Tissue expression of squamous cellular carcinoma antigen and Ki67 in hepatocellular carcinoma-correlation with prognosis: A historical prospective study

Background Squamous cellular carcinoma antigen (SCCA) is overexpressed in hepatocellular carcinoma (HCC) tissue and in sera of HCC patients. Our aim was to assess hepatic SCCA immunostaining in a series of HCCs and to correlate its presence with cell proliferation, apoptosis and clinical outcome. Methods Sixty-one HCC patients were included. Liver specimens were obtained either by biopsy (n = 17) or surgically (resection 27, transplantation 17). Immunostaining for AFP, Ki-67, SCCA and TUNEL assay were performed. Results SCCA staining was detected in 83.6% of specimens. A statistical significant correlation was found between negative SCCA staining and mortality (p = 0.026) and a higher immunostaining score for Ki67 (p = 0.017). Positive SCCA staining was associated with well and moderate differentiated tumors (p = 0.022). Using multiple logistic regression analysis, Ki67 and TUNEL assay were found to be significant independent predictors of negative SCCA immunostaining. The area under the receiver operator characteristic curve was 0.87. Kaplan-Meier survival analysis revealed a significant difference between the patient group with positive versus negative SCCA immunostaining relating to survival time (p = 0.0106). Cox proportional hazard regression analysis demonstrated that Ki67 immunostaining and liver transplantation or resection were independently associated with mortality. Conclusions SCCA is overexpressed in HCC. SCCA status is associated with cell proliferation, apoptosis and survival. SCCA and Ki67 staining can predict survival. Our study results support a potential association of negative SCCA expression with other markers of poor outcome in HCC. More studies are needed to clarify the role of SCCA in HCC and expand the knowledge of the SCCA antigen in HCC patients.


Background
Hepatocellular carcinoma (HCC) is a major health problem [1,2]. Its incidence is increasing [3] and it has become the leading cause of death amongst cirrhotic patients [4]. The predominant risk factors are chronic hepatitis B and chronic hepatitis C [5]. Llovet et al reported that 80% of HCC cases developed in a cirrhotic liver. Cirrhosis is the strongest predisposing factor [6].
A number of serum markers have been proposed as a method of detecting HCC. However, alpha fetoprotein (AFP), des-c-carboxy-prothrombin (DGCP) and AFP-L3 fraction are not sufficiently accurate to predict its early diagnosis [7].
Dysregulation of the balance between proliferation and cell death represents a pro-tumorigenic principle in human hepatocarcinogenesis. The Ki-67 protein is associated with active cell proliferation and expressed in all phases of the cell cycle, except G0, with the highest expression seen in G2/M.
In a study of patients who had undergone resection for HCC, higher levels of expression of Ki-67 in tumor tissue were found to be associated with a higher tumor grade [8] and early disease recurrence [9]. The balance between death and survival is dysregulated in HCC mainly due to overactivation of anti-apoptotic pathways such as Bcl-XL, Mcl-1, c-IAP1, XIAP or survivin, which are over-expressed in HCC cells [10].
Squamous cellular carcinoma antigen (SCCA) is a member of the high molecular weight family of serine protease inhibitors (serpins) [11]. High levels have been reported in cancer of the head and neck tissue and other epithelial cancers [12]. It has also been reported to overexpress in HCC tissue and in serum from HCC patients [13]. SCCA has been reported to overexpress in tumoral compared to peritumoral tissue, suggesting a role as a potential marker for histological detection of HCC [14].
The aim of this study was to assess SCCA immunostaining in a series of HCCs of different etiologies and characteristics and correlate its presence with cell proliferation, apoptosis and survival.

Patients' cohort
This study was simultaneously conducted in two large transplant medical centers in Israel: Beilinson Hospital, Rabin Medical Center, Petah Tiqwa (affiliated with the Sackler School of Medicine, Tel Aviv University, Tel Aviv) and Hadassah-Hebrew University Hospital, Jerusalem. A search was performed in the pathology departments of both hospitals for patients diagnosed with HCC. These patients were followed up at the Liver Institute, Beilinson Hospital or the Liver Unit, Hadassah-Hebrew University Hospital from January 2004 through December 2010, inclusive. Exclusion criteria were either insufficient liver tissue on the biopsy specimen for extra analysis (9) or insufficient clinical data regarding patient outcome (3). These last patients were non-Israeli citizens and either had returned home to their countries after obtaining a diagnosis or had been treated. The survival period was calculated from the point that tissue diagnosis was performed until either the patient's death or termination of follow up (April 2011) (whichever came first). Mortality data were retrieved from the computer system of the Ministry of Interior affairs via the hospital computers.
Serum AFP was measured in all cases. HCC diagnosis was confirmed by the presence of a focal lesion > 2 cm detected by liver ultrasound (US), computed tomography and magnetic resonance imaging, when indicated. Liver specimens, formalin-fixed and paraffinembedded, used to diagnose HCC were obtained from all patients either by US-guided biopsies or surgically in patients who had undergone resection or transplantation.

Serum AFP Level
Immunoassay was used to analyze serum sample AFP levels (Roche Diagnostics) using the Modular Analytics E170.

Histological Evaluation
The SCCA was analyzed by immunohistochemistry in paraffin sections. A score was obtained by assessing entire sections in each case and each nodule in cases of multiple malignant nodules. The highest score was used. For SCCA detection, a novel polyclonal rabbit antibody (Hepa-Ab, Xeptagen, Italy), raised against recombinant SCCA1 and affinity purified on a Sepharose-SCCA1 column, was used.
Epitope mapping studies with SCCA1 fragments obtained by SCCA1 enzymatic digestion or chemical synthesis indicated that the affinity-purified polyclonal antibody recognized several epitopes located in the Nterminal, C-terminal and the central portion of SCCA1 (determined by an enzyme-linked immunosorbent assay (ELISA) and Western blot). The concentration level of Hepa-Ab used was 4 mg/mL. Sections were incubated with primary antibodies for 30 minutes after blocking endogenous peroxidase activity with 3% hydrogen peroxide. The slides were heated in 10 mM sodium citrate in a microwave oven, blocking nonspecific protein binding in normal goat serum. Biotinylated goat anti-rabbit (Dako, Copenhagen, Denmark) was then added for 30 minutes.
Samples were incubated with avidin-biotin-peroxidase and stained with a mixture of 3,30-diamino-benzidine tetrahydrochloride (Dako) and hydrogen peroxide. As a negative control, sections were incubated, omitting a primary antibody. In each case, a diluent or appropriate nonimmune IgG was substituted. Antibody specificity was confirmed using human skin specimens for SCCA as positive controls.
The percentage of stained cells in each specimen was scored on a scale of 0-3: 0 denoted negative staining; 1positivity in 1-30% of hepatocytes; 2-positivity in 31-50%; and 3-in > 50%. Distribution of immunoreactivity was noted and classified as diffuse patterns, clustered or scattered cells.
In all cases, SCCA semiquantitative immunoreactivity was independently evaluated by two pathologist experts in the field. Intra and interobserver differences were < 5%. Discordant cases were simultaneously re-evaluated by the two.
In all immunohistochemical analyses, necrotic areas and edges of tissue sections were not included in the counting as to avoid possible false positivity.

AFP Immunohistochemistry Staining
Immunohistochemistry on paraffin embedded sections with alpha feto protein antibody was performed using a fully automated system (Ventana Benchmark, United States). Sections were pretreated with protease enzymes and incubated with a Rabbit Anti-Human Alpha-1-Fetoprotein antibody (A008 DAKO DK-2600 Glostrup Denmark) in a concentration of 1:300 for 40 minutes.
Only distinct nuclear staining of carcinoma cells was used for scoring via the light microscope, determined semiquantitatively as nil (no immunostaining), low (10% or less immunopositivity) or high (> 10% immunoreactive cells), respectively. Assessment was carried out on the entire tumor represented in the section.

Terminal Deoxynucleotidyl Transferase dUTP Nick End Labeling (TUNEL) Assay
Apoptotic cell nuclei were identified by applying an ApoTag in situ apoptosis peroxidase detection kit using the TUNEL assay (Intergen Co., New York, NY). This assay is designed to specifically detect the fragmented DNA of apoptotic cells by catalytically incorporating fluorescein-12-2'-deoxyuridine 5'-triphosphate at the 3'-OH DNA ends using the terminal deoxynucleotidyl transferase enzyme, thus forming a polymeric tail. The fluorescein-12-2'-deoxyuridine 5'-triphosphate-labeled DNA can subsequently be directly visualized by fluorescence microscopy. Hepatocytes with nuclear positivity for DNA fragmentation were counted in 50 ± 5 fields (original magnification ×10).
All practices conformed to institutional guidelines for the protection of human subjects.

Statistical Analysis
Descriptive statistics are given as median, mean and standard deviation (SD) for continuous variables and frequency distribution for categorical variables. Logarithmic transformation was applied to the serum AFP variable due to skewed distribution of the data. Chi-square or Fisher's exact test were used for categorical variables; the two-sample t-test for continuous variables. A multiple logistic regression model was applied to assess parameters independently associated with SCCA. A receiver operator characteristic (ROC) curve was drawn and the area under the curve (C parameter) was used to quantify the predictive ability of the final regression model. The Kaplan-Meier method was used to evaluate survival rate. Differences in survival between groups were analyzed by the Log-rank test. The Cox proportional hazards regression model was also applied. Results are presented as hazard ratios (HR), together with their 95% confidence intervals. Statistical analyses were performed using SAS for Windows 9.2.

SCCA Immunostaining
SCCA expression in HCC is shown in Figure 1. SCCA was detected in 51 (83.6%) specimens; 17 (27.87%) from transplant specimens; 27 (44.26%) from resection and 17 (27.87%) from fine-needle ultrasound guided biopsies; undetected in 6 surgically obtained normal human livers. A positive signal was clearly detectable in 50% of the cases, with a scored of 1, 19.7% with a score of 2 and 20.3% with a score of 3, Table 1.
SCCA was detected in the cell cytoplasm with a prevalent diffuse pattern. In poorly differentiated tumors, a typical clustered coarse pattern was observed (Figure 1). A statistically significant correlation was found between SCCA staining and: 1) mortality: negative SCCA staining was associated with 100% mortality vs a 64.71% in positive SCCA staining (p = 0.026); 2) tumor differentiation: positive SCCA staining was associated with well and moderate differentiated tumors in 95.85% of specimens (p = 0.022); 3) Ki67 immunostaining: negative SCCA staining was associated with a higher Ki67 score (p = 0.017). However, no such correlation was detected in age, sex, etiology of chronic liver disease, tumor diameter and number, serum AFP, AFP and TUNEL staining (borderline), type or level of SCCA staining ( Table 2).

AFP Immunostaining
AFP staining was negative in 70.7% of the specimens. No correlation was detected between AFP staining and SCCA staining or any clinical parameter.

Independent Predictors of Negative SCCA Immunostaining
Ki67 and the TUNEL staining were significant independent predictors of negative SCCA staining. Table 3 represents the multiple logistic regression analysis for these variables.
ROC curve analysis ( Figure 2) revealed that Ki67 and TUNEL staining were excellent predictors for SCCA staining. The C parameter was 0.87.

Kaplan-Meier Survival Analysis
A significant difference was noted between the two patient groups (positive and negative SCCA staining) in survival time (p = 0.0106) (Figure 3). The mean survival time for negative SCCA staining patients was 19.13 months vs 39.78 months for positive staining patients.

Variables Independently Associated with Mortality
The following variables were found to be independently associated with mortality: Ki67 staining, HBV, liver transplantation and resection. Table 4 presents Cox proportional hazard regression analysis for each parameter.

Discussion
Several studies have reported a high SCCA expression in HCC tissues [13,14]. Giannelli et al [13] measured serum SCCA in 120 HCC patients, 90 cirrhotics, and 41 healthy subjects. SCCA levels were significantly elevated in HCC patients compared to cirrhotic or normal subjects. The sensitivity and specificity for SCCA in HCC diagnosis were 84% and 46%, respectively. Guido et al [15] demonstrated aberrant expression of SCCA in HCC and provided evidence that SCCA overexpression is an early event in liver cell carcinomatous transformation was consistently detected in all considered dysplastic nodules.
The high SCCA expression in HCC tissue seems remarkable since the liver does not possess squamous epithelial cells, although hepatocytes share a common embryogenic origin. Furthermore, the role of serpins in neoplastic cells indicates that SCCA expression makes cancer cells resistant to several killing mechanisms by inhibition of apoptosis [16]. The SCCA antigen represents a marker for more advanced organ tumors such as the cervix, lung and oropharynx [17]. A significant association was also noted between SCCA overexpression and histological grade 2/3 HCC, suggesting SCCA as a marker of cancer aggressiveness [15].
However, in our HCC patient cohort, positive SCCA staining was associated with an improved survival rate. We found that negative SCCA staining was associated with a higher Ki67 score. In multiple logistic regression analysis, the Ki67 and the TUNEL scores were significant independent predictors of negative SCCA staining. A previous study established that Ki-67 immunostaining of HCC lesions was associated with higher mitotic activity [18]. D'Errico et al [9] demonstrated that higher levels of Ki-67 expression in HCC tissue were associated with a higher tumor grade. Nakanishi et al [8] noted that higher levels of Ki-67 expression were associated with early disease recurrence. Insufficient apoptosis has been associated with the development and progression of liver tumors. [10] In HCC, the balance between death and survival is mainly disrupted due to the overactivation of antiapoptotic signals. Therefore, liver cancer cells might express stronger requirements of these intracellular pathways to survive.
Our study results are in accordance with Trerotoli et al [19] who found that stronger staining was associated with a smaller HCC. They speculated that the higher ratio in smaller rather than larger hepatic nodules, suggests that SCCA is produced and released at different times, likely during the earlier events in HCC progression.      There are several limitations to our study. This is a historical study. Due to the retrospective nature of our study, variable treatment modalities of HCC were included. The patients' cohort was composed of only 61 patients. The group size became even smaller when divided by the SCCA positive and negative staining. Nonetheless, the multiple logistic regression model was applied to assess parameters independently associated with SCCA. The ROC curve was drawn and the C parameter used to quantify the predictive ability of the final regression model.
Up to 40% of all HCC's do not produce AFP [5]. In our study, 70.7% of the specimens did not stain for AFP. No significant correlation was detected between the serum level of AFP, AFP staining and SCCA staining.
Our study results support an invasive diagnostic approach for a suspected malignant hepatic lesion since the SCCA expression in the lesion might provide important information regarding patient prognosis. We feel that if only imaging studies are performed in cirrhotic patients with a suspicious hepatic lesion, valuable prognostic information might be missed.

Conclusions
SCCA is overexpressed in HCC. SCCA expression is associated with tumor differentiation, cell proliferation and apoptosis. Our study results confirm a potential association of negative SCCA expression with other markers of poor outcome in HCC.
More studies are needed to clarify the role of this molecule and to further expand knowledge of the SCCA antigen in HCC patients.