Commentary (Cerfolio): Endoscopic Ultrasound Fine-Needle Aspiration in the Staging of Non-Small-Cell Lung Cancer

October 1, 2006

Precise mediastinal staging of non-small-cell lung cancer is extremely important, as mediastinal lymph node metastases generally indicate unresectable disease. Reliance on computed tomography (CT) and positron-emission tomography (PET) alone to stage and determine resectability is limited by false-positive results. Whenever possible, pathologic confirmation of metastases is desirable. Mediastinoscopy and transbronchial fine-needle aspiration are widely established but imperfect modalities. Endoscopic ultrasound fine-needle aspiration (EUS-FNA) has emerged as a diagnostic and staging tool because of its safety, accuracy, and patient convenience. We reviewed 13 prospective studies evaluating the comparative performance of EUS for staging lung cancer. We conclude that EUS is a valuable staging modality. Further studies of the role of EUS compared to other modalities such as integrated PET/CT and endobronchial ultrasound (EBUS) are forthcoming.

Dr. Ogita and colleagues have presented a well balanced review of some of the current literature concerning the art of endoscopic ultrasound and fine-needle aspiration (EUS-FNA) in the staging of patients with non-small-cell lung cancer (NSCLC). The authors have accurately presented the literature, and they should be congratulated for their work. However, they have omitted several concepts that, for the sake of completeness, should be included.


First, the authors begin their abstract with an incorrect statement, noting that "mediastinal lymph node metastases generally indicate unresectable disease." In fact, mediastinal lymph node metastases most commonly represent N2, not N3 disease. If mediastinal nodes harbor malignancy and are on the same side of the primary tumor, they are defined as N2. Lymph nodes that are involved with metastatic cancer and are on the opposite side are N3. An N2 classification, which represents stage IIIA disease, is much more common than N3 disease.

Moreover, N2 disease represents a heterogeneous group of patients. Some patients have microscopic N2 disease, and the results of surgical resection alone in this select group are quite good. Those with large bulky multistation N2 disease have poor results, even with neoadjuvant chemoradiotherapy followed by complete resection. Yet most patients with N2 disease have a prognosis somewhere in between, and they are best treated with neoadjuvant chemotherapy or chemoradiotherapy and then resection if they are downstaged. Thus, the statement that mediastinal lymph node metastases indicate unresectable disease is not correct.

Role of EUS-FNA

EUS-FNA plays a critical role in patients with N2 disease, since it is easier and more accurate to perform repeat EUS-FNA to restage patients with N2 disease after induction therapy then it is to perform repeat mediastinoscopy. When confronted with a patient with suspected N2 disease in both the anterior mediastinal lymph nodes (2R, 4R, 2L, or 4L) as well as the posterior mediastinal lymph nodes (level 7, 8, or 9), we use EUS-FNA first and reserve the mediastinoscopy for after the completion of induction therapy. This holds true unless we suspect N3 disease; then we may perform mediastinoscopy first, since it can assess N3. In a patient with biopsy-proven N2 disease determined via EUS-FNA, we always perform repeat EUS-FNA after the completion of induction therapy to ensure that the pathologically proven N2 node has been downstaged (sterilized) prior to thoracotomy and resection.

The authors do not describe some of the many other and diverse functions of EUS-FNA in the staging of patients with NSCLC. For example, we have used the procedure to rule out T4 lesions in selected patients when invasion of the esophagus, central pulmonary artery, or inferior or superior pulmonary veins is questioned based on review of the integrated 18F-fluorodeoxyglucose positron-emission tomography (FDG-PET)/computed tomography (CT) or the CT scan alone. We have also used it as the procedure of choice for biopsy of a suspicious left adrenal gland. We have had over 30 patients with biopsy-proven stage IV NSCLC to the left adrenal proven via EUS-FNA. Recently we have also used this minimally invasive, safe modality that avoids a general anesthetic and an incision for a handful of patients undergoing right adrenal gland biopsy if the gland is close to the duodenal loop. EUS-FNA has also allowed us to biopsy selected lesions in the right lobe of the liver as well.

The authors state that "no prospective study has been conducted to evaluate the use of EUS-FNA in patients with a negative PET scan." However, we recently completed a prospective series and presented our data at the Western Thoracic Surgical Association meeting, held June 2006 in Sun Valley, Idaho; our report is to be published in Chest at the end of this year.

This study evaluated 153 NSCLC patients, 136 of whom were clinically staged N0 and 17 of whom were staged N1 after integrated PET/CT and 5-mm CT scan. Of the 136 patients who were staged as N0, only 5 (3.7%) had a positive EUS-FNA and 4 (2.9%) had a positive mediastinoscopy. Of the remaining 127, 6 (4.7%) had N2 disease after resection. Seventeen patients were clinically staged as N1 by integrated PET/CT: 4 (23.5%) had a positive EUS-FNA, 3 (17.6%) had a positive mediastinoscopy, and none of the remaining 10 had N2 disease after resection. Patients with unsuspected N2 disease were twice as likely (relative risk [RR] = 2.1; 95% confidence interval [CI] = 1.24-2.51; P = .02) to have a maximum standardized uptake value (maxSUV) > 10 and poorly differentiated cancer (RR = 2.1; 95% CI = 1.14-2.38; P = .03).

Based on this study, we do not recommend routine mediastinoscopy or EUS-FNA in patients who are clinically staged as N0 after both integrated PET/CT and CT, but do if patients are staged as N1 after PET/CT, and/or in those with adenocarcinoma, upper lobe tumors, or tumors with a maxSUV of 10 or greater.

Importance of Pathology

Finally, although I share the authors' optimism about the future of EUS-FNA and endobronchial ultrasound (EBUS), I disagree with their statement that these procedures "will eventually obviate much of the need for surgical staging." In fact, neither modality can or should replace pathologic staging. If they mean that EUS-FNA and EBUS have become part of the armamentarium of surgical staging, then I partially agree with their statement. I consider procedures such as EUS-FNA and EBUS, which provided cytologic and/or pathologic verification of suspected disease, to be surgical staging procedures-only no incisions are made and no anesthetic is necessary, which potentially makes them even better. But these tests are complementary to mediastinoscopy and not a replacement for it. False-positive results are more likely with cytology than with pathology. Sometimes larger tissue samples are needed, as in patients with lymphoma or equivocal EUS and EBUS specimens.

We have performed a prospective study that clearly outlines the great disparity between clinical staging (suspected on integrated FDG-PET/CT or on multicut CT scan) and pathologic staging. Despite continued warnings to medical oncologists, radiation oncologists, and surgeons not to treat patients based on PET scan findings, many continue to do so without obtaining pathologic confirmation.


Almost all patients can undergo minimally invasive procedures to prove or disprove the presence of N2 or N3 lymph node disease prior to the induction of chemotherapy or chemoradiotherapy. In this way the patient's true pathologic stage and not the probable clinical stage is ascertained, and thus an appropriate treatment strategy instituted. Perhaps the only exception to this rule may be the aortopulmonary window nodes (level 5 and/or 6) in a patient with a large left upper lobe mass. Obtaining tissue from mediastinal lymph nodes in patients with NSCLC is an ideal example of how surgeons, gastroenterologists, and pulmonologists can work together using mediastinoscopy, EUS-FNA, and EBUS to better pathologically stage patients with cancer.

-Robert J. Cerfolio, MD


The authors have no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.