Pancoast Tumor Imaging 

Pancoast tumors are neoplasms of pulmonary origin located at the apical pleuropulmonary groove (superior sulcus). Pancoast tumors and their associated morbidities are presented in images below.

By direct extension, Pancoast tumors typically involve the lower trunks of the brachial plexus, intercostal nerves, stellate ganglion, adjacent ribs, and vertebrae.

More than 95% of Pancoast tumors are non–small cell carcinomas, most commonly squamous cell carcinomas (52%) or adenocarcinomas and large cell carcinomas (approximately 23% for each subtype).^[1]Small cell carcinomas are seen in fewer than 5% of cases.^[2]

Staging of Pancoast tumor involves the tumor, node, and metastasis (TNM) classification system, in which T indicates site and size of the primary tumor, N is related to nodal involvement according to site, and M indicates the presence or absence of distant metastases. These tumors are, at a minimum, T3N0M0 (T3 for chest wall invasion, stage IIB), and they are considered T4 lesions if the brachial plexus, mediastinal structures, or vertebral bodies are involved at the time of presentation. When supraclavicular nodes are involved, they are designated as N3 nodes, although they may be the first nodal station involved. Metastatic tumor in the ipsilateral nonprimary-tumor lobe of the lung or metastases to other organ systems is considered M1.^[3]

Preferred examination

Compared with other examinations, magnetic resonance imaging (MRI) is more accurate in identification of the extent of tumor involvement; it is superior to computed tomography (CT) scanning in the detection of invasion of adjacent organs (eg, vertebral bodies, brachial plexus, subclavian vessels).^{[4, 5]}

Histologic diagnosis is made in 95% of the cases by means of percutaneous transthoracic needle biopsy with fluoroscopic, ultrasonographic, or CT scan localization.^{[6, 7]}

Among other considerations, CT scanning or MRI of the brain is recommended in the initial evaluation, because distant metastases to the brain are not infrequent, and diagnosis of these is necessary for staging.

Limitations of techniques

Noninvasive preoperative evaluation of the mediastinum with CT or MRI is limited by substantial false-positive and false-negative results (30-40%, depending on the criteria used to define enlarged lymph nodes and the patient population). Positron emission tomography (PET) scanning and, possibly, surgical assessment of the mediastinum with lymph node sampling should be strongly considered before curative surgery is attempted.

Intervention

Preoperative radiation therapy at doses of 2000-6500 cGy, followed by surgical resection, is the most common form of treatment for Pancoast tumors. However, the advantage of preoperative radiation therapy has not been definitively demonstrated, and the dose of radiation has not been clearly established. The overall 5-year survival rate in patients treated with preoperative radiation therapy and surgery is reported to be 20-35%.^{[1, 8, 9]}

Radiation therapy at a dose of 6000 cGy or greater has been used as a primary treatment modality for inoperable tumors, with successful palliation of pain in as many as 90% of patients. The reported 5-year survival rate is 0-29% in these patients, which is likely a result of extensive disease involvement at initial presentation.

The routine use of intraoperative and postoperative radiation therapy is not currently recommended, except in patients in whom unresectable tumors are found at the time of surgery.

Posteroanterior (PA) chest radiographs show unilateral apical opacity (as seen in the image below) or just asymmetry of the apices of greater than 5 mm. Local rib destruction can sometimes be observed. Lordotic chest views can be beneficial, but the findings can also be misleading.

In the early stages, Pancoast tumors are difficult to detect on PA chest radiographs because of the difficulty in interpreting overlying shadows at the apices.

CT scanning is best in demonstrating bony destruction. MRI appears to be superior in demonstrating chest wall invasion. Also, the anatomy above the lung apex is better demonstrated on multiplanar MRI, because the nerves of the brachial plexus and blood vessels follow a horizontal and parallel course, meeting above the apex of the lung.

In an older study of 31 patients with superior pulmonary sulcus tumors,^[4] CT scanning had a sensitivity of 60% and a specificity of 65%, with an overall accuracy of 63% in the evaluation of the extent of disease.

MRI provides superior delineation of the normal anatomy of the brachial plexus because of its multiplanar capabilities. The absence of streak artifact from bone and accurate identification of vessels are some of the advantages of MRI. It also has superior soft-tissue contrast, and it is more accurate than other methods in documenting or excluding brachial plexus involvement by the tumor. (See the images below.)

Compared with other techniques, MRI is more accurate in the evaluation of extension to the vertebral body, spinal canal, brachial plexus, and subclavian artery. This advantage is important, because vertebral body, spinal canal, and upper brachial plexus invasion are contraindications to surgical resection.

In an older study of 31 patients with Pancoast tumors, MRI had a sensitivity of 88%, a specificity of 100%, and an overall accuracy of 94%.^[4]

Some have suggested that all patients with Pancoast tumors should undergo ultrasonographic examination of the ipsilateral scalene area and that percutaneous biopsy should be performed on nodes with a transverse diameter of greater than 1 cm. The purpose of these studies is to assist in staging of the disease. The use of a sector ultrasonographic unit with a supraclavicular approach has been useful in guiding needle aspirations, with a yield for pathologic diagnosis in 91% of cases.

PET scanning is a promising modality for the detection of distant metastases and mediastinal involvement.

On occasion, subclavian artery angiography may be indicated to rule out local invasion of these vessels by the tumor.