EUS and EBUS in Non–Small Cell Lung Cancer

Key Points

•

Endoscopic ultrasonography (EUS)-guided fine-needle aspiration (EUS FNA) and endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) are the techniques of choice in case mediastinal nodal sampling is indicated for the diagnosis and staging of lung cancer.
•

Combined endosonographic nodal staging (EUS and EBUS) improves nodal staging versus either procedure alone.
•

Incorporation of endosonography in the staging algorithm for non–small cell lung cancer improves locoregional staging, reduces both the number of mediastinoscopies and unnecessary thoracotomies, and is cost-effective.
•

EUS performed with the EBUS scope (EUS-B) seems similar to conventional EUS for mediastinal nodal staging.
•

EUS and EBUS can diagnose intrapulmonary tumors, in case they are located adjacent to the esophagus and major airways, respectively.

Introduction

Transesophageal endoscopic ultrasonography (EUS)-guided fine-needle aspiration (FNA) and endobronchial ultrasound (EBUS)-guided transbronchial needle aspiration (TBNA) are minimally invasive techniques for the diagnosis and staging of lung cancer. The worldwide incidence of lung cancer exceeds 1 million, and one-third of these patients present with mediastinal metastases. The stage of disease determines prognosis and treatment. In case the lung malignancy is confined to the lung and hilar lymph nodes (stage I or II, [N0/N1 disease]), surgery or stereotactic radiotherapy (SABR) with curative intent is the treatment of choice. However, when mediastinal lymph nodes are involved (stage III, [N2/N3 disease]), combined chemoradiotherapy is usually indicated. Patients presenting with distant metastases (stage IV) will be treated with chemotherapy or targeted therapies. Therefore accurate mediastinal staging is crucial for optimal staging and treatment planning. Computed tomography (CT) of the chest and positron emission tomography (PET) are commonly used techniques in the initial characterization of lung tumors and for the search for metastases, but these tests are insufficient for accurate mediastinal staging. Additional tissue confirmation is recommended in patients with enlarged (short axis >10 mm) and/or fluorodeoxyglucose (FDG) avid intrathoracic lymph nodes and in patients with a normal mediastinum who have an increased risk of mediastinal involvement (centrally located tumor or a primary tumor size >3 cm). Starting mediastinal nodal tissue sampling by endosonography has been proven to be superior over initial surgical staging, and therefore guidelines recommend endosonography as the technique of choice when mediastinal staging is indicated. In this chapter the role of EUS FNA and EBUS TBNA for the diagnosis and staging of lung cancer will be evaluated. How to perform EBUS and EUS in the mediastinum will be explained with an emphasis on mediastinal nodal anatomy.

The indications for both methods are addressed ( Table 7.1 ), as well as the concept of combined (esophageal and endobronchial) echoendoscopic staging of the mediastinum. In particular, the single scope, single operator staging approach: performing EBUS and EUS with only an EBUS scope (EUS-B) by introducing the EUS-B in the esophagus following an EBUS examination will be discussed. The position of endosonography in lung cancer staging algorithms will be addressed. In addition, tips and trips regarding training and setting up an EBUS EUS service will be provided.

TABLE 7.1

Indications for Endosonography for the Diagnosis and Staging of Lung Cancer

Mediastinal Lymph Nodes	EUS (B) FNA	EBUS TBNA
Paratracheal to the left	++	++
Paratracheal to the right	−	++
Aortopulmonary window	+	−
Subcarinal	++	++
Lower mediastinum	++	−
Hilar	−	++
Mediastinal restaging	+	+
FDG PET uptake in lymph node within reach	++	++
Lung tumor located adjacent to the esophagus	++	−
Lung tumor located adjacent to the trachea or main bronchi	−	++
Suspected left adrenal metastasis	++	−

++, Strong evidence; +, moderate evidence; −, no evidence; EBUS, endobronchial ultrasound; EUS, endoscopic ultrasonography; FDG, fluorodeoxyglucose; FNA, fine-needle aspiration; PET, positron emission tomography; TBNA, transbronchial needle aspiration.

Endoscopic Ultrasonography-Guided Fine-Needle Aspiration for the Diagnosis and Staging of Lung Cancer

Procedure

Conventional mediastinal EUS is performed with gastrointestinal (GI) linear echoendoscopes with the patient in a left lateral position using midazolam or propofol sedation as previously described (see Chapter 6 ). In addition to a mediastinal nodal evaluation, the left adrenal gland (LAG) (a predilection site of distant metastases) and primary lung tumors (in case located adjacent to the esophagus) can be evaluated. A systematic investigation is advised. Lymph nodes and abnormalities are related to specific (vascular) landmarks such as aorta, pulmonary artery, left atrium, and the liver ( Fig. 7.1 EUS for lung cancer staging).

Intrapulmonary Tumors (T) and Endoscopic Ultrasonography

Intrapulmonary tumors that are located adjacent to or near the esophagus can be visualized by EUS ( Fig. 7.2 ). After the primary tumor has been identified, real-time EUS-guided biopsy of the intrapulmonary lesion is possible (see Fig. 7.2 ). A meta-analysis showed that the average sensitivity of EUS FNA for diagnosing malignant intrapulmonary tumor is 92%. The risk of a pneumothorax is almost negligible.

After the primary tumor has been identified, mediastinal tumor invasion (T4), (defined as invasion in the mediastinum, centrally located large vessels, or vertebrae), can be assessed ( Figs. 7.3 and 7.4 ). CT has limited sensitivity and specificity (<75%) for mediastinal invasion, and PET has no value in detecting T4 tumors because of its limited anatomic resolution. A large study evaluated the diagnostic value of EUS in the assessment of mediastinal/vascular tumor invasion (T4) in non–small cell lung cancer (NSCLC) patients whom additionally underwent surgical-pathologic staging. The sensitivity, specificity, negative predictive value (NPV), and positive predictive value (PPV) of EUS for diagnosing mediastinal/great vessel invasion were 42%, 95%, 83%, and 73%, respectively. The sensitivity, specificity, NPV, and PPV of chest CT for assessing T4 status were 76%, 61%, 88%, and 41%, respectively. Importantly, the combination of EUS and CT had an excellent specificity and PPV and NPV.

In conclusion, intrapulmonary tumors can be visualized and sampled safely by EUS FNA provided the tumors are located adjacent to the esophagus. In addition to establishing a tissue diagnosis, EUS can detect mediastinal tumor invasion, especially of vascular structures.

Mediastinal Nodal (N) Staging and Endoscopic Ultrasonography

Evaluation of the mediastinum by EUS should be performed in a standardized fashion (see Chapter 6 ) to examine all mediastinal lymph node stations that can be detected from the esophagus (see Fig. 7.1 ). We strongly favor this approach above the so-called hit-and-run approach, in which only a single PET avid or enlarged node is targeted. The EUS Assessment Tool (EUSAT) can be helpful for a structural assessment. This assessment tool scores performance on the following domains: scope introduction, nodal anatomy, and safe needle handling and tissue sampling. Lymph nodes should be described in relation to the anatomic (vascular) landmarks and given a number according to the tumor, node, metastasis (TNM) classification. After an initial orientation, enlarged (short axis >10 mm) or sonographically suspicious nodes should be sampled for biopsy, starting with contralateral (N3) nodes before ipsilateral (N2) lymph nodes in order to prevent upstaging.

Diagnostic Reach of Endoscopic Ultrasonography

Lymph nodes that lie adjacent to the esophagus or centrally located vessels can be visualized by EUS. These lymph nodes are located in the following regions: low paratracheal on the left (station 4 left; Fig. 7.5 ), aortopulmonary window (station 4 left and 5; Fig. 7.6 ), para-aortal (station 6; Fig. 7.7 ), subcarinal (station 7; Figs. 7.8 and 7.9 ), lower paraesophageal (station 8), and pulmonary ligamentum (station 9; Fig. 7.10 ). Lymph nodes located in the aortopulmonary window can be detected by EUS but sampled only in selected cases due to the interposition of the pulmonary artery. Para-aortal nodes are located on the other side of the aorta and can be well visualized by EUS (see Fig. 7.7 ). In carefully selected cases, these lymph nodes can be aspirated either transaortally or by a long approach (7 to 8 cm) from the proximal esophagus to obtain a tissue diagnosis. Otherwise this region can be reached only by surgical methods such as mediastinotomy or video-assisted thoracoscopy (VATS). EUS has limitations in its diagnostic reach because air in the trachea and main bronchi inhibits visualization of the upper paratracheal lymph node (station 2R) and the lower paratracheal station on the right (station 4R) in most cases.

Endoscopic Ultrasonography Versus Endoscopic Ultrasonography–Fine-Needle Aspiration

Several ultrasonographic features of mediastinal lymph nodes, such as size (short axis >10 mm), round shape, heterogeneous echogenicity, sharp distinctive margin, and coagulation necrosis sign are associated with malignant involvement. However, EUS in combination with FNA is more accurate than EUS imaging alone. Therefore FNA is always required before a lymph node can be designated as malignant ( Figs. 7.11 and 7.12 and Video 7.1 ). For this reason, curved linear, not radial, ultrasound probes are required for mediastinal staging of NSCLC. Of the different needle sizes (19, 22, and 25 gauge) available for nodal staging, the 22 gauge is regarded as the standard size. Elastography is a technical application that depicts the mechanical properties of tissue during endosonography. An accuracy of 85% for differentiating benign from malignant mediastinal nodes has been reported. The results are good for a noninvasive technique but remain inferior to the success rate of EUS-guided FNA. The value of elastography is investigational, and this technique may be helpful in selecting target lymph nodes for biopsy.

Video 7.1

Endoscopic Ultrasonography-Guided Fine-Needle Aspiration of the Subcarinal Lymph Node in a Patient With Non–Small Cell Lung Cancer

The advised number of biopsies per lymph node range from three to five passes to obtain an optimal yield. The optimal sampling technique (suction/stylet use) is still under discussion. No benefit in diagnostic yield has been correlated with the position of the needle in the lymph node (central vs. peripheral) nor the application of suction. In addition to conventional cytology evaluation, cell blocks can be made of EUS fine-needle aspirates on which immunohistochemistry and molecular analysis can be performed. EUS FNA of mediastinal lymph nodes is safe, and complications such as a mediastinitis are rare.

Accuracy of Mediastinal Staging (N) by Endoscopic Ultrasonography

In a meta-analysis of 18 studies of EUS FNA for the mediastinal staging of lung cancer, sensitivity was 83% (95% confidence interval [CI], 78% to 87%) and specificity was 97% (95% CI, 96% to 98%). In those patients with enlarged lymph nodes on CT, sensitivity was 90% (95% CI, 84% to 94%). Although PPVs were reported in most studies, tumor-positive findings were verified by surgical-pathologic staging in only one study. Although false-positive EUS FNA findings have seldom been reported, they are possible when the primary tumor is located immediately adjacent to a lymph node, a situation in which the EUS images can be misinterpreted. Most studies are performed in selected patients with enlarged (>1 cm) mediastinal lymph nodes at CT, and therefore the results apply only to patients in that category. Few studies have focused specifically on small (short axis ≤10 mm) nodes; sensitivity has varied between 35% and 93%. The pooled sensitivity in a meta-analysis for this subgroup was 58% (95% CI, 39% to 75%).

Distant Metastases (M1) and Endoscopic Ultrasonography

Of the common locations of distant lung cancer metastases, lesions located in the left liver lobe and (left) adrenal gland can be identified ( Figs. 7.13 and 7.14 ) and sampled by EUS-B ( Video 7.2 ). In patients with a LAG suspected for distant metastasis on imaging, pathologic confirmation by EUS FNA is advised because benign LAG enlargement is common, also in a lung cancer setting. Traditionally, adrenal masses have been sampled by percutaneous biopsy. This technique has a sensitivity and NPV for adrenal biopsy of 73% and 60%, respectively, and has a considerable risk of complications. In lung cancer patients with a suspected LAG on imaging, the sensitivity of EUS FNA for LAG metastasis is at least 86%. Although technically more demanding, EUS-B has similar outcomes in comparison to conventional EUS for LAG analysis (87% vs. 83%). Patients with disseminated lung cancer often present with liver metastases. The standard procedure for the detection of liver metastases is transabdominal ultrasonography. Investigators have reported that liver metastases can be assessed by EUS FNA using a transgastric approach.

Video 7.2

Endoscopic Ultrasonography-Guided Fine-Needle Aspiration of the Left Adrenal Gland in a Patient With Metastatic Non–Small Cell Lung Cancer

Endoscopic Ultrasound Using the Endobronchial Ultrasound Scope

EUS can also be performed with an EUS-B. Typically, this is done following an endobronchial (EBUS) procedure in the same setting by the same operator, with patient in supine position. After the endobronchial procedure the EUS-B is retracted from the airways above the vocal cords and subsequently gently introduced into the esophagus. As during a conventional EUS procedure, a systematic evaluation of the lymph nodes adjacent to the esophagus, liver, and LAG can be performed ( Video 7.4 ) using the EUSAT. Even though the conventional EUS scope is more stable as a result of the increased tube diameter and has a wider scanning angle (field of view 120 to 180 degrees compared with 60 degrees of the EBUS), there seems to be no difference in yield between conventional EUS and EUS-B for mediastinal nodal staging ( Fig. 7.15 ). In addition, the LAG can be detected and sampled with the EUS-B. By using only the EUS-B, complete mediastinal, hilar, and LAG staging can be performed in a single endoscopy procedure with one operator. This single scope single operator staging strategy is likely to reduce patient burden and seems to be cost-effective.