![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
2
*
Department of Epidemiology,
Department of Nutrition, University of North Carolina at Chapel Hill, NC, USA. 27599
2 Correspondence to: Lenore Arab, Ph.D., University of North Carolina, Department of Epidemiology, CB# 7400, Chapel Hill, NC 27599. tel: 919-966-7450; fax: 919-966-2089; email: lenore{at}unc.edu.
ABSTRACT
Animal and in vitro studies provide evidence of an anticarcinogenic effect of active ingredients in garlic. This review of the epidemiologic literature on garlic consumption addresses cancers of the stomach, colon, head and neck, lung, breast and prostate. Nineteen studies reported relative risk estimates for garlic consumption and cancer incidence. Site-specific case-control studies of stomach and colorectal cancer, in which multiple reports were available, suggest a protective effect of high intake of raw and/or cooked garlic. Cohort studies confirm this inverse association for colorectal cancer. Few cohort and case-control studies for other sites of cancer exist. Garlic supplements, as analyzed in four cohort studies and one case-control report, from two distinct populations, do not appear to be related to risk. Low study power, lack of variability in garlic consumption categorization within studies and poor adjustment for potential cofounders may limit the reliability of any conclusions regarding garlic supplements. However, an indication of publication bias was also found by visual inspection of a funnel plot and in a log-rank test (P = 0.004). Evidence from available studies nevertheless suggests a preventive effect of garlic consumption in stomach and colorectal cancers. The study limitations indicate the need for more definitive research and improved nutritional epidemiologic analyses of dietary data.
KEY WORDS: • garlic • stomach cancer • colorectal cancer • Allium • review
Numerous scientific reports imply that vegetable intake may affect
cancer incidence. In reviews of epidemiologic studies, there is
convincing evidence that high consumption of certain vegetables and
vegetable groups reduces the risk of colorectal, stomach, lung and
esophageal cancers; in addition, there is probable evidence for cancers
of the breast and bladder (World Cancer Research Fund, American Institute for Cancer Research 1997
). The Allium
genus of vegetables includes garlic, onions, leeks, scallions, chives
and shallots. These vegetables are characterized by a composition that
is high in flavonols and organosulfur compounds. Furthermore, animal
and in vitro studies provide evidence of the anticarcinogenic potential
of several bioactive compounds in Allium vegetables
(Wargovich et al. 1996
).
The pharmacologic mechanism by which garlic may inhibit gastric cancer
continues to be researched actively. An antibacterial effect,
particularly against Helicobacter pylori, has been
demonstrated (Jonkers et al. 1999
). The effect, which is
attributed in part to garlic’s thiosulfinate concentration,
has recently been quantified. Sivam et al. (1997)
reported that an extract of garlic cloves, standardized for its
thiosulfinate concentration, inhibited H. pylori growth at a
minimum concentration of 40 µg/mL (Sivam et al. 1997
).
More general anticarcinogenic effects may derive from the same
organosulfur compounds in garlic that are responsible for its odor and
flavor (Hughes and Lawson 1991
). Garlic is rich in
flavanols, particularly kaempferol, which aid in the detoxification of
carcinogenic compounds (Bilyk and Sapers 1985
,
Hertog et al. 1992
). The anticarcinogenic effects of
sulfur-containing compounds, such as diallyl, have been
demonstrated in animals (Reddy et al. 1993
). A recent
experiment conducted by Schaffer et al. (1997)
demonstrated that garlic powder (allyl sulfur compounds) protected
against induced mammary epithelial cell DNA adduct formation. This
finding is consistent with an earlier animal experiment wherein
selenium-enriched garlic inhibited early-stage mammary
carcinogenesis in Sprague-Dawley rats (Ip et al. 1996
). Motivated by this foundation in animal experiments, this
paper reviews the published epidemiologic reports presenting results on
raw and/or cooked garlic and garlic supplements and human cancers.
MATERIALS AND METHODS
Epidemiologic studies included in this review were identified
through a literature search conducted in July 1998. The MEDLINE
database was searched for manuscripts published between January 1, 1966
and July 1, 1998. The reference terms, "garlic,"
"Allium vegetable," "vegetable," "diet,"
"nutrition," with "cancer" and "neoplasm," as well as
individual cancer subtypes, were used as both keyword and subject
terms. In MEDLINE, the search was limited to human studies published in
English. In addition, journal articles cited in the primary-search
manuscripts were collected and added to the review. Of the 
300
studies examining vegetables and cancer incidence, 19 analyzed and
reported a relative risk
(RR)3
estimate for garlic intake and cancer incidence. The RR estimate,
confidence interval (CI), and/or P-value, year of
publication, country, subjects, garlic categorization and covariates
were abstracted from these references. Details of the studies are
presented in Tables 1
2
3
4
by cancer site and study design.
|
|
|
|
). The relative log risk estimates are plotted
against the estimated inverse of the squared standard error of the
ratio of the highest level of garlic consumption to the reference group
(Greenland 1998
). Details of this procedure are
described elsewhere (Begg and Berlin 1989
, Begg and Mazumbar 1994
); in brief, the procedure assumes that larger
studies are more prone to yield results that are closer to the true
value than are smaller reports. Larger studies, with smaller standard
error and increased power to detect a true effect, will form the apex
of an inverted funnel. Smaller studies with increased variance will
tend to disperse, forming the base. Gaps that appear in this graph are
an indication of publication bias. This is particularly likely with
high standard error and in the implausible direction of the
hypothesized effect. In addition, P-values were obtained by
the log rank test of Begg and Mazumdar (1994)
. Data are
presented in Figure 3
for all cancers.
|
RESULTS
Garlic and stomach cancer
Three case-control studies and one cohort study have reported
estimates for garlic consumption and risk of stomach cancer. You et al. (1989)
, in a case-control study of 564 incident
gastric cancers in China, reported risk of gastric cancer in relation
to frequency of consumption of several Allium vegetables.
The study boasted high participation rates, with 82% cases and 99% of
the population controls enrolled and interviewed. From an eligible pool
of 685 incident cases reported in the region, 111 were either ill, had
died or refused to participate. Of the interviewed cases, 50% were
histologically confirmed; the remaining cases were diagnosed through
surgery or endoscopy (32%) or radiological or clinical grounds (17%).
The controls were selected through census rosters and frequency matched
for age and sex. The interview assessed vegetable intake several years
before the report (1980) and before the Cultural Revolution in China (1965). The garlic-consuming population was divided into thirds on
the basis of yearly frequency of consumption. Their intakes were
converted from frequency per week to kilograms garlic per year. After
adjusting for age, sex, family income and consumption of other
Allium vegetables, significant trends were shown for garlic,
other specific Allium vegetables (scallions and Chinese
chives) and total Allium vegetable consumption. The odds
ratio (OR) and 95% CI for the highest compared with the lowest garlic
consumption third was 0.7 (0.4–1.0, P = 0.03). The
authors found these effects of garlic to be independent of protective
associations found for consumption of other fresh vegetables, vitamin
C, calcium and carotene.
Buiatti et al. (1989)
reported on a case-control
study conducted in Italy in which 1016 histologically confirmed gastric
cancer cases and 1159 population controls were interviewed. The
participation rates were 83 and 81%, respectively. Controls were
sampled from a municipal computerized list of residents and the
National Health Service files believed to represent 100% of the
7-center region. They were frequency matched by gender and 5-y age
intervals. Subjects were asked about their habitual diet for the 12-mo
period beginning 2 y before the interview. Exposures were
categorized into tertiles (high, medium and low) on the basis of
distribution in controls. It is impossible to distinguish between the
effects of garlic or onions as condiments because these two exposures
are assessed jointly in the study. Information on cut-points,
presented in Table 1
, was obtained via personal correspondence with Dr.
Buaitti. Reduced risk of gastric cancer was observed for the highest
tertile of onion/garlic as condiments (OR = 0.8, trend
P = 0.04). Raw garlic consumption was too infrequent
for analysis. Cooked garlic intake information was added later to the
questionnaire in a subset (27%) of the participants and showed a
significant decreasing trend across tertiles (1.0, 0.6, 0.4,
P < 0.01).
Hansson et al. (1993)
in a population-based,
case-control study of diet and gastric cancer in Sweden, reported risk
estimates from 338 of the 456 eligible, histologically confirmed
gastric cancer cases and 679 population-based controls. Controls
were sampled from population registries and were frequency matched for
age and sex. Participation rates for cases and controls were 74 and
77%, respectively. Diet was assessed with a food-frequency
questionnaire (FFQ), which focused on two periods of life, adolescence
(when the interviewee was 15–18 y of age) and 20 y before the
interview. The authors found garlic to be consumed infrequently during
adolescence. Results were based on the time period 20 y before the
interview only. Garlic consumption was analyzed dichotomously
(never/ever). This crude measure limits the ability to detect a
relationship between higher levels of garlic intake and stomach cancer,
but was necessary due to the low numbers of garlic consumers in this
population. The adjusted risk estimate of 0.89 (0.64–1.24) supports a
slight protective effect. Its wide CI reflects the relatively small
number of subjects in the non-zero (ever) consumption group.
The only cohort study to examine garlic and stomach cancer focused on
intake of garlic supplements, rather than raw and/or cooked garlic
(Dorant et al. 1996b
). The study utilized a
case-cohort approach, recruiting a total of 152 subjects who
developed stomach cancers during the 3.3-y follow-up period. Garlic
supplement use in this group was compared with a random sample of 3340
subjects from the entire cohort who did not develop any type of cancer
other than skin cancer. Garlic supplement users were defined as having
reported daily intake for at least one full year during the 5 y
before baseline inclusion. Two analyses were reported. The first
compared exclusive garlic supplement consumers with nonsupplement
users. The second compared garlic supplement users who also consumed
other supplements to those subjects who used any type of supplement
other than garlic. The adjusted risk ratios indicated that garlic
supplements did not confer a protective effect; on the contrary, a
slight, nonsignificant increase in risk was reported. The estimates
should be interpreted with caution given that only 5.9% of the total
population (cases and controls) consumed garlic supplements exclusively
and 3.3% consumed garlic supplements in combination with any other
supplements. As a result of small sample sizes in the exposure groups,
the estimates are unstable and may not have afforded the opportunity to
detect an association.
Garlic and gastric cancer
Case-control studies (Buaitti et al. 1989
,
Hansson et al. 1993
, You et al. 1989
)
suggested a protective effect of raw and/or cooked garlic for gastric
cancer (Table 1
, Fig. 1
). The cohort study found essentially no association between garlic
supplements and stomach cancer (Dorant 1996b
).
Hansson et al. (1993)
used a crude categorization of
garlic consumption (2 categories) that reduced the chances of finding a
meaningful relationship. Similarly, the inability to distinguish the
effect of garlic in the onion/garlic analysis in the report of
Buiatti et al. (1989)
also prevented accurate
interpretation. The inclusion of nonhistologically confirmed cancer
cases by You et al. (1989)
may have increased the
potential for misclassification by case status, which resulted in
underestimation of effect. However, each of these case-control
studies did adjust for three possible risk factors for stomach cancer,
i.e., age, sex and income, as well as several other potential
cofounders. Bias due to residual confounding cannot be dismissed due to
inadequately measured covariates or unmeasured factors.
|
demonstrates a funnel graph designed to detect publication bias
(Begg and Berlin 1989
). Publication bias is evidenced by
visual inspection of the funnel plot and in the log rank test by
Begg and Mazumdar (1994)
(P = 0.004). On
the basis of the aforementioned case-control studies, no convincing
claims can be made with regard to the protective effect of garlic on
stomach cancer. There does appear to be a consistent decreased risk,
however, in the highest level of raw and/or cooked garlic consumption
across all three studies compared with the respective referent groups.
The lone study of garlic supplement use, by contrast, indicates no
association. Garlic supplements are a diverse mixture, many of which
are dried powders and extracts. Evidence suggesting differences between
garlic supplements and raw/cooked garlic is accumulating.
|
The association between garlic consumption and colorectal cancer,
which has garnered slightly more published results than stomach cancer,
was examined in three case-control studies and three cohort
studies. The relationship with colorectal polyps was examined in one
case-control study. Collectively, these studies (American Cancer Society 1998
, Giovannucci et al. 1994
,
Hu et al. 1991
, Iscovich et al. 1992
,
Le Marchand et al. 1997
, Steinmetz et al. 1994
, Witte et al. 1996
) are described below and
presented in Table 2
.
Iscovich et al. (1992)
reported on a case-control
study conducted in Argentina of 110 newly diagnosed colon cancer cases
and 220 age-, sex- and residence-matched, neighborhood controls.
The incident cancer cases recruited for this report constituted 75% of
the total incidence in the study area during the recruitment time
period; 92% of these cases participated. Of the eligible controls,
86% were interviewed. Frequency of consumption of raw/cooked garlic
multiplied by seasonality enabled translation into indices of annual
consumption. These were divided into tertiles on the basis of the
distribution in the controls. Garlic was combined with onions and
peppers in the analysis. A conditional univariate analysis yielded a
strong protective effect for both the second tertile, 0.42 (95% CI
= 0.19–0.91) and the highest level of garlic, onion and pepper
consumers 0.22 (95% CI = 0.095–0.51). Multivariate model
results, in which garlic, onion and peppers were incorporated into one
of seven food categories did not differ from the univariate analyses
for the individual items included in the vegetable group (pulses, low
fiber, green leafy, other and garlic). The independent effects of
garlic cannot be analyzed in the univariate and multivariate models.
Le Marchand et al. (1997)
recently reported on a
population-based, case-control study of 1192 ethnic cases and 1192
age-, sex- and ethnicity-matched controls in Hawaii. The
participation rate for cases was 66%, with nonparticipation
attributable primarily to death (15%) and refusal (11%). The
participation rate among controls was 71%. Subjects completed a FFQ
regarding the 3-y period before the onset of symptoms in the cases and
the 3-y period before the interview in the controls. A nonsignificant
dose-response trend was observed for both men and women. The
highest quartile, exceeding 1.8 g/d (cut-point supplied through
personal conversation with Dr. Le Marchand) of garlic intake, after
adjustment for numerous potential cofounders, yielded weak evidence of
an inverse association with colon cancer (OR = 0.8, 95% CI = 0.5–1.1) for men and no effect for women (OR = 0.9, 95% CI
= 0.6–1.4).
In 1991, Hu et al. (1991)
presented results from a
hospital-based, case-control study of 116 men and 109 women with
rectal cancers and an equal number of sex-, age- and
residence-matched controls in China. Results for garlic consumption
were presented only for women and classified as a dichotomous variable
(consumers vs. nonconsumers), precluding dose-response analysis or
effect threshold estimation. The authors did report P-values
for annual consumption of garlic and 111 colon cancers (P
> 0.05) and 116 rectal cancers in men (P > 0.1). A strong inverse relationship with rectal cancer in women was
observed for garlic consumers (OR = 0.21, 95% CI = 0.04–1.0). The association, however, is based on only nine cases and
two controls in the nonconsumer category. Furthermore, the authors did
not adjust for potential cofounders in this univariate analysis.
The only study to examine colorectal polyps was performed by
Witte et al. (1996)
who interviewed 488 pairs matched
for age, sex, date of sigmoidoscopy and center in southern California.
Members of an HMO who elected to undergo sigmoidoscopy in 1991–1993
were eligible for study inclusion. The participation rate was 77 and
70% for cases and controls, respectively. Subjects were administered a
FFQ concerning the year before the procedure. The highest level of raw
and/or cooked garlic consumption (
3 times/wk) was inversely
associated with risk, with an OR = 0.66 (95% CI = 0.43–1.01). Although the latency for polyps is shorter than colon
cancer incidence, it is unclear whether the year before screening is
adequate to assess temporality. For example, cases in which polyps were
identified by sigmoidoscopy may have altered their diet as a result of
their pre-existing condition. This study deserves attention because
90% of colon neoplasms arise from colorectal polyps (American Cancer Society 1998
).
The Iowa Women’s Health Study (Steinmetz et al. 1994
)
included an initial cohort of 41,837 women aged 55–69 y who completed
a 127-item FFQ. From this cohort, 212 (80%) eligible, incident colon
cancers were identified after 5 y of follow-up. Due to large
numbers of nonconsumers (60%), tertiles of garlic consumption were
created on the basis of the distribution in all respondents. Results
were adjusted for numerous potential cofounders, including body mass
index, total energy, and history of polyps or colitis. The
strongest association among all vegetables was for garlic. The highest
level of intake (1 serving/wk) compared with the lowest (never)
yielded an adjusted RR of 0.68 (95% CI = 0.46–1.02). The garlic
association, in a stratified analysis of the distal colon, was
associated with nearly a twofold reduction in risk for the uppermost
tertile, 0.52 (95% CI = 0.30–0.93). In this report, extreme
levels of garlic consumption could not be analyzed due to the skewed
distribution toward nonconsumers. However, a significant reduction of
risk was reported for the uppermost tertile of consumption.
In the same year, Giovannucci et al. (1994)
published
results from a cohort of 47,949 U.S. male health professionals, aged
40–75 y, who completed a semiquantitative FFQ. After 6 y of
follow-up, 205 incident colon cancer patients were recruited. The
authors compared garlic consumption of the uppermost level (2
servings/wk) to nonconsumers. Of all fruits and vegetables analyzed in
a multivariate age- and energy-adjusted model, garlic was the food
item most linked to a protective effect. The inverse association was
0.77 (95% CI = 0.51–1.16) and for analysis limited to the distal
colon (RR = 0.63, 95% CI = 0.38–1.65). In addition, a
significant trend was reported across categories.
From the same cohort study reporting on garlic supplements and stomach
cancer, Dorant et al. (1996a)
reported on the
relationship between garlic supplements and colon and rectal cancers in
the Netherlands Cohort Study of 120,852 men and women, aged 55–69 y.
Within 3.3 y of follow-up, 293 eligible, incident colon and
150 rectal cancers were recruited. Three garlic supplement exposure
groups were defined as follows: 1) any type of garlic
supplement; 2) exclusive garlic supplements; and
3) garlic supplements taken in combination with other
supplements. The first two exposure groups were compared with
nonconsumers, whereas the third group was compared with any other
supplement (excluding garlic) users. As with results for gastric
cancer, consumption of garlic supplements was not associated with colon
and rectal cancers. The risk ratio for garlic supplement use and colon
cancer was 1.26 (95% CI = 0.84–1.91) and for rectal cancer (RR
= 0.77, 95% CI = 0.41–1.46). However, this finding is
limited in power because only 16 and 9 colon and rectal cancer cases,
respectively, were defined as exclusive garlic supplement users. As
mentioned earlier, the limited length of follow-up time may
preclude detection of an effect.
All three case-control studies showed an inverse association for
the uppermost level of raw and/or cooked garlic consumption and colon
and/or rectal cancers (Fig. 3
). In addition, the relationship between incidence of colorectal
adenomatous polyps and garlic intake suggests a significant protective
effect. Similarly, two large U.S.-based cohort studies reported an
inverse association for the highest levels of garlic consumption and
colon cancer (Fig. 3)
. In contrast, the only study to examine garlic
supplements did not find an association with colon cancer incidence.
This scant information does suggest a protective effect of garlic
consumption and the incidence of colon and rectal cancers. The concern
about publication bias in reports examining stomach and colorectal
cancer reports stems from the numerous epidemiologic articles examining
individual dietary items and cancer incidence. Many of these studies
utilize a FFQ, which often includes garlic and other Allium
vegetables. Furthermore, no reports have published garlic results with
slight-to-moderate positive associations (OR > 1.0). Recognizing
the uncertain role of publication bias and the amount of information
available, there still appears to be a consistent inverse association
between high levels of garlic consumption and colon and rectal cancer
incidence.
Garlic and cancers of the head and neck
Only two case-control studies have reported the relationship
between cancers of the head and neck and garlic consumption. Two other
reports analyzed total Allium vegetable consumption. No
cohort studies have reported results on raw and/or cooked garlic
consumption or garlic supplements and esophageal, laryngeal or nasal
cancers. As a result of limited information and considerable
heterogeneity among disease types, an assessment of the epidemiologic
literature on these cancers is not warranted. However, the reports are
described below and presented in Table 3
.
Hu et al. (1994)
enrolled 196 histollogically confirmed
esophageal cases and 392 age-, sex- and residence-matched controls
into a hospital-based, case-control study in Northeast China. The
case participation rate was 97%; the authors did not report the rate
for the controls. Of the cases, 170 were men and only 26 were women.
Hospital-based controls with diagnoses of acute surgery, hernia,
peptic ulcer disease, urologic disease, heart conditions and trauma
were used. Dietary interviews followed a structured FFQ inquiring into
the consumption of food items in the early 1980s, before the onset of
symptoms in cases, and before the Cultural Revolution. Recall bias is a
concern in this study because the disease symptoms can affect dietary
selection; for example, esophageal symptoms may tend to cause changes
in eating patterns that could bias dietary reports. The adjusted risk
estimate of the uppermost quartile of garlic consumption (>2 kg/y)
compared with the referent group (0 kg/y) was 1.0 (95% CI = 0.6–1.8). The third quartile (0.6–2.0 kg/y) however, was inversely
associated with risk (OR = 0.6; 95% CI = 0.4–1.0).
A much larger population-based, case-control study by Gao et al. (1994)
conducted in Shanghai, China, recruited 902
esophageal cases with a participation rate of 89%; 1552 controls, who
were slightly younger than the cases, participated. Instead of garlic
consumption, a general category of Allium vegetables was
assessed. After adjustment for smoking, alcohol and other factors, it
was found that Allium vegetable consumption was not related
to risk of esophageal cancer in either men and women (Table 3)
.
Zheng et al. (1992a)
published two population-based,
case-controls studies from Shanghai, China. The first study
compared 201 incident laryngeal cases, 76% of the eligible pool, with
414 controls selected randomly from the Shanghai Resident Registry. The
dietary questions concerned the period 10 y before the interview,
ignoring any recent changes. The analysis was limited to men and showed
a protective effect for the highest tertile of garlic consumption (OR
= 0.5). The second study by Zheng et al. (1992b)
reports on 60 incident nasal cancers (95% participation) in Shanghai,
China, and the same 414 controls used in the aforementioned laryngeal
study. This report, similar to the study by Gao et al. (1994)
, examined Allium vegetable intake and found
an inverse, nonsignificant association with risk (OR = 0.6, 95%
CI = 0.3–1.2). The estimate was adjusted solely for age. The wide
CI reflects small case subgroups, in which a maximum of 42 cases
occupied the referent and highest tertile (39 and 30% of cases,
respectively).
In summary, four case-controls studies examined three different cancers, and two of these reports examined garlic consumption specifically. Three of the four Chinese reports were conducted in Shanghai, China, limiting generalizability to Western diets. As a result of such considerable heterogeneity among reports, no conclusions on the relationship between garlic consumption and head and neck cancers can be drawn.
Garlic and other cancers
Two case-control studies have examined the relationship
between garlic consumption and the risk of cancers of the prostate and
breast. In addition, two cohort studies have examined garlic supplement
intake and risk of lung and breast cancers (Table 4)
. Key et al. (1997)
published results from a case-control study of 328
prostate cancers and 328 age-matched population controls in the
U.K. Cases and controls included Caucasian men <75 y old who completed
a FFQ inquiring about consumption of food items during the last 5 y. Controls were selected from patient lists from the cases’ general
practitioner (World Cancer Research Fund, American Institute for Cancer Research 1997
). In an attempt to eliminate selection
bias, controls were excluded for history of prostate cancer or
prostatectomy. Garlic consumption was divided into quartiles on the
basis of distribution among controls. The estimate (adjusted for social
class) for the uppermost quartile (2 servings/wk) of garlic intake to
the lowest (never) was 0.64 (95% CI = 0.38–1.09). When garlic
supplements were factored into the model, the estimate did not change
(OR = 0.68; 95% CI = 0.41–1.1), implying that garlic
supplements were not related to risk.
One paper has published results for the association of garlic and
breast carcinoma. Levi et al. (1993)
reported on 107
histologically confirmed breast cancer cases and 318 hospital-based
controls in Vaud, Switzerland. Controls were ineligible if admitted for
acute, nonhormone-related, gynecological, metabolic or neoplastic
disorders. A self-reported subjective score was used to assess the
level of garlic intake, and tertiles were designed on the basis of case
and control distribution. A nonsignificant decreasing trend was
observed across tertiles (
2 trend = 1.7), in which
the highest tertile compared with the lowest demonstrated an inverse
association (OR = 0.6). The estimate was adjusted solely for age,
which may introduce a bias due to residual confounding for known
hormonal and nondietary risk factors for breast cancer.
The two cohort studies to examine garlic supplements and lung and
breast cancer are reports based on data from the Netherlands Cohort
(Dorant et al. 1994
and Dorant et al. 1995
). The methodology is identical to the previously
mentioned stomach and colorectal Netherlands cohort studies. In each of
these reports, garlic supplements were found not to be associated with
risk of either cancer. A positive association was shown between
exclusive garlic supplement users compared with nonsupplement users and
risk of lung carcinoma (RR = 1.78; 95% CI = 1.08–2.92. In
contrast, no association was found in the comparison of combined garlic
and any other supplement users and consumers of any supplements other
than garlic (RR = 0.93, 95% CI = 0.46–1.86). In the breast
cancer study, these equivalent exposure comparisons suggested a
nonsignificant inverse association for exclusive garlic supplement
users (RR = 0.75, 95% CI = 0.4–1.4) and no association for
combined garlic and any other supplements (RR = 1.12, 95% CI
= 0.6–2.0). Exclusive garlic exposure groups in each report
comprised only 5.3 and 5.1% of lung and breast cancer cases,
respectively. Similarly, garlic and other supplement users comprised
only 3.5 and 5.1% of the total lung and breast cancer case
populations, respectively.
The cohort study in the Netherlands suggests a lack of association
between garlic supplements (exclusive garlic supplements and garlic
supplements combined with any other supplements) and risk of cancer for
lung and breast sites. These findings are supported by the study of
Key et al. (1997)
in which the garlic consumption OR
changed little when garlic supplements were added to the model. On the
basis of five published reports from the U.K. (Key et al. 1997
) and the Netherlands (Dorant et al. 1994
,
Dorant et al. 1995
, Dorant et al. 1996a
, Dorant et al. 1996b
),
no convincing evidence supports the claim that garlic supplements have
an effect on human cancer.
DISCUSSION
Potential anticarcinogens in garlic have been identified, and
animal experiments suggest neoplastic inhibition from several
garlic-derived compounds (Bilyk and Sapers 1985
,
Dorant et al. 1993
, Hertog et al. 1992
,
Hughes and Lawson 1991
, Ip et al. 1996
,
Jonkers et al. 1999
, Reddy et al. 1993
,
Schaffer et al. 1997
, Sivam et al. 1997
,
Sundaram and Milner 1996
, Wargovich et al. 1996
). The published epidemiologic evidence suggests that
protection from stomach and colon cancers may be related to consumption
of raw/cooked garlic. Because these are observational studies, many of
which have not controlled for other dietary differences, intervention
studies are required to confirm an effect, especially because garlic
supplements show no similar association. Because there was great
heterogeneity of frequency categories among reports, it is also unclear
what the minimum dose of garlic necessary to elicit a protective effect
might be. It can also be argued that these are biased and suffer from a
lack of variance in exposure levels. In consideration of this, the
strongest claim for a protective effect of raw/cooked garlic
consumption can be made for stomach and colorectal cancers. Among these
two cancers, there were no risk estimates 1.0 for the highest level
of raw/cooked garlic consumption compared with the lowest, suggesting
either a protective association across all reports or a degree of
publication bias. It is possible that residual confounding may account
for some degree of the association because many reports did not control
for total vegetable consumption or other known risk factors for these
diseases. Promising results for the association of raw/cooked garlic
and prostate, laryngeal and breast cancers were observed, but these
were based on only three case-controls studies. There is no
evidence to support a relationship with esophageal cancer; however,
garlic supplements showed no association with risk of cancer, although
a positive relationship was reported for exclusive garlic supplement
users and risk of lung cancer. This result may be either a spurious
association or the outcome of a physiologic or behavioral interaction
with known risk factors for lung cancer, such as smoking.
Alternative hypotheses must be considered. There is epidemiologic
evidence to support a protective effect for total vegetable consumption
and cancers of the stomach and colon (Hensrud and Heimburger 1998
). Total vegetable consumption may be positively correlated
with garlic intake, in which garlic consumers are more likely to have
higher total vegetable intakes. The protective effects may be
attributed to the type of dish in which garlic is prepared, such as
pasta and tomato dishes or residual confounding from other food
sources. Furthermore, publication bias appears to be influencing any
judgement, particularly for stomach and colon cancers for which several
reports are available. The funnel plot can be interpreted to indicate
that the protective effect found between garlic and cancers of the
stomach and colon may be overestimated due to unpublished reports with
null or positive results. In this review of published articles,
however, an inverse association of raw and/or cooked garlic consumption
is consistent among studies of stomach and colorectal cancer. Garlic
supplements, in contrast, on the basis of one case-control study of
prostate cancer and four studies from the Netherlands Cohort of
colorectal, stomach, lung and breast cancers, do not appear to be
related to risk of cancer. Additional studies are required before
definitive conclusions can be drawn about the role of garlic and garlic
supplements in cancer etiology.
ACKNOWLEDGMENTS
The authors are indebted to Charles Poole and Neal Simonsen for preparation and review of this manuscript and Eva Buaitti and Loic Le Marchand for supplying additional exposure classification information.
FOOTNOTES
1 Presented at the conference "Recent Advances on the Nutritional Benefits Accompanying the Use of Garlic as a Supplement" held November 15–17, 1998 in Newport Beach, CA. The conference was supported by educational grants from Pennsylvania State University, Wakunaga of America, Ltd. and the National Cancer Institute. The proceedings of this conference are published as a supplement to The Journal of Nutrition. Guest editors: John Milner, The Pennsylvania State University, University Park, PA and Richard Rivlin, Weill Medical College of Cornell University and Memorial Sloan-Kettering Cancer Center, New York, NY. 
3 Abbreviations used: FFQ, food frequency questionnaire; OR, odds ratio; CI, confidence interval; RR, relative risk.
REFERENCES
1. American Cancer Society www.cancer.org (1998).
2.
Begg C. B., Berlin J. A. Publication bias and dissemination of clinical research. J. Natl. Cancer Inst. 1989;81:107-115
3. Begg C. B., Mazumbar M. Operating characteristics of a rank correlation test for publication bias. Biometrics 1994;50:1088-1101[Medline]
4. Bilyk A., Sapers G. M. Distribution of quercetin and kaempferol in lettuce, kale, chive, garlic, leek, horseradish, red radish and red cabbage tissues. J. Agric. Food Chem. 1985;33:226-228
5. Buaitti E., Palli D., Decarli A., Amadori D., Avellini C., Bianchi S., Biserni R., Cipriani F., Cocco P., Giacosa A., Marubini E., Puntoni R., Vindigni C., Fraumeni J., Blot W. A case-control study of diet and gastric cancer in Italy. Int. J. Cancer 1989;44:611-616[Medline]
6.
Dorant E., van den Brandt P. A., Goldbohm R. A. A prospective cohort study on Allium vegetable consumption, garlic supplement use, and the risk of lung carcinoma in the Netherlands. Cancer Res 1994;54:6148-6153
7. Dorant E., van den Brandt P. A., Goldbohm R. A. Allium vegetable consumption, garlic supplement intake, and female breast carcinoma incidence. Breast Cancer Res. Treat. 1995;33:163-170[Medline]
8.
Dorant E., van den Brandt P. A., Goldbohm R. A. A prospective cohort study on the relationship between onion and leek consumption, garlic supplement use and the risk of colorectal carcinaoma in the Netherlands. Carcinogenesis 1996a;17:477-484
9. Dorant E., van den Brandt P. A., Goldbohm R. A., Hermus R.J.J., Sturmansd F. Garlic and its significance for the prevention of cancer in humans: a critical view. Br. J. Cancer 1993;67:424-429[Medline]
10. Dorant E., van den Brandt P. A., Goldbohm R. A., Sturmans F. Consumption of onions and a reduced risk of stomach carcinoma. Gastroenterology 1996b;110:12-20[Medline]
11. Gao Y. T., McLaughlin J. K., Gridley G., Blot W. J., Ji B. T., Dai Q., Fraumeni J. F. Risk factors for esophageal cancer in Shanghai, China. II. Role of diet and nutrients. Int. J. Cancer 1994;58:197-202[Medline]
12.
Giovannucci E., Rimm E. B., Stampfer M. J., Colditz G. A., Ascherio A., Willett W. C. Intake of fat, meat, and fiber in relation to risk of colon cancer in men. Cancer Res 1994;54:2390-2397
13. Greenland, S. (1998) In: Modern Epidemiology (Rothman, K. & Greenland, S., eds.), pp. 643–673. Lippencott-Raven, Philadelphia, PA.
14. Hansson L. E., Nyren O., Bergstrom R., Wolk A., Lindgren A., Baron J., Adami H. O. Diet and risk of gastric cancer: a population-based case-control study in Sweden. Int. J. Cancer 1993;55:181-189[Medline]
15. Hensrud D. D., Heimburger D. C. Diet, nutrients and gastrointestinal cancer. Gastroeneterol. Clin. N. Am. 1998;27:325-346
16. Hertog M.G.L., Hollman C. H., Katan M. B. Content of potentially anticarcinogenic flavanoids of 28 vegetables and 9 fruits commonly consumed in the Netherlands. J. Agric. Food Chem. 1992;40:2379-2383
17.
Hu J., Liu Y., Zhao T., Liu S., Wang Q. Diet and cancer of the colon and rectum: a case-control study in China. Int. J. Epidemiol. 1991;20:362-367
18. Hu J., Nyren O., Wolk A., Bergstrom R., Yuen J., Adami H. O., Guo L., Li H., Huang G., Xu X., Zhao F., Chen Y., Wang C., Qin H., Hu C., Li Y. Risk factors for oesophageal cancer in northeast Cina. Int. J. Cancer 1994;57:38-46[Medline]
19. Hughes B. G., Lawson L. D. Antimicrobial effects of Allim sativum L. (garlic), Allium ampeloprasum L. (elephant garlic), and Allium cepa L. (onion), garlic compounds and commercial garlic supplement products. Phytother. Res. 1991;5:154-158
20.
Ip C., Lisk D. J., Thompsom H. J. Selenium enriched garlic inhibits the early stage but not the late stage of mammary carcinogenesis. Carcinogenesis 1996;17:1979-1982
21. Iscovich J. M., L’Abbé K. A., Castelleto R., Calzona A., Bernedo A., Chopita N. A., Jmelnitzsky A. C., Kaldor J. Colon cancer in Argentina. I: Risk from intake of dietary items. Int. J. Cancer 1992;51:851-857[Medline]
22.
Jonkers D., van den Broek E., van Dooren I., Thijs C., Dorant E., Hageman G., Stobberingh E. Antibacterial effect of garlic and omeprazole on Helicobacter pylori. J. Antimicrob. Chemother. 1999;43:837-839
23. Key T.J.A., Silcocks P. B., Davey G. K., Appleby G. K., Bishop D. T. A case-control study of diet and prostate cancer. Br. J. Cancer 1997;76:678-687[Medline]
24. Klipstein-Grobusch K., Kroke A., Voss S., Boeing H. Influence of lifestyle on the use of supplements in the Brandenburg nutrition and cancer study. Z. Ernaehrwiss. 1998;37:38-46[Medline]
25. Le Marchand L., Hankin J. H., Wilkens L. R., Kolonel L. N., Englyst H. N., Lyu L.-C. Dietary fiber and colorectal cancer risk. Epidemiology 1997;8:658-665[Medline]
26. Levi F., La Vecchia C., Gulie C., Negri E. Dietary factors and breast cancer risk in Vaud, Switzerland. Nutr. Cancer 1993;19:327-334[Medline]
27.
Reddy B. S., Rao C. V., Rivenson A., Kelloff G. Chemoprvention of colon carcinogenesis by organosulfur compounds. Cancer Res 1993;53:3493-3498
28. Schaffer E. M., Liu J. Z., Milner J. A. Garlic powder and allyl sulfur compounds enhance the ability of dietary selenite to inhibit 7,12-dimethylbenz[a]anthracene-induced mammary DNA adducts. Nutr. Cancer 1997;27:162-168[Medline]
29. Sivam G. P., Lampe J. W., Ulness B., Swanzy S. R., Potter J. D. Helicobacter pylori—in vitro susceptibility to garlic (Allium sativum) extract. Nutr. Cancer 1997;27:118-121[Medline]
30.
Steinmetz K. A., Kushi L. H., Bostick R. M., Folsom A. R., Potter J. D. Vegetables, fruit and colon cancer in the Iowa Women’s Health Study. Am. J. Epidemiol. 1994;139:1-13
31. Sundaram S. G., Milner J. A. Diallyl disulfide suppresses the growth of human colon tumor cell xenografts in athymic nude mice. J. Nutr. 1996;126:1355-1361
32. Wargovich M. J., Uda N., Woods C., Velasco M., McKee K. Allium vegetables: their role in the prevention of cancer. Bioact. Compounds Foods 1996;24:811-813
33.
Witte J. S., Longnecker M. P., Bird C. L., Lee E. R., Frankl H. D., Haile R. W. Relation of vegetable, fruit and grain consumption to colrectal adenomatous polyps. Am. J. Epidemiol. 1996;144:1015-1025
34. World Cancer Research Fund, American Institute for Cancer Research Food, Nutrition and the Prevention of Cancer: A Global Perspective. AICR, Washington, DC 1997;
35.
You W., Blot W. J., Chang Y., Ershow A. G., Yang Z., An Q., Henderson B., Xu G., Fraumeni J. F., Wang T. Allium vegetables and reduced risk of stomach cancer. J. Natl. Cancer Inst. 1989;81:162-164
36. Zheng W., Blot W. J., Shu X. O., Diamond E. L., Gao Y. T., Ji B. T., Fraumeni J. F. A population-based case-control study of cancers of the nasal cavity and paranasal sinuses in Shanghai. Int. J. Cancer 1992a;52:557-561[Medline]
37.
Zheng W., Blot W. J., Shu X. O., Gao Y. T., Ji B. T., Ziegler R. G., Fraumeni J. F. Diet and other risk factors for laryngeal cancer in Shanghai, China. Am. J. Epidemiol. 1992b;136:178-191
This article has been cited by other articles:
|
|
 |
|
  G. Filomeni, G. Rotilio, and M. R. Ciriolo Molecular Transduction Mechanisms of the Redox Network Underlying the Antiproliferative Effects of Allyl Compounds from Garlic J. Nutr., November 1, 2008; 138(11): 2053 - 2057. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Zhang, M.-L. Noordine, C. Cherbuy, P. Vaugelade, J. M. Pascussi, P.-H. Duee, and M. Thomas Different activation patterns of rat xenobiotic metabolism genes by two constituents of garlic Carcinogenesis, October 1, 2006; 27(10): 2090 - 2095. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. H. Kushi, T. Byers, C. Doyle, E. V. Bandera, M. McCullough, T. Gansler, K. S. Andrews, M. J. Thun, and The American Cancer Society 2006 Nutrition and Phy American Cancer Society Guidelines on Nutrition and Physical Activity for Cancer Prevention: Reducing the Risk of Cancer With Healthy Food Choices and Physical Activity CA Cancer J Clin, September 1, 2006; 56(5): 254 - 281. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. J. Greenblatt, R. A. Leigh-Pemberton, and L. L. von Moltke In Vitro Interactions of Water-Soluble Garlic Components with Human Cytochromes P450 J. Nutr., March 1, 2006; 136(3): 806S - 809S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Ishikawa, T. Saeki, T. Otani, T. Suzuki, K. Shimozuma, H. Nishino, S. Fukuda, and K. Morimoto Aged Garlic Extract Prevents a Decline of NK Cell Number and Activity in Patients with Advanced Cancer J. Nutr., March 1, 2006; 136(3): 816S - 820S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Tanaka, K. Haruma, M. Yoshihara, G. Kajiyama, K. Kira, H. Amagase, and K. Chayama Aged Garlic Extract Has Potential Suppressive Effect on Colorectal Adenomas in Humans J. Nutr., March 1, 2006; 136(3): 821S - 826S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Milner Preclinical Perspectives on Garlic and Cancer J. Nutr., March 1, 2006; 136(3): 827S - 831S. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Matsuura, Y. Miyamae, K. Yamane, Y. Nagao, Y. Hamada, N. Kawaguchi, T. Katsuki, K. Hirata, S.-I. Sumi, and H. Ishikawa Aged Garlic Extract Inhibits Angiogenesis and Proliferation of Colorectal Carcinoma Cells J. Nutr., March 1, 2006; 136(3): 842S - 846S. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Hosono, T. Fukao, J. Ogihara, Y. Ito, H. Shiba, T. Seki, and T. Ariga Diallyl Trisulfide Suppresses the Proliferation and Induces Apoptosis of Human Colon Cancer Cells through Oxidative Modification of {beta}-Tubulin J. Biol. Chem., December 16, 2005; 280(50): 41487 - 41493. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-W. Tsai, J.-J. Yang, H.-W. Chen, L.-Y. Sheen, and C.-K. Lii Garlic Organosulfur Compounds Upregulate the Expression of the {pi} Class of Glutathione S-Transferase in Rat Primary Hepatocytes J. Nutr., November 1, 2005; 135(11): 2560 - 2565. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Berges, M.-H. Siess, I. Arnault, J. Auger, R. Kahane, M.-F. Pinnert, M.-F. Vernevaut, and A.-M. le Bon Comparison of the chemopreventive efficacies of garlic powders with different alliin contents against aflatoxin B1 carcinogenicity in rats Carcinogenesis, October 1, 2004; 25(10): 1953 - 1959. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Druesne, A. Pagniez, C. Mayeur, M. Thomas, C. Cherbuy, P.-H. Duee, P. Martel, and C. Chaumontet Diallyl disulfide (DADS) increases histone acetylation and p21waf1/cip1 expression in human colon tumor cell lines Carcinogenesis, July 1, 2004; 25(7): 1227 - 1236. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Huard, N. Druesne, D. Guyonnet, M. Thomas, A. Pagniez, A.-M. Le Bon, P. Martel, and C. Chaumontet Diallyl disulfide (DADS) enhances gap-junctional intercellular communication by both direct and indirect mechanisms in rat liver cells Carcinogenesis, January 1, 2004; 25(1): 91 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. K. Brown, T. Byers, C. Doyle, K. S. Courneya, W. Demark-Wahnefried, L. H. Kushi, A. McTiernan, C. L. Rock, N. Aziz, A. S. Bloch, et al. Nutrition and Physical Activity During and After Cancer Treatment: An American Cancer Society Guide for Informed Choices CA Cancer J Clin, September 1, 2003; 53(5): 268 - 291. [Abstract] [Full Text]
|
|
|
|
 |
|
 A. W. Hsing, A. P. Chokkalingam, Y.-T. Gao, M. P. Madigan, J. Deng, G. Gridley, and J. F. Fraumeni Jr. Allium Vegetables and Risk of Prostate Cancer: A Population-Based Study J Natl Cancer Inst, November 6, 2002; 94(21): 1648 - 1651. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Fisher and J. E. Morley Editorial: Antiaging Medicine: The Good, the Bad, and the Ugly J. Gerontol. A Biol. Sci. Med. Sci., October 1, 2002; 57(10): M636 - 639. [Full Text] [PDF]
|
|
|
|
|
|
D. Guyonnet, C. Belloir, M. Suschetet, M.-H. Siess, and A.-M. Le Bon Mechanisms of protection against aflatoxin B1 genotoxicity in rats treated by organosulfur compounds from garlic Carcinogenesis, August 1, 2002; 23(8): 1335 - 1341. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Wargovich, C. Woods, D. M. Hollis, and M. E. Zander Herbals, Cancer Prevention and Health J. Nutr., November 1, 2001; 131(11): 3034S - 3036. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Milner A Historical Perspective on Garlic and Cancer J. Nutr., March 1, 2001; 131(3): 1027S - 1031. [Abstract] [Full Text]
|
|
| ||||||||||||||
