CHEMOPREVENTION

Lung cancer, like many other cancers, appears to be the final
consequence of dysregulation involving varied pathways.
However, this results in a common “phenotypic” outcome.
Population-based lung cancer screening has yet to evolve into
an established format. Looking forward, population-based
screening could evolve as a multifaceted public health strategy,
including successful youth-directed tobacco control, more effective
smoking cessation for existing smokers, and integrated
early detection, management, and chemoprevention for current
and former smoker.
The conceptual basis for chemoprevention arises from the
consideration of the long evolution of an epithelial cancer. This
opportunity is best demonstrated by considering the colon
cancer model. Multiple steps cumulatively define a continuum
at the cellular level. These steps then served as watershed
points at which more aggressive histopathologic progression
could be identified and thereby mark the point at which cancer
had evolved. In general, it is believed that lung cancer evolves
by two potential pathways. The first involves the proliferation
of aberrant cells that continue to devolve into carcinoma by
sequential progression from hyperplasia through metaplasia
and dysplasia. The second pathway involves “redifferentiation”
or the dedifferentiation of bronchial epithelial cells that eventually
redifferentiate into carcinoma. 54 It is conceivable that
combinations of both pathways at various points ultimately
results in the phenotypic presentation of lung cancer.
The term chemoprevention was introduced by Sporn
et al. 55 in 1976 and was used to describe interventions used
to slow or reverse progression of premalignant lesions to frank
cancer. Targeting the “at-risk” population is crucial to the success
of population-based chemoprevention. In lung cancer,
chemoprevention strategies are targeted at the following three
main groups:
1. Primary—chemoprevention among individuals at high risk
who are otherwise healthy
2. Secondary—chemoprevention among individuals at high
risk with premalignant conditions
3. Tertiary—chemoprevention among individuals with a
known history of malignancy aimed at preventing second
malignancies 56
Chemoprevention agents encompass a tremendous number
of potential natural and synthetic agents. These can be
broadly subdivided into pharmaceutical, nutritional, and molecular
targets. Each one of these broad categories can be f urther
subdivided.
Chemoprevention Strategies Experimental work in
rodents suggests that specific drugs can mitigate the progressive
development of lung neoplasias after the exposure to specific
tobacco-carcinogen exposures. 57,58 Additional impetus
arises from the clinical experience with antiestrogens in the
chemoprevention of breast cancer. 59,60 However, the clinical
development of these drugs is fraught with a number of challenges,
including the protracted time and profound monetary
resources required for such commercial development efforts.
Defining objective surrogates of long-term outcomes has also
been an unresolved challenge. 61 As previously mentioned, it is
largely held that a multistep carcinogenesis pathway exists from
the normal bronchial epithelium to the frankly malignant tissue
(Fig. 15.1). Along this pathway, changes from hyperplasia,
atypia, metaplasia, and dysplasia are believed to occur and these
changes are being utilized as surrogate end point biomarkers
rather than frank carcinoma. 62–64
Diet and Nutrition The literature is replete with articles
that espouse or refute the benefits of a specific type of diet
on the risk associated with lung cancer. There are no randomized
controlled trials evaluating the advantages or disadvantages
of the consumption of any particular food group and its
putative cause of lung carcinogenesis. The conclusion that the
consumption of meat and fish plays a role in the development
of lung cancer is controversial at best. Some studies have demonstrated
that consumption of meat defined as pork, chicken,
or red meat is associated with a protective effect 65,66 whereas
others have demonstrated the opposite effect. 67
Case-control studies have shown an inverse relationship
between the amount of fruits and vegetables consumed and the
risk of lung cancer among both smokers and nonsmokers. 68,69
Similar results have been observed in other studies, however,
with less profound findings. 70
Vitamin supplementation, on the other hand, has yet to
be associated with any significant chemopreventative impact.
Its use has not been shown to positively influence lung cancer
mortality. 71,72
Pharmaceutical Interventions
CYCLOOXYGENASE INHIBITORS. The inflammatory pathway
in which arachidonic acid is metabolized to prostaglandins,
prostacyclins, and leukotrienes by the cyclooxygenase
(COX) enzyme is one of the mechanisms by which lung cancer
is thought to develop. There are two isoforms of this COX
enzyme, COX-1 and COX-2, of which COX-2 is thought
to involved in carcinogenesis, primarily through a variety of
downstream effects. 91 Naturally, inhibition of either one or
both COX enzymes is perceived as a theoretical means by
which lung cancer may be prevented.
Although some case-control studies have suggested that
aspirin (ASA) use is associated with a decreased risk of lung
cancer, 73–76 others have either failed to demonstrate a protective
effect, demonstrated only a statistically marginal benefit, or
have revealed only a nonstatistically significant beneficial trend
of ASA use with lung cancer. 77–82 When examining the population
with a positive tobacco history, several investigators have
been able to show a decreased risk of lung cancer. 74,75,83 Harris
et al. 74 demonstrated a risk reduction among heavy smokers.
While Moysich et al. 75 did not identify a dose-d ependent relationship
with ASA in smokers, they found a risk reduction
with greater frequency of use.
Population-based cohort studies have demonstrated that
nonsteroidal anti-inflammatory drugs (NSAIDs) for greater
than 1 year in duration can be associated with a decreased
relative risk in the development of lung cancer. 84 Among eversmokers,
a protective effect of NSAID use has been demonstrated.
83 Among high-risk heavy smokers, case-control studies
have shown a decreased risk of lung cancer also. 74 Other studies
have demonstrated that there is no protective benefit to NSAID
use in the development of lung cancer, but these epidemiologic
studies have been limited by the fact that confounding information
is present precluding a definitive conclusion. 85,86
The potential role of inhibiting the COX-2 pathway has
been supported by successful results in the animal models that
blocked this enzyme specifically. 87,88 Manipulation of the
prostaglandins distal to the COX-2 pathway in other animal
models further demonstrated the antineoplastic potential of
manipulating this pathway. 89 Several investigators have extensively
reported on the use of specific cycloxygenase-2 inhibition
with celecoxib to modulate several surrogate end point
biomarkers in human bronchial epithelium. 90–92 Presently,
clinical trials studying the role COX-2 inhibition in lung cancer
chemoprevention are ongoing. 72
STATINS. Recent evidence has suggested that 3-hydroxy-
3-methylglutaryl coenzyme A reductase inhibitors, otherwise
known as statins , may be protective against the development
of lung cancer. 93 In their retrospective case-control study,
Khurana et al.93 reported that statin use of greater than
6 months was associated with a significant reduction in
lung c ancer by 55%. This decrease was noted to increase
to 77% among those using statins for greater than 4 years.
Unfortunately, the retrospective nature of this study did not
allow for the detection of any meaningful interaction between
statin use and other significant factors associated with lung
cancer, including smoking history.
CAROTENOIDS. However, particularly recently, a large
body of evidence based on randomized controlled trials has
established -carotene at best as having a conflicting role in
the prevention of lung cancer. In two large randomized control
trials, the -tocopherol -carotene (ATBC) and the -carotene
and -tocopherol (CARET) studies, there was no protective
benefit noted among the overall population in the development
of lung cancer. In fact, in both studies, -carotene intake
was found to be associated with a significantly increased
lung cancer incidence and mortality. 94,95 In the ATBC trial,
the participants were all male current smokers, 95 whereas the
participants in the CARET trial were male and female current
and former smokers. 94 In the Physician’s Health Study, randomization
of the participants into the -carotene versus placebo
groups was performed in an attempt to evaluate a positive
effect in reducing the incidence of malignancies or cardiovascular
outcomes. Of the participants, 50% were either current
or former smokers. Ultimately, there was no difference in the
incidence of lung cancer between the two groups. 96 Similar
findings were observed in the Women’s Health Study. In that
study, no differences in lung cancer rates were observed among
current, former, and never-smokers who took -carotene compared
to those receiving placebo, and thus no beneficial effect
of -carotene could be demonstrated. 97 This confusing situation
highlights the challenge in trying to find effective chemoprevention
agents.
RETINOIDS/VITAMIN A. There is a profound wealth of data
supporting the role of vitamin A in the prevention, attenuation,
or regression of carcinogenesis. 98–100 Since the aforementioned
CARET study intimately tied together the use of -carotene
and vitamin A, the conclusion that -carotene did not have any
chemopreventative role was only half of the story. The other
half reads similarly, in that the administration of vitamin A also
had no beneficial effect and, in fact, was associated with a significantly
increased relative risk of lung cancer. 94
Tertiary chemoprevention trials also have not demonstrated
any benefit of retinoids in the prevention of second
primary lung cancers. Individuals who have been successfully
treated for stage I NSCLC by operative resection, has not been
shown to realize benefits in terms of recurrence, second primary
tumors, or mortality rates when given oral vitamin A.
In fact, secondary analysis demonstrated an earlier time to recurrence
and mortality among current smokers. 101 In Europe,
the European Study on Chemoprevention with Vitamin A and
N-Acetylcysteine (EUROSCAN) showed similar results with
retinyl palmitate having no effect on the rate of survival or
second primary tumor among patients who had been treated
with curative intent prior to this chemoprevention trial. 102 It
was surmised that the reason for this result was that retinyl
palmitate was not efficiently converted into all -trans retinoic
acid (the active form of the drug).
The negative results of the aforementioned trials may have
been due to the establishment of inadequate vitamin A levels
within the lung tissue itself. In the trial by Lippman et al., 101
only a homeopathic dose of retinyl palmitate was employed and
this may have resulted in lower lung tissue concentrations. In
the original work by Hong et al.103 demonstrating a beneficial
effect of vitamin A in carcinogenesis, there was also an a ssociated
increase in significant side effects associated with the administration
of high doses of vitamin A. Not surprisingly, while other
trials since theirs have demonstrated less toxic side effects, there
has also been an absence of beneficial effects observed with lower
doses of vitamin A. The benefits of increased local concentrations
of vitamin A in lung tissue is further supported by experimental
work with animal models and vitamin A deficiency
demonstrating that only aerosolized vitamin A in animals shows
a consistent beneficial effect. 58 Clinical trials in humans have yet
to establish this effect with consistency, although an aerosolized
trial from Germany was able to show regression of bronchial
dysplasia. 104 It is believed that the aerosol form of vitamin A is
successful because it functions directly at the bronchial epithelium
where vitamin A exists and is depleted.
TOCOPHEROLS/VITAMIN E. In the double-blind, randomized,
placebo-controlled ATBC trial mentioned earlier, there was
a 2% reduction in lung cancer among the participants that were
given vitamin E. This reduction however, was not significant
during the 5- to 8-year follow-up period. Interestingly, during
this same follow-up period the overall mortality rate among recipients
receiving -tocopherol was slightly higher than among
those not receiving -tocopherol. In the randomized, placebocontrolled
Women’s Health Study, women receiving vitamin E
did not have a statistically significant difference in lung cancer
rates than those receiving placebo. 105 In Heart Outcomes
Prevention Evaluation (HOPE) trial and its continuation study,
the HOPE-The Ongoing Outcomes (HOPE-TOO) trial the
incidence of lung cancer was no different between the participants
given vitamin E versus the placebo as identified. The median
follow-up for the 7030 participants that went on to the
HOPE-TOO continuation was 7 years. 106
BUDESONIDE/FLUTICASONE. Lam et al.107 conducted a
phase II study to determine whether inhaled budesonide had
a beneficial effect in smokers with known bronchial dysplasia.
Ultimately, the authors showed that inhaled budesonide had no
effect in causing regression of the previously identified dysplasia
nor in the prevention of new lesions. In another trial, patients at
risk of developing lung cancer had a decrease in the number of
indeterminate pulmonary nodules on CT scan when given fluticasone.
108 Although encouraging, this was a secondary analysis
of patients given fluticasone as part of a chemoprevention trial
that ultimately demonstrated its administration was not associated
with an alteration of the natural course of premalignant lesions.
This study was not designed to look at cancer end points
and as such did not look at the actual cancer rates. 109 More recent
studies have demonstrated a dose-dependent decreased risk
of lung cancer associated with inhaled corticosteroids in patients
with COPD after adjusting for various confounding factors. 110
SELENIUM. In 1996, the results of the Nutritional Prevention
of Cancer (NPC) trial were published. The initial objective of this
study was to evaluate the role of selenium in decreasing the rate
of skin malignancies in 1312 high-risk participants. Although this
study demonstrated that there was no decrease in skin malignancies,
a secondary end point identified was a substantial, but not
a significant, 26% decrease in the incidence of lung cancer. 111
Presently, there are ongoing clinical trials evaluating selenium in
both secondary and tertiary chemoprevention protocols.
OTHER AGENTS. Currently, there are several other investigational
agents being evaluated in their role in chemoprevention
such as the organosulfurs Oltipraz and Anethole
Dithiolethione (ADT). 65 Other agents may be evaluated in
their role as chemopreventative agents as more information is
learned from their role as biomarkers in early detection (see succeeding
d iscussion). 56,65,112–115