Diagnostic value of human epididymis protein 4 in malignant pleural effusion in lung cancer

Abstract

OBJECTIVE:

This study aimed to assess the diagnostic value of human epididymis protein 4 (HE4) in the pleural effusion of lung cancer patients.

METHODS:

HE4 protein in the pleural effusion of 60 lung cancer patients was measured by electrochemiluminescence, in parallel with those from 56 patients with benign lung disease, and the association with malignant pleural effusion was evaluated.

RESULTS:

The level of HE4 in samples from lung cancer patients was significantly higher than the level for those with benign lung lesions ($P=$ 0.001) and patients with lung adenocarcinoma showed significantly higher levels of HE4 than those with squamous cell carcinoma and small cell carcinoma ($P=$ 0.002 and $P=$ 0.034, respectively). Using an optimal threshold of 652.2 pmol/L, the HE4 level distinguished malignant lung cancer from benign lesions with a sensitivity of 78.3% and a specificity of 75.0%. Moreover, the HE4 level differentiated adenocarcinoma from benign lesions with a sensitivity of 75.9% and a specificity of 85.7% when a threshold of 744.05 pmol/L was used. However, there was no significant difference in the 2 year survival rates of lung cancer patients with high and low HE4 concentrations in pleural fluid ($P=$ 0.882). In addition, there was no significant difference in HE4 levels between tuberculous and inflammatory pleural effusions ($P=$ 0.309).

CONCLUSION:

HE4 in the pleural fluid of lung cancer patients can be valuable in the diagnosis of malignant pleural effusion; however, it does not correlate with the prognosis of patients.

1. Introduction

Lung cancer has had the highest incidence and mortality of all malignancies in the world for the past few decades [1]. In China alone, 73,000 new cases are estimated to be diagnosed annually, accounting for one-third of the worldwide total [2]. Malignant pleural effusion (MPE) is a common symptom of advanced stages of cancer [3]. It is most commonly observed in lung cancer, with about one-third of the patients developing MPE [4]. At present, the most reliable approach to diagnose MPE is pathological tissue cytometry, but the positive rate is very limited [5]. Traditional tumor markers, such as carcinoembryonic antigen (CEA), carbohydrate antigen 125 (CA125), carbohydrate antigen 19-9 (CA19-9), and cell keratin 19 fragment (CYFRA21-1) are also used in the diagnosis of MPE, with a high specificity but a low sensitivity. Thus, these tumor markers are often used together to improve sensitivity [5]. A number of new tumor markers for the diagnosis of MPE have been identified in recent years, such as a high level of vascular endothelial growth factor (VEGF), which mediates MPE generation by increasing vascular permeability; however, the diagnostic efficacy needs to be improved [4].

Human epididymis protein 4 (HE4) is encoded by the gene WFDC2, which is located in the long arm of chromosome 6. It is a member of the protease inhibitor family and is usually expressed in the epithelia of normal reproductive and respiratory tracts. HE4 is frequently upregulated in gastric [6], endometrial [7], and peritoneal cancer [8] and in other malignancies. Of clinical importance, a high serum level of HE4 is an ideal marker for the early screening and differential diagnosis of epithelial ovarian cancer [9] as well as an independent prognostic marker for ovarian cancer patients [10].

HE4 is highly expressed in lung cancer [11], particularly in adenocarcinoma, which shows a significantly higher level of HE4 than other histopathological types. HE4 expression is often associated with a poor prognosis [12]. Thus, a high level of HE4 in the serum of patients has been used as a marker for the early diagnosis of lung cancer [13]. Importantly, increased HE4 in the serum also differentiates lung cancer from tuberculosis and pneumonia [14]. In addition, a high serum level of HE4 in patients with non-small cell lung cancer (NSCLC) correlates with tumor node metastasis staging of the tumor and a poorer 5-year survival than that for patients with low serum HE4. Therefore, the serum level of HE4 has been used as an independent prognostic marker for NSCLC patients [15]. In patients with advanced NSCLC, a higher serum level of HE4 is associated with a poorer response to chemotherapy and a poorer outcome, compared to patients with a lower HE4 level [16]. Similarly, using the HE4 level in the serum increases the sensitivity and specificity of the diagnosis of small cell lung cancer (SCLC), and improves the diagnostic efficiency when used in combination with other tumor indexes [17]. So far, research on the relationship between HE4 and lung cancer has been mainly focused on serology, and little is understood about the clinical significance of HE4 in pleural effusion. Therefore, in this study, HE4 in the pleural effusion of lung cancer patients was measured and its value in the diagnosis of MPE was determined.

2. Materials and methods

2.1. Patient recruitment

This study was approved by the Institute Research Ethics Committee of the Cancer Center of Nanjing Medical University, Jiangsu, China. Informed consent was obtained from individual participants before their pleural effusion samples were used.

From October 2015 to May 2017, 116 patients with pleural effusion, including 60 malignant and 56 benign cases, were recruited from the Affiliated Jiangning Hospital of Nanjing Medical University, China. Malignancy was confirmed by pleural effusion shedding cytology, fiberoptic bronchoscopy pathological biopsy, percutaneous lung puncture cytology, and other approaches, and patients with pleural effusion caused by other malignancies were excluded. The 56 benign cases were confirmed by pathology, cytology, or pathogenic microbiological examination.

2.2. Sample collection and analysis

Pleural effusion samples were collected prior to cancer treatment by conventional aseptic puncture, and 15 mL of each sample was centrifuged at 2,000 $\times$g for 10 min at 4${}^{\circ }$C. The supernatant was collected and stored at $-$80${}^{\circ }$C prior to analysis. The samples were processed using an Elecsys HE4 Kit purchased from Roche Diagnostic (Shanghai, China), and the concentration of HE4 was measured on a Cobas e601 analyzer (Roche, Basel, Switzerland) following the manufacturer’s instruction. CEA and CA 125 were also determined using Elecsys CEA Kits and Elecsys CA125 II Kits (Roche Diagnostic), respectively.

2.3. Statistical analysis

Data were processed using the Statistical Package for the Social Sciences (Version 23.0), and presented as mean $\pm$ standard deviation. The significance of differences between two groups was assessed by $t$-test, and the significance of variations between multiple groups was assessed by single factor analysis of variance (ANOVA).  0.05 was considered statistically significant.

3. Results

3.1. Clinicopathological features of the patients

Sixty lung cancer patients with MPE were recruited for this study, together with 56 patients with benign pleural effusion caused by tuberculosis and inflammation. The clinicopathological characteristics of all patients are summarized in Table 1. The 60 lung cancer patients comprised 43 men and 17 women aged between 27 and 91 years with an average age of 66.71 $\pm$ 3.12 years. Forty-one of these patients were diagnosed with adenocarcinoma while seven and 12 patients were diagnosed with squamous cell carcinoma and small cell carcinoma, respectively. The benign group consisted of 43 men and 13 women aged 18 to 79 years with an average age of 64.28 $\pm$ 1.5 years, who were diagnosed with pleural effusion caused by tuberculosis (31 cases) and inflammation (25 cases).

Table 1.

Clinicopathological characteristics of the patients ($n$ denotes the number of patients)

Diagnosis	$n$	Age		Sex		Smoking status		HE4
		$⩽$ 65	$>$ 65	Male ($n$)	Female ($n$)	Yes ($n$)	No ($n$)	(pmol/)L
Benign pleural effusion	56	32	24	43	13	26	30	486.47 $\pm$ 31.27
Tuberculous	31	19	12	23	8	15	16	457.67 $\pm$ 43.90
Pneumonia	25	13	12	20	5	11	14	522.19 $\pm$ 43.99
Lung cancer	60	18	42	43	17	28	32	1004.56 $\pm$ 56.98
Adenocarcinoma	41	12	29	28	13	20	21	1130.25 $\pm$ 66.99
Non-adenocarcinoma	19	6	13	15	4	8	11	752.42 $\pm$ 101.61
Squamous cell carcinoma	7	5	2	5	2	1	6	772.17 $\pm$ 100.57
Small cell carcinoma	12	1	11	10	2	7	5	710.67 $\pm$ 111.68

3.2. HE4 was elevated in MPE and associated with tumor histopathology

Electrochemiluminescence analysis revealed an average of 1004.56 $\pm$ 56.98 pmol/L of HE4 protein in MPE, more than two-fold higher than that in the benign group (486.47 $\pm$ 31.27 pmol/l) (Table 2, Fig. 1). Statistical analysis confirmed the difference was significant ( 0.001). The HE4 level in the pleural effusion of lung cancer patients was also associated with tumor histopathology. Patients with lung adenocarcinoma demonstrated significantly higher levels of HE4 in pleural effusion compared to those with squamous cell carcinoma ($P=$ 0.002) and small cell carcinoma ($P=$ 0.034) (Table 3, Fig. 2). No significant variation was found in the level of HE4 in pleural effusions from patients with squamous cell carcinoma and small cell carcinoma ($P=$ 0.716).

Table 2.

Baseline characteristics of patients with malignant and benign pleural effusion

Characteristics	Benign pleural effusion	Lung cancer	$P$
Age (y)	64.28 $\pm$ 1.5	66.71 $\pm$ 3.12	0.720
Male	43 (71.67%)	43 (76.79%)	0.812
HE4 (pmol/L)	486.47 $\pm$ 31.27	1004.56 $\pm$ 56.98	0.001

Figure 1.

The level of HE4 in the pleural effusion of lung cancer patients was significantly higher than that in pleural effusion originating from tuberculosis and inflammation ( 0.001).

Table 3.

Correlation of serum HE4 level with the clinicopathological characteristics of the patients

Characteristics	$n$	HE4 (pmol/L)	$P$
Age			0.230
$⩽$ 65	18	1109.62 $\pm$ 110.69
$>$ 65	42	959.53 $\pm$ 65.82
Gentle			0.181
Male	43	956.37 $\pm$ 66.48
Female	17	1126.44 $\pm$ 107.78
Pathological type			0.012
Adenocarcinoma	41	1130.25 $\pm$ 66.99
Squamous cell carcinoma	7	772.17 $\pm$ 100.57
Small cell carcinoma	12	710.67 $\pm$ 111.68
Smoke status			0.289
Yes	28	974.29 $\pm$ 71.36
No	32	1049.95 $\pm$ 95.11
CEA			0.002
Positive	42	1114.99 $\pm$ 65.20
($>$ 10 ng/L)
Negative	18	746.88 $\pm$ 89.75
( 10 ng/L)
CA125			0.804
Positive	49	1011.33 $\pm$ 65.49
($>$ 650.8 u/L)
Negative	11	974.40 $\pm$ 112.54
( 650.8 u/L)

Figure 2.

Patients with lung adenocarcinoma showed significantly higher levels of HE4 in pleural effusion than those with squamous cell carcinoma ( 0.001) and small cell carcinoma ( 0.05) and there was no significant difference between patients with squamous cell carcinoma and those with small cell carcinoma ($P>$ 0.05).

HE4 level in MPE was unrelated to the sex and age of the patients ($P=$ 0.230 and $P=$ 0.181, respectively). In the benign group, HE4 levels in tuberculous and inflammatory pleural effusion samples were 457.67 $\pm$ 43.90 pmol/L and 522.19 $\pm$ 43.99 pmol/L, respectively, and there was no significant difference between the two subgroups ($P=$ 0.309) (Table 4, Fig. 3).

Table 4.

HE4 level ($x\pm s$) in patients with benign lung lesions and clinicopathological parameters

Variables	$n$	HE4 (pmol/L)	$P$
Benign pleural effusion
Tuberculous	31	457.68 $\pm$ 43.90	0.309
pleural effusion
General	25	522.19 $\pm$ 43.99
inflammatory
pleural effusion
Age
$⩽$ 65	32	479.44 $\pm$ 45.35	0.798
$>$ 65	24	495.86 $\pm$ 41.86
Gender
Man	43	468.89 $\pm$ 34.78	0.311
Woman	13	544.63 $\pm$ 70.22
History of smoking
Yes	26	536.69 $\pm$ 49.42	0.136
No	30	442.96 $\pm$ 38.67

Figure 3.

There was no significant difference in HE4 level in pleural effusion resulting from tuberculosis and inflammation ($P>$ 0.05).

3.3. HE4 level differentiates malignant from benign pleural effusion

To evaluate the diagnostic efficiency of the HE4 concentration for MPE, a receiver operating characteristic (ROC) curve was plotted and is shown in Fig. 4. The area under the curve (AUC) for the HE4 level was 0.831 and the confidence interval (CI) was 0.758–0.904. When the HE4 level was 652.2 pmol/L, the sensitivity was 78.3% and the specificity was 75%. Considering CEA and CA125 are two commonly used biomarkers for lung cancer, we also examined the levels of CEA and CA125 in pleural effusion in parallel with that of HE4 to assess their association and compare their diagnostic efficiencies. As shown in Table 3, the concentration of HE4 was associated with that of CEA but not CA125. The sensitivities of CEA and CA125 for MPE were 70.0% and 81.67%, respectively, comparable to that of HE4.

Figure 4.

ROC curve of HE4 to predict malignant pleural effusion.

3.4. Diagnostic value of HE4 for lung adenocarcinoma

To further evaluate the value of HE4 in the diagnosis of lung adenocarcinoma, the ROC for lung adenocarcinoma against benign pleural effusion was plotted (Fig. 5), and the AUC value was found to be 0.883. When the HE4 level was 744.05 pmol/L, the sensitivity was 75.9 % and the specificity was 85.7%. To assess the correlation of the HE4 level with the prognosis, lung cancer patients were classified into two subgroups using the HE4 value of 652.2 pmol/L at the maximum Youden index as a threshold and the survival rates of the groups were compared (Fig. 6). The survival periods for the patients with low and high HE4 were 13.33 $\pm$ 2.70 months and 12.49 $\pm$ 1.92 months, respectively. There was no statistically significant difference in the survival time of the two subgroups of patients.

Figure 5.

ROC curve of HE4 to predict lung adenocarcinoma.

Figure 6.

Kaplan-Meier analysis of the accumulative survival of lung cancer patients with high and low levels of HE4 in pleural effusion, $P>$ 0.05.

4. Discussion

It remains challenging to diagnose MPE and common biomarkers are still elusive. This necessitates further efforts toward the identification of markers or marker panels that can improve the sensitivity and specificity of the diagnosis of MPE. In this study, the level of HE4 in pleural effusion resulting from lung cancer was analyzed in comparison with those from tuberculosis and inflammation. The association of the HE4 level with the malignancy of the pleural effusion and the histopathological features of lung cancer was assessed to evaluate the potential of this protein in the diagnosis of lung cancer.

Our results demonstrated a significant increase of HE4 in MPE, differentiating it from pleural effusion resulting from tuberculosis and inflammation. This is mirrored by its elevation in the serum of lung cancer patientss [18, 19]. The results suggest that the HE4 level in pleural effusion can be used as a biochemical marker to identify patients with MPE. Plotting the ROC curve for malignant against benign pleural effusion demonstrated an AUC value of 0.831. When an HE4 level of 652.2 pmol/L was used, the sensitivity and specificity were 78.3% and 75%, respectively. Noticeably, this threshold is considerably lower than the previously recommended value [20]. This is not surprising as the cutoff values for MPE can be variable [20].

Furthermore, patients with lung adenocarcinoma demonstrated significantly higher levels of HE4 in pleural effusion than those with other histopathological types of lung cancer, highlighting the correlation of HE4 level in pleural effusion with the histopathology of lung cancer. This result is in accord with previous observations using serum samples, which demonstrated a significant correlation of serum HE4 level with histological type, TNM stage, and lymph node metastasis [16]. The ROC curve of lung adenocarcinoma against benign pleural effusion showed an AUC value of 0.881. At an HE4 level of 744.05 pmol/L, the sensitivity was 75.9% and the specificity was 85.7%, suggesting using the HE4 level in pleural effusion could significantly improve the diagnosis of lung adenocarcinoma. On the contrary, in pleural effusion caused by tuberculosis and inflammation, no significant difference was found in HE4 level. In addition, in pleural fluid with benign and malignant origins, the levels of HE4 are unrelated to sex, age, and smoking history.

In this study, the efficiency of the HE4 level in pleural effusion in predicting the prognosis of lung cancer was also investigated. Kaplan-Meier survival analysis showed no significant differences in the survival of patients with high and low levels of HE4. This is inconsistent with the results of a previous study using patient serum in which a higher HE4 level ($>$ 140 pmol/L) predicted a poorer overall survival [16]. This is likely due to the clinico-pathological features of patients recruited in this study. There were more patients with advanced stage lung cancer in this study, in which all patients had developed pleural effusion.

In summary, our data suggest that patients with MPE have a higher level of HE4 than those with pleural effusion from tuberculosis and inflammation. In particular, patients with lung adenocarcinoma tend to have a higher level of HE4 in pleural effusion than those with squamous cell carcinoma and small cell carcinoma. Our results suggest that HE4 in pleural effusion can be employed as a biomarker in the diagnosis of malignant lung cancer.