The expression and clinical significance of GADD45A in breast cancer patients

Background

Breast cancer has become the most common malignant tumor threatening women’s health whether in the developed countries or in developing countries. In the United States, there are an estimated 252,710 female cases diagnosed with breast cancer in 2017 (Siegel, Miller & Jemal, 2017). Meanwhile the incidence of breast cancer in China presents the trend of rising rapidly, over 3.9% annual percentage change from 2000 to 2011 (Chen et al., 2016). However, the pathogenesis of breast cancer, possibly related to disorder of cell cycle and genetic abnormalities, is not clear yet. Therefore, understanding the connection of biological characteristics and clinicopathological significance is an important aspect for investigating the pathogenesis and identifying therapeutic targets.

Growth arrest and DNA-damage-inducible protein 45 alpha (GADD45A) is a downstream target gene of p53 and BRCA1 (breast cancer susceptibility gene 1), which are two of the most important genes maintaining the stability of the genome and suppressing the development of tumor through various mechanisms (Li et al., 2017b). As a member of the GADD45 family of genes that are known as stress sensors, GADD45A modulates the cellular response to a variety of stress conditions, including genotoxic and oncogenic stress (Cretu et al., 2009). Also, some research showed GADD45A can promote or suppress breast carcinoma development depending on different signaling pathways (Carrier et al., 1999; Miki et al., 1994).

Additionally, the correlation of GADD45A expression and clinicopathologic factors is not clear in breast cancer thus far. The aim of the current study is to investigate the significance of GADD45A in breast cancer, including the correlation with clinicopathologic factors and long-term survival situation.

Materials and Methods

Results

Analysis of clinical data

The color reaction/positive staining of GADD45A was observed in 419 tumor tissue samples and 116 adjacent non-neoplastic tissue samples (Fig. 1). Among these 419 tumor tissue samples, 365 (87.1%) were invasive ductal carcinoma (IDC) and 54 (12.9%) were non-IDC. As for molecular types, there were 74 (17.7%) for luminal A, 206 (49.2%) for luminal B, 70 (16.7%) for Her-2 and 69 (16.5%) for triple-negtive (TN) subtype breast cancer. As for histological grades, there were 291 (69.5%) for grade I–II and 128 (30.5%) for III. In addition, there were 113 (27.0%) for TNM I, 233 (55.6%) for TNM II, and 73 (17.4%) for TNM III, respectively.

Survival analysis

Kaplan–Meier curves and log-rank tests were used to perform survival analysis to identify the correlation between survival and GADD45A expression. The last follow-up was carried out in July 2017. Table 4 showed GADD45A expression was not related to survival when all types as total (P = 0.803). However, in a survival analysis for TN subtype patients, the low GADD45A expression group had a mean survival period of 119.61 months (SE: 10.81, 95% CI [98.43–140.79]). However, the high GADD45A expression group had a mean survival period of 88.73 months (SE: 7.78, 95% CI [73.48–103.97]) (P = 0.041). Moreover, the 5-year and 10-year survival rate were 94.7% and 80.5% for the high GADD45A expression group while 79.7% and 68.5% for the low GADD45A expression group, respectively. Kaplan–Meier survival curves were performed in Fig. 2.

Table 4:

Univariate analyses of OS and GADD45A expression of different types patients.

Analyses for GADD45A	χ2	P value
All types as total	0.062	0.803
Luminal A	2.011	0.156
Luminal B	3.536	0.060
HER2	1.565	0.211
TN	4.157	0.041

DOI: 10.7717/peerj.5344/table-4

In the univariate analysis, the TNM stage and GADD45A were found to be significant independent predictors of OS, respectively (Table 5). However, in the multivariate analysis, only the TNM stage was identified the significant independent predictor of OS in the TN subtype patients (Table 5).

Table 5:

Univariate analyses and multivariate analyses overall survival for TN breast cancer patients.

Characteristics	Univariate analyses	Multivariate analyses
	P value	HR	95% CI	P value
Age (>50 vs. ≤50)	0.158	NA	NA	NA
TNM stage (III vs I/II)	0.001	3.720	1.637–8.452	0.002
Pathology type (non-IDC vs IDC)	0.445	NA	NA	N
Histological grade (III vs I/II)	0.251	NA	NA	NA
Ki-67 (>20% vs. ≤20%)	0.060	NA	NA	NA
GADD45A (Positive vs. Negative)	0.041	2.792	0.967–8.060	0.058

DOI: 10.7717/peerj.5344/table-5

Discussion

As GADD45A plays a role in genomic stability and tumorigenesis, some studies explored the role of GADD45A in various cancers, such as prostate cancer (Reis et al., 2015), ovarian cancer (Yuan et al., 2015), esophageal cancer (Ishiguro et al., 2016) and malignant gliomas (Cui et al., 2017), which revealed different GADD45A expressions in various carcinoma tissues. In research on esophageal squamous cell carcinoma (ESCC) (Wang et al., 2012) and another article on breast cancer (Tront et al., 2013), GADD45A protein levels were increased significantly in tumor tissues than those in normal tissues. In addition, a recent study showed that the expression of GADD45A was upregulated in human non-small-cell lung cancer (NSCLC) cells through a microRNA-138 related post-transcriptional pathway (Li et al., 2017a).

Compared with the previous study for GADD45A in breast cancer (Tront et al., 2013), our study had a larger sample analysis for the correlations between clinicopathological factors and GADD45A expression. We also showed the correlation with PR and ER, respectively, while Tront’s study only analyzed tissues of ER+/PR+. Moreover, our study was the first study estimating the prognostic significance of GADD45A in triple negative breast cancer. We found that GADD45A expression was upregulated in breast cancer tissues. DNA damage was a major source of potentially mutagenic lesions leading carcinogenesis (Gao, Herman & Guo, 2016). GADD45A was shown to participate actively in DNA repair mechanisms (Wingert & Rieger, 2016). Therefore, its upregulation may be due to extensive DNA repair in breast cancer tissues. Beyond that, a study showed that GADD45A promotes Myc-driven breast carcinogenesis by negatively regulating matrix metalloproteinase 10 (MMP10) through GSK-3β (Glycogen synthase kinase 3 beta)/β-catenin signaling, resulting in increased tumor vascularization and growth (Tront et al., 2010). In our opinions, this mechanism may result for the high GADD45A expression in breast cancer tissues.

In our study, ER negative and high Ki-67 index were demonstrated the significant association with GADD45A expression. ER-negative breast cancer, which had less treatment options such estrogen-blocking endocrine therapy, had more malignancy behaviors than ER-positive tumors. Ki-67 index was a cellular marker strictly associated with cell proliferation and emerged as a key discriminative feature of malignant phenotype in breast cancer (Pathmanathan et al., 2014). The above factors all suggested high malignant degree. From this, we considered that GADD45A may be a biomarker for the malignancy grade of breast cancer tissue. Our results were opposite to the results presented by Tront et al. (2013) which showed that high GADD45A levels were associated with luminal A and luminal B subtypes but not in triple negative tissues. The discrepancy could be generated by the higher number of tissues analyzed in our study compared to the previous one (419 vs 45). In addition, we explored the correlation with ER and PR status separately while (Tront et al., 2013) analyzed only tissue where they were both positive, as mentioned above. As a result of these advantages, our study was a more credible and detailed exploration for the correlations between clinicopathological factors and GADD45A expression compared previous study.

For further evaluation, we assessed the effect of GADD45A expression on OS. We demonstrated that patients with higher GADD45A expression levels had a poor long-term prognosis for TN subtype based on Kaplan–Meier survival curves. However, GADD45A expression was not an independent prognosis factor in TN type breast cancer in multivariate analysis. The finding may suggest that GADD45A expression affect the prognosis through the effects of other factors but not the independent action, such as the effect of Ki-67 index, which was reported as an independent variable associated with prognosis in TN subtype patients (Adamo et al., 2017; Pan et al., 2017; Ricciardi et al., 2015). In addition, we found that TNM stage independently influenced the prognosis of TN breast cancer patients. The similar conclusion was determined in previous studies (Han et al., 2017; Yu et al., 2013; Yu et al., 2014).

The limitations of current study included the deficiency of follow-up in chemotherapy regimens of patients. It was difficult to analyze the influence of chemotherapy regimen on prognosis of breast cancer. The relationship between different chemotherapy treatments and prognosis in breast cancer should be the focus of the future study and further analyses in biological mechanisms about tumor progression are need. Moreover, the possible selective bias due to the small sample size and inclusion of only postoperative patients may exist in TN breast cancer group. Despite the limitations, the correlation between GADD45A expression and certain clinicopathological factors in breast cancer can be verified.

Conclusions

In present study, we explored the expression and clinical significance of GADD45A in breast carcinoma. High GADD45A was observed in breast cancer tissues compared with adjacent non-neoplastic tissues. ER negative and high Ki-67 index were correlative with high GADD45A expression. Patients with higher GADD45A expression levels had a poorer long-term prognosis in TN type breast cancer. However, only the TNM stage was the independent prognosis factor in TN type breast cancer. Our study suggested that GADD45A may play a role in breast cancer pathogenesis and may eventually help in understanding the biological mechanisms affecting tumor progression.

Institutional Abbreviations

GADD45A

DNA-damage-inducible protein 45 alpha

BRCA1

Breast cancer susceptibility gene 1

TMA

Tissue microarray

OS

Overall survival

IDC

Invasive ductal carcinoma

TN

Triple-negtive

ER

Estrogen receptor

PR

Progesterone receptor

HER2

Human epidermal growth factor receptor-2

ESCC

Esophageal squamous cell carcinoma

NSCLC

Non-small-cell lung cancer

MMP10

Matrix metalloproteinase 10

GSK-3β

Glycogen synthase kinase 3 beta

Supplemental Information