Prognostic impact of p53 and/or NY‐ESO‐1 autoantibody induction in patients with gastroenterological cancers

Abstract Background and Aim We evaluated the clinicopathological and prognostic significance of serum p53 (s‐p53‐Abs) and serum NY‐ESO‐1 autoantibodies (s‐NY‐ESO‐1‐Abs) in esophageal squamous cell carcinoma (ESCC), gastric cancer and hepatocellular carcinoma (HCC). Patients and Methods A total of 377 patients, 85 patients with ESCC, 248 patients with gastric cancer, and 44 patients with HCC were enrolled to measure s‐p53‐Abs and s‐NY‐ESO‐1‐Abs titers by the enzyme‐linked immunosorbent assay before treatment. The clinicopathological significance and prognostic impact of the presence of autoantibodies were evaluated. Expression data based on the Cancer Genome Atlas and the prognostic impact of gene expression was also examined for discussion. Results The positive rates of s‐p53‐Abs were 32.9% in ESCC, 15% in gastric cancer, and 4.5% in HCC. The positive rates of s‐NY‐ESO‐1‐Abs were 29.4% in ESCC, 9.7% in gastric cancer, and 13.6% in HCC. The presence of s‐p53‐Abs was not associated with tumor progression in these three cancer types. On the other hand, the presence of s‐NY‐ESO‐1‐Abs was significantly associated with tumor progression in ESCC and gastric cancer. The presence of s‐p53‐Abs and/or s‐NY‐ESO‐1‐Abs was significantly associated with poor prognosis in gastric cancer but not in ESCC nor HCC. Conclusions The presence of s‐p53‐Abs and/or s‐NY‐ESO‐1‐Abs was associated with tumor progression in ESCC and gastric cancer. These autoantibodies might have poor prognostic impacts on gastric cancer (UMIN000014530).


| INTRODUC TI ON
The immunoglobulin G (IgG) autoantibodies against tumor antigens are known to appear in the serum of patients with cancer 1 even in the early stages of tumor development. Therefore, such autoantibodies have potential as tumor markers for early detection. [1][2][3] We have previously screened autoantibodies using the serological identification of antigens by recombinant expression cloning method and reported on their usefulness. 2,4 Among various autoantibodies, serum p53 antibodies (s-p53-Abs) and serum NY-ESO-1 antibodies (s-NY-ESO-1-Abs) are reported to show high positive rate in various cancer types. 3,5 The P53 gene is mutated in the majority of solid cancers, leading to the common expression of mutant p53 gene and protein in such diseases. s-p53-Abs appear in the cancer patients with this mutant p53 protein. In our previous studies, we also reported that s-p53-Abs are present in many cancer types. 3,6 In recent years, a high positive rate of s-NY-ESO-1-Abs was reported not only in esophageal cancer, but also in other cancer types. 5,7,8 Although several reports showed clinicopathological significance of these two serum autoantibodies, prognostic impact has not been reported in the same patient group among gastroenterological cancers.
Therefore, we focused on the presence of s-p53-Abs and/or s-NY-ESO-1-Abs to examine the positive rate, clinicopathological significance, and prognostic impact on ESCC, gastric cancer, and HCC. Moreover, expression and prognostic impact of p53 and NY-ESO-1 gene expression were also discussed.

| Detection of serum antibodies by enzymelinked immunosorbent assay
Serum samples from patients and healthy controls were analyzed via the enzyme-linked immunosorbent assay, as previously described. 2 The signals of s-p53-Abs and s-NY-ESO-1-Abs were evaluated by calculating the differences in absorbance between the wells containing antibodies and the wells containing phosphate-

| Gene expression analysis and survival analysis of p53 and NY-ESO-1
Aside from antibody expression, UALCAN (an interactive web portal for detailed analysis of TCGA gene expression data; available online: http://ualcan.path.uab.edu/) was used. 9 The same web portal was used to analyze survival based on p53 and NY-ESO-1 gene expression level. Survival analysis with P < .05 was considered statistically significant. High-expression patients show expression value >3rd quartile.

| Statistical analysis
All analyses were performed using SPSS version 17.0 (SPSS,

| Positive rate of P53 and NY-ESO-1 Abs in each carcinoma type
The positive rates of s-p53-Abs of each cancer were 32.9% in ESCC, 15% in gastric cancer, and 4.5% in HCC, respectively ( Figure 1A)

| Clinicopathological features and autoantibody status
In ESCC, the presence of s-p53-Abs was not associated with clinicopathological factors ( Table 2). In gastric cancer, the presence of s-p53-Abs was significantly associated with men, lymph node metastases, and carcinoembryonic antigen. The presence of s-NY-ESO-1-Abs was also significantly associated with more advanced tumor invasion, distant metastasis, and advanced cases (Table 3). However, in HCC, s-p53-Abs and s-NY-ESO-1-Abs were not significantly associated with clinicopathological factors (Table 4).

| Prognostic impact of autoantibodies
Overall survival rates for each cancer type were compared between the autoantibody-positive and -negative groups ( Figure 3).
In ESCC, there was no significant difference between s-p53-Abs negative and positive groups ( Figure 3A). Similarly, although s-NY-ESO-1-Abs negative group showed slightly better prognosis, the difference was not statistically significance ( Figure 3B). In gastric cancer, the s-p53-Abs negative group showed significantly better prognosis than did the positive group (

| D ISCUSS I ON
In this study, we evaluated the positive rate, clinicopathological significance, and prognostic impact of s-p53-Abs and s-NY-ESO- were correlated with prognosis in gastric cancer, but not in ESCC or HCC.
In this study, the positive rate of autoantibodies was highest in ESCC. On the other hand, we found that the presence of autoantibodies was associated with poor overall survival only in gastric cancer. We could not confirm the prognostic impact of s-p53-Abs in ESCC previously reported by Suzuki et al. 10 The difference can be partly explained by different cut-off values and/or different assay systems. In the present study, s-NY-ESO-1-Abs showed constantly high positive rates in the three types of gastroenterological cancers.
The presence of s-NY-ESO-1-Abs was also found to be a predictor of poor prognosis in gastric cancer. Based on these results, additional treatment might be necessary for patients who are positive for these antibodies. However, in order to prove it, it is necessary to conduct a large-scale detailed study such as multicenter research.
It has been suggested that the process resulting in antibody induction after gene alteration is complicated. For example, although  be evaluated. We found that except for the expression of NY-ESO-1 in ESCC, p53 and NY-ESO-1 gene expression levels were higher in patients with all cancer types than in healthy subjects ( Figure S1).
Notably, for NY-ESO-1, there is no correlation between protein expression in cancer tissue and serum antibody levels. 12 In addition, the NY-ESO-1 low-expression group tended to have a better prognosis than did the high-expression group in HCC ( Figure S2B). On the other hand, no association was found between gene expression and prognosis in gastric cancer or ESCC. On the other hand, gene expression only in HCC was significantly associated with prognosis. Actually, the gene expression data were not based on the same groups as were the serum autoantibody analysis data in this present study.
Limitations of this study were the small number of included cases and that all cases originated at a single institution. In particular, prognostic evaluation was difficult for HCC, because the antibody-positive rate was the lowest and the number of cases was also the lowest among the examined cancer types. In addition, since serum sample was collected only once at the first consultation, the antibody titer over time could not be evaluated. For this reason, we have not been able to examine changes in antibody titers due to cancer progression or treatment effects.
To demonstrate these results, it may be necessary to conduct a larger sample size and collect samples multiple times, for example, to conduct clinical trials in collaboration with other institutions.
Regarding the discussion on gene expression analysis, TCGA data is mainly based on data from overseas, and most of the data is from Caucasians. It is necessary to examine gene expression in the same patient at the same time for accurate comparison with the expression data, however, there is no remaining sample and it cannot