Heterophile carbohydrate antigen N‐glycolylneuraminic acid as a potential biomarker in patients with hepatocellular carcinoma

Abstract Background and Objectives Hepatocellular carcinoma (HCC) has a high recurrence rate even after radical hepatectomy. More optimal biomarkers may help improve recurrence and prognosis. Methods We investigated whether the oncological properties of N‐glycolylneuraminic acid (NeuGc) can participate in the prognosis of HCC. We evaluated the NeuGc antigen (Ag) expression in the HCC tissues and measured the preoperative anti‐NeuGc IgG antibodies (Abs) in the sera of the patients with HCC. We compared the clinical characteristics and survival rate in the hepatectomized patients (initial; n = 66, recurrent; n = 34) with and without the NeuGc Ag or Abs. Results Multivariate analyses showed positive expression of NeuGc Ag in HCC tissues (Odds ratio; initial = 6.3, recurrent = 14.0) and higher titers of preoperative anti‐NeuGc Ab (Odds ratio; initial = 4.9; recurrent = 3.8), which could be the predictive factors related to early recurrence. Both the NeuGc Ag‐positive and Ab‐positive groups in the initial hepatectomized patients exhibited significantly shorter recurrent free survival compared to those in the negative groups. Conclusions Our findings suggested that anti‐NeuGc Ab titers and NeuGc Ag expression in the HCC tissues can be used as the predictive factors for the postoperative recurrence and prognosis of HCC.

Therefore, the natural anti-NeuGc immunoglobulin (Ig) G antibodies (Abs) were reportedly detectable in 85% of healthy humans. 4 Furthermore, the sugar chains expressed on the cell surface do not only qualitatively and quantitatively change during tumor development but also play an important role in cancer cell infiltration and metastasis.
Recently, NeuGc antigen (Ag) was discovered to be expressed in lung, breast, colon cancer, melanoma, retinoblastoma, and hepatocellular carcinoma (HCC) patients. [5][6][7][8] One study found that non-small cell lung cancer patients with NeuGc Ag expression have a significantly low progression-free survival rate. 5 There is a similar paper reporting the possibility of NeuGc containing glycans as a diagnostic biomarker for patients with ovarian cancer. 9 However, no studies have analyzed the oncological effects of NeuGc Ag expression and anti-NeuGc Abs to predict HCC prognosis. NeuGc has been reported to be useful as a prognostic and diagnostic biomarker in other carcinomas, and may also be useful in hepatocellular carcinoma. Thus, we aimed to determine whether the NeuGc Ag expression in hepatectomized tissues and anti-NeuGc Ab titers in the serum may affect HCC prognosis separately for initial and recurrent hepatectomized patients. Our findings may provide new insights to the current research on predicting the recurrence and prognosis in HCC patients. Prediction of the risk of recurrence of HCC may lead to improvement of prognosis of HCC if adjuvant treatment such as molecular target therapy can be given to the high-risk group for recurrence.

| Study patients
Between April 2014 and December 2015, 100 patients (of these 100 cases, the initial (n = 66) and recurrent (n = 34) cases were not overlapped) were hepatectomized for HCC at Hiroshima University Hospital. As a control group, 50 healthy living liver-transplant donors were preoperatively diagnosed as tumor-free by computed tomography. This study was performed following the Declaration of Helsinki and approved by the Institutional Review Board of Hiroshima University (approval number: E 1580). Informed consent was obtained from all participants.

| Blood and liver tissue samples
Blood samples were preoperatively collected from HCC patients and healthy living liver transplant donors. The resected liver tissues were macroscopically divided into normal and tumor tissues, and then were cryopreserved.

| Immunohistochemistry assay
Chicken anti-NeuGc monoclonal IgG Abs (HU/Ch2-7; Pharma Foods International Co., Ltd., Kyoto, Japan) were used for the detection of NeuGc Ag in the hepatectomized liver tissues. 10,11 HU/Ch2-7, which reacted strongly with NeuGc Ag, was specifically directed toward an antigenic epitope containing NeuGc. 11 The sections were incubated with HU/Ch2-7 in phosphatebuffered saline (PBS) overnight. Biotin-conjugated goat anti-chicken IgY (H&L) polyclonal Ab (Abcam, Cambridge, USA) in PBS was added and incubated for 30 min. The sections were incubated with horseradish peroxidase-conjugated streptavidin (Histofine SAB-PO Kit; Nichirei Biosciences, Tokyo, Japan) for 30 min. Peroxidase activity was analyzed by staining the sections with 3,3 0 -diaminobenzidine (Muto Pure Chemicals, Tokyo, Japan) for 10 min. 12 The sections were examined and photographed at 400X magnification using a BZ-9000 microscope (Keyence, Osaka, Japan). The quantification of fluorescence intensity was performed using the Hybrid cell count BZ-H2C software (Keyence). Based on the percentage of immunoreactive cells within the microscopic field of view, NeuGc Ag levels were classified into two groups: negative (0%) and positive (>0%) group. Four of the 66 initial hepatectomy samples were excluded because the quantification of the immunostaining Ag was impossible owing to poor storage conditions.

| Reverse transcription-quantitative polymerase chain reaction (RT-qPCR)
The sections of the hepatectomized liver tissues were immediately stored in RNAlater ® Tissue Protect Tubes (Qiagen, Hilden, Germany).
After crushing the tissue with TissueLyser II (Qiagen), total RNA was extracted using the single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform (Qiagen). Reverse transcription of total RNA was performed to generate the first-strand cDNA. To determine the mRNA expression levels of human CMAH (CMAHP), the forward (5 0 CCAGTCAGGAAGTC-3 0 ) (the lesion upper 92-bp deletion) and reverse (5 0 -GGTTGGAGGACCAG-3 0 ) (the lesion lower 92-bp deletion) primers were designed according to the procedures for identifying the gene. 1,3 RT-qPCR was performed with the following amplification profile: preincubation and initial activation at 94 C for 15 min, followed by 30 cycles of denaturation at 94 C for 1 min, primer annealing at 54 C for 1 min, and elongation at 72 C for 2 min. With β actin expression as reference, the average ΔCt value of the whole group was determined. The relative quantification of mRNA expression was performed using the 2 -ΔΔ Ct method. 13

| Enzyme-linked immunosorbent assay (ELISA)
The anti-NeuGc Abs in the sera of patients with HCC and healthy volunteers were measured by ELISA, according to previously reported methods. 14,15 ELISA microplates (Costar 3591; Corning, New York, NY, USA) were coated with NeuGc-and NeuAc-conjugated polyacrylamide (GlycoTech, Gaithersburg, MD, USA) overnight. We added peroxidase-conjugated goat anti-human IgG (Southern Biotech, Birmingham, AL, USA) or peroxidase-conjugated mouse anti-human IgG subclasses (Southern Biotech). The optical density was measured at 490 nm using the MTP 310 microplate reader (Corona Electric Co., Ltd., Ibaraki, Japan). We decided to analyze x100 and x500 dilution data based on the preliminary data (n = 16, Figure S1) in which the serum was diluted in 4 steps.

| Statistical analysis
All statistical analyses were performed using the JMP ® 14 software (SAS Institute Inc., Cary, NC, USA). The receiver operator characteristic (ROC) curve analysis was performed to determine the optimal cutoff values of anti-NeuGc IgG Abs in the blood sera of the HCC patients and healthy volunteers and the area under the curve value was calculated. A multivariable logistic regression analysis was performed and the odds ratio (OR) and corresponding 95% confidence interval were calculated. Statistical analysis of the categorical data was performed using the Student's t-test and Kruskal-Wallis test for continuous variables and the Pearson's chi-squared test for categorical variables. Kaplan-Meier plots and log-rank analysis were performed to determine the difference in overall survival (OS) and recurrent free survival (RFS) rate between the patient groups, with Pvalues <.05 considered as statistically significant. CMAHP, which encodes a key enzyme in the synthesis of the sialic acids Neu5Ac and Neu5Gc in other mammals, was rendered a pseudogene in humans through an inactivating microdeletion and the subsequent fixation of the inactive allele in early human populations. 16 The NeuGc expression in HCC tissues is likely due to the aberrant functionalization of CMAHP, leading to NeuGc biosynthesis.
To address this possibility, we evaluated CMAHP mRNA expression normalized against β-actin in the normal and tumor portions of the resected liver tissues from the hepatectomized HCC patients (n = 100). RT-qPCR results revealed that the tumor tissues exhibited significantly higher CMAHP mRNA expression than the normal tissues (P < .01; Figure 1B).

| Serum titers of anti-NeuGc IgG Ab tended to be higher in the HCC patients than in the healthy volunteers
The preoperative anti-NeuGc IgG Ab titers in the sera of the HCC patients (n = 100) or healthy donors (n = 50) were measured by ELISA. The average of anti-NeuGc IgG Ab titers tended to be higher in the sera of the HCC patients than in the healthy volunteers; however, the difference was not statistically significant (P = 0.089; Figure 2A). The titers of anti-NeuGc IgG Ab subclasses were also measured to investigate the probable difference in the opsonic capacity of anti-NeuGc IgG in the HCC patients and healthy volunteers. The results revealed that the level of anti-NeuGc IgG1, which has high opsonizing activity (Ab-dependent cellular cytotoxicity (ADCC) or complement-dependent cytotoxicity (CDC)), was significantly higher in the sera of the HCC patients than in the healthy volunteers (P < .01; Figure 2B). By contrast, no significant differences were observed in other IgG subclasses. Using ROC analysis, the optimal cutoff value of anti-NeuGc IgG Ab between the HCC patients and healthy volunteers was calculated to be 0.04. Thus, an absorbance value of 0.04 or higher was defined as anti-NeuGc Ab positive  F I G U R E 2 Serum titer of anti-NeuGc IgG Ab in the HCC patients and healthy volunteers. (A) The average of anti-NeuGc immunoglobulin (Ig) G antibody (Ab) titers tended to be higher in the sera of the HCC patients than in the healthy volunteers but no significant difference was observed between both groups (P = 0.089). (B) The titers of preoperative anti-NeuGc IgG Ab subclasses in the HCC patients and healthy volunteers. The level of anti-NeuGc IgG1 was significantly higher in the sera of the HCC patients than in the healthy volunteers (P < .01). (C) The optimal cutoff value of anti-NeuGc IgG Ab between the HCC patients and healthy volunteers was calculated to be 0.04 by the receiver operator characteristic curve analysis.
The clinical characteristics of the initial patient group are shown in Table 1. No significant differences were observed in the underlying diseases, tumor diameter, number of tumors, tumor marker levels of alpha-fetoprotein (AFP), pathological vascular invasion, and the degree of differentiation between NeuGc Ag-positive and -negative groups. However, protein induced by vitamin K absence-II (PIVKAII) level, NeuGc Ab titer, and the recurrence rate within 2 years after hepatectomy in the NeuGc Ag-positive group were significantly higher than those in the negative group (P < .05, <.01, and <.05, respectively).

| NeuGc expression in the HCC tissues was closely related to serum anti-NeuGc IgG titers in the initial patient group
We compared the titers of preoperative anti NeuGc IgG Ab between the patients with (n = 28) and without (n = 38) HCC recurrence within 2 years after hepatectomy for initial HCC. The titers of anti-NeuGc Ab were significantly higher in the patients with recurrence than in those without recurrence (P < .01; Figure 3A). We also compared the titers of anti NeuGc IgG Ab subclasses and found no significant difference between the patients with and without HCC recurrence ( Figure 3B). Next, we investi- CMAHP mRNA expression was significantly higher in the tumor tissues of the NeuGc Ag-positive group than in the normal tissues (P < .05; Figure 3D). Consistently, the tumor tissues had significantly higher levels of CMAHP mRNA than the normal tissues in the anti-NeuGc Ab-positive group (P < .01; Figure 3E), whereas no significant difference was observed between the normal and tumor tissues in the NeuGc Ag and Ab-negative group.
3.5 | Among patients hepatectomized for recurrent HCC, those with positive NeuGc Ag expression and higher preoperative anti-NeuGc IgG Ab titers had a higher recurrence rate after the surgery As NeuGc Ag and Ab were shown to be involved in the earlier recurrence of HCC, we investigated whether NeuGc Ag or Ab is a factor that contributes to repeated recurrence only in cases that were hepa-  Table 4-6. PIVKAII level and the recurrence rate within 2 years after hepatectomy were significantly higher in the NeuGc Ag-positive group than in the negative group (P < .05) ( Table 4). The anti-NeuGc Ab positive group exhibited a significantly higher AFP-L3 level and earlier recurrence rate after hepatectomy (P < .05) (  Comparison of the preoperative titers of anti-NeuGc IgG Ab in the patients with (n = 21) and without (n = 13) HCC recurrence within 2 years after hepatectomy revealed that the anti-NeuGc IgG Ab titers were significantly higher in the patients with HCC recurrence (P < .01; Figure 4A). We further compared the titers of anti NeuGc IgG Ab subclasses between the patients with and without HCC recurrence ( Figure 4B). The IgG2 titer, which has a low opsonizing capacity, was significantly higher in the patients with early recurrence than in the patients without early recurrence (P < .05  (1)  results for the initial HCC patient group, the NeuGc Ag-positive group had significantly increased titers of anti-NeuGc IgG Ab (P < .05; Figure 4C).
We compared the relative CMAHP mRNA expression in the recurrent HCC patient group. The NeuGc Ag-positive group exhibited a significantly higher CMAHP mRNA expression in the tumor tissues than that in the normal tissues (P < .05) ( Figure 4D). The CMAHP mRNA expression in the anti-NeuGc Ab-positive group was also significantly higher in the tumor tissues than in the normal tissues ( Figure 4E).
3.6 | Positive NeuGc Ag expression and the higher titers of preoperative anti-NeuGc Ab were associated with significantly reduced RFS of the patients with initial hepatectomy We compared the OS rate in the initial patient group between the NeuGc Ag-positive and -negative groups. The positive group exhibited shorter OS (not significant, Figure 5A) and RFS (P < .05; Figure 5C) than the negative group. Next, we observed that the anti-NeuGc Abpositive group exhibited a significantly reduced OS (P < .01; Figure 5B) and RFS (P < .01; Figure 5D) compared with the negative group.
We also compared the OS of the recurrent patient group between the NeuGc Ag-positive and -negative groups. The results showed that the OS and RFS rate did not show a significant difference. However, the positive group tended to have shorter OS and RFS than the negative group ( Figure 5E and G). The OS and RFS of patients hepatectomized for recurrent HCC were not significantly different between the anti-NeuGc IgG Ab-positive and -negative groups, whereas the positive group tended to exhibit a reduced OS and RFS rate ( Figure 5F and H).

| DISCUSSION
Despite optimal treatment, HCC has a high recurrence rate even after curative surgery. It recurs in 50%-80% of patients following resection, with a majority of recurrences developing within 2 years. 17 Previous studies have shown that the early recurrence of HCC after resection, which is associated with increased tumor size, number of tumors, and portal vein invasion, results in low OS. [18][19][20][21] Several studies have reported that serological biomarkers are useful in predicting HCC recurrence risk. However, the serum biomarkers for HCC exhibit inadequate sensitivity for predicting recurrence. 22 Sialic acids play fundamental roles in cell-cell and cellmicroenvironment interactions such as immune responses. 23 The most common sialic acids in mammals, NeuAc and NeuGc, have a structural difference of a single oxygen atom at the C5 position. 24 In general, normal human cells do not express NeuGc because of the inactivation of functional CMAH caused by a mutation in CMAH. 25 Figure 5E) and RFS ( Figure 5G) rates in the recurrent HCC patient group, in both the positive and negative NeuGc Ag expression groups. The positive group exhibited shorter OS than the negative group without any significant difference. The OS ( Figure 5F) and RFS ( Figure 5H) rate in the recurrent HCC patient group between anti-NeuGc Ab-positive and -negative groups. No significant difference was observed between both groups.
concluded that the reduced consumption of NeuGc-containing food may have a major effect in preventing the development of malignant diseases. 28 Further studies on the biology of CMAHP and NeuGc Agexpression in HCC tissues and circulating anti-NeuGc Abs are required to completely elucidate the mechanism of HCC carcinogenesis.
Regardless of the mechanism, the intra-tumoral expression of NeuGc These results indicated that the serum anti-NeuGc Ab can be used as a risk factor for predicting HCC recurrence at an early stage after hepatectomy; however, its potential benefit as a tumor marker monitored regularly after surgery for speculating HCC recurrence remains entirely unknown. We additionally found that NeuGc Ag expression in the HCC tissues and NeuGc Ab titers in the perioperative sera of the HCC patients were correlated ( Figure 3C and 4C weak affinity to all FcγRs. 29 As shown in Figure 2B, the healthy volunteers exhibited anti-NeuGc IgG predominantly in IgG2, whereas the HCC patients exhibited a significantly increased IgG1 level. In the patients with recurrent HCC, the IgG2 subclass was significantly promoted in subsequent recurrent cases than in the cases without another recurrence ( Figure 4B). Thus, anti-NeuGc Abs switch to a subclass with weak opsonizing effects on HCC recurrence is. Further studies are needed to clarify how HCC recurrence is associated with class switching of Abs that respond to cancer glycan antigens.
In this study, we investigated the relationship between the NeuGc Ag on the resected specimen and the NeuGc Ab titer at the time of resection. Therefore, measuring the serum NeuGc Ab may predict early recurrence and prognosis even in non-hepatic resection cases. On the other hand, since we have not investigated temporal kinetics in NeuGc Ab titers after hepatic resection, it may not be a biomarker in the true sense.

| CONCLUSIONS
Our findings suggested that preoperative anti-NeuGc Ab titers and positive NeuGc Ag expression in resected HCC tissues can predict the early recurrence and prognosis of HCC after curative hepatectomy.
Therefore, these factors may serve as potential predictive factors for future studies on improving HCC prognosis and therapeutics.

CONFLICT OF INTEREST STATEMENT
The authors declare that they have no competing interests.

DATA AVAILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were created or analyzed in this study.

ETHICS STATEMENT
The protocol for this research project has been approved by a suitably