Diagnostic utility of LunXmRNA in peripheral blood and pleural fluid in patients with primary non-small cell lung cancer

Lung cancer is one of the leading causes of cancer death and has become an increasingly urgent worldwide health problem. Progress in lung cancer treatment is hampered by a lack of diagnostic markers useful in clinical practice. Tumor markers, including carcinoembryonic antigen (CEA), neuron-specific enolase (NSE), squamous cell carcinoma (SCC) antigen, cytokeratin 19 (CK19), vascular endothelial growth factor-C (VEGF-C), heterogeneous ribonuclear proteins A2/B1 (hnRNP A2/B1), muc1, BJ-TSA-9, KS1/4 and lung-specific X protein (LunX), have been investigated for their putative diagnostic and prognostic value for lung cancer [1‐7]. On the basis of RT-PCR analysis, CEA mRNA in blood cells or in lymph nodes and CK19 mRNA in mediastinal lymph nodes have been suggested as promising tools for the detection of micrometastatic cells in patients with lung cancer [8‐11]. KS1/4 was shown to be the most sensitive marker for detecting metastatic NSCLC in mediastinal lymph nodes using real-time RT-PCR [6]. Determining CEA and NSE in pleural fluid could enhance the diagnostic yield for malignant effusion associated with lung cancer [12]. Additionally, accumulating evidence suggests that hnRNP B1 expression may be useful for the early diagnosis of lung cancer, provided that expression levels can be accurately quantified [5]. LunX, a novel human lung-specific gene, has been reported to be a superior diagnostic marker for the detection of micrometastases in lymph nodes and peripheral blood of NSCLC patients [1, 6, 13]. Although lots of studies have provided suggestive results, a definitive assessment of the relative value of these molecular markers for lung cancer is lacking.

A demonstration of the diagnostic utility of these various tumor markers requires a detailed, direct comparison using reliable, sensitive methodologies. Quantitative real-time RT-PCR is a development of the RT-PCR procedure, which is simple, rapid and automated. Most importantly, real-time RT-PCR analysis can yield accurate estimates of gene expression levels, differentiating between baseline levels of gene expression in normal tissue and increased levels in cancer cells [14, 15]. Molecular diagnosis using RT-PCR technique can detect tumor marker-expressing cells undetectable by other means in patients with localized or metastatic cancer, and may offer the most effective solution for detecting micrometastases at the molecular level in various types of cancer patients [16].

The purpose of this study was to evaluate the known molecular markers,LunX, CK19, CEA, VEGF-C and hnRNP A2/B1, for their expression in lung cancer cells in peripheral blood and pleural fluid using real-time RT-PCR, with the ultimate goal of establishing a more reliable molecular diagnostic method as an adjunct to clinical decision-making.

The current staging evaluation for lung cancer is based on the presence or absence of disease in lymph nodes (hilar and mediastinal) and distant organs (principally bone, brain, adrenal glands, and liver) [19]. Approximately 35% of patients are diagnosed at early stage and, as such, are candidates for curative lung resection. However, 50% of these patients will develop metastases and die from their disease. Furthermore, about one fourth of patients at the earliest stage of NSCLC (pathologically confirmed stage I) die of tumor recurrence after radical surgery, indicating that undetected metastases are present at the time of surgery and demonstrating that conventional staging techniques lack the sensitivity necessary to properly characterize patients [20, 21]. Cancer cells can be released from a primary site and spread via the bloodstream to form a micrometastatic deposit in distant organs [1]. However, due to their extremely low concentration, these circulating tumor cells in peripheral blood are difficult to detect. Thus, developing sensitive and specific detection methods for cancer cells in peripheral blood may have important diagnostic, prognostic and therapeutic implications [22].

In this study, we have developed quantitative real-time RT-PCR procedures to detect potentially diagnostic tumor markers (Figure 1). The detection of metastatic cancer cells by RT-PCR is possible because cancer cells continue to express genetic markers specific to the tissue from which they originate, but which are not normally expressed in tissue compartments that frequently harbor metastatic foci [23, 24]. Using this RT-PCR approach, which is ideal for the detection of genes expressed at low levels [13], we have assessed the expression of the known molecular markers,LunX, CK19, CEA, VEGF-C and hnRNP A2/B1, in lung cancer cells in peripheral blood and pleural fluid. Although LunX has been previously reported to be the most sensitive marker among the five genes LunX, muc1, KS1/4, CEA and CK19, for detecting circulating NSCLC cells by real-time RT-PCR in a study distinguishing patients with NSCLC from healthy volunteers, the specificity of LunX for lung cancer cells has not been tested [13]. Ours is the first study to directly compare the expression of LunX with other biomarkers in peripheral blood and pleural fluid, not only from NSCLC patients but also from patients with other epithelial cancer or benign lung disease and healthy volunteers. We have found that LunX mRNA is the most specific marker for lung cancer cells in peripheral blood (Figure 2, Table 4) and pleural fluid (Figure 4, Table 6). Compared with LunX mRNA, CK19 and CEA mRNA were over-expressed in other epithelial cancers, such as breast and esophagus cancer (Figure 2, Table 4), thus limiting their specificity for detection of lung cancer cells in peripheral blood. Because of lack of specificity, VEGF-C and hnRNP A2/B1mRNA were found to be similarly ineffective as genetic markers for lung cancer cells in our quantitative real-time RT-PCR assay (Figure 2, Table 4).

Further, it was demonstrated that when one lung cancer cell (A549 cell) was added into 3 ml peripheral blood, LunX mRNA could be detectable as a positive case in our established method (data not shown). For NSCLC patients, the positive detection rate of LunX mRNA in peripheral blood was high, almost as high as that of CK19 mRNA, and much higher than that of CEA mRNA (Table 4). Using RT-PCR, KS1/4 was previously reported to be the most sensitive marker among CEA, CK19, KS1/4, LunX, muc1 and PDEF for the detection of metastatic NSCLC in mediastinal lymph nodes, and LunX was with the second highest sensitivity distinguishing lung cancer from lung benign disease [6]. However,KS1/4 encodes a glycoprotein that is expressed on epithelial cells and is also present on epithelial cancers, thus, like CEA and CK19, KS1/4 is not specific to lung tissue. Another novel tumor-specific gene BJ-TSA-9 was reported to be a marker for circulating cancer cells in lung cancer patients, but BJ-TSA-9 alone was not sensitive enough to detect disseminated cancer cells in peripheral blood, and a combination of BJ-TSA-9 with LunX and SCC was required [7]. BJ-TSA-9 also suffers from the same tissue specificity problem that plagues CEA and CK19. In contrast, expression of the human LunX gene is lung-specific, and mRNA could be detected at a concentration of 10^-4 μg cancer RNA in 1 μg normal lymph node RNA [1].

A malignant pleural effusion may be the initial presentation of cancer in 10–50% of patients [25]. Cytology is the standard method for the diagnosis of malignant effusion, but the sensitivity of cytology was not good enough [26]. Although an aggressive diagnostic technique thoracoscopy can be used to establish the diagnosis with a higher sensitivity (~90%), this procedure may not be available at all facilities and/or may be too invasive for many patients [27]. The evaluation of tumor markers in pleural fluid thus represents an alternative method for establishing the diagnosis of malignant pleural effusion. In this study, quantitative real-time RT-PCR was performed, for the first time, on pleural fluid. As was the case with peripheral blood, LunX mRNA was the most specific marker in malignant pleural fluid, showing a high positive detection rate in lung cancer (13 of 14, 92.9%), compared with the other gene markers CK19, CEA, VEGF-C and hnRNP A2/B1 mRNA (Figure 4, Table 6). Because determining the presence of circulating cancer cells in the peripheral blood of NSCLC patients is significant for early disease diagnosis and clinical therapy, a test for LunX mRNA that is able to reveal small amounts of lung cancer cells in peripheral blood with high specificity and sensitivity would be a great benefit in the clinical management of lung cancer. The potential value of differential LunX mRNA expression in pleural fluid for diagnosing malignant effusion reinforces this view.

Recent mass spectroscopy studies indicate that the human LunX gene product is also expressed in normal adult nasal lavage fluid. Further, this expression may be up-regulated in response to certain airway irritants, such as cigarette smoke and dimethylbenzylamine [28, 29]. Nasal lavage fluid contains a large number of proteins that altogether comprise a potential source for detecting and characterizing biochemical alterations associated with airway diseases. In the present study, nasal lavage fluid samples were not investigated, so the element of smoking was not addressed in comparisons between NSCLC patients, other epithelial cancer patients, benign lung disease patients and healthy volunteers. Furthermore, as shown in Figure 2 and Table 4, LunX gene was not detectable in the peripheral blood of healthy volunteers, whether smokers or non-smokers. To date, the functional role of LunX protein remains unknown; however, LunX mRNA expression is known to be significantly enhanced in NSCLC tumors compared with corresponding cancer-free lung tissues [1]. In our study, we found that the expression level of LunX mRNA in peripheral blood correlated with the pathologic stage of NSCLC (Table 5, Figure 3). The more severe the disease was, the higher the expression level of LunX mRNA. Thus, the expression level of LunX mRNA in peripheral blood might be a valuable tool for staging NSCLC patients clinically. Furthermore, the expression of LunX mRNA in peripheral blood was sensitive to be influenced by the treatment of NSCLC patients (Figure 5), indicating that LunX mRNA expression might be associated with lung cancer progression.

In this study, we demonstrate that the detection of LunX mRNA in peripheral blood and pleural fluid by quantitative real-time RT-PCR provides a specific and sensitive indication of the presence of lung cancer cells. To our knowledge, LunX mRNA is the most specific NSCLC gene marker currently identified, and has tremendous clinical potential as an NSCLC diagnostic tool. Because LunX mRNA levels correlate with clinical severity and change in response to treatment protocols, this marker may prove valuable for NSCLC patients in the clinical decision-making process.