Epidemiologic risk factors in LC

Smoking
Cigarette smoking
Cigarette smoke contains >80 carcinogens evalu-
ated by the International Agency for Research on
Lung Cancer, 3rd edition. Edited by Jack A. Roth, James D. Cox,
and Waun Ki Hong. c  2008 Blackwell Publishing,
ISBN: 978-1-4051-5112-2.
Cancer (IARC) [6] with “sufficient evidence for
carcinogenicity” in humans or lab animals. While
smokers are at higher risk of LC than those who
have never smoked, there is substantial variation
in LC risk among smokers [7,8]. Peto et al. [7]
related UK national trends in smoking, smoking
cessation, and LC to the contrasting results from
two case–control studies of smoking and LC in the
United Kingdom[9]. Results showed large increases
in cumulative LC risk among continuing smokers
in 1990 data, reflecting prolonged smoking expo-
sure. Among bothmen and women in 1990, former
smokers had lower LC rates than continuing smok-
ers, with the reduction increasing substantially for
increased time since quitting. This study stresses the
importance of quitting smoking at an earlier age
to avoid subsequent risk of LC. The lifetime risk
of developing LC remains high for former smok-
ers, no matter how long the period of abstinence;
however, longer duration of abstinence is associated
with greater reductions in risk [7]. The relationship
between ETS exposure and LC has been extensively
evaluated in many epidemiologic studies. Results
from several meta-analyses report a positive asso-
ciation [10,11].
Family history
There have been a number of published studies
showing familial aggregation of LCs in first-degree
relatives of probands with LC [12–18]. Schwartz
et al. [17] showed that the LC risk in a first-degree
relative was associated with a 7.2-fold (95%
confidence interval (CI), 1.3–39.7) increased risk
among nonsmokers with early age at onset (40–
59 year old group). The association between an in-
creased risk of LC among first-degree relatives has
been confirmed in other studies [19,20]. On the
other hand, Kreuzer et al. [21] reported no evi-
dence of familial risk. Familial aggregation only pro-
vides indirect evidence for the genetic influence,
and could be due to common genetic profiles among
the familymembers, shared smoking patterns, or by
a combination of both factors.
Prior respiratory diseases
LC risk may be modified by a prior history of res-
piratory diseases such as asthma, bronchitis, em-
physema, and hay fever. It has been reported
that there was a significant protective effect in
the association between hay fever and LC (odds
ratio (OR) = 0.58; 95% CI, 0.48–0.70), and a
significantly increased risk associated with prior
physician-diagnosed emphysema (OR = 2.87; 95%
CI, 2.20–3.76) [22]. A significantly lower frequency
of hay fever was observed among patients with ma-
lignancies of lung, colon, bladder, and prostate as
compared to controls [23]. It was suggested that
the protective effects were attributed to enhanced
immune surveillance resulting from better detec-
tion and destruction of malignant cells [16,23–27];
also possibly, anti-inflammatory agents used to treat
hay fever might contribute to this protection. In
contrast, Talbot-Smith et al. [24] and Osann [16]
found no association between hay fever and LC
risk.
In another large case–control study comprised of
2854 cases and 3116 controls from seven different
European countries, a history of eczema was in-
versely associated with LC risk with an OR of 0.61
(0.5–0.8) [28]. In a meta-analysis, asthma was a
significant risk factor for LC among never smok-
ers with a pooled risk ratio (RR) of 1.9 (1.4–2.5)
when adjusted for ETS exposure [29]. Currently,
there is no consensus on the role of prior respiratory
diseases, other than emphysema, in LC and the em-
pirical evidence, which is not entirely consistent,
has been largely derived from observational epi-
demiologic data.
Environmental and occupational
exposures
Asbestos, arsenic, bischloromethylether,chromium,
nickel, polycyclic aromatic compounds, radon, and
vinyl chloride have all been implicated in LC etiol-
ogy, and have been reviewed extensively before.
Nutrition and dietary patterns
Fruits and vegetables
Observational studies strongly suggest that in-
creased vegetable and fruit intake is associated
with reduced risk of LC [30–33]. In the European
Prospective Investigation into Cancer and Nutrition
(EPIC), with data collected from 478,021 subjects, a
significant inverse association between LC risk and
fruit consumption was observed after adjusting for
smoking and other confounders (RR = 0.60; 95%
CI, 0.46–0.98 for the highest quintile compared to
the lowest); however, there was no such associa-
tion with vegetable consumption [33]. In a large
prospective Danish cohort study comprising 54,158
participants, the incidence rate of LC was highest
in the lowest quartile of plant food intake (fruit,
vegetable, legumes, and potatoes) [34]. Neuhouser
et al. [35] in a pooled analysis of eight prospective
studies with a total of 3206 incident LC cases oc-
curring among 430,281 individuals followed for 6–
16 years reported that compared to the lowest quin-
tile of consumption, the RRs of the highest quintile
consumption for total fruits, total fruits and vegeta-
bles, and total vegetables were 0.77 (0.67–0.87; p <
0.001), 0.7 (0.69–0.90; p = 0.001), and 0.88 (0.78–
1.00; p = 0.12), respectively. They concluded that
elevated fruit and vegetable consumption, mostly
due to fruit intake, is associated with a modest re-
duction in LC risk [35]. Overall, the association
between high intake of fruit and vegetables and re-
duced LC risk appears conclusive but what subtypes
of fruits and vegetables and which micronutrients
contribute to this protection remain controversial.
Carotenoids
Carotenoids are red and yellow fat-soluble pig-
ments found in many fruits and vegetables.
Numerous studies have shown that a diet high
in total carotenoids is protectives, but results are
inconsistent in the association between individual
carotenoids and LC. In a pooled analysis of seven
cohort studies in North America and Europe based
on a follow-up of 7–16 years with 3155 incident LC
cases among 399,765 participants, Mannisto et al.
[36] reported that only β-cryptoxanthin intake
was inversely associated with LC risk with a RR of
0.76 (0.67–0.86) even after controlling for intake of
vitamin C, folate, other carotenoids, multivitamin
use, and smoking status. In the Alpha-Tocopherol,
β-carotene Cancer Prevention Study (ATBC) [37],
with 1644 incident LC cases, during 14 years
of follow-up, lower risks were observed for the
highest versus the lowest quintiles of lycopene
(28% reduction), lutein/zeaxanthin (17%), β-
cryptoxanthin (15%), total carotenoids (16%),
serumβ-carotene (19%), and serumretinol (27%),
while intakes of β-carotene, α-carotene, and retinol
were not associated with significant reduction. In
a pooled analysis of the Nurse’s Health Study and
the Health Professional Follow-up Study (HPFS),
Michaud et al. [38] reported that only α-carotene
and lycopene intakes were significantly associated
with lower risk of LC. In overall analyses of all
carotenoids combined, LC risk was significantly
lower in subjects with high total carotenoid intake
with RR of 0.68 (0.49–0.94). Inadequate adjust-
ment for confounding, especially smoking factors,
and the lack of consideration of multicollinearity
between individual carotenoids may be responsible
for inconsistent results across studies.
Data generated by the Beta-Carotene and Retinol
Efficacy Trial (CARET) and ATBC trials failed to con-
firm the protective role of β-carotene in smokers
[35,37]. Contrary to the expectation and observa-
tional epidemiologic evidence, supplementation of
β-carotene resulted in a surprisingly increased over-
all LC incidence and higher total mortality among
current smokers [39–41]. Debate has been focused
on dosage, duration of trials, and the difference
between dietary intake and supplement use [42].
Preclinical data provide biologic plausibility for this
adverse interaction between cigarette smoking and
β-carotene [43]. Recently, in a cohort of French
Women [44], high intake of β-carotene was signifi-
cantly protective against LC among never smokers,
but was associated with increased LC incidence in
ever smokers. Tests for interaction between smok-
ing and β-carotene intake were significant [44].