Lung cancer is the most common cause of cancer deaths worldwide. In the United States, lung cancer accounted for more than 250,000 cases of cancer and greater than 150,000 deaths from cancer in 2013. Chapter 52 is dedicated to the epidemiology of lung cancer but some of the information found therein is worth repeating. For example, the number of lung cancer deaths each year exceeds the number of cancer deaths from breast, colon, and prostate cancer combined. A common misconception among the general public is that breast cancer accounts for more cancer deaths in women. However, lung cancer is now the greatest cancer killer of women and will account for 25% of cancer-related deaths among women in the United States. Particularly alarming is the fact that young women are in the fastest rising demographic of new cigarette smokers in the United States. Many have clearly made the association between weight control and cigarette smoking. This will have effects on the prevalence of lung cancer in the decades to come.
One of the more disturbing trends in lung cancer is the explosion in rates of lung cancer in countries of the developing world. In 1985, it was estimated that there were 921,000 lung cancer deaths worldwide—an increase of 17% from 1980. In 2011, lung cancer accounted for 13% of cases (1.6 million) and 18% of deaths (1.4 million) worldwide. The International Agency for Research on Cancer in France found that the rates of lung cancer in Africa in the early 1990s were similar to those in the United States in the 1930s, at about 5 per 100,000. By 1999, the rate of lung cancer in males in developing countries was 14 in 100,000 and on the rise, compared with a rate of 71 in 100,000 in developed countries, which continues to decline. These rates may actually be underestimates of the true rates of lung cancer, because many cases may go undiagnosed or underreported in areas where health care is not readily available. The seriousness of this problem is exemplified in China, where it is estimated that nearly 800,000 Chinese men died of lung cancer in 1998. It remains imperative that the medical community devotes much of its educational efforts and resources to the elimination of cigarette smoking worldwide, which would nearly eliminate the development of lung cancer (see also Chapter 46 ).
This chapter reviews the current strategies for the diagnosis, reviews the staging system, and describes the current treatment of lung cancer. As much as possible, evidence-based review of the best currently available literature has been included. The role of the pulmonologist is described for every aspect of lung cancer care, from diagnosis to staging to caring for the complications of the disease itself and the complications of cancer treatment.
Screening for Lung Cancer
The U.S. Preventive Services Task Force now recommends screening for lung cancer with low dose computed tomography (CT) in high-risk patients (see Chapter 18 ). Before 2013, there was insufficient evidence to support screening for lung cancer. For example, the 2004 U.S. Preventive Services Task Force position was based on the results of five randomized, controlled trials that suggested that neither chest radiography nor sputum cytology satisfies the primary criterion of a beneficial screening test: a reduction in lung cancer mortality. One deficiency was that most of these studies did not include a “no screening” arm. Others arguably had an inadequate sample size. In 2011, the results of the lung arm of the National Cancer Institute’s Prostate, Lung, Colorectal, and Ovarian Trial demonstrated that, in a randomized trial of chest radiography versus no screening in a low-risk population of both genders, there was no reduction in mortality from lung cancer in the chest radiography screened group compared with usual care.
Chest CT has been shown to be much more sensitive for detecting pulmonary nodules than the standard chest radiograph. There have been a number of single-arm screening trials utilizing low-dose computed tomography (LDCT), defined as a single breath-hold scan that exposes the patient to a five to six times smaller radiation dose than a standard CT scan, with or without sputum cytology. Although these trials consistently showed that chest CT detects more lung cancers than chest radiography, they were not designed to provide information on mortality benefit. Trials without control arms can be subject to different potential sources of bias. Lead-time bias is the detection of tumors earlier in their course; length-time bias is the greater detection of less aggressive, slower-growing tumors than more aggressive, faster-growing tumors; and overdiagnosis is the detection of tumors that would otherwise never cause symptoms or death.
The National Lung Cancer Screening Trial (NLST) is the first large-scale randomized trial to demonstrate a convincing mortality benefit for LDCT lung cancer screening in high-risk individuals. The trial included 33 centers across the United States. Eligible participants were between ages 55 and 74 years at the time of randomization, had a history of at least 30 pack-years of cigarette smoking, and, if former smokers, had quit within the past 15 years. A total of 53,454 persons were enrolled; 26,722 were randomly assigned to screening with LDCT and 26,732 to screening with chest radiography. Any noncalcified nodule found on LDCT measuring at least 4 mm in any diameter and chest radiographic images with any noncalcified nodule or mass were classified as positive. The LDCT-screened group had a substantially higher rate of positive screening tests compared with the radiography group (round 1: 27.3% vs. 9.2%; round 2: 27.9% vs. 6.2%; round 3: 16.8% vs. 5%). Overall, 39.1% of participants in the LDCT group and 16% in the radiography group had at least one positive screening result. Of those with a positive screening result, the false-positive rate was 96.4% in the LDCT group and 94.5% in the radiography group.
In the LDCT group, 649 cancers were diagnosed after a positive screening test, 44 after a negative screening test, and 367 among participants who either missed the screening or received the diagnosis after the completion of the screening phase. In the radiography group, 279 cancers were diagnosed after a positive screening test, 137 after a negative screening test, and 525 among participants who either missed the screening or received the diagnosis after the completion of the screening phase. There were a total of 356 deaths from lung cancer in the LDCT group and 443 in the chest radiography group, with a relative reduction in the rate of death from lung cancer of 20% with LDCT screening. Overall mortality was reduced by 6.7%. The number needed to screen with LDCT to prevent 1 death from lung cancer was 320, which is comparable to the numbers in studies on screening mammography for breast cancer in women older than 50 years.
One of the concerns with using LDCT for lung cancer screening is the high rate of positive test results necessitating further workup. Investigators from the NELSON study demonstrated that this can be overcome by using semiautomated volumetry software to measure diameter and volume doubling time. Growth was defined as a change in volume between the first and the second scan of 25% or greater. Nodules meeting growth criteria were then classified into three categories based on volume doubling time (<400 days; 400–600 days; >600 days). This approach to nodule management resulted in a decrease in the rate of positive test results at baseline from 30% to 2%. The final results regarding the reduction in mortality from lung cancer in this trial are pending.
Despite the impressive results from NLST in high-risk adults, the ability to generalize the results to other populations has been questioned. Participants in the NLST were enrolled in urban tertiary care hospitals with expertise in all aspects of cancer care. LDCT studies were interpreted by dedicated chest radiologists with expertise in characterizing nodules and providing appropriate recommendations for follow-up. As a result, few patients required further invasive testing, and radiographic follow-up was sufficient for many.
In contrast, community practice gives rise to the potential for considerable variation in the management of solitary pulmonary nodules identified by screening LDCT. One study demonstrated a twofold variation among geographic regions in use of CT-guided biopsy, ranging from 14.7 to 36.2 per 100,000 adults. This substantial variation in the management of solitary pulmonary nodules may lead to an increased number of invasive procedures with risk of harm. For instance, complications from transthoracic biopsies include a 1% rate of bleeding (with one third of affected patients requiring transfusion) and a 15% rate of pneumothorax. More than 6% of CT-guided biopsies result in a pneumothorax requiring chest tube drainage, a clinically important complication that results in pain, serial imaging with radiation exposure, and hospitalization. Older patients and those with chronic obstructive pulmonary disease (COPD) have an increased risk of biopsy-related complications that can result in longer length of hospital stay, contributing both to increased cost and higher rates of respiratory failure that can affect long-term health.
Another difference between the results of the NLST and community practice is that the mortality from lung cancer surgery was 1% in the NLST, whereas the national average is between 3% and 5% for a lobectomy. Whereas NLST participants were allowed to choose where they had their evaluation and management for screen-detected nodules, many were managed at an NLST site with high volume and dedicated thoracic surgery support, both of which have been shown to have better outcomes. Even though 70 years is the average age at lung cancer diagnosis, only 9% of the NLST study population was older than age 70. The patients enrolled in NLST were younger and healthier than persons who would participate in a broad-based lung cancer screening. Participants had to be medically fit to undergo surgery. Those screened in the NLST were also less likely to be current smokers, less ethnically diverse, and more educated than the general U.S. population. These differences follow the healthy volunteer effect of screening trials in which there is a self-selection of persons who are better educated and more health-conscious and who have better access to medical care.
The evidence of the NLST contributed to a systematic review that serves as the basis for a multisociety recommendation for screening in those people meeting the NLST entry criteria. The caveat to the recommendations is that screening should only be done in those centers with multidisciplinary groups capable of providing comprehensive care as was present in the trial. Based largely in part on the results of the NLST, in 2013, the U.S. Preventive Services Task Force published a draft recommendation giving lung cancer screening a grade B recommendation (moderate certainty that annual screening for lung cancer with LDCT is of moderate net benefit in high-risk asymptomatic persons). The U.S. Preventive Services Task Force found there to be adequate evidence to screen asymptomatic patients aged 55 to 79 years with significant tobacco use history. Their assessment was that the moderate net benefit of screening depends on two factors: (1) the accuracy of image interpretation would be comparable to that in the NLST and (2) most false-positive could be handled without invasive procedures.
There may be potential barriers to lung cancer screening, especially in current smokers. In one study, current smokers were less likely to believe that early cancer detection would result in a good chance of survival. Current smokers are also less likely to consider CT screening for lung cancer (71.2%) than are never smokers (87.6%). In addition, only half of the current smokers surveyed would opt for surgical resection of a screening-diagnosed cancer. Finally, it is significant that smokers make up 31% of the population below the poverty line compared with 20% of those at or above the poverty line ; as a result, smokers are likely to be a target population more difficult to reach for large-scale screening in the community.
Unfortunately, the symptoms of lung cancer can be nonspecific and variable, thereby delaying diagnosis and frequently leading to an advanced stage at the time of diagnosis. The focus of an initial patient evaluation should include signs and symptoms related to the following: local tumor effects, extension of disease into the thoracic cavity, radiologic correlation, paraneoplastic syndromes, and distant metastatic disease. Whereas only 40% of patients with lung cancer in a screened high-risk outpatient population had symptoms, 98% of patients in a hospitalized population presented with symptoms. In general, only approximately one fourth of patients are asymptomatic at the time lung cancer is diagnosed, and these patients are more likely to have less advanced disease. Table 53-1 displays some of the common symptoms associated with the presentation of lung cancer. Most are nonspecific; however, some clues can be gained from the history, thus raising the clinician’s suspicion that lung cancer is present.
|Symptoms and Signs||Frequency, %|
|Superior vena cava obstruction||0–4|
|Wheezing and stridor||0–2|
Although many smokers cough, lung cancer patients usually admit to a change in the character of their cough. The cough can increase in frequency or strength, or may not be relieved with local measures. Chest pain can be present in 25% to 50% of patients at the time of presentation for evaluation for lung cancer. The pain is generally dull in nature, tends to be persistent, remains in the same location, and is not relieved with local measures. Chest pain is usually related to involvement of the pleura but can be related to extension into the mediastinum or chest wall. However, chest pain in and of itself does not preclude the patient from consideration for surgery with curative intent. Dyspnea is frequently a complaint of patients who present with bronchogenic carcinoma, noted in half of all new patients at presentation. A partial list of the reasons for dyspnea related to lung cancer includes pulmonary embolism, superior vena cava (SVC) syndrome, deconditioning, reactive airway disease, endobronchial obstruction with tumor, prior obstructive pneumonia, hemoptysis, hemorrhage, malignant pleural effusion, and extrinsic compression of the airway by tumor.
Hemoptysis in a smoker should raise suspicion of lung cancer. Hemoptysis can present as blood streaking of the sputum and can be noted over a lengthy period of time before presentation to the physician’s office because the patient attributes it to smoking-related bronchitis. The clinician should not be led astray, even if the chest radiograph is normal, because up to 5% of patients with hemoptysis and a smoking history and a normal radiograph can harbor lung cancer. Because of the vascular nature of lung cancer, patients can also present with massive hemoptysis.
Weight loss, a nonspecific symptom, in the right clinical setting should raise the suspicion of both lung cancer and metastatic disease. Weight loss alone has been correlated with an advanced presentation and poor outcome in lung cancer cases.
In summary, patients with lung cancer can present asymptomatically or with relatively nonspecific symptoms of underlying pulmonary disease. There are often clues in the history that should alert the clinician that lung cancer is a possibility and further investigation is warranted.
Lung Cancer Staging
Perhaps the most critical role of the pulmonologist in the management of lung cancer is in the diagnostic and staging evaluation of the patient. Accurately staging patients with newly diagnosed lung cancer is critical because staging dictates the patient’s treatment options and predicts survival. It is intuitive that early-stage disease has a much better survival than late-stage disease. What may not be so obvious is that simple staging procedures are available to the diagnostician that can help stage patients accurately. The treatment options for lung cancer have now evolved so that treatment for patients in different stages is vastly different. In general, stage I (early-stage lung cancer) is treated with surgery alone. Stage II lung cancer (a less common stage, intermediate between early and locally advanced) is treated with surgery followed by adjuvant chemotherapy. Stage IIIA and B (locally advanced lung cancer) is treated with a combination of chemotherapy and radiotherapy, and stage IV (metastatic disease) is treated with chemotherapy alone. However, there are important exceptions to these general rules that are discussed later in this chapter.
The staging of non–small cell lung cancer (NSCLC) using the tumor-node-metastasis (TNM) classification underwent a major revision in 2007. The new staging system is remarkable in that it is based on more than 100,000 cases of lung cancer from 23 institutions, 12 countries, and 3 continents. The data are robust, internally validated, and externally validated against the Surveillance Epidemiology and End Results cancer registry. Tables 53-2 and 53-3 show the current TNM descriptors and stage groupings.
|T (PRIMARY TUMOR)|
|TX||Primary tumor cannot be assessed, or tumor proven by the presence of malignant cells in sputum or bronchial washings but not visualized by imaging or bronchoscopy|
|T0||No evidence of primary tumor|
|Tis||Carcinoma in situ|
|T1||Tumor ≤ 3 cm in greatest dimension, surrounded by lung or visceral pleura, without bronchoscopic evidence of invasion more proximal than the lobar bronchus (i.e., not in the main bronchus) *|
|T1a||Tumor ≤ 2 cm in greatest dimension|
|T1b||Tumor > 2 cm but ≤ 3 cm in greatest dimension|
|T2||Tumor >3 cm but ≤ 7 cm or tumor with any of the following features (T2 tumors with these features are classified T2a if <5 cm) |
Involves main bronchus, ≥ 2 cm distal to the carina
Invades visceral pleura
Associated with atelectasis or obstructive pneumonitis that extends to the hilar region but does not involve the entire lung
|T2a||Tumor >3 cm but ≤ 5 cm in greatest dimension|
|T2b||Tumor >5 cm but ≤ 7 cm in greatest dimension|
|T3||Tumor > 7 cm or one that directly invades any of the following: chest wall (including superior sulcus tumors), diaphragm, phrenic nerve, mediastinal pleura, parietal pericardium; or tumor in the main bronchus < 2 cm distal to the carina * but without involvement of the carina; or associated atelectasis or obstructive pneumonitis of the entire lung or separate tumor nodule(s) in the same lobe|
|T4||Tumor of any size that invades any of the following: mediastinum, heart, great vessels, trachea, recurrent laryngeal nerve, esophagus, vertebral body, carina; separate tumor nodule(s) in a different ipsilateral lobe|
|N (REGIONAL LYMPH NODES)|
|NX||Regional lymph nodes cannot be assessed|
|N0||No regional lymph node metastasis|
|N1||Metastasis in ipsilateral peribronchial and/or ipsilateral hilar lymph nodes and intrapulmonary nodes, including involvement by direct extension|
|N2||Metastasis in ipsilateral mediastinal and/or subcarinal lymph node(s)|
|N3||Metastasis in contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s)|
|M (DISTANT METASTASIS)|
|MX||Distant metastasis cannot be assessed|
|M0||No distant metastasis|
|M1a||Separate tumor nodule(s) in a contralateral lobe; tumor with pleural nodules or malignant pleural (or pericardial) effusion †|
† Most pleural (and pericardial) effusions with lung cancer are due to tumor. In a few patients, however, multiple cytopathologic examinations of pleural (pericardial) fluid are negative for tumor, and the fluid is nonbloody and is not an exudate. Where these elements and clinical judgment dictate that the effusion is not related to the tumor, the effusion should be excluded as a staging element and the patient should be classified as T1, T2, T3, or T4.
|T and M Descriptor (6th edition)||T and M Descriptor (7th edition)||N0||N1||N2||N3|
|T1 (≤2 cm)||T1a||IA||IIA||IIIA||IIIB|
|T1 (>2 to 3 cm)||T1b||IA||IIA||IIIA||IIIB|
|T2 (≤5 cm)||T2a||IB||IIA||IIIA||IIIB|
|T2 (>5 to 7 cm)||T2b||IIA||IIB||IIIA||IIIB|
|T2 (>7 cm)||T3||IIB||IIIA||IIIA||IIIB|
|T4 (same lobe nodules)||T3||IIB||IIIA||IIIA||IIIB|
|M1 (ipsilateral lung)||T4||IIIA||IIIA||IIIB||IIIB|
|T4 (pleural effusion)||M1a||IV||IV||IV||IV|
|M1 (contralateral lung)||M1a||IV||IV||IV||IV|
There are several important changes adopted in the 2007 revision. The main modifications are in the T and M classification; the N status remains the same. Within the T classification, tumor size was found to be an important prognosticator and the T factor was subdivided based upon five different size criteria. Because survivorship was better than previously thought, a primary tumor with satellite nodules in the same lobe was reclassified from T4 to T3 and a tumor with additional nodules in a different lobe of the ipsilateral lung was moved from an M1 designation to T4. This change in classification and stage allows more patients to be considered for surgery. Malignant pleural effusion was reclassified from T4 (or stage IIIB) disease to M1 disease because the survival of patients in this group was found to resemble the survival of those with metastatic disease more closely than those with locally advanced disease. Another significant change is that the M status is now split into M1a (metastatic disease confined to the chest) and M1b (extrathoracic metastatic disease) because survival was found to be better in those patients with metastatic disease confined to the thorax compared with those with extrathoracic metastases.
Whereas the TNM staging system is applied to NSCLC, a more simplified version is employed for patients with small cell lung cancer (SCLC). In this classification, patients are classified as having limited or extensive disease. Limited disease (LD) is disease limited to one hemithorax, although it can include supraclavicular and mediastinal lymphadenopathy; extensive disease (ED) is any disease outside of the hemithorax. The implication in this classification is that LD is treated with chemotherapy and radiotherapy and ED is treated with chemotherapy alone. Malignant pleural effusion can technically be categorized as LD in the staging classification for SCLC if the patient otherwise meets criteria. However, for all intents and purposes, patients with malignant pleural effusions and SCLC have the same characteristics as those with ED, and the large cooperative group trials have treated them as such.
Some important nuances to staging are described in detail later in this chapter. However, there are certain tenets of staging that must be emphasized. Before classifying a patient within a certain stage, the clinician should make every effort to verify any noninvasive radiologic findings with tissue confirmation of malignancy. This is particularly important when surgical resection would be precluded based on noninvasive radiologic tests. Thus a patient who would otherwise be resectable except for a single abnormality suspicious for metastasis should have tissue confirmation of that abnormality before being deemed unresectable. As is pointed out later in this chapter, no noninvasive radiologic study is infallible. In studies of the mediastinum, false-positive findings range from 12% in positron emission tomography (PET) scans to nearly 20% in CT scans ( eFigs. 53-1 and 53-2 ) with lower false-positive findings using integrated PET-CT, at the cost of lower sensitivity. Therefore reliance on the scan alone to predict malignancy is simply not appropriate.
Staging can be accomplished by a number of noninvasive and invasive studies. The choice of the most appropriate study rests with the clinician and is based on how the patient presents. Some patients—for example, those with isolated solitary pulmonary nodules—may be referred for immediate surgical resection as both a diagnostic and a therapeutic maneuver. Others, such as those with extremely poor performance status or those suspected of widely metastatic disease, may undergo no testing at all. The next sections present a discussion of the attributes of each of the staging options available, divided into noninvasive and invasive techniques.
Noninvasive Staging Techniques
Many lung cancers are detected initially by plain chest radiograph. In certain situations, chest radiography may be sufficient to detect spread to the mediastinum. For example, the presence of bulky lymphadenopathy in the superior or contralateral mediastinal areas may be considered adequate evidence of metastatic disease to preclude further imaging evaluation of the chest. This may be particularly true if the patient is too ill or unwilling to undergo treatment of any kind. Still, most patients should undergo a chest CT scan unless they are so debilitated that no further evaluation or treatment is planned. The chest radiograph is simply too insensitive a measure of mediastinal lymph node involvement with lung cancer and thus further noninvasive or invasive assessment is usually necessary.
Chest Computed Tomography
The vast majority of patients who present with lung cancer will undergo chest CT, which can confirm the suspicion of lung cancer or raise the suspicion of an alternate diagnosis. CT is helpful in defining the size, location, and characteristics of the primary mass (e.g., circumscribed, spiculated, calcified), the presence or absence of lymphadenopathy, and, if performed through the adrenal glands, the presence of abnormalities in the liver and adrenal glands. The bony structures of the thoracic cavity can also be evaluated by chest CT.
Chest CT is the most widely available and commonly used noninvasive modality for evaluation of the mediastinum in lung cancer. Numerous studies of CT have been performed comparing clinical staging by CT with the “gold standards” of mediastinoscopy or surgery. The results demonstrated that, regardless of the lymph node size used as a threshold for defining malignant adenopathy, CT findings in isolation could not be considered as conclusive evidence that lymph nodes were malignant. In other words, in all studies, there are meaningful numbers of false-positive cases detected by CT (see eFigs. 53-1 and 53-2 ). The vast majority of reports evaluating the accuracy of CT for mediastinal lymph node staging have employed the administration of intravenous contrast material. Although contrast is not absolutely necessary in performing chest CT for this indication, it is helpful to distinguish vascular structures from lymph nodes as well as to delineate mediastinal invasion from centrally located tumors. The most widely accepted criterion for an abnormal lymph node on CT is a lymph node with a diameter of 1 cm or greater across the short axis.
The American College of Chest Physicians (ACCP) compiled the studies assessing the performance characteristics of CT for staging the mediastinum in a meta-analytic format. Thirty-five studies were identified, comprising 5111 evaluable patients. The pooled sensitivity of CT for staging the mediastinum was 51% (95% CI, 47% to 54%) and the pooled specificity was 86% (95% CI, 84% to 88%). The corresponding positive and negative likelihood ratios were 3.4 and 0.6, respectively, confirming that CT has a limited ability to either confirm or exclude mediastinal metastasis. However, because CT guides the selection of nodes for biopsy by mediastinoscopy or transbronchial, transthoracic, or transesophageal needle aspiration, it remains an important diagnostic tool in lung cancer. The limitation of CT-based mediastinal lymph node evaluation is evident in that 5% to 15% of patients with clinical T1N0 lesions will be found to have positive lymph node involvement by surgical lymph node sampling ( eFig. 53-3 ). Perhaps the most important message in evaluating the accuracy of CT is that approximately 40% of all nodes deemed malignant by CT criteria are actually benign (see eFigs. 53-1 and 53-2 ), depending on the patient population. Specificity can be affected by clinical factors such as the presence of obstructive pneumonitis. There is no node size that can reliably determine stage and operability. When CT criteria for identification of a metastatic node are met, the clinician must still prove beyond reasonable doubt by biopsy or resection that the node is indeed malignant. Given the limitations of the imperfect sensitivity and specificity of CT, it is usually inappropriate to rely solely on CT to determine mediastinal lymph node status. Nonetheless, CT continues to play an important and necessary role in the evaluation of patients with either a known or suspected lung cancer who are eligible for treatment.
CT can also be helpful in the evaluation of pleural effusion in patients with lung cancer. The CT scan can indicate the presence or absence of fluid, the contour of the pleural space, and whether or not nodules or masses are present on the pleural surface ( eFig. 53-4A and B ). However, the clinician should interpret these findings with caution because pleural disease can predate cancer and the presence of pleural effusion does not guarantee that the cytology will be positive. This is an important staging issue because the finding of malignant pleural effusion in NSCLC is considered evidence of metastatic disease (stage IV). If the pleural fluid has benign cytology (e.g., it represents fluid from obstructive pneumonia), then the patient may still be considered for surgical resection. To resolve this issue, recommendations have been made to perform thoracentesis with cytology on two separate occasions, followed by thoracoscopy to evaluate the pleural surface directly ( eFig. 53-5 ). If the patient remains cytology-negative, then the patient should be considered to be at the lower, nonmetastatic stage and be treated accordingly. Thoracoscopy can be helpful in differentiating the extent of the primary tumor involvement into or through the pleura but, at times, open thoracotomy is needed to sort out this issue. (See Chapters 24 and 82 .)
Positron Emission Tomography
Perhaps the single most notable addition to the staging armamentarium for the evaluation of lung cancer is PET (see Chapter 21 ). Because the image is created by the biologic activity of neoplastic cells, PET is a metabolic imaging technique based on the function of a tissue rather than on its anatomy. Lung cancer cells demonstrate increased cellular uptake of glucose and a higher rate of glycolysis when compared with normal cells. The radiolabeled glucose analogue 18 F-fluorodeoxyglucose (FDG) undergoes the same cellular uptake as glucose but, after phosphorylation, is not further metabolized and becomes trapped in cells. Accumulation of the isotope can then be identified using a PET detector. Specific criteria for an abnormal PET scan are either a standard uptake value of greater than 2.5 or uptake in the lesion that is greater than the background activity of the mediastinum (see eFig. 53-1B and C , eFig. 53-3B, E , and F , and Fig. 21-1F-J ). It has proved useful in differentiating neoplastic from normal tissues. In two well-performed studies that evaluated the use of PET in the preoperative setting for lung cancer, nearly 20% of patients were staged differently after evaluation by PET (see eFigs. 53-1, 53-2 , and 53-3 , and Fig. 21-3 ), However, the technique is not infallible because certain nonneoplastic processes, including granulomatous ( eFig. 53-6 ) and other inflammatory diseases and infections may also demonstrate positive PET imaging. Furthermore, size limitations are also an issue, with the lower limit of resolution of the study being approximately 7 to 8 mm depending on the intensity of uptake of the isotope in the abnormal cells ( eFig. 53-7A-D ). One should not rely on a negative PET finding for lesions less than 1 cm on CT.
A burgeoning number of studies in the last several years have reported on the utility of PET in the assessment of the mediastinum in patients with lung cancer. Increasing availability of the technology now allows PET to be used widely as a diagnostic tool. PET is primarily a metabolic examination and has limited anatomic resolution. It is possible for PET to identify lymph node stations but not individual lymph nodes. CT provides much more anatomic detail but lacks the functional information provided by PET. The third edition of the ACCP guidelines on lung cancer reviewed the complexity surrounding PET. While PET provides information about the primary tumor, mediastinal lymph nodes, and sites of distant metastasis, the contribution that it makes to the stage evaluation is influenced by a multitude of factors. These include the probability of cancer, likelihood of metastasis, and the extent that the investigation for metastases has been completed by other modalities.
To date, there have been five randomized controlled trials to evaluate the role of PET. The variation in results among the studies was likely due to the significant differences among the patients involved, the prior evaluations, and risk for advanced disease. While two of the studies demonstrated a reduction in the number of noncurative resections from 40% to 20%, another detected no difference in the number of thoracotomies performed or of sites of distant metastasis. This latter study primarily involved stage I patients with extensive preenrollment imaging explaining why there was no difference detected between the PET and conventional staging arms.
Population-based studies using the U.S. National Cancer Data Base and the Surveillance, Epidemiology and End Results registry suggest that the use of PET has had a positive impact on stage migration from stage III to stage IV classification ( eFig. 53-8 ). Conversely, PET adds little to the staging of patients with clinical stage I cancers. PET is also of limited value in those with ground-glass opacities with or without a solid component because these patients are at low risk for nodal and distant metastatic disease.
Compared with conventional staging, staging by PET is, on average, 20% more correct in identifying nodal or distant metastasis in randomized controlled trials. Confirmation of PET findings, however, is imperative because, as is known for CT, PET can be wrong. PET also carries the possibility of incorrectly upstaging a patient. Whereas a negative mediastinal PET may obviate the need for mediastinoscopy before thoracotomy in certain situations, a positive mediastinal PET should not negate further evaluation or the possibility of resection. In the latter case, lymph node sampling should still be pursued because the possibility of a false-positive PET scan cannot be ignored.
Where PET is available, a PET scan should be obtained during the staging evaluation for lung cancer. Newer technology includes CT-PET fusion, a single machine that incorporates CT and PET during the same scan. This allows the clinician to obtain anatomic (CT) and functional (PET) images simultaneously. Studies suggest an improvement in the number of patients correctly staged with this modality over CT or PET alone. The future of PET in lung cancer may also include its use to evaluate response to treatment. Due to potential false positive images, the optimal timing for PET following treatments will require further evaluation, especially when chest radiotherapy is used. For example, it may be more useful later in the post- stereotactic body radiation therapy (SBRT) period (i.e., after the first year) rather than early in the period (i.e., within the first 3 to 6 months).
Magnetic Resonance Imaging
There are very few circumstances in which magnetic resonance imaging (MRI) is a useful tool in staging lung cancer. However, MRI can be useful in evaluating superior sulcus tumors, especially for possible invasion of the brachial plexus, and for evaluating vertebral invasion.
Search for Metastatic Disease
The purpose of extrathoracic scanning in NSCLC is usually to detect metastatic disease at common metastatic sites, such as the adrenal glands, liver, brain, and skeletal system, thereby sparing the patient fruitless surgical intervention. CT of the chest ( eFig. 53-9A ), CT ( eFig. 53-9C ) or MRI with contrast of the brain, and 99m Tc nuclear imaging of the skeletal system are the conventional staging studies when the clinician needs to evaluate for metastatic disease. The use of whole-body PET and PET-CT (see eFig. 53-8 ) for extrathoracic staging has advanced the field of metastatic disease evaluation. Studies demonstrate that PET and PET-CT outperform conventional staging tests in the evaluation of metastatic disease to key specific distant sites including adrenal glands, liver, and bone (see eFig. 53-8 ). PET reveals unsuspected metastases in 6% to 37% (see eFig. 53-8 and Fig. 21-4 ). This allows for more accurate TNM staging, stage shift, and changes in management including more appropriate consideration of surgical candidacy. Detection of brain metastases is a problem for PET because the high background brain FDG uptake can mask the small size of most brain metastases, which can be either hypermetabolic or hypometabolic. There is some evidence to suggest that integrated PET-CT has an accuracy that nears diagnostic brain CT (see eFig. 53-8B ).
The initial clinical evaluation may reveal abnormalities, such as abnormal symptoms, physical findings, and routine blood tests, that then lead to an expanded clinical evaluation ( Table 53-4 ). If a patient has abnormalities on these clinical evaluations, scans will be abnormal in around 50% of cases. The third iteration of the ACCP guidelines for NSCLC staging recommends PET to evaluate for metastasis (except for metastasis to the brain) for patients with a normal clinical evaluation and no suspicious extrathoracic abnormalities on chest CT being considered for curative-intent treatment. Advanced thoracic lesions and mediastinal lymphadenopathy are important variables because these are associated with more scan abnormalities. This is particularly true for N2 disease, in which asymptomatic metastases have been documented at a higher rate than would have been expected (see eFig. 53-8 ). Although several studies have documented a higher incidence of brain metastases in patients with adenocarcinomas than in those with squamous cell cancers, the largest single series of patients with stage I and II lung cancer found no difference.
|SYMPTOMS ELICITED IN HISTORY|
|Constitutional—weight loss >10 lb (>4.5 kg)|
|Musculoskeletal—focal skeletal pain|
|Neurologic—headaches, syncope, seizures, extremity weakness, recent change in mental status|
|SIGNS FOUND ON PHYSICAL EXAMINATION|
|Lymphadenopathy (>1 cm)|
|Hoarseness, superior vena cava syndrome|
|Hepatomegaly (>13-cm span)|
|Focal neurologic signs, papilledema|
|Soft tissue mass|
|ROUTINE LABORATORY TESTS|
|Hematocrit < 40% in males|
|Hematocrit < 35% in females|
|Elevated alkaline phosphatase, GGT, AST, calcium|
Several important caveats must be considered in the search for metastatic disease. First and foremost, there is the problem of false-positive scans. Adrenal adenomas (present in 2% to 9% of the general population), hepatic cysts, degenerative joint disease, old fractures, and a variety of nonmetastatic space-occupying brain lesions are present in the general population. When clinically indicated, additional imaging studies, biopsies, or both are performed to establish the diagnosis; however, complications and costs resulting from such subsequent investigations have received less attention. There is also the problem of false-negative scans, scans that fail to detect actual metastases. For example, in a CT study of the adrenal glands, Pagani found metastatic NSCLC by percutaneous biopsy in 12% of radiologically normal adrenal glands. An additional problem is that the studies in this area often fail to specify exactly which elements make up the prescan clinical evaluation or that the studies use differing clinical indicators. For example, organ-specific findings such as headache and non–organ-specific complaints such as weight loss are both important. In addition, in many studies, abnormal findings on scans have not been pursued with biopsies to prove metastatic disease. Finally, there are few prospective randomized trials and outcome studies to guide appropriate use and interpretation of extrathoracic scanning. It is hoped that careful studies will improve the workup of patients with lung cancer.
Adrenal and Hepatic Imaging
It is relatively common to encounter adrenal masses on routine CT, but many of these lesions are probably unrelated to the malignant process. A unilateral adrenal mass in a patient with NSCLC is more likely to be a metastasis than a benign lesion according to some studies, but not others. In the setting of clinical T1N0 NSCLC, adenomas predominate, whereas in the setting of large intrathoracic tumors or other extrathoracic metastases, adrenal metastases are more common. Many studies suggest that the size of a unilateral adrenal abnormality on CT is an important predictor of metastatic spread, but this is not a universal finding. Lesions larger than 3 cm are more likely to signify metastases (see eFig. 53-9A ), but benign disease is still possible.
For adrenal masses, CT, MRI, PET, percutaneous biopsy, and even adrenalectomy can be used to help distinguish benign from malignant disease. Well-defined, low-attenuation (fatty) lesions with a smooth rim on unenhanced CT are more likely to be benign adenomas, but the CT appearance of many lesions is insufficiently distinctive. Follow-up scanning with repeat CT, serial ultrasonography, MRI (especially with chemical shift and dynamic gadolinium-enhanced techniques ), or 6-β-iodo- 131 I-methyl-norcholesterol scanning can sometimes help with the critical distinction between metastatic disease and adenoma. One study demonstrated the utility of PET-CT in differentiating between benign and metastatic malignant adrenal masses in patients with lung cancer. In 110 adrenal masses ranging in size from 0.5 to 6.3 cm, the sensitivity, specificity, and accuracy for detecting metastatic disease were 97% (74 of 76), 94% (31 of 34), and 95% (105 of 110), respectively ( eFig. 53-10 ). The positive predictive value was 95% (74 of 77), and the negative predictive value was 94% (31 of 33).
Percutaneous adrenal biopsy is a relatively safe and effective means of achieving a definitive diagnosis in doubtful cases and is especially important when the histology of the adrenal mass will dictate subsequent management. However, this procedure may be nondiagnostic or infeasible due to anatomic constraints. When insufficient material results from a biopsy, repeat aspiration or even adrenalectomy should be considered.
Most liver lesions are benign cysts and hemangiomas, but CT with intravenous contrast (or ultrasound) is often required to establish a likely diagnosis. Percutaneous biopsy can be performed when diagnostic certainty is required. PET can detect liver metastases with a diagnostic accuracy ranging from 92% to 100% ( eFig. 53-11 ) with rare false-positive results; however, the data in NSCLC are currently limited.
In most studies, the yield of CT/MRI of the brain in NSCLC patients with negative clinical examinations is 0% to 10%, possibly rendering the test cost-ineffective. The negative predictive value of the clinical evaluation in this setting is 95% (range, 91% to 96%).
An association of positive findings between brain metastases and N2 disease in the chest and adenocarcinoma has been described. The false-negative rate, wherein patients return with brain metastases within 12 months of the original scan, is reported to be 3%. False-positive scans can be a problem in up to 11% of cases due to brain abscesses, gliomas, and other lesions ; therefore biopsy may be essential in cases in which management is critically dependent on the histology of the brain lesion.
MRI is more sensitive than CT of the brain and picks up more lesions and smaller lesions but, in some studies, this has not translated into a clinically meaningful difference in terms of survival. Although studies show that MRI can identify additional brain lesions in a particular patient, there are no studies that show that MRI is better than CT in its ability to identify additional patients with brain metastases from lung cancer. Therefore CT is an acceptable modality for evaluating patients for metastatic disease. However, MRI remains the preferred modality at many centers.
False-positive abnormalities in radionuclide bone scintigraphy are common concerns because of the frequency of degenerative and traumatic skeletal damage and the difficulty in obtaining a definitive diagnosis via follow-up imaging or biopsy. With a negative clinical assessment, the negative predictive value for radionuclide bone imaging is 90%. PET has excellent performance characteristics for determining bone metastases ( eFig. 53-12 ) with a specificity, sensitivity, negative predictive value, and positive predictive value all greater than 90%. The accuracy of PET was superior to radionuclide bone scanning in direct comparison studies.
The noninvasive clinical staging of lung cancer relies on the clinical evaluation and a number of readily available staging studies. The clinician must be wary of abnormal scans that may falsely suggest metastatic disease to the mediastinum and distant sites. Tissue confirmation by whatever means necessary is the rule rather than the exception before deciding on correct stage and the most appropriate treatment. If the patient has clinical findings indicative of metastatic disease, further evaluation is necessary because nearly 50% of the time the patient will have metastases. Even if the clinical evaluation is normal and imaging does not demonstrate suspicious extrathoracic abnormalities, PET imaging should be done where available in those being considered for curative-intent treatment.
Invasive Diagnostic and Staging Techniques
There are a myriad of methods that can be used to diagnose and stage patients with lung cancer. In certain circumstances, this is accomplished with a single test. For example, a positive percutaneous biopsy (or endoscopic ultrasound-guided fine-needle aspiration) of the adrenal gland performed as a first test in a patient with a lung mass will provide both a diagnosis and a stage (stage IV) simultaneously. Every effort should be made to use the least invasive, most accurate procedure to expedite the patient’s treatment, to minimize patient discomfort and inconvenience, and to ensure that the most appropriate treatment is rendered.
Sputum cytology is the least invasive method for obtaining a diagnosis of lung cancer. Its accuracy depends on the expertise of the health care team in obtaining the sample (three samples are required), the preservation technique, and the size and location of the lesion. Central lesions are more likely to yield positive cytologic results than are peripheral lesions. Sputum cytology should be obtained in all patients with central lesions who are at risk for more invasive biopsy techniques and considered in those with hemoptysis with or without a mass on chest radiography. Previously published systematic reviews have summarized the performance characteristics for sputum cytology for the diagnosis of suspected lung cancer. The ranges for sensitivity and specificity were 42% to 97% and 68% to 100%, respectively. The accuracy of sputum cytology is highly variable, so, in patients suspected of having lung cancer with negative sputum cytology, further testing should be performed.
Transthoracic Needle Aspiration
Transthoracic needle aspiration (TTNA), usually under ultrasound (see Fig. 19-3 ), CT, or fluoroscopic guidance, is an expedient and relatively safe way to diagnose the primary tumor mass and establish a diagnosis of lung cancer (see Figs. 19-1 and 19-6 ). As a general rule, if a lesion is less than 3 cm in size and lateral to the mid-clavicular line ( eFig. 53-13 ), TTNA should be considered if tissue diagnosis is necessary. One important point about TTNA or other nonsurgical biopsy techniques for peripheral pulmonary lesions is that they afford no preoperative benefit because they do not eliminate the need for surgery in most cases. For a patient presenting with a solitary pulmonary nodule suspicious for malignancy (e.g., noncalcified, upper lobe, spiculated lesion in a long-term smoker), the diagnosis, stage, and therapy can be accomplished simultaneously with a thoracotomy and surgical resection. Thus TTNA may be essential only in certain situations: patients who are poor surgical candidates but who require tissue diagnosis prior to treatment, patients in whom a noncancerous lesion is strongly suspected (see Fig. 19-7 ), patients who request that a diagnosis of cancer be confirmed before considering surgery, and patients with high likelihood of metastatic disease (see Fig. 19-2 ). The sensitivity and specificity of TTNA are 90% and 97%, respectively (see Chapter 19 ).
One drawback of TTNA is the risk of pneumothorax. Several investigations have reported a 15% to 45% risk of pneumothorax for CT-guided TTNA. Although pneumothorax may lead to hemodynamic compromise without therapeutic tube thoracostomy, in most cases of pneumothorax secondary to TTNA, treatment is not required. The primary factors shown to increase the risk or incidence of pneumothorax are the presence of emphysema, a smaller lesion size, and a greater depth of needle penetration from the pleural surface to the edge of the lesion.
More than 50% of patients with advanced-stage lung cancer will have involvement of the central airways either by bulky endobronchial disease, extension into the airways, or extrinsic compression of the airways by the tumor or by lymphadenopathy. Patients with known or suspected lung cancer may have symptoms due to endobronchial involvement that require airway inspection with bronchoscopy: shortness of breath, unilateral wheezing, hemoptysis, and cough. Endobronchial lesions can be visualized easily and biopsied through a flexible bronchoscope. The yield with three or more biopsies should approach 100% for centrally located lesions. Data from 4507 patients revealed that central endobronchial biopsies provide the highest sensitivity (74%), followed by brushings (61%) and washings (47%). The combination provides a diagnosis in 88% of cases. Endobronchial needle aspiration may be helpful especially when a rim of necrotic debris surrounds an endobronchial malignancy, because deeper tissue penetration may access viable tumor cells. The addition of endobronchial needle aspiration to forceps biopsies and brushings may improve sensitivity to 95% for the diagnosis of endobronchial cancer.
Submucosal and Peribronchial Lesions
When lung cancer presents with submucosal infiltration or extrinsic compression from peribronchial disease, endobronchial forceps biopsy has a lower yield (55%) than transbronchial needle aspiration (TBNA) (71%). In these situations, normal mucosal markings are often obscured and the surface is replaced with bronchial collateral vessels and firmer surface tissue, which may have to be penetrated to reach malignant cells. In addition, peribronchial tumor may be inaccessible to biopsy forceps. TBNA can be more effective if the lesion is close enough to the tracheobronchial tree to be encountered with a 1.3- to 1.5-cm-long needle. Of note, in cases like this when sampling error may be high, diagnostic yields may be improved by combining different methods or by using with endobronchial ultrasound (EBUS)-TBNA, which offers the advantages of real-time ultrasound and an adjustable needle length up to 4 cm.
Navigational bronchoscopy provides a novel option for the diagnosis of peripheral lung lesions (see Chapter 22 ). It has a lower pneumothorax risk than TTNA and a higher diagnostic yield for peripheral lesions than traditional bronchoscopy. In general, there are three types of navigational bronchoscopy: (1) radial probe endobronchial ultrasonography, which is conventional radial EBUS using guide sheaths that allow the use of biopsy tools after successful navigation to the target; (2) virtual bronchoscopy, which creates a CT-based “road map” overlaid on endoscopic real-time images; and (3) electromagnetic navigational bronchoscopy which uses a virtual navigation system with steerable devices. A combination of navigational techniques has been shown to augment the diagnostic yield (88%) when compared to either radial probe endobronchial ultrasonography or electromagnetic navigational bronchoscopy alone. A meta-analysis reported the accuracy and side effect profile of all available guided bronchoscopy procedures. In 39 studies, which together included more than 3000 patients, the pooled diagnostic yield was approximately 70% (with wide variation) and the pneumothorax rate was less than 2% (need for chest tube insertion less than 1%). In addition, navigation may be used for placement of fiducial markers for stereotactic radiation treatments.
Bronchoscopy for Staging Lung Cancer
Initially, the role of bronchoscopy in staging lung cancer was limited to the determination of T (tumor) status. Now bronchoscopy has a crucial role in determining the presence of metastatic deposits of tumor in mediastinal lymph nodes, thus contributing to an accurate and a minimally invasive staging method for lung cancer.
The use of traditional TBNA in staging lung cancer has been reported to be both sensitive and specific in diagnosing spread of cancer to lymph nodes. The overall sensitivity of TBNA for NSCLC is 78% and the specificity is 99%. The standard method of performing TBNA starts with a CT scan of the chest to guide needle aspirations toward the most involved group of lymph nodes. Lesions localized by CT can be accessed with bronchoscopy by measuring the number of CT slices above or below the carina (or other airway landmarks) and placing the needle the required distance above or below the landmark corresponding to that number of CT slices. When performing TBNA, the question invariably arises about the number of negative passes to perform before stopping. For various reasons, such as patient comfort and safety, need for sedation, and time spent by medical staff, it is imperative to manage the time for bronchoscopy. It has been shown that a plateau in yield for malignancy is achieved after seven passes with the needle through a lymph node. The importance of having a qualified and experienced cytopathologist on site cannot be overemphasized. Thorough interpretation by such individuals who are available for rapid on-site evaluation has been shown to enhance yield from TBNA. With the assistance of these individuals, the adequacy of sampling can be rigorously assessed. All samples should contain a preponderance of lymphocytes to define true nodal sampling. Specimens without lymphocytes should be deemed unsatisfactory, and the presence of respiratory epithelium should raise concerns about contamination.
TBNA allows for minimally invasive sampling of the mediastinum and hilar lymph nodes and potentially avoids more invasive procedures such as mediastinoscopy, mediastinotomy, and open thoracotomy. There is no doubt that the combined use of TBNA and CT can improve not only the diagnostic but also the staging evaluation of lung cancer. However, thus far, there has been wide variability in training and in usage of this helpful procedure. It is also operator-dependent, with certain techniques allowing for higher yields. See Chapter 22 for an in-depth discussion of bronchoscopy.
Endoscopic ultrasound (EUS) is another modality that has significantly impacted lung cancer staging, primarily due to its superior ability to sample the posterior mediastinum through the esophageal wall. Currently, EUS with fine-needle aspiration is performed using real-time ultrasound. In pooled analysis of 2433 patients with lung cancer and mediastinal adenopathy, EUS had a sensitivity and specificity of 89% and 100%, respectively. In patients with lung cancer who have no adenopathy seen on CT, EUS has been shown to sample nodes as small as 3 mm in diameter. This is useful given the high incidence of metastasis found in normal-sized lymph nodes in lung cancer. Based on surgical studies, it may be possible to predict the location of mediastinal lymph node metastases at certain levels based on the location of the tumor. This relationship may influence the use of EUS in certain patients without adenopathy on chest CT. Lymphatic pathways favor spread to aortopulmonary window nodes from left upper lobe tumors and to subcarinal nodes from left and right lower lobe lesions. EUS has been studied in patients with known lung cancer without enlarged mediastinal lymph nodes on CT, and it has detected mediastinal involvement (stage III or IV disease) in up to 42% of cases.
In addition, EUS has the advantage of being able to stage lung cancer from locations outside the mediastinum. The left lobe of the liver, a substantial part of the right lobe of the liver, and the left (but not the right) adrenal gland can be identified and sampled in 97% of patients. In addition, left pleural effusions can be visualized and sampled during an EUS procedure. EUS is increasingly being combined with EBUS for minimally invasive staging of lung cancer.
Perhaps the greatest addition in the armamentarium for staging lung cancer is endobronchial ultrasound with transbronchial needle aspiration (EBUS-TBNA). EBUS-TBNA is indicated for the assessment of mediastinal and hilar lymph nodes and diagnosis of lung and mediastinal tumors. It can be used to sample the highest mediastinal (station 1), the upper paratracheal (station 2R, 2L), the lower paratracheal (station 4R, 4L), the subcarinal (station 7), as well as the hilar (station 10) and the interlobar (station 11) lymph nodes ( Fig. 53-1A ). The para-aortic (station 6), aortopulmonary window or subaortic (station 5), paraesophageal (station 8), and pulmonary ligament (station 9) lymph node stations are usually not accessible by this technique (see Fig. 53-1B ). Pooled analysis of 2756 patients shows that EBUS-TBNA has a sensitivity and specificity of 89% and 100%, respectively. If a patient presents with a lung mass and mediastinal lymphadenopathy in accessible lymph node stations, EBUS-TBNA should be considered as the test of first choice because this modality can provide a diagnosis and stage simultaneously. In addition, the role of EBUS in lung cancer has expanded to include preoperative mediastinal staging and tissue acquisition for molecular analysis in addition to immunohistochemical staining.
The combination of EUS and EBUS has shown better yield than either technique alone. A pooled analysis from 7 studies and 811 patients showed a sensitivity and specificity of 91% and 100%, respectively. These complementary procedures provide near-complete access to the mediastinum for staging, even in the radiologically normal mediastinum. Among patients with (suspected) NSCLC, a staging strategy combining endosonography and mediastinoscopy compared with mediastinoscopy alone was shown to have greater sensitivity for mediastinal nodal metastases and led to fewer unnecessary thoracotomies.
For further discussion of EBUS, see Chapter 22 .
Mediastinoscopy is the historical gold standard for invasively staging the mediastinum in patients with known or suspected lung cancer; however, if local expertise is available, ultrasound-guided needle techniques (EBUS-TBNA, EUS, or their combination) are now recommended as the best first tests. If there is mediastinal lymph node enlargement regardless of FDG uptake on PET or if there is FDG uptake in a lymph mediastinal node regardless of its size, a surgical mediastinal procedure should be performed before thoracotomy despite a negative needle technique. Mediastinoscopy is most often used to sample nodes of the paratracheal (station 4), and anterior subcarinal (station 7) region (see Fig. 53-1A ). Because the subcarinal area is more difficult to sample, mediastinoscopy has a lower yield for lymph nodes in this area. An extended cervical mediastinoscopy can be carried out to reach aortopulmonary and para-aortic lymph nodes (stations 5 and 6) by using the same cervical incision as mediastinoscopy but dissecting into a different fascial plane. Alternatively, an anterior mediastinotomy (the so-called Chamberlain procedure) may be needed to sample lymph nodes in these aortopulmonary and para-aortic locations (stations 5 and 6) (see Fig. 53-1B ). Overall, mediastinoscopy has a reported sensitivity of 78%, with a specificity of 100%. Mediastinoscopy may also differentiate between stage IIIA and IIIB mediastinal involvement, which may be important for prognosis and potential therapy. As with any surgical procedure, mediastinoscopy has risks and limitations. It requires general anesthesia, with a morbidity of 2% and a mortality of 0.08%.
Treatment of Lung Cancer
The overall 5-year survival for patients diagnosed with lung cancer is a dismal 14%. This figure has not changed substantially since the 1980s. The survival curves vary by stage, with earlier stage lung cancer patients enjoying a much better survival than patients with later stage disease. Treatment is based on the stage of the disease and the patients’ performance status at the time therapy is initiated. In general, early stage disease is surgically managed, locally advanced disease is managed with chemotherapy and radiotherapy, and advanced disease is managed with chemotherapy with supportive care or supportive care alone. This paradigm has shifted toward more multimodality therapy (surgery, chemotherapy, and radiotherapy). This raises the issue of how best to manage patients with newly diagnosed lung cancer through their diagnosis, staging, and therapy. The ACCP guidelines on lung cancer recommend the use of a multidisciplinary lung cancer setting wherein patients can be evaluated by the major disciplines involved in the care of these patients, namely the pulmonologist, the thoracic surgeon, and the medical and radiation oncologists. A “tumor board” that includes the aforementioned specialties, with the addition of chest radiology, pathology, nursing, and social work, should review all new cases to ensure that patients receive optimal treatment and are considered for enrollment in clinical trials.
In addition to surgery, treatment such as chemotherapy and or radiotherapy can be applied in a neoadjuvant or adjuvant fashion. Neoadjuvant therapy indicates therapy given before the main treatment; it has the potential to reduce tumor volume, treat micrometastases, and improve outcomes. Adjuvant therapy is therapy given after the main treatment; it is aimed at treating any residual tumor or micrometastases with the aim of preventing tumor recurrence. The efficacy of treatments can be assessed by median survival time (MST) or by progression-free survival (PFS), the length of time a patient lives with lung cancer before it progresses.
Prognostic Factors for Lung Cancer
Based on an analysis of large databases of inoperable lung cancer cases, the strongest predictors of survival are good performance score (Karnofsky scale), lower extent of disease (stage), age, and absence of weight loss. Some reports have shown female gender to be a predictor of better survival, but this varies between studies. Performance score and the presence or absence of symptoms are predictors of outcome even with resectable early-stage disease. For example, in stage I NSCLC patients undergoing curative resection, those who were symptomatic at presentation had worse survival than those who were asymptomatic. Although individual reports have noted superior survival for patients with one cell type of NSCLC versus another, the general consensus in the literature is that histologic subtypes of NSCLC are not a major predictor of survival. Absence of smoking or smoking cessation has been associated with improved survival. The maximal standard uptake value of the primary tumor on PET has been inversely correlated with survival.
In more recent years, there have been numerous reports that various molecular markers are associated with outcome. Some of the best known markers include KRAS, epithelial growth factor receptor (EGFR), EML4-ALK translocation, p53, p16, and BCL2. However, in many instances, the results are conflicting about the prognostic significance of these individual molecular markers, perhaps due to the types of cases under review and to individual laboratory variations in techniques for measuring these molecular markers. In a meta-analysis, KRAS mutations were associated with poorer survival, especially in adenocarcinoma in which the hazard ratio was 1.6 (95% CI, 1.3 to 2). Testing the genetic profile of samples obtained from metastatic lymph nodes or malignant pleural effusions in those patients with stage III and IV disease has become standard practice, because this allows for the selection of targeted drug therapies specific for the mutation as first-line chemotherapy.
Non–Small Cell Lung Cancer Treatment by Stage
This section presents a discussion of the treatment of NSCLC by stage and cell type, followed by a discussion of the treatment of SCLC. Table 53-5 presents an overview of treatment strategies for NSCLC based on stage.
|I and II||1st line||Adjuvant—stage IIA, IIB||2nd line||No||Survival improvement with adjuvant therapy (= 5%) |
Radiotherapy for inoperable patients
|IIB (T3N0M0) Pancoast||1st line||No||No||1st line—neoadjuvant||Neoadjuvant chemoradiotherapy improves survival in this subset of stage IIB|
|IIIA||1st line||Adjuvant treatment—in totally resected IIIA||Controversial||1st line||Combined chemoradiotherapy followed by surgery is feasible, but more data are needed to recommend routinely|
|IIIB Unresectable||No||No||No||1st line||Treatment similar to unresectable stage IIIA|
|IV||No||1st line *||No||No||Radiotherapy is used for palliation only |
All stage IV should have mutational analysis, including EGFR, EML4-ALK , and KRAS .
* The targeted therapies are indicated as first-line treatment for those with advanced NSCLC and documented EGFR mutations or EML4-ALK fusion. Erlotinib is FDA-approved as first-line treatment of metastatic NSCLC for patients whose tumors have EGFR exon 19 deletions or exon 21 (L858R) substitution mutations.
In the most recent staging system for NSCLC, stage I NSCLC is broken down into stage IA (tumors ≤ 2 cm [T1a] ( eFig. 53-14A ) and tumors between 2 and 3 cm [T1b]) (see eFig. 53-14B ) and stage IB (tumors between 3 and 5 cm [T2a]) (see eFig. 53-15 ). All stage I tumors are completely surrounded by lung parenchyma greater than 2 cm away from the carina and do not invade the chest wall or parietal pleura ( Fig. 53-2 ). Stage I lung cancer does not include patients who have malignant lymph node disease or patients with metastatic disease. Thus the TNM classification is either T1aN0M0, T1bN0M0 (stage IA), or T2aN0M0 (stage IB). The differences between the two are the size of the primary tumor and the survival after surgical resection. Although stage I disease offers the best chance for long-term survival, the sad fact is that only 15% of all lung cancers present with stage I disease.
The current treatment for stage I lung cancer is surgery alone. The surgical procedure of choice is a lobectomy or pneumonectomy with mediastinal lymph node sampling. It should be recognized that the patient must be a reasonable surgical candidate. The 5-year survival for surgically resected stage IA lung cancer is 73%, whereas the survival for stage IB lung cancer is 58%. Local postoperative radiation for stage I and II lung cancer, after either complete or incomplete resection of the tumor, has not been found to be of any benefit. Postoperative adjuvant chemotherapy has not been shown to improve survival for those with resected stage I disease. A further discussion of postoperative adjuvant therapy is presented later (see “Stage IIIA”).
Some patients are surgically resectable but medically inoperable, usually because the patient does not have the pulmonary reserve to tolerate a lobectomy. These patients, particularly those with T1 tumors, may be able to tolerate a wedge resection or segmentectomy of their tumor as opposed to a lobectomy or pneumonectomy. In such cases, the local recurrence rate is higher than that of a complete resection, but the overall 5-year mortality is no different. However, for patients with smaller peripheral stage I lung cancers, anatomic segmentectomy may offer local control, the opportunity for prolonged disease free survival, and overall survival that is comparable to lobectomy. Still, whenever possible, a complete anatomic resection is preferred over a minimal resection. For discussion of radiofrequency ablation and other nonsurgical treatments, see Chapter 19 .
Patients who are “close calls” for surgery should be thoroughly evaluated by both a pulmonologist and a thoracic surgeon before making a decision on operability. As previously stated, there is a difference between a patient who is resectable and a patient who is operable. Chapter 27 is devoted to the preoperative evaluation of patients with lung disease, but a brief review is warranted here. Patients with a postoperative percent predicted forced expiratory volume in 1 second or diffusion capacity less than 40% will have a higher morbidity and mortality following lung cancer surgery. Patients with borderline values preoperatively should be referred for a differential ventilation-perfusion scan for a better prediction of postoperative function. When there is still a question, a cardiopulmonary exercise test can be obtained. Compared with those with an oxygen consumption greater than 20 mL/kg/min, patients with an oxygen consumption between 11 and 19 mL/kg/min have a higher morbidity but no greater mortality and those with an oxygen consumption less than 10 mL/kg/min have both a higher predicted morbidity and mortality.
For patients who either refuse surgery or are deemed medically unfit for surgery, primary radiotherapy for cure can be considered. This approach was evaluated in a meta-analysis of 1 randomized and 35 nonrandomized trials. The studies were heterogeneous, and the 5-year cancer-specific survival ranged between 13% and 39%. The authors concluded that, even in patients with severe emphysema, radiation therapy can be tolerated if careful planning with three-dimensional conformal techniques is undertaken. Conformal indicates that the radiation is designed in 3 dimensions to match the shape of the tumor allowing maximal targeting to the tumor with minimal injury to adjacent normal tissue.
More recently, stereotactic body radiation therapy has been introduced as a highly precise and accurate delivery method of highly conformal and dose-intensive radiation to small-volume targets. This method is also referred to as stereotactic ablative body radiotherapy and stereotactic radiosurgery. It is a more aggressive dose intensification than could previously be achieved using conventional radiotherapy methods.
In the Radiation Therapy Oncology Group trial (ROTG 0236), patients with biopsy proven stage I NSCLC deemed medically inoperable received stereotactic body radiation therapy with 60 Gy in three fractions. Primary tumor control of 98%, regional control of 87%, and overall survival of 56% at 3 years were achieved. This study and others (ROTG 0618) have demonstrated a dose-response relationship that favors more intensive regimens with biologically equivalent doses of more than 100 Gy consistently resulting in more than 90% primary tumor control for T1 tumors and overall survival greater than 50%. A meta-analysis demonstrated an overall survival improvement with stereotactic body radiation therapy compared with conventionally fractionated radiation therapy.
Stage II NSCLC lung cancer is divided into stage IIA and stage IIB. Stage IIA is defined as a T1a-T2aN1M0 and T2bN0M0 disease and stage IIB includes T2bN1M0 and T3N0M0. The 5-year survivals for stage IIA and IIB are 46% and 36%, respectively.
Stage IIA lung cancer is quite uncommon, representing between 1% and 5% of patients treated in several surgical series. Stage IIB cancer may represent up to 15% of surgically resected cases. For stage IIA and IIB cancer, surgical therapy is the treatment of choice. There is no benefit to postoperative radiotherapy. The value of adjuvant chemotherapy after surgery is discussed later (see “Stage IIIA”) but, in short, adjuvant chemotherapy is recommended for all patients with resected stage II disease. With chest wall invasion (T3N0M0) ( eFig. 53-16 ), an en bloc resection of the tumor and chest wall is the treatment of choice. A specific discussion of the evaluation and treatment of Pancoast tumor is discussed later in the chapter.
The outcome of lung cancer surgery is improved when the surgery is performed at hospitals with a higher volume of procedures. It is also important that the surgery be performed by a thoracic surgeon; when lobectomy is performed by a thoracic surgeon compared to a general surgeon, mortality is nearly halved.
Stage IIIA NSCLC represents a heterogeneous group of patients with N2 disease ( Fig. 53-3 ) and includes T3N1 patients. In addition, within the new staging system, patients with T4N0-1 have been down-staged to IIIA from their previous classification.
For carefully selected T4N0-1M0 patients, surgery may be indicated with or without neoadjuvant chemotherapy or neoadjuvant chemoradiotherapy (superior sulcus tumors) ( eFigs. 53-17 and 53-18 ). Individuals with T4N0-1 disease due to main carinal involvement have been treated with carinal resection with or without pulmonary resection. Carinal resection carries an operative mortality of 10% to 15% and 5-year survival of approximately 20% in carefully selected series. Patients who are T4N0 solely due to tumor nodules within the ipsilateral nonprimary lobe ( eFig. 53-19 ) have a 5-year survival of around 20% with surgery alone.
There is substantial debate over what constitutes resectable IIIA (N2) disease. However, there is no debate that T3N1 disease is best treated with surgical resection. At surgery, when a complete resection of the lymph nodes and the primary tumor is possible, some patients are found to have occult N2 metastasis. If so, these patients are best served by a resection of all known disease and then by consideration for adjuvant chemotherapy.
It is less certain how best to treat patients when N2 (single or multiple station) metastases are documented before thoracotomy. The new ACCP guidelines used the groupings of infiltrative stage III (N2/N3) tumors and stage III with discrete N2 involvement. In infiltrative stage III with N2/N3 involvement, the mediastinal nodes can no longer be clearly distinguished and measured ( eFig. 53-20 ). These patients have extensive tumor infiltration in the mediastinum, which partially surrounds the major structures. In patients with infiltrative stage III (N2/N3), good performance score, and minimal weight loss (≤10%), the recommended treatment with curative intent is with concurrent chemoradiotherapy.
Two multicenter trials have concluded that concurrent chemoradiotherapy is superior to sequential therapy with chemotherapy followed by thoracic radiotherapy. Concurrent therapy, however, is associated with a higher rate of severe esophagitis than is sequential therapy. The recommended chemotherapy is a platinum doublet and the most common agents are etoposide and cisplatin. Three cooperative group trials of definitive chemoradiotherapy for unresectable stage III had an MST of 19 to 22 months with 2-year survivals of 40% to 45% and a 5-year survival of approximately 20%.
For patients with discrete N2 involvement, the ACCP guidelines recommend that the treatment plan be discussed by a multidisciplinary team. Either definitive chemoradiotherapy or induction therapy followed by surgery is recommended over either surgery or radiation alone. These recommendations were largely influenced by the two cooperative group trials that evaluated the role of surgery in patients with N2 disease. The European trial randomized patients with histologically proven N2 disease to radiotherapy or surgery after initial induction therapy with three cycles of a cisplatin doublet chemotherapy. The MST was 16 months in the surgery arm and 17 months in the radiotherapy arm. The 5-year survival in both arms was 16% and 14%, respectively, and was not significantly different. The North American trial also required pathologic proof of N2 disease and randomized patients to completion radiotherapy or surgery after induction of two cycles of etoposide and cisplatin with concurrent thoracic radiotherapy of 45 Gy in 25 fractions over 5 weeks. The MST was 24 and 22 months in the surgery and radiotherapy arms, respectively, with 5-year survivals of 27% and 20% (hazard ratio 0.87, P =0.24). These survival differences were not statistically significant. However, the PFS favored the surgery arm. A subset analysis in those having only a lobectomy after induction therapy did better than a matched group receiving only chemoradiotherapy. This was an unplanned subset analysis that weakens the strength of this analysis. Accordingly, the role of surgery for stage III with discrete N2 disease has not been definitively answered.
In patients with stage II or IIIA totally resected NSCLC, adjuvant chemotherapy with four cycles of a cisplatin-based doublet chemotherapy is recommended. The Lung Adjuvant Cisplatin Evaluation (LACE) meta-analysis evaluated all stages of totally resected disease and observed a 5.4% overall 5-year survival benefit with adjuvant chemotherapy. The benefit was the greatest for patients with stage II and III disease and those with better performance status. The role of postoperative radiation therapy for patients with totally resected stage III (N2) remains controversial. Local recurrence varies in reports from 20% to 60%. Some nonrandomized trials suggest possible benefit from postoperative radiation therapy. Accordingly, this option should be discussed with fit patients. A randomized phase III Adjuvant Radiotherapy (ART) trial is underway to evaluate postoperative radiation therapy in these patients.
Stage IIIB is also a heterogeneous group and includes T4N2M0 and any T N3M0 patients. There are no phase III randomized trials to date that demonstrate that neoadjuvant chemoradiotherapy followed by surgery for stage IIIB disease results in prolonged survival compared with chemoradiotherapy alone.
Patients with unresectable stage IIIB NSCLC are treated the same as those with unresectable IIIA disease. The MST is generally 19 to 22 months, with a 5-year survival of 10% to 20%. The randomized trials of chemoradiotherapy included both stage IIIA and IIIB disease participants so it is not possible to separate out the survival of stage IIIB patients specifically. Concurrent chemoradiotherapy is recommended for both unresectable stage IIIA and IIIB disease. Trials have evaluated multiple daily fractions of thoracic radiotherapy, but there are no convincing data that hyperfractionated thoracic radiotherapy (the same total dose of radiation therapy split into two treatments in the same day) is superior to standard once-daily treatment.
Stage IV NSCLC is generally considered to be incurable with 5-year survivals of 1% to 3%. The goal of therapy is to try to control the disease and palliate symptoms. Major response rates with current chemotherapy regimens are 10% to 30%. Patients who respond to chemotherapy may gain an additional 3 to 9 months of life on average but eventually relapse and die of their disease. In previous trials in the 1970s and 1980s, patients were randomized to best supportive care or systemic chemotherapy. A meta-analysis evaluated eight of these randomized trials, including more than 700 patients. Each of these trials used a cisplatin-based chemotherapy versus supportive care. With best supportive care, the MST was 4 months and the 1-year survival was 15%; with chemotherapy, there was an increase in the MST of 1.5 months and an increase in 1-year survival of 10%. In the 1990s, a number of new chemotherapy agents were introduced, including paclitaxel, docetaxel, irinotecan, vinorelbine, and gemcitabine. Phase III trials have incorporated these newer agents in combination with cisplatin or carboplatin.
Trials comparing single-agent chemotherapy to a chemotherapy doublet containing a platinum compound have shown that the chemotherapy doublet is superior. Treatment with three drugs has not been shown to be superior to that with two drugs. Large randomized trials have tried to identify the optimum chemotherapy doublet. The results were uniformly similar, with response rates of 20% to 30% and MST of 7 to 9 months. No one platinum doublet has been shown to be superior. The chemotherapy combinations did have different toxicity profiles.
Histology influences response to certain chemotherapeutic agents. A large phase III trial randomized patients to cisplatin and pemetrexed or gemcitabine and cisplatin. Whereas there was no difference in the overall survival for the 847 patients with adenocarcinoma, the survival was significantly better with the pemetrexed regimen (MST 12.6 months vs. 10.9 months). Conversely, squamous cell cancers had superior survival with the gemcitabine regimen (MST 10.8 months vs. 9.4 months). The ACCP guidelines recommend that chemotherapy for stage IV NSCLC should be guided by histology. Pemetrexed should be limited to patients with nonsquamous NSCLC. In patients with a good performance score, a platinum-based doublet chemotherapy regimen is recommended.
In the Eastern Oncology Group Trial (E4500), patients with nonsquamous histology were randomized to treatment with carboplatin and paclitaxel with or without bevacizumab, a monoclonal antibody that inhibits a vascular endothelial growth factor. Bevacizumab was continued as maintenance therapy after six cycles of therapy in those responding to treatment or with stable disease. Patients in the bevacizumab arm had a higher response rate (35% vs. 15%) and better survival (MST 12.3 months vs. 10.3 months; 2-year survival 44% vs. 15%). A meta-analysis of four randomized trials of NSCLC patients treated with bevacizumab demonstrated increased PFS and overall survival in patients treated with combination chemotherapy and bevacizumab than in patients treated with chemotherapy alone. In selected patients with nonsquamous histology and good performance scores, the addition of bevacizumab is recommended. Hemoptysis, uncontrolled brain metastasis, deep venous thrombosis, and anticoagulation treatment are contraindications to using bevacizumab.
In patients with stable or responding disease whose initial therapy included bevacizumab, maintenance treatment with bevacizumab until progression is generally recommended. For patients with an initial platinum-based doublet chemotherapy and responding or stable disease, maintenance treatment with pemetrexed has been shown to prolong survival. A randomized Eastern Oncology Group Trial is currently evaluating maintenance treatment with bevacizumab versus pemetrexed versus the two-drug combination. This is likely to be a definitive study on maintenance therapy.
In 2004, investigators identified activating mutations in the tyrosine kinase domain of EGFR that predicted response to novel tyrosine kinase inhibitors (TKIs) in selected patients with NSCLC (see Chapter 51 ). This activating mutation was found to be a “driver” mutation because it was causal in tumor development; the mutation activated EGFR signaling thereby driving and sustaining the tumor. Driver mutations, as opposed to the much more common passenger mutations, thereby render a tumor vulnerable to blockade of that single pathway.
Multiple studies have demonstrated that the phenotype most associated with response to the EGFR-TKIs, gefitinib and erlotinib, includes adenocarcinoma, never-smoker, female, and East Asian descent. Subsequent reports have shown that these are the groups most likely to harbor the activating EGFR mutations. Emerging data have demonstrated that KRAS and EGFR mutations are almost always mutually exclusive and that patients with KRAS mutations do not respond to EGFR-TKIs. Erlotinib has been shown in a phase III trial in previously treated patients with NSCLC to result in superior survival versus placebo (6.7 vs. 4.7 months). On this basis, erlotinib was approved by the U.S. Food and Drug Administration (FDA) for second-line treatment. A phase III trial of gefitinib in second-line therapy failed to show a significant improvement in survival and approval for second-line therapy was withdrawn by the FDA in North America. In a landmark phase III trial in East Asia, patients with untreated stage IV disease were randomized to gefitinib or platinum-based chemotherapy. Of the 437 patient samples tested, 60% were positive for an activating mutation of EGFR . This subgroup had a significantly better response and PFS after gefitinib than after chemotherapy. The subgroup negative for mutations responded better and had a better PFS after chemotherapy.
Since that key study of first-line treatment with an EGFR-TKI or with chemotherapy, there have been multiple additional trials in patients with EGFR activating mutations. A meta-analysis of the use of EGFR-TKI in a first-line (13 trials) or in a second-line setting (7 trials) observed a PFS advantage (HR 0.43) in those patients with an activating EGFR mutation treated with an EGFR-TKI in the first-line setting. Results were similar when the EGFR-TKI was used in the second-line setting in those with an activating EGFR mutation (HR 0.34 for PFS). For patients whose tumor contains an activating EGFR mutation, the response rate to an EGFR-TKI is 60% to 80% with a median PFS of 9 to 12 months and an overall survival of approximately 2 years. As a result of these and other studies, erlotinib is FDA-approved as first-line treatment of metastatic NSCLC for patients whose tumors have EGFR exon 19 deletions or exon 21 (L858R) substitution mutations.
The anaplastic lymphoma kinase ( ALK ) fusion gene is the second most common driver mutation for which there is an effective TKI. The efficacy of the first generation ALK-TKI, crizotinib, was demonstrated in a phase I study that was expanded and resulted in a rapid FDA approval of the drug. Of 143 patients with stage IV NSCLC with an ALK fusion, the objective response rate was 61% and a median PFS was 9.7 months. The estimated 12-month survival was 75% and the MST had not been reached at the time of publication.
In a randomized phase III trial, advanced stage NSCLC patients with an ALK fusion who failed one prior platinum-based treatment were treated with crizotinib (ALK-TKI) or single agent pemetrexed or docetaxel. The median PFS was 7.7 months in the crizotinib group compared with 3 months in the chemotherapy group (HR 0.49). The response rate favored crizotinib treatment (65% vs. 20%), even though there was no significant difference in survival between the two treatment arms. The FDA has approved ceritinib for the treatment of NSCLC patients with an ALK fusion who were previously treated with crizotinib, to which some cancers develop resistance.
The Lung Cancer Mutation Consortium consists of 14 cancer centers in the United States and is sponsored by the National Cancer Institute. These centers have tested more than 1000 lung adenocarcinomas to look for driver mutations in 10 genes with a goal of determining treatment based on the molecular subtypes. A driver mutation was detected in 63% of 733 fully genotyped cases. The driver mutations identified were KRAS (25%), activating EGFR (15%), ALK rearrangements (8%), BRAF (2%), HER2 (2%), PIK3CA (1%), and MET amplification (1%). Results were used to select targeted therapy or targeted therapy trials in 279 patients with a driver mutation (28% of 1007). This study demonstrated the potential for multiplex genomic testing in patients with advanced NSCLC and is certainly the direction for the future.
Recently, joint guidelines by the College of American Pathologists, International Association for the Study of Lung Cancer, and the Association of Molecular Pathology have recommended that all patients with advanced stage adenocarcinomas should be tested for EGFR and ALK mutations and patients should not be excluded from testing based on clinical characteristics. These results should be used to select patients for targeted therapy with EGFR-TKIs or ALK-TKIs. Ideally, this testing should be performed before initial treatment of advanced stage lung cancer. In addition to targeted therapy based on mutation analysis, immunotherapy is a promising future direction in therapy for NSCLC. There are several large clinical trials underway using vaccines and checkpoint inhibitors which may provide additional therapeutic options.
Small Cell Lung Cancer
SCLC accounts for approximately 15% of all lung cancers. This cell type has the strongest association with cigarette smoking and is rarely observed in a never-smoker. It is the cell type most commonly associated with paraneoplastic syndromes such as the syndrome of inappropriate (excessive) antidiuretic hormone secretion (SIADH), ectopic corticotropin secretion, Lambert-Eaton myasthenic syndrome (LEMS), and sensory neuropathy.
SCLC usually presents as a centrally located mass in the hilum on chest radiograph ( eFig. 53-21 ) and may be associated with obstructive pneumonia. In 5% or fewer cases, SCLC may present as a solitary pulmonary nodule/mass ( eFig. 53-22 ). SCLC is generally staged according to the Veterans Administration Staging System, and classified as limited disease (LD) or extensive disease (ED). LD is confined to one hemithorax, the mediastinum, and the ipsilateral supraclavicular lymph nodes, and the disease can be encompassed adequately in a safe radiation portal. ED is any disease spread beyond these limits. Malignant pleural effusion or disease extending to the contralateral supraclavicular or hilar lymph nodes is generally considered to be ED. More recently, it has been proposed that the new TNM staging system (7th edition) should also be used for small cell lung cancer. The clinical stage groupings of I to IV were predictive of overall survival and the findings were validated in a cohort from the Surveillance Epidemiology and End Results registry and the California Cancer Registry. The survival rate for patients with LD with pleural effusions was intermediate between those with ED and LD without pleural effusions. The TNM staging is most helpful in potentially resectable patients with T1-2NO disease. Accordingly, it would be advisable to use both the TNM for tumor registries and clinical trials to define patients with minimal disease.
After establishing the histologic diagnosis of SCLC, patients are usually staged with MRI of the brain, CT of the chest (through the adrenal glands), and bone scan or PET. In an evidence-based review of the literature, staging using PET was compared with conventional staging using non-PET imaging. Of 267 patients with LD by conventional imaging, 16% were upstaged by PET. Of 199 patients with ED, PET resulted in 11% being down-staged. In total, staging with PET improves the accuracy of initial staging and radiotherapy planning. If a PET scan is obtained, then a bone scan may be omitted. In the unusual case when SCLC presents as a peripheral nodule, the treatment of choice is surgical resection followed by adjuvant chemotherapy and possibly sequential thoracic radiotherapy. Careful preoperative staging should be performed in these individuals to rule out metastatic disease. Pre-resection mediastinoscopy should also be performed in all patients being considered for resection with curative intent. If there are mediastinal node metastases, then surgery should be abandoned, and the patient treated with concurrent chemoradiotherapy as outlined later. The 5-year survival for peripheral SCLC that is treated with surgery and adjuvant therapy is approximately 40% to 50%. Approximately one third of patients have LD at diagnosis. LD-SCLC has a response rate of 70% to 80% with standard chemotherapy and thoracic radiotherapy ( eFig. 53-23 ), and a complete clinical response of 50% to 60%. In a meta-analysis of trials with chemotherapy alone versus combined chemotherapy and thoracic radiotherapy, survival was significantly better with combined modality therapy. A meta-analysis evaluated the timing of thoracic radiotherapy. Early thoracic radiotherapy (<9 weeks from the start of chemotherapy) resulted in a 5.2% increase in 2-year survival. The interval from start of treatment to the end of radiotherapy was also identified as an important predictor of outcome. Accordingly, data suggest that thoracic radiotherapy should begin early and be completed quickly. Chemotherapy usually consists of a platinum-based regimen. The two most commonly used regimens are etoposide and cisplatin or etoposide and carboplatin. Chemotherapy beyond four to six cycles has not been shown to prolong survival.
The National Comprehensive Cancer Network and the ACCP guidelines recommend treatment with four to six cycles of a platinum-based chemotherapy with either cisplatin or carboplatin plus either etoposide or irinotecan. Multiple trials have evaluated a number of the novel molecularly targeted agents but to date none have shown improved outcomes when added to standard treatment of small cell lung cancer. A number of trials evaluating insulin-like growth factor receptor inhibitors, antiangiogenesis, and immunomodulatory drugs are under evaluation.
If a patient with SCLC achieves a complete remission, then there is a 50% chance of development of cranial metastasis within the next 2 years. A meta-analysis of seven randomized trials of prophylactic cranial irradiation (PCI) versus no PCI for patients in complete remission reported an observed beneficial effect after PCI, with a 5.4% increase in absolute survival (20.7% vs. 15.3%) at 3 years. The major questions raised by the meta-analysis concern the optimal dose of PCI and the neuropsychological sequelae. PCI is recommended in patients who achieve a complete remission with initial therapy. PCI has also been shown to result in a survival advantage in patients with ED-SCLC who achieve a complete or partial response to initial therapy. PCI at 25 Gy in 10 fractions is the standard dose fractionation used in SCLC.
When patients relapse after initial therapy, the median survival is 3 to 4 months. There are no cures with second-line therapy. If a patient has been off treatment for 6 months or longer, then it is reasonable to use the same agents that he or she received initially. If initial therapy did not include a platinum agent, then second-line therapy should be with a platinum-containing doublet. Currently, the only drug approved for second-line treatment of SCLC by the FDA is single-agent topotecan. Other single agents such as oral etoposide, paclitaxel, docetaxel, irinotecan, gemcitabine, and amrubicin are active but are not yet approved by the FDA for second-line treatment of SCLC. Amrubicin is commercially available and approved for second-line treatment in Japan but is not available in the United States. No molecularly targeted therapy has been approved for SCLC in either the first-line or second-line setting.