KPNA2 is a potential diagnostic serum biomarker for epithelial ovarian cancer and correlates with poor prognosis

Abstract

This study aimed to determine whether serum karyopherin alpha 2 levels can be used as a diagnostic biomarker for epithelial ovarian carcinoma. Karyopherin alpha 2 protein was detected by enzyme-linked immunosorbent assay in serum samples from 162 epithelial ovarian carcinoma patients and 48 healthy controls. Serum karyopherin alpha 2 levels in epithelial ovarian carcinoma patients were significantly higher than in healthy controls (p < 0.001). When a karyopherin alpha 2 serum level of 2.52 µg/mL was used as a cut-off, the sensitivity and specificity of the assay for diagnosing epithelial ovarian carcinoma were 71.4% and 81.2%, respectively. High serum karyopherin alpha 2 levels (>485 µg/mL) correlated with International Federation of Gynecology and Obstetrics stage (p < 0.0001), lymphatic metastasis (p = 0.045), overall survival (p = 0.001), and disease-free progression (p = 0.006). Serum karyopherin alpha 2 represents a potential diagnostic biomarker for epithelial ovarian carcinoma.

Introduction

Epithelial ovarian carcinoma (EOC) accounts for 90% of all ovarian malignancies and is the most fatal gynecological malignancy in women worldwide—21,980 new cases and 14,270 deaths were estimated to occur in 2014. ¹ Late diagnosis of the condition is often associated with an advanced stage and high rate of metastasis. Therefore, understanding the molecular pathogenesis of EOC and discovering molecular biomarkers to facilitate its early detection are crucial for improving patient survival.

Deregulation of the cellular transport machinery often occurs in tumors. In accordance with this phenomenon, elevated expression of karyopherin alpha 2 (KPNA2), a protein involved in nucleocytoplasmic transport, has been shown in a variety of malignancies.^2–9 KPNA2 expression was also shown to be higher in the serum of lung cancer ⁵ and esophageal squamous cell carcinoma, ¹⁰ suggesting that KPNA2 can be secreted into the serum. Since KPNA2 overexpression is a common phenomenon in different types of cancer, it is possible that KPNA2 serum levels may also be upregulated in other types of cancer such as EOC.

We previously showed that KPNA2 is overexpressed in human EOC cell lines and tissues. ¹¹ Furthermore, KPNA2 overexpression correlates with poor prognosis in both EOC ¹¹ and ovarian malignant germ cell tumors. ¹² However, the serum levels of KPNA2 in EOC patients, and their correlation with patient prognosis, had not been investigated. Therefore, in this study, we analyzed the serum KPNA2 levels in EOC patients and their correlation with prognosis and clinicopathological features of the disease.

Materials and methods

Clinical specimens

Before treatment, serum samples were collected from 162 EOC patients (median age = 50 years; range = 10–80 years) and 48 healthy control volunteers (median age = 45 years; range = 23–67 years) from Sun Yat-sen University Cancer Center (SYSUCC, Guangzhou, China) between January 2008 and October 2012. The inclusion criteria are as follows: the pathological diagnosis of ovarian cancer; serum samples were collected before operation/radiotherapy/or chemotherapy; and have detailed clinical record data. The tumor stage was classified according to the International Federation of Gynecology and Obstetrics (FIGO) staging system. All patients were regularly followed up, with a median observation period of 46 months (range = 2–158 months). Other clinical and pathological variables are shown in Table 1. Control subjects did not display any type of cancer for at least 6 months after serum sample collection. Venous blood (10 mL) was collected and centrifuged at 4°C for serum collection. Serum samples were stored at −80°C until further analysis. The study was approved by the medical ethics committee of SYSUCC.

Table 1.

The association of the concentration of KPNA2 in serum with clinicopathological features in epithelial ovarian carcinoma.

Clinicopathological	Number of cases	Concentration of KPNA2				p value
		Low	%	High	%
Age (years)
⩽45	49	35	71.4	14	28.6
>45	112	74	66.1	38	33.9	0.504
FIGO stage
I	31	27	87.1	4	12.9
II	26	23	88.5	3	11.5
III	69	47	68.1	22	31.9
IV	17	3	17.6	14	82.4	<0.0001
Differentiation
Well	7	6	85.7	1	14.3
Moderate	22	17	77.3	5	22.7
Poor	74	47	63.5	27	36.5	0.278
Lymphoid node
Negative	128	91	71.1	37	28.9
Positive	34	18	52.9	16	47.1	0.045
ALP (µg/L)
⩽110	143	98	68.5	45	31.5
>110	17	11	64.7	6	35.3	0.749
LDH (µg/L)
⩽245	101	73	72.3	28	27.7
>245	59	36	61.0	23	39.0	0.140

ALP: alkaline phosphatase; LDH: lactic dehydrogenase; FIGO: International Federation of Gynecology and Obstetrics.

Bold indicates significant values.

Enzyme-linked immunosorbent assay

The KPNA2 and CA125 levels in human serum were measured using a commercially available enzyme-linked immunosorbent assay (ELISA) kit (USCN Life Science Inc., Wuhan, Hubei, China), according to the manufacturer’s instructions. Briefly, 100 µL of diluted serum sample (1:500 dilution) was added to each well and incubated at 37°C for 2 h. Then, 100 µL reagent A was added for an additional hour. After five washes, reagent B was added for 30 min. The amount of proteins was determined by adding 3,3′,5,5′-tetramethylbenzidine (TMB) substrate, and the plate was incubated at 37°C to allow for color development. The absorbance was measured at 450 nm using a Model 680 microplate reader.

Statistical analysis

Statistical analysis was performed using the SPSS software package (version 16.0; SPSS, Chicago, USA). The relationship between the concentration of KPNA2 in serum and clinicopathological characteristics was assessed using Pearson’s chi-squared test. Survival curves were estimated by the Kaplan–Meier method. The log-rank test was used to estimate the statistical differences between survival curves. A p value of <0.05 was considered statistically significant.

Results

Serum KPNA2 levels are higher in EOC patients than healthy controls

We assessed the serum levels of KPNA2 in EOC patients (n = 162) and healthy controls (n = 48) by ELISA. The serum concentrations of KPNA2 ranged from 0.13 µg/mL to 864.67 µg/mL in EOC patients and from 0 µg/mL to 21.52 µg/mL in healthy controls. KPNA2 levels were significantly higher in EOC patients compared to healthy controls (p < 0.001; Figure 1(a)). Serum KPNA2 levels showed promising capacity for use as a diagnostic marker: when a KPNA2 serum level of 2.52 µg/mL was used as the cut-off value, its sensitivity and specificity as a diagnostic marker for EOC were 71.4% and 81.2%, respectively (Figure 1(b)). In addition, using a combination of serum KPNA2 and CA125 (a known tumor biomarker for ovarian cancer) displayed higher diagnostic capacity than CA125 alone (area under the curve (AUC) = 0.96 vs AUC = 0.89, respectively; Figure 1(b)).

Figure 1.

(a) Comparison of serum KPNA2 levels in EOC patients and healthy controls. Serum KPNA2 levels measured by ELISA in EOC patients (n = 162) and healthy volunteers (n = 48). KPNA2 levels were significantly higher in EOC patients compared to healthy controls (p < 0.001). (b) Diagnostic capacity of KPNA2, CA125 (carbohydrate antigen 125, a currently used tumor biomarker), and combination of both (KPNA2 with CA125) as determined using receiver operating characteristic (ROC) curve analysis.

Serum KPNA2 levels correlate with FIGO stage, lymphatic metastasis, and prognosis in EOC patients

To identify the clinical relevance of serum KPNA2 in EOC, we examined the correlation between serum KPNA2 concentration and clinicopathological parameters such as age, FIGO stage, tumor differentiation, lymphatic metastasis, alkaline phosphatase (ALP) levels, and lactic dehydrogenase (LDH) levels (Table 1). Based on receiver operating characteristic (ROC) curves, the 162 EOC patients were further classified into a high serum KPNA2 group (n = 52) and a low serum KPNA2 group (n = 110) using a cut-off value of 485 µg/mL. High serum KPNA2 levels (i.e. >485 µg/mL) were significantly correlated with FIGO stage (p < 0.001; Figure 2(a)) and lymphatic metastasis (p = 0.045; Figure 2(b)), but not other clinical characteristics. In addition, EOC patients with high serum KPNA2 levels (>485 µg/mL) showed significantly poorer overall survival (p = 0.001; Figure 3(a)) and disease-free survival (p = 0.006; Figure 3(b)) than those with low serum KPNA2 levels.

Figure 2.

(a) Serum KPNA2 levels according to FIGO stage (I–IV). KPNA2 expression was significantly higher in tumors of higher FIGO stage than that in tumors of lower FIGO stage. (b) Serum KPNA2 levels in tumors with or without lymphatic metastasis. KPNA2 expression was higher in patients with lymphatic metastasis than that in patients without lymphatic metastasis.

Figure 3.

Kaplan–Meier survival curve analysis in EOC patients with high serum KPNA2 levels and low serum KPNA2 levels. EOC patients with high serum KPNA2 levels showed significantly poorer (a) overall survival (p = 0.001) and (b) disease-free survival (p = 0.006) than those with low serum KPNA2 levels.

Discussion

Epithelial ovarian cancer is one of the major causes of death from gynecological malignancies. Although new treatment modalities have prolonged survival during recent decades, the overall prognosis has not improved. ¹³ The symptoms, such as abdominal pain, are often non-specific until late in disease progression, leading to delayed diagnosis. This study demonstrates that measurement of serum KPNA2 level is a promising tool for preoperative testing to predict patients with EOC. We selected to study KPNA2 as a potential biomarker for EOC as it is upregulated in EOC tissue and can also be secreted into conditioned medium. As blood flows in and out of tumors and circulates tumor-specific protein profiles, the serum is the ideal media for finding biomarkers.

Currently established tumor serum markers for the diagnosis and monitoring of EOC include carbohydrate antigen 125 (CA125) and human epididymis protein 4 (HE4).^14–16 Cancer antigen 125 (CA125, also known as carbohydrate antigen 125), the most widely used serum biomarker for EOC, was discovered over 30 years ago. The major disadvantages of CA125 include poor sensitivity and specificity for EOC, especially for diagnosis of early-stage disease. Furthermore, the use of only a single marker may be inefficient due to low sensitivity or specificity. In this study, we found that the serum concentration of KPNA2 was significantly higher in EOC patients than in healthy controls, which suggests that the upregulation of KPNA2 in tumor tissue is responsible for the serum level of KPNA2. The diagnostic sensitivity of serum KPNA2 for EOC was lower than CA125 among all EOC patients, but combination of serum KPNA2 and CA125 displayed higher diagnostic specificity and accuracy than CA125 alone. Furthermore, we found that most early-stage EOC patients have higher serum KPNA2 levels than healthy controls. Therefore, we believe that serum KPNA2, alone or in combination with CA125, would be useful for the diagnosis of EOC.

The exact pathway responsible for KPNA2 secretion is currently unknown. KPNA2 has been reported to interact with importin beta 1 (KPNB1), ¹⁷ a nucleocytoplasmic shuttle protein detected in the exosomes of cancer cells. ¹⁸ Thus, KPNA2 and KPNB1 may interact in cancer cells, with the resulting KPNA2–KPNB1 complex secreted via exosomes. This might represent a plausible explanation as to why KPNA2 is detectable in the media secreted from cancer cells. Indeed, nucleocytoplasmic transport mechanisms are important for many key cellular processes such as gene expression, cell-cycle progression, and signal transduction. Gathering evidence suggests that these mechanisms contribute to malignant cell transformation, highlighting the potential of these transport proteins as therapeutic targets. ¹⁹

To the best of our knowledge, this is the first study in which serum levels of KPNA2 were analyzed in a relatively large group of untreated patients with EOC. Our findings suggest a potential role for KPNA2 as a tumor marker in the diagnosis of EOC, particularly in combination with CA125 as part of a new diagnostic panel. Further research to validate KPNA2 as a useful biomarker for EOC diagnosis is underway.