Lung cancer is the leading cause of cancer death in men and women
in both the United States and the world. More Americans die each
year of lung cancer than of colon, breast, and prostate cancer combined.
The overall 5-year survival of lung cancer is 15% or less
and has only shown minimal improvement over the past 30 years. The
vast majority (85–90%) of cases of lung cancer
are attributable to smoking, and intensive research efforts have
identified hundreds of carcinogens contained in both mainstream
smoke (smoke directly inhaled by the smoker) and sidestream smoke
(smoke released from burning tobacco between puffs plus smoke exhaled
by the smoker). Although risk for lung cancer decreases significantly
after smoking cessation, overall disease risk reduction takes years
and an individual’s risk never returns to that of a never
smoker (never smoker is defined as fewer than 100 cigarettes in
an individual’s lifetime). Historically, 20 pack-years
of tobacco exposure or more has been considered to contain the highest
risk populations. Due to the large number of former smokers, new
cases of lung cancer in the United States are diagnosed more commonly
in former smokers than current smokers.
Other environmental factors can contribute to the development
of lung cancer. Table 41–1 contains a list of
proven and suspected agents that have been implicated in the development
of lung cancer in current, former, and never smokers. Of particular
note, many of these exposures, when coupled with smoking, lead to
exponential increases in the risk of developing lung cancer. For example,
smokers with asbestos exposure have a 50–100 times increased
risk of developing lung cancer.
Environmental Risk Factors for Developing Lung Cancer. |Favorite Table|Download (.pdf)
Environmental Risk Factors for Developing Lung Cancer.
|Passive/environmental tobacco smoke|
|Metals (chromium, arsenic, iron oxide)|
|Industrial (bischloromethyl ether)|
|Polycyclic aromatic hydrocarbons|
|History of tuberculosis|
There are also disease states that are associated with an increased
risk of lung cancer. First, patients who develop COPD [defined
as chronic bronchitis or emphysema with pulmonary function testing
showing at least mild airflow limitation, forced expiratory volume
in is (FEV1) <70% predicted and/or
FEV1/forced vital capacity (FVC) <0.70] secondary
to smoking have been shown to have higher rates of lung cancer.
Ongoing large epidemiological studies have shown that hazard ratios
for the development of lung cancer are the highest in the lowest
quartile of percentage predicted FEV1. There is also a
genetic predisposition to lung cancer as evidenced by increased
lung cancer mortality rates in nonsmoking relatives of lung cancer
cases when compared with nonsmoking relatives of matched controls.
In addition, females have a higher rate of developing lung cancer
when compared to males and controlled for amount of tobacco exposure.
The genetic contribution, along with common genetic susceptibilities
to developing COPD and lung cancer, is currently an area of intensive
research and may in the future allow better identification of “genetically” high-risk
groups. In addition to COPD, sarcoidosis and pulmonary fibrosis/interstitial
lung disease (ILD) are also associated with increased risk. Lastly,
the highest risk groups are those with a history of a previous lung
cancer or a head and neck cancer.
Cancer develops in a multistep fashion in which cells become
malignant by multiple genetic alterations affecting cellular growth,
differentiation, and survival. This can include the mutation of tumor
suppressor genes (for example p53), the activation of oncogenes
(for example, myc, jun, and fos), and the transformation of apoptotic
genes. The overwhelming majority of cases of lung cancer are due
to cigarette smoking. It is estimated that of cases of lung cancer,
90% in men and 80% in women are smoking related.
Smoke contains hundreds of known carcinogens, including free radical
oxidants and nonradical oxidants, which can damage DNA, proteins,
and lipids. The chronic inflammation accompanying repeated smoke
exposure also leads to genetic alterations in bronchial cells and
contributes to development of lung cancer.
Lung cancer is classified into two main categories, non-small
cell (NSCLC) and small cell (SCLC). Within these two major categories
are four basic histological types that account for over 90% of
the cases. NSCLC has three main types: squamous cell carcinoma (25–35% of
cases) arising from the bronchial epithelium and typically more
central in location; adenocarcinoma (25–35% of
cases) arising from mucous glands and typically more peripheral
in location; and large cell carcinoma (10% of cases), a
heterogeneous group of poorly differentiated tumors that does not
have features of adenocarcinoma, squamous cell, or SCLC. A distinct
subtype of adenocarcinoma is bronchoalveolar cell carcinoma (2% of
cases), which arises from distal airway epithelial cells and typically
presents as an unresolving infiltrate or as multiple nodules. Small
cell carcinoma (20–25% of cases) is of bronchial
origin and typically begins as a central lesion that can often narrow
or obstruct bronchi. Hilar and mediastinal adenopathy as well as
evidence of metastatic disease are often present on initial presentation.
For staging and treatment purposes, NSCLC and SCLC are viewed very
Growth factors and growth factor receptors are also involved
in the pathogenesis and progression of both SCLC and NSCLC. Many
of these factors or receptors are preferentially produced by the tumor
cells and they induce cell-specific growth. Classic SCLC has a neuroendocrine
origin accounting for many of the paraneoplastic syndromes, which
can be seen at presentation or during disease progression. Paraneoplastic
syndromes are also seen in patients with NSCLC due to accumulated
genetic alterations in tumor cells.
The majority of lung cancers are symptomatic at the time of diagnosis,
and clinical presentation is largely dependent on the tumor type,
tumor location, and tumor stage (local or distant spread), and whether
paraneoplastic syndromes are present. The clinical presentation
often contains clues as to tumor type (NSCLC or SCLC).
In its earliest stages, lung cancer is asymptomatic. Primary
lung cancers can reach a large size without causing any symptoms,
although careful history and physical examinations reveal that only
about 5% of lung cancer presentations are truly asymptomatic.
Many of these are solitary pulmonary nodules, a topic discussed
at length later in this chapter. Cough, anorexia, weakness, and
weight loss are the most common presenting symptoms in
patients with undiagnosed lung cancer. Other common presenting symptoms
include new cough or a change in a chronic cough (60+%),
hemoptysis (10–25%), and pain, either local at
a thoracic site or secondary to metastatic disease (25–35%).
Presentation also depends on tumor location, for example, endobronchial obstruction
can lead to postobstructive pneumonia, atelectasis, and pleural
effusions. Enlarging tumor size and/or lymph node involvement
can lead to hoarseness (secondary to recurrent laryngeal nerve injury),
superior vena cava syndrome (ie, supraclavicular venous engorgement,
much more common in SCLC), Horner’s syndrome (ptosis, anhidrosis,
and miosis from inferior cervical ganglion and sympathetic chain
involvement), and dysphagia (secondary to esophageal obstruction
from bulky mediastinal adenopathy). The Pancoast syndrome is shoulder
and upper chest wall pain caused by a tumor in the apex of the lung.
A tumor in this location can also be accompanied by Horner’s
syndrome, brachial plexopathy, and reflex sympathetic dystrophy. Bronchoalveolar
cell tumors may induce copious amounts of “salty” sputum
(a condition termed bronchorrhea).
Symptoms of metastatic disease are also relatively common presentations.
Lung cancer commonly spreads to the adrenal glands, liver, brain,
and bone. Central nervous system (CNS) spread may lead to headache,
nausea, altered mental status, and possibly seizures. SCLC typically
metastasizes at a much earlier time point than NSCLC.
Paraneoplastic syndromes are remote effects of the primary tumor
leading to organ dysfunction. Up to 20% of lung cancer
patients develop paraneoplastic syndromes, but these syndromes may not
necessarily indicate metastatic disease. The more common paraneoplastic
syndromes are listed in Table 41–2.
Lung Cancer Paraneoplastic Syndromes. |Favorite Table|Download (.pdf)
Lung Cancer Paraneoplastic Syndromes.
|Cachexia (anorexia, weight loss, weakness)|
|Lambert–Eaton myasthenic syndrome|
|Hypertrophic pulmonary osteoarthropathy|
|Autoimmune hemolytic anemia|
|Noninfectious thrombotic endocarditis|
|Idiopathic thrombocytopenic purpura|
The diagnosis of lung cancer is completely dependent on a tissue
sample containing malignant cells. There are a variety of ways to
obtain diagnostic tissue, including sputum cytology (best for central
airway lesions); bronchoscopy with endobronchial or transbronchial
biopsies; thoracentesis with cytological examination of the cellular
component; and fine needle aspiration of intrathoracic masses, lymph
nodes, or metastatic foci. The sensitivity of bronchoscopy in making
the diagnosis is variable and depends on the size and location of
the lesion. Transbronchial needle aspiration, video-assisted thoracoscopic
surgery (VATS), mediastinoscopy, and thoracotomy may also be needed
to adequately diagnose and stage patients. Other laboratory abnormalities
may be present due to the paraneoplastic syndromes (listed in Table
Virtually all patients with lung cancer have abnormal chest x-rays
or chest CT scans. Patients may present with a solitary pulmonary
nodule, and this particular clinical scenario is discussed at length
later in the chapter.
Correctly staging patients with lung cancer is crucial in determining
the proper therapeutic approach. One of the most important parts
of staging is a thorough history and physical examination. These
components directly determine blood work and further imaging. All
patients need electrolyte testing, liver function tests [including
alkaline phosphatase and lactate dehydrogenase (LDH)],
and a chest x-ray. Elevated alkaline phosphatase suggests bone metastases.
In the past, patients routinely had head CTs and radionuclide bone
scans as part of the diagnostic workup, but large studies have shown
that these tests should be ordered only if the patient’s
signs and symptoms indicate they are necessary. For example, CNS
symptoms or an abnormal neurological examination necessitate a brain
CT with contrast.
NSCLC and SCLC are staged differently. Due to the high incidence
of micrometastases early in the disease state, SCLC is divided into
two stages: limited disease (25–30%), in which
the tumor is limited to ipsilateral hemithorax (including contralateral
mediastinal nodes), and extensive disease (70–75%),
in which the tumor extends beyond the hemithorax (including pleural
effusions). SCLC is typically treated with chemotherapy and radiation
therapy. NSCLC is staged using the TNM staging system (T is tumor
size, N is nodal involvement, and M is presence or absence of metastases).
Table 41–3 contains the TNM descriptors and staging
for lung cancer.
TNM Descriptors and Staging for Lung Cancer. |Favorite Table|Download (.pdf)
TNM Descriptors and Staging for Lung Cancer.
|T (primary tumor)|
|Tis—carcinoma in situ|
<3 cm (not in mainstem bronchus)|
>3 cm or present in mainstem bronchus but not within 2 cm of carina,
invasion of visceral pleura, associated atelectasis or pneumonitis
extending to hilar region|
of any size that invades the chest wall, diaphragm, mediastinal
pleura, parietal pericardium; tumor <2 cm from carina, associated
atelectasis or pneumonitis of entire lung|
of any size with invasion of mediastinum, heart, great vessels,
trachea, esophagus, vertebral body, or carina; malignant pleural
or pericardial effusion; satellite tumor nodules in ipsilateral
|N (nodal status)|
to ipsilateral peribronchial or ipsilateral hilar region (including
to ipsilateral mediastinal and/or subcarinal lymph nodes|
to supraclavicular or contralateral mediastinal, hilar, or scalene
|M (distant metastases)|
|M0—no distant metastasis|
|0||Tis (carcinoma in situ)|
|IIB||T2N1M0 or T3N0M0|
|IIIA||T3N1M0 or T1–3N2M0 (N2 disease)|
|IIIB||Any TN3M0 or T4, any NM0|
|IV||Any T, any N, M1|
In general, patients with NSCLC at an earlier stage with disease
amenable to surgery have the best chances to be cured. Patients
being considered for surgery must be thoroughly evaluated to determine
if they have resectable disease. The decision for surgical resection
is largely based on tumor invasion and lymph node status, along
with underlying medical comorbidities. CT imaging and, to a larger
extent, positron emission tomography (PET) scanning are important
staging modalities. Chest CT scanning is performed with contiguous
sections through the liver and adrenal glands to aid in staging.
Lymph nodes larger than 1 cm in diameter suggest tumor involvement
and should be surgically sampled at the time of resection, or in
procedures to help determine the stage (ie, mediastinoscopy, transbronchial
needle aspiration, or VATS). CT scanning does have limitations, for
instance, determination of chest wall invasion has a sensitivity
of 38–87% and a specificity of 40–90% and
determination of mediastinal invasion is difficult. Patients should
not be denied surgery based on unproven CT findings. In fact, many
patients have their stage altered based on surgical pathology. A
comprehensive discussion of staging can be found in Mountain (1997).
One new addition to many staging algorithms is 2-[18F]fluoro-2-deoxyglucose
(FDG)-PET scanning. PET scans exploit differences between normal and
neoplastic tissue. Transformed cells exhibit increased glucose metabolism
resulting in increased accumulation of FDG. Sensitivity and specificity
of PET for detecting mediastinal metastases is superior to CT scans.
PET is also used to evaluate patients with multiple nodules, although
it does have limitations in the resolution of nodules <1 cm.
As clinical experience with PET scanning continues, it likely will
attain a more prominent role in staging algorithms.
This is thoroughly discussed in Chapter 4: Laboratory Evaluation. Many patients with
NSCLC have concomitant chronic lung disease that increases the risk
of thoracic surgery. All patients considered for surgery need complete
pulmonary function testing. A predicted postresection FEV1 >800
mL (or >40% predicted FEV1) is indicative of decreased
postoperative complications. A quantitative lung perfusion scan
(Q scan) can be used to improve the estimate of the patient’s
postoperative FEV1. Patients with an estimated postoperative
FEV1 <800 mL, but who otherwise have a favorable performance
status, may be evaluated with cardiopulmonary testing (discussed
in Chapter 4: Laboratory Evaluation).
The remarkable improvements in survival achieved in other common
cancers such as breast and colon cancer have not been realized in
lung cancer. This is due to many factors, most importantly the inability
to create reliable screening methods. Smoking cessation is crucial
to decreasing rates of lung cancer, but even with this intervention
there is still a large at-risk population. Because of the heavy
burden of disease and the large number of current and former smokers,
considerable research has focused on the best way to screen asymptomatic
persons at highest risk. Early detection or screening for lung cancer
has been a highly controversial topic for many years. In the mid-1970s,
several large trials were organized to determine the efficacy of
sputum cytology and chest radiography either in combination or separately
for early detection of lung cancer. The design and results of these
trials have been extensively reviewed, criticized, and debated.
In all of the trials, screening resulted in detection of earlier
stage disease with better lung cancer-specific survival, but no
reduction in lung cancer-specific mortality. Therefore, at present,
screening for lung cancer is not recommended by any of the major
advisory groups. These recommendations may change as knowledge regarding
the epidemiology and biology of premalignant airway lesions advances.
For instance, studies have shown that those with 30 pack-years of
tobacco exposure or greater who develop at least mild airflow limitation
and have cellular atypia on sputum cytology represent the highest
risk cohort. Although many of the early screening studies did not
concentrate on these high-risk patients, studies currently in progress
are doing so.
Advances in imaging technology are also being applied to lung
cancer screening. Recently, there has been considerable interest
in using low-dose, helical CT (spiral CT) as a lung cancer screening
modality. CT is a more sensitive test than chest x-ray and can be
used to accurately identify pulmonary nodules. Large-scale trials
funded by the National Cancer Institute and the American College
of Radiology are underway, and these trials will attempt to address
directly many of the issues raised by earlier studies, including
the question of whether screening will result in a decreased mortality
rate. Improved understanding of the distinct morphological and genetic changes
that occur in the airways will allow the highest risk patients and
those who may benefit from screening and chemoprevention to be targeted.
The indentification of this population may ultimately be based on
a variety of factors (tobacco exposure, pulmonary function testing,
and sputum cytology).
The 5-year survival for lung cancer has shown relatively little
improvement in the past 50 years, predominantly due to the late
stage presentation in the majority of patients. The treatment regimens
discussed below are those currently in use, but they are subject
to change based on large, multicenter trials that are currently
in progress. Predictors of survival are tumor type, stage, and patient’s
SCLC is classically treated with cisplatin and etoposide, with
responsive rates directly related to disease stage. Two-year survival
is around 20% in limited disease and 5% in extensive
disease. Remissions tend to be relatively short, with a median duration
of 7–9 months. Once SCLC recurs, survival is 3–4
months. Radiation therapy, which is used to treat symptomatic metastases,
such as those in bone and the CNS, improves survival in patients
with limited stage disease. Patients with SCLC are not treated surgically,
although in rare instances patients who have solitary pulmonary nodules
resected turn out to have SCLC. These patients tend to have an improved
survival compared to patients with limited disease SCLC.
Cell Lung Cancer
Surgical resection provides the best opportunity for cure. Many
features will preclude resection, including extrathoracic metastases;
malignant pleural effusions; tumors involving the contralateral
mediastinal nodes; and tumors that invade the heart, great vessels,
pericardium, esophagus, or trachea, or are within 2 cm of the main
carina. For those proceeding to surgery, the type of surgery does
affect the outcome. For example, trials comparing lobectomy to “limited
resection” found a higher incidence of local recurrence
in the limited resection group, along with a trend toward decreased
mortality at 5 years in the lobectomy cohort (44% versus
27% survival, p= 0.09).
Neoadjuvant therapy, the administration of chemotherapy or radiation
prior to surgery, is gaining favor as an initial therapy for NSCLC.
Studies suggest there is an improved survival rate in patients with
earlier stage (I and II) disease who receive chemotherapy prior
to surgery. In general, nonadjuvant therapy is being studied in
earlier stage disease (in combination with surgery) and may be one
important avenue to improve the survival rates.
Adjuvant chemotherapy, the administration of chemotherapy after
radiation or surgery, is commonly used in treating NSCLC. In patients
with stage IIIA disease and node-positive stage II disease, adjuvant
chemotherapy probably improves survival, but mostly in those with
a good performance status. Patients with advanced stage disease
(IIIA and IIIB), which comprises the majority of patients diagnosed
with lung cancer, who are not surgical candidates do have an improved
survival when treated with chemotherapy and radiation. Patients
with advanced disease (stages IIIB and IV) have an increased survival
at 1 year with such treatments, provided they exhibit a good performance
status on presentation. Overall, multiple trials have shown that patients
with stage IIIB and stage IV disease who are treated with chemotherapy
and/or radiation have better symptom control and performance
compared to patients receiving only palliative care. Newer chemotherapeutic
agents are continually being evaluated, which may result in some improvements
in survival. It is imperative that patients placed on newer regimens
do so as part of an organized, multicenter protocol that acts to
best evaluate efficacy.
The cumulative 5-year survival rate for lung cancer is 14% based
largely in part on the large percentage of patients who present
with late stage (IIIA or greater) disease. Survival best correlates with
surgical–pathological stage of disease, and can vary from
74% 5-year survival for stage IA NSCLC to roughly 5% 5-year
survival for stage IIIB NSCLC. Most large studies have failed to find
a significant difference in prognosis for the variety of NSCLC types
when adjusted for stage and performance status.
Henschke CI et al: Early Lung Cancer Action Project:
overall design and findings from baseline screening. Lancet 1999;354:86.
(One of the first studies reporting the use
of spiral CT in early detection of lung cancer. Discusses the difficulties
in designing lung cancer screening trials and potential uses for
Jemal A et al: Cancer statistics, 2002. CA Cancer J Clin 2002;52;23.
(Annual publication reporting
the cancer frequency, incidence, mortality, and survival rates for
Mountain CF: Revisions in the International System for Staging
Lung Cancer. Chest 1997;111:1710.
(Original manuscript describing the current TNM system for staging
lung cancer. Database represents all clinical, surgical–pathological,
and follow-up information for 5319 patients treated for primary
Patz EF Jr., Goodman PC, Bepler G: Screening for lung cancer.
N Engl J Med 2000;343:1627.
(A comprehensive review of past screening trials and future directions
in early lung cancer detection.)
Pretreatment Evaluation of Non-Small Cell Lung Cancer. Official
Consensus Statement of the American Thoracic Society. Am J Crit
Care Med 1997;156:320.
review of the literature on screening, clinical evaluation, staging,
and management of lung cancer. Includes discussions of the role
of bronchoscopy, mediastinoscopy, and preoperative evaluation.)
Rom WN et al: Molecular and genetic aspects of lung cancer.
Am J Respir Crit Care Med 2000;161:1355.
(A thorough review of the molecular biology of lung cancer with
a concentration on more recent advances and future directions.)