Lung Cancer: Diagnosis, Staging, and Prognosis

Samir S. Makani and Mark M. Fuster

CLINICAL PRESENTATION

Patients with lung cancer most often present with signs and symptoms heralding local or metastatic disease. Less commonly, an incidental radiographic finding in an asymptomatic patient triggers workup and diagnosis. This may change with more widespread use of low-dose CT-based imaging strategies. Rarely, patients may present with a paraneoplastic syndrome. How an individual presents is more often a function of (1) the site of origin (central versus peripheral airways), (2) the inherent biologic activity of the neoplasm, and (3) the presence of comorbid conditions. Even with early stage disease, signs and symptoms of local tumor growth, including cough, dyspnea, wheezing, or hemoptysis, frequently are encountered. Additionally, patients may present with purulent sputum, fever, and chills from a postobstructive pneumonia. More ominous signs and symptoms of local tumor growth include superior vena cava syndrome, Horner syndrome, dysphagia, odynophagia, hoarseness (recurrent laryngeal nerve involvement), elevated hemidiaphragm (phrenic nerve impingement), dyspnea and chest pain (pleural effusion), and dyspnea and hemodynamic compromise (pericardial involvement).

Metastatic disease usually is accompanied by malaise and anorexia; weight loss also may be present. Other symptoms reflect the site(s) of metastases: supraclavicular and cervical lymph nodes, brain, bone, liver, and adrenal glands are the most common. The presence of a paraneoplastic syndrome does not necessarily imply metastatic disease. The most common systems involved are endocrine-metabolic, neuromuscular, hematologic-vascular, dermatologic, and skeletal/connective tissue.

DIAGNOSIS

A tissue diagnosis is essential to distinguish non–small-cell lung carcinoma (NSCLC) from small-cell lung carcinoma (SCLC), and ideally is carried out with the least invasive means to accurately establish the diagnosis and extent of disease. For solitary pulmonary nodules (defined as lesions <3 cm in diameter), the initial challenge is to distinguish a benign from a malignant lesion. In some cases, the lesion is sufficiently proximal or localized to the central airways and bronchoscopic diagnosis resolves the diagnosis via direct visualization with bronchoscopic biopsy or fluoroscopically guided transbronchial biopsy. The examination of bronchoscopic brushings and washings increases the diagnostic yield. The diagnostic capability of bronchoscopy is well defined: the yield is over 90% for airway-visualized lesions, over 80% for non-visualized lesions greater than 4 cm in size, and greater than 60% for non-visualized lesions 2 to 4 cm in size. Peripheral lung lesions generally require a CT-guided transthoracic needle aspiration for the highest diagnostic yield. Regardless of the modality, a “negative” biopsy should be regarded as non-diagnostic, especially when the pretest probability of malignancy is moderate to high, and cannot be used to exclude the possibility of cancer. Unless a specific diagnosis is obtained, thoracotomy may be indicated for excisional biopsy and definitive anatomic resection in the same operative setting. In exceptional cases, patients who are high-risk for surgery (e.g., due to severe emphysema) may require empiric radiotherapy to treat a lesion that is highly suspicious but otherwise not safely accessible by biopsy.

While techniques such as conventional diagnostic bronchoscopy and transthoracic needle aspiration may often yield the tissue diagnosis, they are generally limited in the accurate evaluation of suspicious mediastinal lymphadenopathy. The latter may first appear on CT scans, though CT scans are limited in sensitivity and specificity in this setting for mediastinal lymphadenopathy greater than 1 cm in size in short axis (approximately 60% and 80%, respectively). In addition to a careful clinical survey for possible distant sites of metastases, two additional modalities have now become valuable additions to our diagnostic capability: these include positron emission tomography (PET) scanning, which relies on the specific uptake of radiolabeled glucose, and endobronchial ultrasound (EBUS), which has become especially useful in the tissue diagnosis of mediastinal lymphadenopathy. While PET scanning and EBUS now serve very important roles in staging of the mediastinum, one may often obtain the first tissue diagnosis of lung cancer through EBUS–fine-needle aspiration (FNA; i.e., positive lymph node[s], primary lesion, or both). For solitary pulmonary nodules, PET achieves approximately 94% sensitivity and 83% specificity, provided nodule size is 8 mm or greater, while for mediastinal lymphadenopathy the values are in the 85% to 90% range. For diagnosing carcinoma involvement of mediastinal lymph nodes, several analyses now combine the use of PET/CT and EBUS, increasing sensitivity to approximately 92%, while specificity approaches 100%.

The use of PET increasingly is combined in one procedure with CT, as a “PET/CT,” and is integrated into the staging algorithm for any primary lesion that is either suspicious or even proven to be lung cancer, as it both strongly modifies the pretest probability of cancer and aids in assessing stage. There are, however, some important diagnostic caveats for PET/CT. First, its sensitivity falls off markedly for lesions less than 0.8 cm in mean diameter; thus, for subcentimeter lesions one cannot rely on modifying the probability of cancer with PET, and other algorithms for radiographic tracking and biopsy decisions such as the Fleischner Radiological Society guidelines, 2005 must be applied for any lesion likely to be lung cancer. Second, one must consider the fact that a highly suspicious primary lesion on chest X-ray or CT imaging that is greater than 0.8 cm may represent a slow-growing carcinoma even in the face of a negative or “cold” PET scan. This may occur, for example, with low-mitotic activity lung adenocarcinomas, including adenocarcinoma in situ (formerly known as “bronchoalveolar carcinoma”) or with carcinoid tumors. Thus, one must be careful not to ignore supracentimeter PET-negative lesions that are otherwise suspicious for carcinoma by CT and (or) have other concerning clinical characteristics.

STAGING

For non–small-cell lung cancer, the international lung cancer staging system is used (see Table 101-1). This relies on TNM definitions that are carefully outlined. Several classification changes have been made since the last revision, including (1) multiple nodules or “satellite” lesions in the same lobe as the primary tumor are defined as T3 (and thus potentially resectable); (2) size matters in terms of prognosis, so T1 and T2 subdivisions (a and b) exist and a lesion greater than 7 cm is considered T3; (3) T4 tumors include satellite lesions found in more than one lobe but confined to one/ipsilateral lung; (4) satellite nodules in both lungs represent M1 disease; (5) malignant pleural effusion is considered M1 disease; and (6) T4N0M0 or T4N1M0 lesions are now classified as stage IIIA. Nodal stage definitions remain unchanged.

To define the T (tumor) characteristics of a lesion, chest radiograph, CT scanning, PET, magnetic resonance imaging (MRI), and bronchoscopy may all be helpful. MRI may be particularly useful in assessing patients with possible T3 or T4 lesions and for specific assessment of brachial plexus involvement (i.e., superior sulcus tumors) and/or chest wall invasion. In some cases, video-assisted thoracoscopic (VATS) techniques may be needed to assess or better define T4 disease.

Only gold members can continue reading. Log In or Register to continue