be potential markers for identifying HCC in patients with chronic liver disease. Indeed, it was necessary for CK-1 and NMP-52 to be combined with other routine markers so that the overlap in their values among patients with F4 and HCC could be diminished and their diagnostic capabilities could be enhanced. Upon incorporating AFP to both of them, the diagnostic accuracy was enhanced giving 0.90 AUC with 80 % sensitivity and 92 % specificity. The effectiveness of HCCDETECT for discriminating HCC from F1–F3 was then performed using ROC curve that showed an AUC of 0.94. Thus, CK-1 and NMP-52 together with AFP were selected as the best combination for the early detection of HCC as previously described in the “Results” section. It is well known that persistently elevated AFP levels are related to the presence of HCC and that its determination can be helpful for a better definition of at-risk patients [37]; hence, AFP is the most widely tested biomarker in HCC. Our results showed that AFP value at a cutoff ≥400 U/L presents an absolute specificity. However, the sensitivity dropped to 33 %. The later result were almost similar to that obtained by Marrero et al. [38] who investigated the diagnostic performance of AFP in discriminating patients with HCC from cirrhotics providing a sensitivity of 34 % and a specificity of 100 %. This cutoff value was chosen because this value is frequently reported to be specific for the diagnosis of hepatocellular carcinoma. The sensitivity achieved by HCC-DETECT is comparable with that produced by ultrasonography (Sn=84 %) but higher than those yielded by computed tomography (79 %) and magnetic resonance imaging (77 %) [39]. Indeed, HCC-DETECT is much more simple and practical score than other scores such as the one developed by Ishida et al. [40] based on AFP, AST, lactate dehydrogenase, hemoglobin, prothrombin time, and male ratio. The sensitivity and specificity achieved by our established score was higher than those produced by Ishida et al. In summary, we showed that a simple diagnostic discriminant equation utilizing CK-1 and NMP-52 together with AFP may improve the detection of HCC with a high degree of accuracy. Further prospective multicenter studies involving a greater number of patients are warranted to validate the usefulness of the produced score in clinical practice.

3.

4. 5.

6.

7. 8.

9. 10.

11.

12. 13.

14.

15.

16. 17.

18.

19.

20.

21. 22. Conflicts of interest None. 23.

References 1. 2.

Bataller R, Brenner DA. Liver fibrosis. J Clin Invest. 2005;115: 209–18. Hernandez-Gea V, Friedman SL. Pathogenesis of liver fibrosis. Annu Rev Pathol. 2011;6:425–56.

24.

25. 26.

Llovet JM, Bruix J. Novel advancements in the management of hepatocellular carcinoma in 2008. J Hepatol. 2008;48 Suppl 1: S20–37. Rampone B, Schiavone B, Confuorto G. Current management of hepatocellular cancer. Curr Oncol Rep. 2010;12:186–92. Gonzalez SA, Keeffe EB. Diagnosis of hepatocellular carcinoma: role of tumor markers and liver biopsy. Clin Liver Dis. 2011;15: 297–306. Malaguarnera G, Giordano M, Paladina I, Berretta M, Cappellani A, Malaguarnera M. Serum markers of hepatocellular carcinoma. Dig Dis Sci. 2010;55:2744–55. Johnson PJ. Screening for hepatocellular carcinoma—answers to some simple questions. Am J Gastroenterol. 2002;97:225–6. Lin DY, Liaw YF. Optimal surveillance of hepatocellular carcinoma in patients with chronic viral hepatitis. J Gastroenterol Hepatol. 2001;16:715–7. Sherman M. Alphafetoprotein: an obituary. J Hepatol. 2001;34: 603–5. Bertino G, Ardiri A, Malaguarnera M, Malaguarnera G, Bertino N, Calvagno GS. Hepatocellualar carcinoma serum markers. Semin Oncol. 2012;39:410–33. Shimizu Y, Nakano I, Katano Y, Shimizu H, Fukuda Y. Clinical implications of measurement of serum nuclear matrix protein levels in patients with liver disease. Hepatogastroenterology. 2005;52: 1809–13. Berezney R, Coffey DS. Identification of a nuclear protein matrix. Biochem Biophys Res Commun. 1974;60:1410–7. Pienta KJ, Partin AW, Coffey DS. Cancer as a disease of DNA organization and dynamic cell structure. Cancer Res. 1989;49: 2525–32. Alix-Panabieres C, Vendrell JP, Slijper M, Pelle O, Barbotte E, Mercier G, et al. Full-length cytokeratin-19 is released by human tumor cells: a potential role in metastatic progression of breast cancer. Breast Cancer Res. 2009;11:R39. Moll R. Cytokeratins as markers of differentiation. Expression profiles in epithelia and epithelial tumors. Veroff Pathol. 1993;142:1– 197. Yilmaz Y. Cytokeratins in hepatitis. Clin Chim Acta. 2011;412: 2031–6. Yilmaz Y, Kedrah AE, Ozdogan O. Cytokeratin-18 fragments and biomarkers of the metabolic syndrome in nonalcoholic steatohepatitis. World J Gastroenterol. 2009;15:4387–91. Yilmaz Y. Systematic review: caspase-cleaved fragments of cytokeratin 18—the promises and challenges of a biomarker for chronic liver disease. Aliment Pharmacol Ther. 2009;30:1103–9. Attallah AM, El-Far M, Abdel Malak CA, Zahran F, Farid K, Omran MM, et al. Evaluation of cytokeratin-1 in the diagnosis of hepatocellular carcinoma. Clin Chim Acta. 2011;412:2310–5. Poynard T, Bedossa P, Opolon P. Natural history of liver fibrosis progression in patients with chronic hepatitis C. The OBSVIRC, METAVIR, CLINIVIR, and DOSVIRC groups. Lancet. 1997;349: 825–32. Bruix J, Sherman M. Management of hepatocellular carcinoma. Hepatology. 2005;42:1208–36. Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature. 1970;227:680–5. Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A. 1979;76:4350–4. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the Folin phenol reagent. J Biol Chem. 1951;193:265– 75. Masuzaki R, Karp SJ, Omata M. New serum markers of hepatocellular carcinoma. Semin Oncol. 2012;39:434–9. Cui R, He J, Zhang F, Wang B, Ding H, Shen H, et al. Diagnostic value of protein induced by vitamin K absence (PIVKAII) and

Tumor Biol.

27.

28. 29.

30.

31. 32. 33.

hepatoma-specific band of serum gamma-glutamyl transferase (GGTII) as hepatocellular carcinoma markers complementary to alpha-fetoprotein. Br J Cancer. 2003;88:1878–82. Gebo KA, Chander G, Jenckes MW, Ghanem KG, Herlong HF, Torbenson MS, et al. Screening tests for hepatocellular carcinoma in patients with chronic hepatitis C: a systematic review. Hepatology. 2002;36:S84–92. Behne T, Copur MS. Biomarkers for hepatocellular carcinoma. Int J Hepatol. 2012;2012:859076. Daniele B, Bencivenga A, Megna AS, Tinessa V. Alpha-fetoprotein and ultrasonography screening for hepatocellular carcinoma. Gastroenterology. 2004;127:S108–12. Apaydin R, Gurbuz Y, Bayramgurler D, Bilen N. Cytokeratin contents of basal cell carcinoma, epidermis overlying tumour, and associated stromal amyloidosis: an immunohistochemical study. Amyloid. 2005;12:41–7. Kim MA, Lee HS, Yang HK, Kim WH. Cytokeratin expression profile in gastric carcinomas. Hum Pathol. 2004;35:576–81. Chu PG, Weiss LM. Keratin expression in human tissues and neoplasms. Histopathology. 2002;40:403–39. D’Errico A, Baccarini P, Fiorentino M, Ceccarelli C, Bonazzi C, Ponzetto A, et al. Histogenesis of primary liver carcinomas: strengths and weaknesses of cytokeratin profile and albumin mRNA detection. Hum Pathol. 1996;27:599–604.

34.

Voorzanger-Rousselot N, Garnero P. Biochemical markers in oncology. Part I: molecular basis. Part II: clinical uses. Cancer Treat Rev. 2007;33:230–83. 35. Moll R, Franke WW, Schiller DL, Geiger B, Krepler R. The catalog of human cytokeratins: patterns of expression in normal epithelia, tumors and cultured cells. Cell. 1982;31:11–24. 36. Hughes JH, Cohen MB. Nuclear matrix proteins and their potential applications to diagnostic pathology. Am J Clin Pathol. 1999;111: 267–74. 37. Tsukuma H, Hiyama T, Tanaka S, Nakao M, Yabuuchi T, Kitamura T, et al. Risk factors for hepatocellular carcinoma among patients with chronic liver disease. N Engl J Med. 1993;328:1797–801. 38. Marrero JA, Su GL, Wei W, Emick D, Conjeevaram HS, Fontana RJ, et al. Des-gamma carboxyprothrombin can differentiate hepatocellular carcinoma from nonmalignant chronic liver disease in American patients. Hepatology. 2003;37:1114–21. 39. Llovet JM, Bruix J, Fuster J, Castells A, Garcia‐Valdecasas JC, Grande L, et al. Liver transplantation for small hepatocellular carcinoma: the tumor‐node‐metastasis classification does not have prognostic power. Hepatology. 1998;27:1572–7. 40. Ishida H, Matsuo S, Inoue Y. Evaluation of diagnostic performance of alpha-fetoprotein (AFP) and des-gamma-carboxy prothrombin (DCP) for HCV related hepatocellular carcinoma developed after long-term follow up. Rinsho Byori. 2010;58:1065–72.