Abstract

INTRODUCTION

Precision medicine needs to select patients with the right target to achieve appropriate treatment approach and prognosis [1]. In the past decades, several biomarkers were discovered, for example, epidermal growth factor receptor (EGFR) in lung cancer patients [2] and hormone receptor (HR), human epidermal growth factor receptor 2 (HER2), estrogen (ER), progesterone receptor (PR) in breast cancer patients [3, 4]. Another potential target in cancer is Jab1. Jab1 was initially identified as c-Jun activation domain-binding protein-1 (Jab1) [5] and subsequently discovered to be the fifth component of the constitutive photomorphogenic-9 signalosome (COPS5). This protein regulates a variety of cellular and developmental processes, including signal transduction, cells proliferation, cell cycle, apoptosis, DNA damage response (DDR) and tumorigenesis [6]. Increasing evidence indicates that dysregulation of Jab1/COPS5 activity contributes to tumorigenesis, which is functionally associated with cancer-related genes including p27 [7], p57 [8], SMAD4 [9], NcoR [10], Trx [11] and so on [12]. Studies on Jab1/COPS5 has focused on identification of tumor types that show Jab1/COPS5 overexpression, including colon, ovarian, lung and breast cancer [6].

Lung cancer is the most common cancer worldwide with a poor five-year survival estimated at 15%. This disease can be divided in NSCLC, contributing to 85% of all lung cancers, and small cell lung cancer (SCLC) [13]. Tumor-targeted therapies have already been extensively studied in lung cancer patients, which resulted in implementation of such therapies, for example, erlotinib, bevacizumab and gefitinib, in the treatment guideline of lung cancer [14]. Although Jab1/COPS5 was found to be overexpressed in lung cancer, its role on therapeutic response remains to be determined.

In breast cancer, treatment is already personalized by detecting molecular markers [4]. The outcome of patients is significantly improved by targeted therapy focused on HR positive, or HER2 positive cancers. However, triple negative breast cancers (TNBC) account for 15% of all breast cancers, defect in either ER/PR or HER2. TNBC has the worst prognosis amongst all breast cancers [15]. Previous studies reported Jab1/COPS5 overexpressed in breast cancers and therefore Jab1/COPS5 could be an interesting target for breast cancer patients [16]. However, the role of Jab1/COPS5 on breast cancer relapse needs to be elucidated.

The aim of the current study is to examine Jab1/COPS5 expression on both lung and breast cancers and to investigate its role on therapeutic response and cancer relapse.

RESULTS

Patient characteristics and demographics

To analyze the Jab1/COPS5 expression patterns, we used 88 lung cancer tissue samples (mean age, 59 years; range, 33–81 years) and 76 breast cancer tissue samples (mean age, 49 years; range, 25–80 years). Characteristics of noncancerous and cancerous patients were summarized in Supplementary Table 1. HE staining of the patients’ tissue is showed in Figure ​Figure11.

Open in a separate window

Figure 1

HE staining of cancer tissue and non-cancerous tissue

(A) Pneumonia, small cell lung cancer (SCLC) and lung squamous carcinoma. (B) Breast hyperplasia, ductal breast cancer and lobular breast cancer.

Association between Jab1/COPS5 and clinicopathological parameters

The expression of Jab1/COPS5 in lung cancer or breast cancer was detected by IHC (Figure ​(Figure2).2). It was demonstrated that COPS5-positive cells in cancer tissue were significantly higher than those in non-cancerous tissue (Figure ​(Figure22 and Figure ​Figure3).3). Correlations between COPS5 positive cases and clinicopathological parameters were summarized in Table ​Table11 and Table ​Table2.2. COPS5 expression was correlated with histological type (p=0.035), tumor stage (p=0.017), lymph node metastasis (p=0.016), distant metastasis (p=0.007) and hydrothorax status (p=0.039) in lung cancer patients (Table ​(Table1).1). However, COPS5 expression was correlated with ER status (p<0.001) and PR status (p=0.013) in breast cancer patients (Table ​(Table22).

Open in a separate window

Figure 2

Immunohistochemical analysis of Jab1/COPS5 in cancer tissues

(A) Pneumonia (left), small cell lung cancer (SCLC) (middle) and lung squamous carcinoma (right). (B) Breast hyperplasia (left), ductal breast cancer (middle) and lobular breast cancer (right).

Open in a separate window

Figure 3

Jab1/COPS5 is overexpressed in cancer tissues

(A) The percentages of positive Jab1/COPS5 cell in pneumonia and lung cancer (left), in lung cancer patients with non-responsive and responsive to chemotherapy (middle). (Right) Jab1/COPS5 gene expression in lung cancer patients with complete response or stable disease was plotted using the Oncomine gene expression tool (https://www.oncomine.com). The clinical data were downloaded from Oncomine Data Portal. (B) The percentages of positive Jab1/COPS5 cell in breast hyperplasia and breast cancer (left), in breast cancer patients without relapse or with relapse (middle). (Right) Jab1/COPS5 gene expression in breast cancer patients without relapse or with relapse was plotted using the Oncomine gene expression tool.

Table 1

Demographic and characteristics of Jab1 in lung cancer patients (n=88)

Variable	Total n=88	Low COPS5	High COPS5	p
Age
<60	39	13(33.3%)	26(66.7%)	0.249
≥60	49	12(24.5%)	37(75.5%)
Sex
Male	67	19(28.4%)	48(71.6%)	0.594
Female	21	6(28.6%)	15(71.4%)
Histological type
SCLC	31	13(41.9%)	18(58.1%)	0.035
NSCLC	57	12(21.1%)	45(78.9%)
Smoking
No	37	11(29.7%)	26(70.3%)	0.500
Yes	51	14(27.5%)	37(72.5%)
Stage
I	9	1(11.1%)	8(88.9%)	0.017
II	9	1(11.1%)	8(88.9%)
III	11	0(0%)	11(100%)
IV	59	23(39.0%)	36(61.0%)
Tumor size
T1	29	6(20.7%)	23(79.3%)	0.186
T2	44	13(29.5%)	31(70.5%)
T3	12	6(50.0%)	6(50.0%)
T4	3	0(0%)	3(100%)
LN metastasis
No	22	2(9.1%)	20(90.9)	0.016
Yes	66	23(34.8%)	43(65.2%)
Distant metastasis
No	37	5(13.5%)	32(86.5%)	0.007
Yes	51	20(39.2%)	31(60.8%)
Ki-67
<14%	2	1(50.0%)	1(50.0%)	0.490
≥14%	86	24(27.9%)	62(72.1%)
Hydrothorax
No	63	14(22.2%)	49(77.8%	0.039
Yes	25	11(40.0%)	14(56.0%)

Open in a separate window

Table 2

Demographic and characteristics of Jab1 in breast cancer patients (n=76)

Variable	Total n=76	Low Jab1	High Jab1	p
Age
<60	64	24(37.5%)	40(62.5%)	0.785
≥60	12	5(41.7%)	7(58.3%)
Histological type
Ductal	70	28(40.0%)	42(60.0%)	0.398
Lobular	6	1(16.7%)	5(83.3%)
Stage
I	9	4(44.4%)	5(55.6%)	0.211
II	46	14(30.4%)	32(69.6%)
III	21	11(52.4%)	10(47.6%)
Grade
G1	6	2(33.3%)	4(66.7%)	0.758
G2	50	18(36.0)	32(64.0%)
G3	20	9(45.0%)	11(55.0%)
Tumor size
T1	8	4(50%)	4(50%)	0.087
T2	46	13(28.3%)	33(71.7%)
T3	22	12(54.5%)	10(45.5%)
LN metastasis
Yes	63	25(39.7%)	38(60.3)	0.547
No	13	4(30.8%)	9(69.2%)
Distant metastasis
No	75	28(37.3%)	47(62.7%)	0.382
Yes	1	1(100.0%)	0 (0.0%)
Ki-67
<14%	37	15(40.5%)	22(59.5%)	0.677
≥14%	39	14(35.9%)	25(64.1%)
HER2
negative	36	12(33.3%)	24(66.7%)	0.411
positive	40	17(42.5%)	23(57.5%)
ER
negative	38	22(57.9%)	16(42.1%)	<0.001
positive	38	7(18.4%)	31(81.6%)
PR
negative	44	22(50.0%)	22(50.0%)	0.013
positive	32	7(21.9%)	25(78.1%)
Family history
Yes	22	12(54.5%)	10(45.5%)	0.060
No	54	17(31.5%)	37(68.5%)
Menopause
Yes	32	15(46.9%)	17(53.1%)	0.182
No	44	14(31.8%)	30(68.2%)

Open in a separate window

Association of Jab1/COPS5 expression with cancer patients

To investigate the therapeutic value of COPS5 in lung cancer patients, ROC curve analysis was performed. As shown in Figure ​Figure4,4, the AUC value of ROC curve was 0.811, suggesting that COPS5 could distinguish the chemo-responsive patients from non-responsive patients. Combing COPS5 and Ki-67 would further enhance the therapeutic value with a higher AUC of 0.865 (Figure ​(Figure4A4A).

Open in a separate window

Figure 4

The diagnostic significance of Jab1/COPS5 was analyzed via establishing ROC curve in cancer

The curve demonstrated that Jab1/COPS5 could discriminate between responder and non-responder of chemotherapy in lung cancer (A), or non-relapse and relapse in breast cancer (B).

The recrudescent value of COPS5 in breast cancer patients was also investigated. It was found the AUC value of ROC curve was 0.642, and the value (0.779) was further enhanced by combing COPS5 with Ki-67, indicating a potential role of COPS5 in distinguish primary tumors from relapse tumors (Figure ​(Figure4B4B).

Survival analysis

The mean follow-up time of breast cancer patients was 66 months (17–120 (OS) and range 5-109 (DFS)). To evaluate the prognostic influence of Jab1/COPS5 expression on breast cancer patients, we performed Kaplan-Meier analyses to compare grouped patients. The survival curves suggested that patients with high levels of COPS5 had a significant association with worse DFS and OS in breast cancer (OS, p=0.029; DFS, p= 0.044; Figure ​Figure5A5A).

Open in a separate window

Figure 5

Jab1/COPS5 predicts survival in lung and breast cancer

(A) Kaplan-Meier analyses of the association between Jab1/COPS5 protein expression and OS (left) and DFS (right) in breast cancer patients. (B) In vivo microarray data and Kaplan-Meier plots from OncoLnc database (http://www.oncolnc.org/) were used to assess correlations between Jab1/COPS5 gene expression and patient survival of lung cancer (left) and breast cancer (right).

We also investigated the relationship between COPS5 and survival of cancer patients in OncoLnc database. The online data was consistent with the IHC data, suggesting high levels of COPS5 was associated with worse survival in either breast cancer (p=0.00555; Figure ​Figure5B)5B) or lung cancer (p=0.0439; Figure ​Figure5B5B).

DISCUSSION

The common development of chemotherapy resistance is a major determinant of poor outcome in lung cancer [17, 18]. Despite increased knowledge about drug resistance mechanisms [19–21], reliable biomarkers for prediction of therapy response or resistance in lung cancer patients have been lacking. Our study demonstrates that Jab1/COPS5 is upregulated in lung cancer tissue compared with non-cancerous tissue. High expression of Jab1/COPS5 protein in biopsy specimens of untreated primary tumors was associated with chemotherapy resistance, defined as non-responder after chemotherapy.

Increased levels of Jab1/COPS5 have previously been associated with shorter overall survival of lung cancer [22]. Our results suggest that Jab1/COPS5 is associated with lung cancer progression, for example, stage, metastasis and malignant hydrothorax. We also found that Jab1/COPS5 was an independent prognostic factor for treatment response and could be useful to identify platinum based chemotherapy resistant tumors prior to first-line systemic therapy. The prognostic value of Jab1/COPS5 was confirmed at the mRNA level by our analysis of independent publicly available microarray datasets, demonstrating a significant correlation between higher Jab1/COPS5 gene expression and reduced overall survival of lung cancer patients.

Derived from our findings that high protein levels of Jab1/COPS5 in primary untreated tumor specimens might help to identify lung cancer patients in whom chemotherapy resistance and the potentially improved efficacy of Jab1/COPS5-inhibition may balance toxicities and side effects of anti-cancer treatment.

In recent years, a number of studies revealed an association between elevated Jab1/COPS5 level and tumor progression/poor prognosis in several types of cancer, including breast cancer [10]. However, it lacks studies of Jab1/COPS5 in cancer patients with regard to chemotherapy response and cancer relapse. This study on breast cancer patients reveals that the changes in Jab1/COPS5 predict the DFS and OS; patients who had higher Jab1/COPS5 level had shorter DFS and OS than patients who had lower Jab1/COPS5 level. In addition, multivariate survival analyses showed that Jab1/COPS5 and histological type are independent prognostic factors for poor survival in patients. Increased Jab1/COPS5 levels in breast cancer can indicate cancer relapse, although the reason for why the elevated Jab1/COPS5 levels correlates with tumor relapse remains to be elucidated.

Accumulating evidence suggests critical roles of Jab1/COPS5 in breast cancer progression. A number of studies have demonstrated that Jab1 is overexpressed in breast cancer [23–25]. The involvement of Jab1 in breast cancer is linked to the ER, EGFR and HER2 pathways by mediating downstream signalling that contributes to the tumour progression [24, 25]. It was also proposed that ligand-induced ERα degradation was regulated by Jab1 in hormone-dependent breast cancer cells [26]. Recently, Lu R et. al., revealed that high levels of Jab1 are significantly enriched in the ERα+ breast cancer patients, implicating that the amplification and overexpression of COPS5 may contribute to the ERα-related signalling and tumour progression [10]. However, the clinical outcomes of the TCGA breast cancer patients with high expression levels of Jab1 appeared to be poorer regardless of tumour subtypes because Jab1 expression was similar in luminal A and B tumors [10]. In addition, Jab1 also interacts with both the progesterone receptor (PR) and the steroid receptor coactivator 1 (SRC-1) and that it stabilizes PR-SRC-1 complexes [27]. In line with these reports, we found Jab1 is overexpressed in human breast cancers, and in particular the ERα-, or PR- subset of breast cancer. However, the relationship between these biochemical indicators need to be further studied.

Our previous studies demonstrated that Stat3 [28] and non-coding RNAs [29] contribute to overexpression of Jab1/COPS5 in cancer, and highly expressed Jab1/COPS5 regulates DNA damage response, which confers chemotherapy and radiotherapy resistance to tumor cells [30]. Our previous animal studies also confirmed that suppression of Jab1/COPS5 signaling induces radio- and chemo-sensitivity in nasopharyngeal carcinoma [30]. The above evidence could support some results of this study that breast cancer patients with different Jab1/COPS5 levels have different relapse incidence.

In this study, we reproduced similar conclusion as previous studies that the non-responders to platinum based chemotherapy, tended to have increased Jab1/COPS5. Therefore, Jab1/COPS5 appears to indicate the response to platinum based chemotherapy. In addition, the Jab1/COPS5 level was also found to be an independent determining factor for DFS and OS. With detecting the Jab1/COPS5 expression, we could classify cancer patients into good and poor prognosis groups, select patients who have a high risk of postoperative relapse, and need strict follow-up and intense treatment.

In conclusion, we demonstrated that therapeutic resistant lung cancers and relapse breast cancers are characterized by distinct elevated of Jab1/COPS5. High intratumoral levels of Jab1/COPS5 might serve as novel predictive biomarkers to identify chemotherapy resistance in lung cancer or to identify relapse in breast cancer. Those cancer patients with high expression of Jab1/COPS5 might particularly benefit from additional target therapy.

MATERIALS AND METHODS

SUPPLEMENTARY MATERIALS TABLE

Abbreviations

Footnotes

REFERENCES