© 2003 American Thoracic Society
From Clinical and Pathologic to Molecular Staging of Lung CancerDivision of Allergy, Pulmonary, and Critical Care Medicine Division of Hematology and Oncology Vanderbilt University Medical Center Nashville, Tennessee Lymph node involvement is a critical predictor of survival in operable nonsmall cell lung cancer (NSCLC) and is increasingly important for determination of the type and timing of therapy. Although mediastinoscopy remains the standard of comparison with 89% sensitivity and close to 100% specificity (1), it does not assess all lymph nodes easily. In addition, mediastinoscopy is an invasive procedure requiring general anesthesia, which is increasingly replaced by less sensitive and specific imaging studies that can detect only macroscopic disease. Although computed tomography alone has a rather poor sensitivity and specificity (63% and 57%, respectively), the combination of computed tomography and positron emission tomography has a sensitivity and specificity for diagnosing N2 disease of 84% and 94%, respectively (2). It is estimated that approximately 10% of patients with negative nodes diagnosed by computed tomography scan will have N1 or N2 disease by mediastinoscopy. Even standard pathologic evaluation of mediastinoscopy biopsies can miss micrometastatic disease. Micrometastases have been shown to have implications in survival of stage I NSCLC (3, 4). False negatives can result in unnecessary or suboptimal timing for surgery, but false positives can falsely categorize a curable patient as incurable, a self-fulfilling prophecy in this disease. In this issue of AJRCCM (pp. 1670–1675), Wallace and coworkers analyze the value of telomerase expression in lymph node aspirates for the staging of NSCLC (5). These authors combined a minimally invasive approach for lymph node sampling, transesophageal endoscopic ultrasound-guided fine-needle aspiration, with telomerase gene expression analysis by quantitative real-time polymerase chain reaction to assess the mediastinal nodes for involvement by tumor. This approach is particularly attractive because successful sampling of nodes by this technique can be achieved without general anesthesia in the vast majority of cases depending on nodal location (levels 2, 4, and 6 are inaccessible) and operator experience. Transbronchial ultrasound-guided fine-needle aspiration is another recently introduced, minimally invasive method of staging lung cancer (6). The obvious limitation to this technique relates to the lack of real-time guidance of nodal station sampling (7). Wallace and coworkers ask two important questions. Can we perform real-time polymerase chain reaction on material obtained by ultrasound-guided fine-needle aspiration? What is the prevalence of telomerase overexpression in mediastinal lymph nodes of patients with NSCLC using this approach? The authors showed that an average of 7.8 µg of high-quality RNA was extracted successfully from 0.5 ml of lymph node aspirates using multiple passes. They were able to establish the prevalence of telomerase gene overexpression in mediastinal lymph nodes by real-time polymerase chain reaction. Telomerase gene was overexpressed in none of the 14 control subjects, 18 of the 57 negative nodes, and 10 of the 16 pathologically positive cases (which implies low sensitivity). Twenty-four patients also underwent surgery with lymph node sampling. A quarter of these patients had false-negative cytology obtained by ultrasound-guided fine-needle aspiration, a number somewhat disappointing but consistent with the previously published false-negative rate for transbronchial needle aspiration (8). One of the most significant findings of the study was that 28% of the patients with negative nodes by histology were found to express telomerase. The number of telomerase-negative lymph nodes at ultrasonography but telomerase positive in surgical samples was not reported as an assessment of the sampling problem associated with fine-needle aspiration. Telomerase expression is a recognized marker of neoplastic transformation and is indicative of telomerase activation. Cancer cells avoid replicative senescence by maintaining telomere length leading to unlimited replication (9). Wallace and coworkers acknowledge that telomerase expression is not absolutely cancer specific. Expression and activity of telomerase gene have been found in bronchial epithelium, peripheral lung, preinvasive lesions (10), and inflammatory conditions. Nevertheless, telomerase expression has been shown to have prognostic value in biopsies of patients with NSCLC (11, 12). In other organ systems, such as in cervical cancer, telomerase expression has not been shown to correlate with survival (13). The added clinical value of knowing telomerase expression in lymph nodes of NSCLC patients with or without documented nodal involvement remains to be tested prospectively. It is, however, unlikely that any single marker will allow optimal detection. A set of markers judiciously selected may ultimately prove clinically useful. From this study and this new area of investigation, many questions arise. To mention a few: Is there any correlation between carinal biopsy from a bronchoscope and lymph node aspirate under ultrasound guidance? What are the correlations between samples obtained by fine-needle aspiration and whole node tissue specimens? Will we be able to confirm that the nodes resected at the time of surgery are the same nodes sampled by fine-needle aspiration under ultrasound guidance? What are the possible effects of inflammatory diseases (e.g., chronic obstructive pulmonary disease) on telomerase gene expression in mediastinal lymph nodes? What is the clinical significance and impact of this observation on survival? Investigations prompted by this interesting study will undoubtedly address these questions in the near future and may establish the utility of molecular staging in NSCLC. REFERENCES
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
