| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Imaging, Diagnosis, Prognosis |
Authors' Affiliations: Departments of 1 Surgery, 2 Biostatistics, and 3 Pathology, University of Pittsburgh; and 4 University of Pittsburgh Cancer Institute, Pittsburgh, Pennsylvania
Requests for reprints: Tony E. Godfrey, Mount Sinai School of Medicine, Box 1079, One Gustave L. Levy Place, New York, NY 10029. Phone: 212-241-8238; Fax: 212-241-5432; E-mail: tony.godfrey{at}mssm.edu.
 |
Abstract |
|---|
 Top Abstract Materials and MethodsResultsDiscussionReferences |
|---|
Improved histologic examination using immunohistochemistry and/or serial sectioning is certainly one way that lymph node staging can be improved. However, despite the fact that most pathologists would acknowledge the improved sensitivity of this approach, the increased labor and cost involved in examining 10 to 20 lymph nodes removed in a typical lung cancer resection remains a limiting factor for clinical acceptance. Furthermore, as the size of a metastatic focus gets smaller, at least one group has reported alarming discordance in the histologic interpretation by different pathologists (9). We believe that well-designed, appropriately controlled molecular assays, such as quantitative reverse transcription-PCR (qRT-PCR), may eventually prove to be a useful adjunct to pathologic examination, resulting in a more sensitive, standardized, and objective approach to lymph node analysis. qRT-PCR is capable of analyzing much more of each lymph node than routine pathology, thus improving sampling, and qRT-PCR is at least as sensitive as immunohistochemistry when using the appropriate markers. Furthermore, qRT-PCR is amenable to automated processing and analysis, therefore potentially reducing both the manual labor involved and the subjectivity of analysis. When the overall patient care and cost is considered, molecular staging of lymph nodes, with directed treatment, may therefore prove superior to current practices.
Before this hypothesis can be tested, the appropriate markers for qRT-PCR in NSCLC need to be identified, rigorously validated, and then applied to pN0 patients to show prognostic ability. Automated processes can then be developed and used in clinical trials to evaluate treatment strategies in patients staged by either routine or molecular methods. In this report, we have identified and analyzed expression of 42 genes with potential as markers of lymph node metastasis in NSCLC. From this set, we have identified and validated three genes [tumor-associated calcium signal transducer gene 1 (TACSTD1), pemphigus vulgaris antigen (PVA), and surfactant protein B (SFTPB)] that, when used in combination, provide 100% correlation with histologic examination, including serial sectioning and immunohistochemistry. Ongoing studies are evaluating these genes for detection and clinical relevance of occult disease in NSCLC patients.
|
Materials and Methods |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
Tissues and pathologic evaluation
Tissue specimens were obtained from tissue banks at the University of Pittsburgh Medical Center through institutional review board–approved protocols. One half of each lymph node, and a small piece of each primary tumor, were snap frozen in liquid nitrogen and later embedded in OCT for frozen sectioning and RNA isolation. The remaining tissue was used for routine diagnostic pathology evaluation. Twenty 5-µm sections were cut from each frozen tissue piece for RNA isolation. In addition, sections were cut and placed on slides for H&E and immunohistochemical analysis at the beginning, middle (between the 10th and 11th sections for RNA), and end of the sections for RNA isolation. All three H&E slides from each specimen underwent pathologic review to confirm presence of tumor, percentage of tumor, and to identify the presence of any contaminating tissues (such as lung tissue on a lymph node). All of the unstained slides were stored at –20°C. Immunohistochemistry evaluation was done using the AE1/AE3 antibody cocktail (DAKO, Carpinteria, CA), and Vector Elite ABC kit and Vector AEC Chromagen (Vector Laboratories, Burlingame, CA). Immunohistochemistry was used as needed to confirm the H&E histology. Only tissues on which diagnostic pathology and frozen section pathology were concordant were included in the screening tissue sets.
Marker screening
The screening was conducted in two phases. All potential markers entered the primary screening phase and expression was analyzed in six primary NSCLC and 10 benign lymph nodes obtained from patients without cancer (five RNA pools with two lymph node RNAs per pool). Markers that showed good characteristics for lymph node metastasis detection (i.e., high expression in primary tumors and low expression in benign lymph nodes) passed into the secondary screening phase. The secondary screen consisted of expression analysis on 21 primary tumors, 21 histologically positive lymph nodes, and 21 benign lymph nodes from 21 patients without cancer (called the screening tissue set). All tumors and positive lymph node samples were from independent patients.
Marker validation
To externally validate the classification accuracy of markers tested in the secondary screen, an independent, validation set of lymph nodes was analyzed. The validation set consisted of 32 benign lymph nodes from independent patients without cancer and 38 pathology-positive lymph nodes obtained from 25 patients undergoing surgery for NSCLC. Diagnostic pathology and frozen section pathology were concordant on all nodes in the validation set.
Analysis of lymph nodes from pN0 NSCLC patients
Three markers were selected from the screening and used to identify occult metastasis in 462 nodes from 68 patients with NSCLC (mean 6.8 nodes per patient, median six nodes per patients). Clinical information for these 68 patients is shown in Table 1
. Expression of markers was determined as described below and was compared with expression in 30 benign nodes previously analyzed in the screening and validation phases. Cancer patient lymph nodes were considered positive for occult metastasis if expression of any marker was higher than the maximum expression observed in the benign nodes.
|
CT methods previously described (11) and with ß-glucuronidase as the endogenous control gene. All assays were designed for use with 5' nuclease hybridization probes although the primary screening was done using SYBR green quantification to save cost. Assays were designed using the ABI Primer Express Version 2.0 software and, where possible, amplicons spanned exon junctions to provide cDNA specificity. In addition, all primer pairs were tested empirically for amplification from 100 ng genomic DNA and primers were redesigned when necessary. With the exception of CK19, cDNA-specific primer sets were successfully identified for all genes in this study. Finally, optimal annealing temperature was determined by testing primers for generation of a single band on gels using cDNA templates and annealing temperatures of 60°C, 62°C, and 64°C. Further details describing our methods for primer design and testing have recently been published (12). PCR efficiency was estimated using SYBR green quantification before use in the primary screen. Further optimization and more precise estimates of efficiency were done with 5' nuclease probes for all assays used in the secondary screen. Briefly, reactions were done with a probe concentration of 200 nmol/L and a 60-second anneal/extend phase at 60°C for ß-GUS, CK19, and SFTPB; 62°C for CK7, PVA, and squamous cell carcinoma antigen (SCCA) 1/2; and 64°C for carcinoembryonic antigen (CEA), lung-specific X protein (LUNX), and TACSTD1. The sequences of primers and probes (purchased from IDT, Coralville, IA) for genes evaluated in the secondary screen are listed in Supplementary Table S1. The primer sequences for markers used in the primary screen will be provided upon request. The details of RNA preparation, reverse transcription, and qRT-PCR have been described in our previous studies (13, 14).
Data analysis
In the primary screen, data from the melt curve was analyzed using the ABI Prism 7700 Dissociation Curve Analysis 1.0 software (Applied Biosystems, Foster City, CA). The first derivative of the melting curve was used to determine the product Tm as well as to establish the presence of the specific product in each sample. In general, samples were analyzed in duplicate PCR reactions and the average Ct value was used in the expression analysis. However, in the secondary screen, triplicate reactions were done for each individual benign node and the lowest Ct value was used to calculate relative expression and thus obtain the highest (most conservative) estimate of background expression for the sample.
Statistical analysis
Generation of prediction rules. Six markers that passed the secondary screen were evaluated individually and in combination with other markers. The characteristics used to evaluate markers were sensitivity, specificity, classification accuracy, and the area under the receiver operating characteristic curve. For individual markers, a cutoff value was determined that maximized the classification accuracy (proportion of lymph nodes correctly classified). In cases where classification accuracy was 100%, the cutoff was set at the midpoint between the highest expressing negative node and the lowest expressing positive node.
Validation of prediction rules. Internal validation of prediction rules was conducted by leave-one-out cross-classification in which each case is classified by a decision rule formulated from the remaining cases. The average sensitivity, specificity, and classification accuracy over all cases is reported. External validation was conducted by applying the most accurate cutoff from the secondary screen (n = 42) to a new set of benign and histologically positive lymph nodes (n = 70).
|
Results |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
When using median expression in the tumors as the numerator, five genes (SFTPB, TACSTD1, CK7, CK19, and CEA) clearly stand out as having tumor/highest benign node ratios >200. When the minimum difference between tumors and benign nodes is calculated as a ratio (lowest tumor/highest benign node), all five of these genes still distinguish tumors from benign nodes very well. Thus, these five genes were selected for analysis in the secondary screen. In addition, whereas PVA and SCCA1/2 were only expressed highly in two (both squamous cell) and three (two squamous, one adenocarcinoma) of six primary tumors, respectively, these genes are primarily squamous cell markers and we would therefore only expect them to show high expression in squamous cell lung tumors. Because 
30% of NSCLCs are squamous cell tumors, we therefore decided to evaluate these two markers in the secondary screen also. Finally, LUNX is frequently mentioned as a lung cancer marker and although it did not show high expression in all tumors, no background expression was detected at all in benign nodes. For this reason, LUNX was also analyzed in the secondary screen.
Secondary screen. In this stage of the screening procedure, histologic evaluation of the 21 primary tumor specimens revealed 2 large cell tumors, 7 squamous cell carcinomas, and 12 adenocarcinomas with a median tumor percentage of 70% (range: 5-95%). Similarly, in the 21 histologically positive nodes, the median tumor percentage was 70% (range: 2-100%; six tumors had 
10% tumors). The histologies of the primary tumor for these specimens were as follows: 1 large and small cell mixed cancer, 3 large cell cancers, 6 squamous cell carcinomas, 10 adenocarcinomas, and 1 adenosquamous cell carcinoma (Supplementary Table S3).
The relative expression profiles of eight markers in the primary lung tumors, histologically positive nodes, and benign lymph nodes are shown in Fig. 1
. With the exception of LUNX and SFTPB, the expression levels observed in positive lymph nodes were very similar to those in primary tumors. LUNX expression was consistently lower in positive nodes that in tumors, whereas SFTPB expression was lower in 50% of positive nodes. Using an absolute cutoff value equal to that of the highest expression in benign lymph nodes (100% specificity), the sensitivity to detect tumor in the histologically positive nodes achieved with each marker was 100% for CK19 and TACSTD1, 95% for CEA, 61.9% for LUNX, 57.1% for PVA, 61.9% for SFTPB, 76.2% for CK7, and 38.1% for SCCA1/2. Although both CK19 and TACSTD1 provided 100% sensitivity in this data set, the minimum difference in expression between positive nodes and benign nodes was only 1.4-fold for CK19, whereas it was 7.9-fold for TACSTD1. Thus, in a single-gene assay, TACSTD1 is likely to be a more robust discriminator than CK19.
|
|
|
10% tumor). The histology of the primary tumors in this set was as follows: 24 adenocarcinomas, 10 squamous cell carcinomas, 3 large cell carcinomas, and 1 large cell and small-cell mixed cancer (Supplementary Table S3). Similar to the screening data, 11 of 24 positive nodes from adenocarcinomas had the highest fold change with SFTPB and the remaining 14 were highest for TACSTD1 (Fig. 2C). All 10 nodes from squamous cell carcinomas were highest with PVA and all four large-cell carcinoma nodes were highest for TACSTD1. The median difference between positive nodes and highest benign lymph node was 489.8-fold, with a minimum of 25.7-fold among this set of 38 positive nodes and 32 benign lymph nodes. Association with histologic subtype. The joint expression levels of TACSTD1, SFTPB, and PVA were plotted in three dimensions (Fig. 3 ). The three-dimensional view clearly shows complete separation for (a) benign nodes versus positive nodes and (b) positive nodes in patients with squamous cell carcinomas and adenocarcinomas. Nodes associated with large-cell lung carcinomas, however, cannot be distinguished by this view. The ability of these three markers to discriminate by histologic origin is primarily due to PVA. Although PVA by itself did poorly in discerning positive lymph nodes (validated classification accuracy of 57%), it was capable of distinguishing positive lymph nodes from squamous versus adenocarcinoma primaries with 98% accuracy. Conversely, TACSTD1 perfectly discriminated between benign and all histologic types of positive nodes but had no ability to discern histologic subtype.
|
|
|
Discussion |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
An alternative approach to improve lymph node staging has been to use modern molecular techniques, such as RT-PCR and mutation or epigenetic alteration detection. There are now hundreds (if not thousands) of such reports in the literature, the vast majority of which have used RT-PCR to detect tumor cells in histologically negative lymph nodes. Almost invariably, these reports indicate that RT-PCR finds positive results in histologically negative nodes and in cases where clinical follow-up is available, many studies show prognostic significance of RT-PCR positivity (17–19). In most studies, however, the frequency of RT-PCR positivity in histologically node-negative patients is much higher than disease recurrence rates and thus this literature is commonly criticized based on poor specificity of the assay. It is now becoming clear that background or ectopic expression of the genes used in these studies is present in normal lymph nodes and that this is responsible for at least some of the false-positive results. Furthermore, despite the large volume of RT-PCR reports, the number of genes used as metastasis markers remains very small with CEA and various cytokeratins (CK19 and CK20 in particular) being by far the most popular in most tumor types except melanoma and prostate. This is probably due to the historical use of these genes as targets for immunohistochemical staining in pathology and is not a result of systematic searches to identify targets specifically for RT-PCR assays. Thus, it is likely that alternative molecular markers could be found and that these markers may provide better specificity than previously obtained. Indeed, we have already reported successful searches for new markers in esophageal adenocarcinoma (13), breast cancer (20), and head and neck cancer (14) and Mitas et al. (21) have done a similar study, also in breast cancer. In lung cancer, however, there is only one study [also by Mitas et al. (23)] looking at six genes in 27 mediastinal lymph nodes of NSCLC patients. For this reason, we designed the current project to perform a large-scale marker survey and identify the best marker or combination of markers for lymph node metastasis detection by RT-PCR in NSCLC.
Of the 42 markers analyzed in the primary screen, eight (CEA, CK7, CK19, LUNX, PVA, SCCA1/2, SFTPB, and TACSTD1) were chosen for more detailed analysis in a secondary screen. Only two of these markers (CK19 and TACSTD1) provided 100% classification accuracy, regardless of primary tumor histology. CK19 is an acidic keratin that is expressed at high levels in many epithelial tissues and their derived tumors. CK19 is one of the keratins targeted by the AE1/AE3 antibody cocktail and, as such, is frequently used by pathologists for immunohistochemical detection or confirmation of lymph node metastases in many cancer types. CK19 has also been used in many RT-PCR-based studies to detect occult lymph node disease in breast cancer (21, 24, 25). Unfortunately, this gene has at least two pseudogenes that make it difficult to develop a cDNA-specific assay. As a result, the utility of this marker may be limited in clinical settings given the potential for false positive results. TACSTD1 is also known as EPCAM, EGP-2, KS1/4, GA-733-2, and MIC-18 and encodes a cell surface antigen that is defined by the monoclonal antibody AUAI. TACSTD1 is also the antigen for the antibody Ber-Ep4 and as such has been used in several studies to detect lymph node micrometastases in a variety of tumor types, including NSCLC (8, 25, 26). The recent studies by Wallace et al. (22) and Mitas et al. (23) also found TACSTD1 (KS1/4 in their article) to be a good RT-PCR-based marker for identification of lymph node metastasis in mediastinal lymph nodes from NSCLC patients. From our study and the Mitas et al. study, TACSTD1 seems to be superior to CK19 for two reasons. First, the observed variation in expression within either positive or benign nodes is much less for TACSTD1 than for CK19. Second, the expression difference between positive and benign nodes is consistently greater for TACSTD1 compared with CK19. In our data, there is at least a 7.9-fold difference between the positive and benign nodes with TACSTD1, with a median fold difference of 90. However, for CK19, the minimum difference is only 1.4-fold with a median of 36. Therefore, although CK19 distinguished positive from negative nodes in our relatively small sample set, CK19 is less likely to stand up as an independent marker in larger data sets. Therefore, of all the markers that underwent the screening process, TACSTD1 showed the most promise as a stand-alone marker for nodal involvement in NSCLC patients.
Overall, the next best marker in our study was CEA, with a sensitivity of 95%. CEA (CEACAM5) is a cell adhesion molecule that is frequently found in adenocarcinomas of endodermally derived digestive system epithelia. Historically, CEA has been the marker of choice for identifying adenocarcinoma of the gastrointestinal tract and was among the first markers to be used for RT-PCR analysis of lymph nodes (27). In lung cancer, D'Cunha et al. (28) analyzed expression of CEA in 53 primary tumors and 232 lymph nodes from 53 patients with stage I NSCLC. In this study, 90.5% of primary tumor samples were positive for CEA mRNA expression as were 25.4% of histologically negative lymph nodes. Fifty-seven percent of the patients had at least one CEA-positive lymph node and would theoretically be upstaged to stage II. Unfortunately, no follow-up was reported in this article so the clinical relevance of upstaging is currently unknown. CEA was also evaluated in the Mitas et al. study but high expression in some control lymph nodes led to relatively poor sensitivity and specificity compared with TACSTD1.
None of the remaining markers in the secondary screen gave very good sensitivity overall (<77%) but on further analysis by histologic subtype of the primary tumor, CK7 and SFTPB seem to be good markers for adenocarcinoma, whereas PVA and SCCA1/2 are good markers for squamous cell carcinoma. CK7 (KRT7) is a type II keratin of simple, nonkeratinizing epithelia and is typically found in tissues, such as the epidermis, bronchus, and mesothelium. In our study, CK7 gave 100% sensitivity (11 of 11) for detecting lymph node metastases from adenocarcinoma tumors; however, based on fold expression difference, TACSTD1 was superior to CK7 in 9 of the 10 cases. SFTPB is one of a family of surfactant molecules secreted by type II alveolar cells. Betz et al. (29) evaluated this gene as a marker for metastatic lung adenocarcinoma using standard gel-based RT-PCR, but discarded it citing expression in nonneoplastic lymphatic tissue. However, in our study, the ability to quantify the expression of mRNA allowed us to distinguish this background expression from expression associated with metastatic cancer cells in 9 of 11 adenocarcinoma cases. Furthermore, in seven of these cases, the fold expression difference was much higher than for either CK7 or TACSTD1. Thus, SFTPB, but not CK7, may be a useful marker in combination with TACSTD1 to improve the detection of adenocarcinoma foci in lymph nodes.
SCCA is a member of the ovalbumin family of serine proteinase inhibitors. The SCCA protein is expressed in neutral and acidic forms, designated as SCCA1 and SCCA2, and is detected in the superficial and intermediate layers of normal squamous epithelium. The expression of SCCA2 in cancer has been associated with an aggressive phenotype and this gene has been used in several studies for detection of squamous cell carcinoma metastases to lymph nodes (30). In our study, SCCA was able to identify metastatic tumor in four of six lymph nodes from patients with squamous cell carcinoma of the lung. PVA (also known as desmoglein 3) is a 130-kDa surface glycoprotein that is the serologic target in the autoimmune skin disease pemphigus vulgaris. Apart from our previous study in head and neck cancer (14), PVA has not been described as a marker for lymph node metastasis detection by RT-PCR. In this lung cancer study, we found that PVA is also an excellent marker for squamous cell lung cancer, providing 100% sensitivity in this histologic subtype. Furthermore, as with SFTPB in adenocarcinoma, PVA is superior to TACSTD1 in five of six cases and could, therefore, be useful in a combination marker assay. The last gene included in our secondary screen, LUNX, was first isolated and reported as a novel lung-specific gene of unknown function by Iwao et al. (31). Using RT-PCR, LUNX expression was detected in all 35 NSCLC tumor samples and in 80% (16 of 20) of histologically positive lymph nodes. In our hands, however, the LUNX gene did poorly as a marker for nodal disease, with a sensitivity of only 61.9% overall and with no obvious improvement in either histologic subtype.
In summary, we have shown that TACSTD1 may be used alone as a molecular marker to detect metastatic lymph node involvement in NSCLC patients and that combining this marker with PVA and SFTPB may provide a more robust assay. Ongoing studies are using these three markers for detection of occult lymph node metastases in pN0 patients and in the mediastinal nodes of pN1 patients. Preliminary data presented here indicates that 7% of histologically negative lymph nodes from NSCLC patients are positive for one or more of these markers and this identifies 30% of patients as potentially having occult lymph node metastasis. These numbers are very similar to the frequencies found in studies using immunohistochemistry but a larger patient series and more follow-up is needed to evaluate the prognostic significance of qRT-PCR positivity.
|
Footnotes |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Note: Supplementary data for this article are available at http://www.mssm.edu/labs/godfrt01/publications/supp.htm.
L. Xi, V.R. Litle, and T.E. Godfrey are currently in Mount Sinai School of Medicine, New York, New York.
Received 9/16/05; revised 11/18/05; accepted 1/19/06.
|
References |
|---|
|
TopAbstractMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  L. Xi, D. G. Nicastri, T. El-Hefnawy, S. J. Hughes, J. D. Luketich, and T. E. Godfrey Optimal Markers for Real-Time Quantitative Reverse Transcription PCR Detection of Circulating Tumor Cells from Melanoma, Breast, Colon, Esophageal, Head and Neck, and Lung Cancers Clin. Chem., July 1, 2007; 53(7): 1206 - 1215. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Cell Growth & Differentiation |
