Mediastinoscopy: An Endangered Species?

Accurate staging is a critical part of the management of newly diagnosed non–small-cell lung cancer (NSCLC), and of the decision to offer patients induction therapy. During the last decade, the introduction into clinical practice of positron emission tomography (PET) and of endoscopic ultrasound biopsy techniques has again raised the question of when mediastinoscopy, long the gold standard for determining the presence and extent of mediastinal nodal metastases, should be performed. In this edition of the Journal of Clinical Oncology, two articles address this issue. PozoRodriguez et al report a prospective study in 132 patients comparing the efficacy of helical computed tomography (CT) and PET in the mediastinal staging of NSCLC. They find that CT and PET have similar accuracies and, when combined, yield a negative predictive probability of .98. Conversely, they note that positive results of either test need to be confirmed pathologically. In the second article, Annema et al evaluated endoscopic ultrasound-guided fine-needle aspiration (EUS FNA) in the diagnosis and staging of lung cancer in 242 consecutive patients for whom mediastinoscopy or exploratory thoracotomy were planned. EUS FNA prevented 70% of scheduled surgical procedures, and the accuracy of this procedure in mediastinal staging was found to be 93%. How does the practicing clinician incorporate this information into daily practice? First, it is important to note that these are not the first studies to highlight the utility of PET and EUS in the staging of NSCLC. A literature search yields a plethora of studies of variable quality on both modalities, especially PET. Table 6 in the article by PozoRodriguez et al summarizes some of the most relevant articles published on CT and PET in NSCLC since 1980. However, the authors correctly point out that their trial was more stringently designed than many previous studies, and benefited from the readings of CT and PET being blinded and independent both of each other and of the reference tests. A long follow-up period allowed clinical confirmation of the study findings. Multiple articles already attest to the safety and accuracy of EUS FNA in detecting mediastinal lymph node metastases and mediastinal tumor invasion in NSCLC. However, Annema et al have performed the largest prospective study to date, even though approximately 15% of their patients had EUS FNA performed for reasons other than the initial staging of NSCLC. Therefore, both of these studies are noteworthy by virtue of their size and careful design. Second, these studies reflect the evolving algorithm of staging of NSCLC that is already reducing the use of mediastinoscopy. In many countries, including the United States, PET (and more recently PET-CT) has become a widely accepted study for the initial extent of disease evaluation of NSCLC. Beyond its utility in mediastinal staging, PET provides important prognostic information and detects distant metastatic disease, obviating the need for additional bone scanning or imaging of the liver and adrenal glands. EUS FNA is not as widely practiced, partly because it is not yet a procedure commonly performed by thoracic surgeons or pulmonary medicine physicians, and partly because it does not provide access anatomically to the right and left paratracheal nodes, which are key to the staging of most NSCLC. However, it is well recognized as an important adjunct in evaluating the periaortic, subcarinal, and periesophageal lymph nodes. Endobronchial ultrasound (EBUS) FNA is just now being introduced into clinical practice outside of Japan and should prove highly effective in diagnosing paratracheal and perhaps even hilar nodal metastases. Thus, the staging algorithm of the not-toodistant future could include a contrast CT of the chest and upper abdomen, a PET-CT, brain imaging (as clinically indicated), and an EBUS FNA with or without an EUS FNA. Mediastinoscopy might only be performed in situations where other staging studies are equivocal or when additional tissue is needed for histologic or molecular evaluation. Initial staging with a combination of EUS, EBUS, and PET may also allow mediastinoscopy to be used more freely for restaging after induction therapy in situations where this is important for decisions about subsequent treatment. The accuracy of EUS and EBUS FNA for restaging will need to be examined in the future. JOURNAL OF CLINICAL ONCOLOGY E D I T O R I A L VOLUME 23 NUMBER 33 NOVEMBER 2