High HDAC5 expression correlates with a poor prognosis and the tumor immune microenvironment in gastric cancer

Introduction

Gastric cancer (GC) is a global health problem, with more than one million new diagnoses worldwide every year. Although incidence and mortality rates have declined over the past 5 years, the latest statistics report that GC ranks fifth in terms of incidence and fourth in terms of mortality among all malignancies (1). Early diagnosis of GC is difficult; currently, the early diagnosis of GC mainly relies on imaging, serum tumor markers, endoscopy, and biopsy pathology (2), which are limited by the cumbersome process or insufficient specificity. Surgery is still the main treatment for GC, but with the combination of chemotherapy, radiotherapy, and targeted therapy, the prognosis of GC has improved significantly (3). However, the clinical efficacy of conventional therapy is limited and the prognosis remains relatively poor. As a recent breakthrough, immunotherapy has become an effective treatment modality after surgery, chemotherapy, radiotherapy, and targeted therapy (4). Thus, there is a pressing need to identify new specific markers and potential targets related to immunotherapy to improve the diagnosis and treatment of GC.

First identified in the mouse genome in 1999, histone deacetylase 5 (HDAC5) is a member of the HDAC class IIa family (5). This protein consists of 1,122 amino acids, has a molecular weight of 121.9 kDa, and has C-terminal deacetylase and N-terminal adapter domains. HDAC5 is known to be expressed in the lung, brain, myocardium, skeletal muscle, and placenta, and many studies have shown that HDAC5 is differentially expressed in different types of tumors. Previous research has confirmed that HDAC5 expression is upregulated in breast cancer (BC) (6), hepatocellular carcinoma (HCC) (7), lung cancer (LC) (8), pancreatic neuroendocrine tumors (pNETs) (9), and colorectal cancer (CRC) (10). It has also been shown that HDAC5 affects cancer cell proliferation, invasion, apoptosis, and cell cycle progression. Zhong et al. demonstrated that overexpression of HDAC5 significantly promotes tumor cell proliferation and invasion and inhibits apoptosis by constructing LC cell lines; meanwhile, knockdown of HDAC5 significantly inhibits tumor cell proliferation and invasion and promotes apoptosis (8). He et al. found that HDAC5 messenger RNA (mRNA) and protein levels were upregulated in human CRC cell lines, and the cell counting kit-8 (CCK-8) assay showed that overexpression of HDAC5 promotes the proliferation of CRC cells. However, knockdown of HDAC5 was observed to inhibit the growth of CRC cells (11). In addition, a study by Peixoto et al. showed that HDAC5 was associated with the active replication of perisynaptic heterochromatin in the late S phase, and the specific depletion of HDAC5 by RNA interference led to structural changes in heterochromatin. This defect in heterochromatin maintenance and assembly was sensed by the DNA damage checkpoint pathway, triggering autophagy and apoptosis in cancer cells (12).

Immune checkpoint inhibitors (ICIs), such as anti-programmed cell death-1 (PD-1) or programmed cell death ligand-1 (PD-L1) monoclonal antibodies, are the new standard of targeted therapy for advanced or metastatic GC and have shown some prognostic improvement in clinical trials (13,14). The tumor immune microenvironment (TIME) is the internal environment of malignant tumor progression and site of the host antitumor immune response and normal tissue destruction. Tumor-infiltrating lymphocytes (TILs) are an important part of the TIME; TILs include CD3⁺ T cells, CD4⁺ T cells, and CD8⁺ T cells, which can reflect the host antitumor immune response (15). HDACs are associated with the immune response, and HDAC5 interacts with the immune system (including immune cells and inflammatory cytokines) during cancer development and progression. HDAC5 is associated with macrophage differentiation in lymphoma cells (16), and depletion of HDAC5 in lymphoma cells via stimulation of nuclear factor-κB (NF-κB) activity reduces the levels of tumor necrosis factor-α (TNF-α) and monocyte chemotactic protein-1 (MCP-1) (17), suggesting a regulatory function of HDAC5 in the proinflammatory response of macrophages. In pancreatic cancer, Zhou et al. revealed an unknown role of HDAC5 in regulating NF-κB signaling pathway and antitumor immune response (18). And in GC, Deng et al. confirmed that HDAC is essential for interferon-γ (IFN-γ)-induced B7-H1 in GC, and suggests the possibility of targeting B7-H1 using small molecular HDAC inhibitors for cancer treatment (19). Hence, HDAC has a role in immunotherapy and the value of HDAC5 in the immune microenvironment of GC needs to be explored.

In this study, we assessed the expression level of HDAC5 in 355 tumor tissues and 300 paracancerous tissues by immunohistochemistry (IHC). Independent factors affecting the prognosis of GC were analyzed by univariate and multivariate analyses. The expression levels of CD3⁺ T cell, CD4⁺ T cell, and CD8⁺ T cell markers and PD-L1 were also measured to compare the correlations between HDAC5 and PD-L1 vs. HDAC5 and TILs. Further analysis of GC samples in The Cancer Genome Atlas (TCGA) was performed to identify differential immune infiltrating cells and jointly investigate the role of HDAC5 in the immune microenvironment of GC. We present the following article in accordance with the REMARK reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-22-4325/rc).

Methods

Patients

This study enrolled 355 patients who were admitted to The Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital) and underwent radical surgery for GC between July 2008 and July 2017. The inclusion criteria were as follows: (I) pathological diagnosis of GC; (II) relatively complete medical records; (III) no preoperative integrated antitumor therapy, such as radiotherapy, targeted therapy, or immunotherapy; and (IV) complete survival follow-up data. The exclusion criteria were as follows: (I) other types of malignant tumors; (II) metastasis from other malignant tumors; and (III) severe cardiopulmonary insufficiency, renal insufficiency, and other underlying diseases.

We collected the data from the inpatient medical records system, including demographic characteristics and clinicopathological features. The pathological stage was determined according to the American Joint Committee on Cancer (AJCC) 8^th edition system. Survival information was obtained by telephone follow-up and medical records, and the last follow-up visit was conducted in August 2021. Overall survival (OS) was defined as the duration from the initial surgery to death or the last follow-up visit. In addition, a dataset containing 375 stomach adenocarcinoma (STAD) tumor tissue samples was downloaded from TCGA.

The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital) (No. IRB-2021-431) and informed consent was taken from all the patients

IHC

The 355 GC tumor tissues and 300 paracancerous tissues were collected, fixed in formalin, and embedded in paraffin. Two veteran pathologists independently selected representative tissues for tissue microarrays (TMAs). The sections were dewaxed separately and rinsed with distilled water; then, antigen repair was performed by washing with phosphate-buffered saline (PBS) for 5 min (three times). Next, the primary antibody (HDAC5: 16166-1-AP; CD3: ab16669; CD4: ab133616; CD8: ab17147; PD-L1: SK006) was added, incubated overnight at 4 ℃, and washed with PBS for 5 min (three times). Subsequently, goat anti-rabbit immunoglobulin G (IgG) H&L (SP-9000, ZSGB-BIO Corp., Shanghai, China) was added to the TMAs (dilution ratio 1:1,000), incubated for 30 min, and washed with PBS for 5 min (three times). 3,3'-diaminobenzidine (DAB) color development and hematoxylin restraining of cell nuclei were then performed using a DAB color development kit. Finally, the TMAs were dehydrated and closed with neutral gel.

IHC assessment

IHC staining of HDAC5 was interpreted separately by two pathologists using the H-score system. The formula for the H-score system was as follows: H score = (∑IS × AP), where IS indicates the staining intensity and AP indicates the percentage of positively stained cells. IS was determined by the cell staining: 0 for no staining; 1 for weak staining; 2 for moderate staining; and 3 for strong staining. AP was recorded as follows: 0 for 0% stained cells; 1 for 1–25% stained cells; 2 for 26–50% stained cells; 3 for 51–75% stained cells; and 4 for 76–100% stained cells. A H-score =6 was set as the cutoff value, and the patients were divided into groups according to HDAC5 expression (positive vs. negative).

PD-L1 expression was recorded based on the combined positivity score (CPS) score, CPS = [number of PD-L1-positive cells (tumor cells, lymphocytes, macrophages)/total tumor cells] × 100 for evaluation, where a CPS ≥10 was considered positive. TILs were quantified by pathologists who observed and recorded the total number of corresponding lymphocytes in the entire magnification field and divided the samples into high and low expression groups using the median as the cutoff value.

Statistical analyses

Statistical analyses were performed using IBM SPSS Statistics for Windows, version 25.0 (IBM Corp, Armonk, NY, USA) and GraphPad Prism for Windows, version 8.3.0 (GraphPad Software, San Diego, CA, USA). Counting data were expressed as frequencies and percentages, and measurement data were expressed as x¯±s. The correlation between HDAC5 expression levels and the clinicopathological features and immune-related factors was determined by the Mann-Whitney test or chi-square test. Survival curves were plotted by the Kaplan-Meier method, and independent factors affecting the prognosis of patients with GC were determined by univariate and multivariate Cox regression analyses. Factors in multivariate analyses were selected according to the importance of clinical information. The hazard ratios (HRs) and their corresponding 95% confidence intervals (CIs) were also calculated. TCGA data were analyzed by single sample gene set enrichment analysis (ssGSEA) using the gene set variance analysis (GSVA) package to identify differentially infiltrated immune cells (Wilcoxon rank-sum test), and heat maps and violin plots were generated according to the gene expression level of HDAC5 (high vs. low). P<0.05 was considered to indicate a statistically significant difference.

Results

Clinicopathological features of the 355 GC patients

The mean age of the 355 GC patients included in the study was 63.74 years, with a median age of 64 years and an age range of 28–91 years. Of these patients, 256 (72.11%) were men, while only 99 (27.89%) were women. Adenocarcinoma was the main pathological type among the included patients (n=322, 90.70%), and most tumors were undifferentiated or hypodifferentiated (47.89% collectively). In terms of the tumor site, most tumors were considered distal GC (226 cases, 63.66%), and only 116 cases (32.68%) were proximal GC. According to the pTNM stage, most of the assessed tumors were stage III (72.68%), with stages I, II, and III accounting for 4.79%, 14.37%, and 6.20% of tumors, respectively. Detailed clinicopathological information is shown in Table 1.

HDAC5 is highly expressed in GC tissues and predicts a poor prognosis

We found that HDAC5 was expressed in both the cytoplasm and nucleus by IHC staining. Representative immunohistochemical plots (×200-fold) and specific H-scores are shown in Figure 1A,1B. Among the 355 tumor TMAs, 268 had different levels of HDAC5 expression, and the HDAC5 expression rate was 75.49%. Eighty-seven patients were negative for HDAC5 expression, accounting for 24.51% (Table 2). In this study, we defined H-score =6 as the cutoff value, H-score ≥6 as the HDAC5-positive expression group, and H-score <6 as the HDAC5-negative expression group. The results showed that 138 of 355 GC tissues (38.87%) exhibited high HDAC5 expression, while only 44 of 300 paracancerous tissues (14.67%) exhibited high HDAC5 expression, which indicated that HDAC5 was upregulated in GC tissues compared with the paracancerous tissues (P<0.001; Table 3).

Figure 1 HDAC5 is highly expressed in GC tumor tissues, and a high level of HDAC5 predicts a worse prognosis in GC patients. (A) Representative images of HDAC5 staining by IHC (×200-fold). (B) Differential expression of HDAC5 in tumor and paracancerous tissues of GC. (C) The Kaplan-Meier OS curves of GC patients with different HDAC5 levels in tumor tissues (log-rank test). (D) The Kaplan-Meier OS curves of GC patients with different HDAC5 levels in paracancerous tissues (log-rank test). HDAC5, histone deacetylase 5; IHC, immunohistochemistry; OS, overall survival; GC, gastric cancer.

The effect of HDAC5 on prognosis has been confirmed in other types of tumors. To investigate its effect on the prognosis of GC, we used the Kaplan-Meier method to plot survival curves. The prognosis of patients with high HDAC5 expression levels in tumor tissues was found to be worse than that of those with low HDAC5 expression in tumor tissues (5-year OS: 44.7% vs. 60.6%, P=0.007; Figure 1C). However, there was no significant correlation in the paracancerous tissues (5-year OS: 52.9% vs. 60.1%, P=0.227; Figure 1D), implying that the expression level of HDAC5 in tumor tissues is negatively correlated with prognosis.

To investigate the independent factors affecting the prognosis of GC, we included important clinicopathological data, such as sex, age, family history, tumor location, PD-L1, and TILs, in a univariate Cox regression model. The results (Table 4) revealed that HDAC5 expression levels (P=0.008), CD4⁺ T cells (P=0.029), CD8⁺ T cells (P=0.040), family history (P=0.002), pT stage (P=0.010), pN stage (P<0.001), pM stage (P<0.001), pTNM stage (P<0.001), carcinoembryonic antigen (CEA) (P=0.004), carbohydrate antigen (CA)199 (P=0.025), and CA50 (P=0.035) had an impact on the prognosis of GC. Subsequently, a multivariate Cox regression model was constructed, and subsequent analysis revealed that HDAC5 expression levels (P=0.036; HR =1.581; 95% CI: 1.031–2.426), CD4⁺ T cell levels (P=0.012; HR =0.539; 95% CI: 0.334–0.872), CD8⁺ T cell levels (P<0.001; HR =0.288; 95% CI: 0.144–0.577), and pTNM stage (P<0.001; HR =3.757; 95% CI: 1.790–7.886) were independent factors affecting GC prognosis (Table 5). Thus, both univariate and multivariate analyses showed that HDAC5 was an independent prognostic factor for GC.

The expression of HDAC5 is correlated with age, gastric history, Lauren type, and pM stage

To further investigate the correlation between HDAC5 expression levels and clinicopathological characteristics, we analyzed the correlation between groups using the chi-square test. The results (Table 6) revealed that the positive expression rate of HDAC5 was higher in patients aged 65 years or older vs. those aged less than 65 years (44.19% vs. 33.88%, P=0.046) and in patients who smoked vs. those who did not (43.59% vs. 38.41%, P=0.001). The Lauren type was also found to be closely related to the expression level of HDAC5. Specifically, the expression level of HDAC5 was significantly higher in mixed-type patients than in intestinal-type, and diffuse-type patients (P=0.042), with positive expression rates of 39.09%, 31.48%, and 53.49% in intestinal-type, diffuse-type and mixed-type patients, respectively. In addition, the expression level of HDAC5 was significantly correlated with the pM stage, and the expression level of HDAC5 was higher in the pM1 stage than in the pM0 (P=0.012). This suggests that the expression level of HDAC5 is correlated with age, smoking history, Lauren type, and pM stage. However, other indicators, including sex, family history, tumor location and size, Borrmann type, grade of differentiation, pT stage, pN stage, CEA, and other common tumor markers, were not significantly correlated with HDAC5 expression.

HDAC5 expression regulates the GC TIME

To explore the status of the TIME of GC patients, we performed an IHC assessment of 355 patients to determine the expression of TILs (CD3⁺ T cells, CD4⁺ T cells, CD8⁺ T cells) and PD-L1 in tumor tissues. The median numbers of CD3⁺ T cells, CD4⁺ T cells, and CD8⁺ T cells were used as the cutoff value to divide patients into high and low groups (Figure 2A-2C).

Figure 2 TILs and PD-L1 were correlated with the prognosis of GC. Representative images (×200-fold) of CD3⁺ T cells (A), CD4⁺ T cells (B), CD8⁺ T cells (C) and PD-L1 (D) staining by IHC. The Kaplan-Meier OS curves of GC patients with different CD4⁺ T cell (E), CD8⁺ T cell (F), CD3⁺ T cell (G) and PD-L1 (H) levels in tumor tissues (log-rank test). OS, overall survival; PD-L1, programmed cell death ligand-1; TILs, tumor-infiltrating lymphocytes; GC, gastric cancer; IHC, immunohistochemistry.

Patients were divided into positive and negative PD-L1 expression groups based on the CPS score (Figure 2D). Kaplan-Meier survival curves revealed that the numbers of CD4⁺ T cells and CD8⁺ T cells had a prognostic effect, and patients who exhibited high levels of CD4⁺ T cells or CD8⁺ T cells had a better prognosis (5-year OS: 60.6% vs. 48.9%, P=0.027; 60.6% vs. 48.6%, P=0.038; Figure 2E,2F). However, the CD3⁺ T cell levels and PD-L1 expression levels had no significant effect on the prognosis of GC patients (5-year OS: 55.2% vs. 54.5%, P=0.994; 55.2% vs. 53.5%, P=0.981; Figure 2G,2H). This finding suggests that the levels of CD4⁺ T cells and CD8⁺ T cells are positively correlated with the prognosis of GC patients.

Furthermore, we further analyzed the relationship between HDAC5 expression and TILs and PD-L1 in GC (Table 7). The correlations between HDAC5 and the levels of CD3⁺, CD4⁺, and CD8⁺ T cells were analyzed (Figure 3A-3C); we found that the levels of CD3⁺ T cells (292.23±14.88 vs. 197.62±18.16, P<0.001), CD4⁺ T cells (56.83±4.62 vs. 44.83±7.13, P<0.001), and CD8⁺ T cells (177.08±10.22 vs. 108.79±10.30, P<0.001) were negatively correlated with the expression level of HDAC5. The analysis of PD-L1 expression between the high and low HDAC5 expression groups (Figure 3D) showed that the positive rate of PD-L1 expression was higher in the high HDAC5 expression group (20.7% vs. 37.0%, P=0.001), indicating that the expression of PD-L1 was positively correlated with the level of HDAC5.

Figure 3 HDAC5 expression was highly negatively correlated with TILs levels and positively correlated with PD-L1 expression. (A) Correlation between HDAC5 expression and CD3⁺ T cell levels. (B) Correlation between HDAC5 expression and CD4⁺ T cell level. (C) Correlation between HDAC5 expression and CD8⁺ T cell levels. (D) Correlation between HDAC5 expression and PD-L1 expression. **, P<0.01; ***, P<0.001. HDAC5, histone deacetylase 5; PD-L1, programmed cell death ligand-1; TILs, tumor-infiltrating lymphocytes.

In addition, we performed ssGSEA on the 375 STAD tissues downloaded from TCGA to identify their differential immune infiltrating cells. We plotted heat maps and violin plots (Figure 4A,4B) and observed that the expression level of HDAC5 was significantly correlated with the levels of activated CD4 T cells (P<0.0001), activated CD8 T cells (P<0.0001), activated dendritic cells (P<0.001), CD56 bright natural killer cells (P<0.01), central memory CD8 T cells (P<0.05), gamma delta T cells (P<0.01), neutrophils (P<0.001), plasmacytoid dendritic cells (P<0.05), type 17 T helper cells (P<0.01), and type 2 T helper cells (P<0.05). The expression level of HDAC5 was negatively correlated with the levels of immune infiltrating cells, except for plasmacytoid dendritic cells.

Figure 4 The expression of HDAC5 was closely related to TILs levels according to TCGA. Heat map (A) of HDAC5-associated infiltrating cells and violin plot (B) of HDAC5-associated differentially infiltrating immune cells. *, P<0.05; **, P<0.01; ***, P<0.001; ****, P<0.0001. HDAC5, histone deacetylase 5; TILs, tumor-infiltrating lymphocytes; TCGA, The Cancer Genome Atlas.

HDAC5^low + CD4^high status and HDAC5^low + CD8^high status predict a better prognosis

Based on the correlation between HDAC5 and TILs and PD-L1, and Kaplan-Meier analysis confirming that TILs and PD-L1 influence the prognosis of GC, we next explored the prognostic impact of HDAC5 on GC in combination with TILs or PD-L1. We divided the patients into four groups according to their CD4⁺ T cell levels and HDAC5 expression levels (HDAC5^low + CD4^low group; HDAC5^low + CD4^high group; HDAC5^high + CD4^low group; HDAC5^high + CD4^high group). Kaplan-Meier survival analysis (Figure 5A) demonstrated that the HDAC5^low + CD4^high group had the best prognosis (5-year OS: 69.5%), the HDAC5^high + CD4^high group had the worst prognosis (5-year OS: 39.9%), and the HDAC5^low + CD4^low and HDAC5^high + CD4^low groups had similar prognoses (5-year OS: 49.0% vs. 48.4%). The overall prognosis was significantly different among the four groups (P=0.004). The prognostic value analysis of HDAC5 combined with CD8⁺ T cells (Figure 5B) revealed that the HDAC5^low + CD8^high group had the best prognosis (5-year OS: 62.4%), the HDAC5^high + CD8^low group had the worst prognosis (5-year OS: 40.6%), and the remaining two groups had similar prognoses (5-year OS: 57.2% vs. 53.6%); the difference in prognosis among the groups was statistically significant (P=0.023).

Figure 5 HDAC5^low + CD4^high and HDAC5^low + CD8^high status predict a better prognosis. (A) Kaplan-Meier OS curves of GC patients with different levels of HDAC5 and CD4⁺ T cells (log-rank test). (B) Kaplan-Meier OS curves of GC patients with different levels of HDAC5 and CD8⁺ T cells (log-rank test). (C) Kaplan-Meier OS curves of GC patients with different levels of HDAC5 and CD3⁺ T cells (log-rank test). (D) Kaplan-Meier OS curves of GC patients with different levels of HDAC5 and PD-L1 (log-rank test). HDAC5, histone deacetylase 5; OS, overall survival; PD-L1, programmed cell death ligand-1; GC, gastric cancer.

Although the survival analysis of CD3⁺ T cells and PD-L1 showed no significant correlation with prognosis, based on the correlation between these factors and HDAC5, we next performed an integrated prognostic value analysis of HDAC5 expression coupled with CD3⁺ T cells (Figure 5C). The prognosis was relatively good in the HDAC5^low + CD3^low and HDAC5^low + CD3^high groups, with similar 5-year OS rates (60.6% and 60.5%, respectively). Also, the HDAC5^high + CD3^low (5-year OS: 48.3%) and HDAC5^high + CD3^high (5-year OS: 37.9%) groups had worse prognoses, and there was a significant difference in the OS among the four groups (P=0.031). Finally, a survival analysis was performed according to the expression of HDAC5 and PD-L1 (Figure 5D) and found that the HDAC5^low + PD-L1⁻ and HDAC5^low + PD-L1⁺ groups had good and similar prognoses (5-year OS: 59.4% vs. 64.9%), while the HDAC5^high + PD-L1⁻ and HDAC5^high + PD-L1⁺ groups had relatively worse prognoses (5-year OS: 46.5% vs. 42.6%), and there was a statistically significant OS difference among the groups (P=0.046).

Discussion

Histone acetylation and deacetylation are among the most common post-translational modifications. HDACs maintain a dynamic balance between acetylation and deacetylation (20), thereby regulating cell proliferation, apoptosis, metastasis, and cell cycle progression and affecting histone properties and their biological functions (21,22). Previous studies have demonstrated that HDAC5 is differentially expressed in tumor tissues. Patani et al. performed RNA extraction and reverse transcription in 127 BC tissues and 33 normal tissues and used quantitative real-time PCR (qRT-PCR) to determine the transcription levels of HDAC genes and investigate the expression differences. The expressions of HDACs, including HDAC5, were found to be significantly different in BC tissues compared with normal tissues, and HDAC5 expression was significantly upregulated in BC tissues (23). A study by Fan et al. reported that the mRNA and protein levels of HDAC5 were determined in HCC tissues and cells using qRT-PCR and protein blotting, and similarly, both the mRNA and protein levels of HDAC5 were found to be upregulated (24).

The expressions of HDACs in GC have also been investigated, and some studies have reported that the expression levels of HDAC1, HDAC2, and HDAC4 are upregulated in cancer tissues (25-27), while some other family members, such as HDAC3, have been shown to exhibit decreased expression in cancer tissues (28). However, studies on HDAC5 in GC are lacking, and the results of available studies are highly variable; therefore, further investigation is needed. The expression of HDAC5 in GC was first reported by Orenay-Boyacioglu et al., who assessed the expressions of HDACs by qRT-PCR in 28 GC tumor tissues and 20 normal tissues. They reported that the expression level of HDAC5 was downregulated in tumor tissues compared with control tissues (29). However, Chen et al. used gene expression profiling interaction analysis (GEPIA) to explore the mRNA levels of HDACs and found that the expression levels of HDAC5 were not significantly different in GC tissues compared with normal tissues (30). Interestingly, we determined the expression levels of HDAC5 in tumor tissues (n=355) and paracancerous tissues (n=300) by IHC and observed that the expression levels of HDAC5 were significantly increased in tumor tissues (P<0.001). This suggests that the value of HDAC5 in GC still needs to be further confirmed by more studies.

HDAC5 plays an important role in cancer development and is a potential prognostic marker (31), and has been shown to have an impact on the prognosis of different tumors. Zhou et al. showed that overexpression of HDAC5 adversely affected the OS and progression-free survival (PFS) of ovarian cancer patients (32). Similarly, Klieser et al.’s study in pNET confirmed HDAC5 as a predictor of poor clinical outcomes (9). However, Zhang et al. investigated HDAC5 in astrocytoma and found that HDAC5 expressed at high levels was indicative of a better prognosis (33), suggesting that this gene exhibits different prognostic values for different cancer types. There are few studies on the prognosis of HDAC5 in GC patients, and only Chen et al. have studied the effect of HDAC5 on the prognosis of GC; their study confirmed that a high HDAC5 expression level was closely associated with poor prognosis (30). Similarly, we performed a survival analysis for high and low HDAC5 expression and identified a negative correlation between GC prognosis and HDAC5 expression level (P=0.007). Meanwhile, the univariate and multivariate analyses confirmed that HDAC5 was an independent factor affecting the prognosis of patients.

Currently, there are few studies demonstrating a correlation between HDAC5 expression and clinicopathological features. Only Chen et al. used GEPIA to explore the mRNA levels of HDACs in GC and reported that the expression levels of HDAC5 in GC were correlated with Lauren type, clinical stage, lymph node status, treatment, and human epidermal growth factor receptor 2 status (30). Our study reached some of the same conclusions, as we found a significant correlation between HDAC5 expression levels and Lauren type (P=0.042) and pM stage (P=0.012). In addition, we observed that HDAC5 expression levels were also markedly correlated with age (P=0.046) and smoking status (P=0.001). These results suggest that HDAC5 is more likely to be highly expressed in seniors, smokers, those with mixed Lauren type, and those with distant metastases.

A growing number of studies have confirmed that the activation of HDACs can affect PD-L1 expression in various types of cancer. In pancreatic ductal adenocarcinoma, Zhou et al. analyzed the correlation between HDAC5 and PD-L1 expression using TMAs and found that PD-L1 expression was negatively correlated with HDAC5 expression (P=0.0028) (18). Woods et al. demonstrated that HDAC inhibitors can alter immunogenicity and enhance antitumor immune responses in melanoma, and that class I HDAC inhibitors can upregulate PD-L1 (34). Thus far, no studies have reported on the effect of HDAC5 expression on PD-L1 expression in GC. We measured the expressions of HDAC5 and PD-L1 in 355 GC tissues by IHC and found that the expression of HDAC5 was positively correlated with PD-L1 (P=0.001). Further clarification of the role of TILs might contribute to a comprehensive understanding of the TIME, which may help guide personalized immunotherapy, and TILs are currently a hot topic in cancer immunotherapy research (35).

The relationship between HDAC expression and immune cell infiltration remains debatable. Xiao et al. designed experiments showing that HDAC5-negative mice attenuated the suppressive function of regulatory T cells (Treg cells), while the silencing of HDAC5 inhibited the switch from CD4⁺ T cells to Tregs and suppressed IFN-γ production in CD8⁺ T cells (36). We identified a correlation between the HDAC5 expression level and the numbers of CD3⁺ T cells, CD4⁺ T cells, and CD8⁺ T cells, and determined that all of these were negatively correlated with the HDAC5 expression level. We also analyzed the TIME of 375 GC tissues in TCGA and observed that the expression level of HDAC5 was negatively correlated with activated CD4 T cells, activated CD8 T cells, and other types of immune cells. This finding indicated that HDAC5 may play an important regulatory role in the TIME of GC.

Previous research has shown that TILs (including CD3⁺, CD4⁺, and CD8⁺ T cells) are associated with a good prognosis in GC (37-39). However, the prognostic value of PD-L1 in GC is still controversial. Some studies have confirmed that PD-L1 expression is associated with a good prognosis (40-42), while others have confirmed that PD-L1 expression is either associated with a poor prognosis or does not have a prognostic value (43,44). We investigated the prognostic significance of PD-L1 and TILs and found that high expression levels of CD4⁺ T cells and CD8⁺ T cells were associated with a good prognosis. Our univariate and multivariate analyses confirmed that CD4⁺ T cells and CD8⁺ T cells were independent factors affecting the prognosis of patients. However, PD-L1 and CD3⁺ T cells did not exhibit a significant effect on prognosis in this study.

Finally, through combined survival analysis, we also found that the HDAC5^low + CD4^high and HDAC5^low + CD8^high groups had the best prognosis, with 5-year OS rates of 69.5% and 62.4%, respectively. Based on the negative correlation between HDAC5 expression level and prognosis and the positive correlation between high expression levels of CD4⁺ T cells and CD8⁺ T cells, we can conclude that the joint action of HDAC5 with CD4⁺ T cells and CD8⁺ T cells has some influence on the prognosis of GC. This confirms that HDAC5 may be involved in the regulation of the GC tumor microenvironment, but the specific mechanism still requires further investigation. In conclusion, HDAC5 can be considered a potential diagnostic marker for GC and a potential target for immunotherapy.

Conclusions

HDAC5 is highly expressed in tumor tissues and is an independent factor affecting the prognosis of GC. Additionally, HDAC5 can regulate the TIME of GC and is a potential target for immunotherapy.

Acknowledgments

Funding: This study was supported by Natural Science Foundation of Zhejiang Province (No. HDMY22H160008); Medical Science and Technology Project of Zhejiang Province (Nos. 2022KY114, WKJ-ZJ-2104); Chinese Postdoctoral Science Foundation (No. 2022M713203); Program of Zhejiang Provincial TCM Sci-Tech Plan (No. 2022ZQ020); National Natural Science Foundation of China (Nos. 82074245, 81973634); Zhejiang Provincial Research Center for Upper Gastrointestinal Tract Cancer (No. JBZX-202006).

Footnote

Reporting Checklist: The authors have completed the REMARK reporting checklist. Available at https://atm.amegroups.com/article/view/10.21037/atm-22-4325/rc

Data Sharing Statement: Available at https://atm.amegroups.com/article/view/10.21037/atm-22-4325/dss

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-22-4325/coif). All authors report that this study was supported by Natural Science Foundation of Zhejiang Province (No. HDMY22H160008), Medical Science and Technology Project of Zhejiang Province (Nos. 2022KY114, WKJ-ZJ-2104), Chinese Postdoctoral Science Foundation (No. 2022M713203), Program of Zhejiang Provincial TCM Sci-Tech Plan (No. 2022ZQ020), National Natural Science Foundation of China (Nos. 82074245, 81973634), Zhejiang Provincial Research Center for Upper Gastrointestinal Tract Cancer (No. JBZX-202006). The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the Ethics Committee of Cancer Hospital of the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital) (No. IRB-2021-431) and informed consent was taken from all the patients.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. Necula L, Matei L, Dragu D, et al. Recent advances in gastric cancer early diagnosis. World J Gastroenterol 2019;25:2029-44. [Crossref] [PubMed]
3. Sexton RE, Al Hallak MN, Diab M, et al. Gastric cancer: a comprehensive review of current and future treatment strategies. Cancer Metastasis Rev 2020;39:1179-203. [Crossref] [PubMed]
4. Li K, Zhang A, Li X, et al. Advances in clinical immunotherapy for gastric cancer. Biochim Biophys Acta Rev Cancer 2021;1876:188615. [Crossref] [PubMed]
5. Verdel A, Khochbin S. Identification of a new family of higher eukaryotic histone deacetylases. Coordinate expression of differentiation-dependent chromatin modifiers. J Biol Chem 1999;274:2440-5. [Crossref] [PubMed]
6. Oltra SS, Cejalvo JM, Tormo E, et al. HDAC5 Inhibitors as a Potential Treatment in Breast Cancer Affecting Very Young Women. Cancers (Basel) 2020;12:412. [Crossref] [PubMed]
7. Gu H, Fang Z, Cai X, et al. Highly expressed histone deacetylase 5 promotes the growth of hepatocellular carcinoma cells by inhibiting the TAp63-maspin pathway. Am J Cancer Res 2018;8:462-75. [PubMed]
8. Zhong L, Sun S, Yao S, et al. Histone deacetylase 5 promotes the proliferation and invasion of lung cancer cells. Oncol Rep 2018;40:2224-32. [Crossref] [PubMed]
9. Klieser E, Urbas R, Stättner S, et al. Comprehensive immunohistochemical analysis of histone deacetylases in pancreatic neuroendocrine tumors: HDAC5 as a predictor of poor clinical outcome. Hum Pathol 2017;65:41-52. [Crossref] [PubMed]
10. Stypula-Cyrus Y, Damania D, Kunte DP, et al. HDAC up-regulation in early colon field carcinogenesis is involved in cell tumorigenicity through regulation of chromatin structure. PLoS One 2013;8:e64600. [Crossref] [PubMed]
11. He P, Liang J, Shao T, et al. HDAC5 promotes colorectal cancer cell proliferation by up-regulating DLL4 expression. Int J Clin Exp Med 2015;8:6510-6. [PubMed]
12. Peixoto P, Castronovo V, Matheus N, et al. HDAC5 is required for maintenance of pericentric heterochromatin, and controls cell-cycle progression and survival of human cancer cells. Cell Death Differ 2012;19:1239-52. [Crossref] [PubMed]
13. Högner A, Moehler M. Immunotherapy in Gastric Cancer. Curr Oncol 2022;29:1559-74. [Crossref] [PubMed]
14. Takei S, Kawazoe A, Shitara K. The New Era of Immunotherapy in Gastric Cancer. Cancers (Basel) 2022;14:1054. [Crossref] [PubMed]
15. Lane AN, Higashi RM, Fan TW. Metabolic reprogramming in tumors: Contributions of the tumor microenvironment. Genes Dis 2020;7:185-98. [Crossref] [PubMed]
16. Baek YS, Haas S, Hackstein H, et al. Identification of novel transcriptional regulators involved in macrophage differentiation and activation in U937 cells. BMC Immunol 2009;10:18. [Crossref] [PubMed]
17. Poralla L, Stroh T, Erben U, et al. Histone deacetylase 5 regulates the inflammatory response of macrophages. J Cell Mol Med 2015;19:2162-71. [Crossref] [PubMed]
18. Zhou Y, Jin X, Yu H, et al. HDAC5 modulates PD-L1 expression and cancer immunity via p65 deacetylation in pancreatic cancer. Theranostics 2022;12:2080-94. [Crossref] [PubMed]
19. Deng R, Zhang P, Liu W, et al. HDAC is indispensable for IFN-γ-induced B7-H1 expression in gastric cancer. Clin Epigenetics 2018;10:153. [Crossref] [PubMed]
20. Halasa M, Adamczuk K, Adamczuk G, et al. Deacetylation of Transcription Factors in Carcinogenesis. Int J Mol Sci 2021;22:11810. [Crossref] [PubMed]
21. Li G, Tian Y, Zhu WG. The Roles of Histone Deacetylases and Their Inhibitors in Cancer Therapy. Front Cell Dev Biol 2020;8:576946. [Crossref] [PubMed]
22. Telles E, Seto E. Modulation of cell cycle regulators by HDACs. Front Biosci (Schol Ed) 2012;4:831-9. [PubMed]
23. Patani N, Jiang WG, Newbold RF, et al. Histone-modifier gene expression profiles are associated with pathological and clinical outcomes in human breast cancer. Anticancer Res 2011;31:4115-25. [PubMed]
24. Fan J, Lou B, Chen W, et al. Down-regulation of HDAC5 inhibits growth of human hepatocellular carcinoma by induction of apoptosis and cell cycle arrest. Tumour Biol 2014;35:11523-32. [Crossref] [PubMed]
25. Jiang Z, Yang H, Zhang X, et al. Histone deacetylase-1 as a prognostic factor and mediator of gastric cancer progression by enhancing glycolysis. Hum Pathol 2019;85:194-201. [Crossref] [PubMed]
26. Wagner T, Brand P, Heinzel T, et al. Histone deacetylase 2 controls p53 and is a critical factor in tumorigenesis. Biochim Biophys Acta 2014;1846:524-38. [PubMed]
27. Wang Z, Qin G, Zhao TC. HDAC4: mechanism of regulation and biological functions. Epigenomics 2014;6:139-50. [Crossref] [PubMed]
28. Xu G, Zhu H, Zhang M, et al. Histone deacetylase 3 is associated with gastric cancer cell growth via the miR-454-mediated targeting of CHD5. Int J Mol Med 2018;41:155-63. [PubMed]
29. Orenay-Boyacioglu S, Kasap E, Gerceker E, et al. Expression profiles of histone modification genes in gastric cancer progression. Mol Biol Rep 2018;45:2275-82. [Crossref] [PubMed]
30. Chen L, Fei Y, Zhao Y, et al. Expression and prognostic analyses of HDACs in human gastric cancer based on bioinformatic analysis. Medicine (Baltimore) 2021;100:e26554. [Crossref] [PubMed]
31. Li A, Liu Z, Li M, et al. HDAC5, a potential therapeutic target and prognostic biomarker, promotes proliferation, invasion and migration in human breast cancer. Oncotarget 2016;7:37966-78. [Crossref] [PubMed]
32. Zhou L, Xu X, Liu H, et al. Prognosis Analysis of Histone Deacetylases mRNA Expression in Ovarian Cancer Patients. J Cancer 2018;9:4547-55. [Crossref] [PubMed]
33. Zhang K, Zhao H, Zhang K, et al. Chromatin-regulating genes are associated with postoperative prognosis and isocitrate dehydrogenase mutation in astrocytoma. Ann Transl Med 2020;8:1594. [Crossref] [PubMed]
34. Woods DM, Sodré AL, Villagra A, et al. HDAC Inhibition Upregulates PD-1 Ligands in Melanoma and Augments Immunotherapy with PD-1 Blockade. Cancer Immunol Res 2015;3:1375-85. [Crossref] [PubMed]
35. Schmidt M, Heimes AS. Immunomodulating Therapies in Breast Cancer-From Prognosis to Clinical Practice. Cancers (Basel) 2021;13:4883. [Crossref] [PubMed]
36. Xiao H, Jiao J, Wang L, et al. HDAC5 controls the functions of Foxp3(+) T-regulatory and CD8(+) T cells. Int J Cancer 2016;138:2477-86. [Crossref] [PubMed]
37. Zhang N, Cao M, Duan Y, et al. Prognostic role of tumor-infiltrating lymphocytes in gastric cancer: a meta-analysis and experimental validation. Arch Med Sci 2020;16:1092-103. [Crossref] [PubMed]
38. Zhang D, He W, Wu C, et al. Scoring System for Tumor-Infiltrating Lymphocytes and Its Prognostic Value for Gastric Cancer. Front Immunol 2019;10:71. [Crossref] [PubMed]
39. Ni Z, Xing D, Zhang T, et al. Tumor-infiltrating B cell is associated with the control of progression of gastric cancer. Immunol Res 2021;69:43-52. [Crossref] [PubMed]
40. Qiu Z, Du Y. Clinicopathological and prognostic significance of programmed death ligant-1 expression in gastric cancer: a meta-analysis. J Gastrointest Oncol 2021;12:112-20. [Crossref] [PubMed]
41. Wang Y, Zhu C, Song W, et al. PD-L1 Expression and CD8+ T Cell Infiltration Predict a Favorable Prognosis in Advanced Gastric Cancer. J Immunol Res 2018;2018:4180517. [Crossref] [PubMed]
42. Pereira MA, de Castria TB, Ramos MFKP, et al. Cytotoxic T-lymphocyte-associated protein 4 in gastric cancer: Prognosis and association with PD-L1 expression. J Surg Oncol 2021;124:1040-50. [Crossref] [PubMed]
43. Fang W, Chen Y, Sheng J, et al. Association between PD-L1 Expression on Tumour-Infiltrating Lymphocytes and Overall Survival in Patients with Gastric Cancer. J Cancer 2017;8:1579-85. [Crossref] [PubMed]
44. Gu L, Chen M, Guo D, et al. PD-L1 and gastric cancer prognosis: A systematic review and meta-analysis. PLoS One 2017;12:e0182692. [Crossref] [PubMed]

(English Language Editor: A. Kassem)