Research paper 1
Exploring the potential carcinogenic role of arsenic in
gallbladder cancer
Nivetha Ganesana, Kathryn Bambinob, Paolo Boffettaa and Ismail Labgaac
Gallbladder cancer (GBC) is an aggressive malignancy,
associated with dismal outcomes. Although several risk
factors including age, sex, and gallstones have been
postulated, epidemiologic determinants of the disease
remain largely uncovered. Moreover, the implication of
environmental toxicants as possible risk factors is
increasingly suspected. Arsenic (As), an established human
carcinogen, is a natural contaminant of groundwater and
has a geographic distribution similar to GBC incidence.
This, combined with As metabolites being partially excreted
in bile, raised the hypothesis that As may represent a
carcinogenic hazard for the gallbladder. We conducted an
analysis of the association between As concentration in
groundwater and incidence rates of GBC worldwide in 52
countries. The USA, India, and Taiwan were selected on the
basis of availability and quality of data for further
investigation at a county-level. Relationships between As
levels and GBC incidence were assessed using
multivariable linear regression analyses. Analyses revealed
significant associations between high As concentrations in
groundwater and increased GBC incidences. Among
women, correlations were observed worldwide
(Spearman = 0.31, P = 0.028), in Taiwan (Spearman = 0.57,
P = 0.005) and in India (R2 = 0.23, P = 0.006). In men, a
Introduction
Gallbladder cancer (GBC) is characterized by aggressive
biology. It is one of the rare malignancies showing an
increasing mortality during the last two decades (Sawyers et al.,
2013; Are et al., 2017) and it is foremost associated with dismal
outcomes as 5-year survival rates of ~ 5% (Levy et al., 2001;
Randi et al., 2006). GBC typically only becomes symptomatic
at an advanced stage where curative treatments can no longer
be offered. Improvements of outcomes have been reported
because of more aggressive surgical resections and combined
neoadjuvant therapies; nonetheless, progress remained mod-
est (Hundal and Shaffer, 2014). Hence, GBC is a condition
where prevention is warranted to improve outcomes, but this
approach relies on the identification of etiologic factors (Jaffee
et al., 2017).
Epidemiologically, GBC shows striking disparities in its geo-
graphic and ethnic distributions (Lazcano-Ponce et al., 2001;
Misra et al., 2003; Hundal and Shaffer, 2014). While risk factors
including age, sex, gallstones, and ethnicity have been
Supplemental Digital Content is available for this article. Direct URL citations
appear in the printed text and are provided in the HTML and PDF versions of this
article on the journal's website, www.eurjcancerprev.com.
0959-8278 Copyright © 2019 Wolters Kluwer Health, Inc. All rights reserved.
Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
correlation was observed in India (R2 = 0.26, P = 0.009) and
a modest correlation was identified in the USA
(Spearman = 0.14, P = 0.026). These results provide some
support to the hypothesis of an association between high
exposures to As-contaminated water on GBC, which
appeared more prominent in women. Further observational
and molecular studies, conducted at the individual level, are
required to confirm this association and decipher its
nature. European Journal of Cancer Prevention 00:000–000
Copyright © 2019 Wolters Kluwer Health, Inc. All rights
reserved.
European Journal of Cancer Prevention 2019, 00:000–000
Keywords: arsenic, biliary cancer, cancer epidemiology, heavy metals,
toxicology
a
Tisch Cancer Institute, bDepartment of Environmental Medicine and Public
Health, Icahn School of Medicine at Mount Sinai, New York City, New York, USA
and cDepartment of Visceral Surgery, Lausanne University Hospital CHUV,
Lausanne, Switzerland
Correspondence to Paolo Boffetta, MD, MPH, Tisch Cancer Institute, Icahn
School of Medicine at Mount Sinai, One Gustave L. Levy Place, Box 1130,
New York, NY 10029, USA
Tel: + 1 212 824 7378; fax: + 1 212 849 2566; e-mail: paolo.boffetta@mssm.edu
Received 27 November 2018 Accepted 29 March 2019
identified, they explain a modest proportion of the burden of
this disease, and there is accumulating evidence to suspect the
implication of environmental agents (Wistuba and Gazdar,
2004). Heavy metals are relevant candidates given their role in
the development of diseases including cancers (Landrigan
et al., 2018). Among them, arsenic (As) is a human carcinogen
with demonstrated associations with bladder, kidney, and lung
cancers (Straif et al., 2009), but data on GBC remain scarce
(Chhabra et al., 2012). Although exposure to As occurs from
industrial and environmental sources, consumption of con-
taminated groundwater of natural origin represents an impor-
tant route of exposure for many populations worldwide (Straif
et al., 2009). Metabolized by the liver, its metabolites are
excreted in bile and urine (Argos et al., 2010; Ponomarenko
et al., 2017). Thus, one may expect an oncogenic role of As for
both urogenital and biliary cancers. While an association
between As and urinary cancer has been established, data on
its potential impact in hepatobiliary malignancies remain scant
(Hopenhayn-Rich et al., 1998).
GBC incidence and concentration of As in groundwater
show similar patterns across countries and regions as
exemplified in Chile, Bangladesh, or India. We tested
DOI: 10.1097/CEJ.0000000000000521
 2 European Journal of Cancer Prevention 2019, Vol 00 No 00
the hypothesis that exposure to As-contaminated water
may be a risk factor in GBC, by conducting a series of
ecologic analyses on the correlation between As in
groundwater and GBC incidence.
Materials and methods
Study design
We performed an observational ecological study using
publicly available data sets to investigate the correlation
between As concentration in groundwater and GBC
incidence.
Data collection
Worldwide
Worldwide GBC incidence rates (ICD10 code, C23) in 52
countries were compiled using GLOBOCAN 2012
(Ferlay et al., 2013), run by the International Agency for
Research on Cancer that includes estimates of worldwide
cancer incidence rates (Curado et al., 2007; Forman et al.,
2013). A published data set was utilized to extract global
information on As level in groundwater (Ravenscroft
et al., 2009). After a detailed assessment of data available
in individual countries, a more thorough analysis of
county and/or district-level data was performed for the
USA, Taiwan, and India. These countries were selected
on the basis of the availability and quality of data.
USA
County-level sex-stratified GBC incidence rates for years
2000–2014 were calculated using SEER-Stat Software
(version 8.3.4; Surveillance Research Program, NCI,
Bethesda, Maryland, USA) [Surveillance Epidemiology
and End Results (SEER) Program, 2018]. Rates are
expressed as per 100 000 person-years, age-standardized
according to the 2000 US Standard population. The
SEER database is composed of 18 registries in 13 states
that covers 629 counties used for county-level analyses.
A data set containing 20 042 measurements of As col-
lected between the years of 1976–1999 from groundwater
sources in 1375 counties in 49 states was obtained from
the US Geological Survey (USGS) [United States
Geological Survey (USGC), 2001]. Sampling and analysis
were performed by USGS through its district offices and
national programs. The measurements were sorted by
county using the Federal Information Processing
Standards code and by year.
Counties with unavailable data for either As or GBC
incidence were excluded. Additional demographics and
health risk factors were considered and integrated into
our model. As an increased prevalence of GBC is typi-
cally reported among US Hispanics (Jaruvongvanich
et al., 2019), females (Lazcano-Ponce et al., 2001; Randi
et al., 2006) and obese patients (Larsson and Wolk, 2007),
county-level proportions of Hispanics and obesity pre-
valence, stratified for sex, were obtained from NCI’s
county attributes files (SEER-Stat Software, version
Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
8.3.4; National Cancer Institute, Surveillance Research
Program, NCI) [Surveillance Epidemiology and End
Results (SEER) Program, 2018] and from the Division of
Diabetes Translation in the Centers for Disease Control
[Centers for Disease Control and Prevention (CDC),
Division of Diabetes Translation (2016)], respectively.
Taiwan
Data on age-standardized, sex-stratified GBC incidence in 25
counties were courteously provided by the Taiwan Cancer
Registry Center for the period 2003–2009 (Taiwan Cancer
Registry Center, 3–19 September 2017; personal commu-
nication). Annual rates, expressed as per 100 000 persons/year,
were averaged for an overall incidence rate on county-basis.
A data set containing 12 831 As measurements in groundwater
for each of the 25 counties between 1993 and 2007 were
obtained from the Taiwan Environmental Protection Agency
(Taiwan Environmental Protection Agency, 2010). Sample
collection and analysis were conducted by the Environmental
Analysis Laboratory, a branch of the Taiwan EPA.
The data set was sorted by year and county and was
matched to GBC rates in their respective regions. The
final data set used for analysis included 22/25 counties. As
Chinese-Han primarily makes up 98% of the Taiwanese
population (Spain, 1984) and data on obesity prevalence
rates were limited; no demographics or other potentially
confounding factors were considered.
India
Data from Population-Based Cancer Registries were
published in consolidated 3-year reports through the
National Cancer Registry Programme [National Cancer
Registry Programme (NRCP), 2013]. The most recent
report was used to collect sex-stratified GBC incidence
rates for 27 districts between 2012 and 2014. Of note, the
reported rates involved all cancers of the biliary tract
(ICD10 code C23 and C24). Therefore, GBC incidence
was inferred by applying ratios calculated using data from
CI5 Volume XI (Bray et al., 2017), which reported rates
with only C23 and rates with C23 and C24 combined.
Two of the 27 districts, which were in the north, were not
reported in the CI5 data and had to be excluded from the
analysis. Thus, a total of 25 districts were included in the
final analysis.
Indian As data for the period 1974–2008 were primarily
extracted from the Central Ground Water Board Central
Ground Water Board (CGWB) (2015), a government
branch of the Ministry of Water Resources which
annually monitors the quality of groundwater in 15 640
wells throughout the major part of Indian Territory. To
also analyze northeast states (Sikkim, Manipur,
Arunachal Pradesh, Mizoram, and Manipur), we utilized
data collected by the North Eastern Regional Institute of
Water and Land Management (Singh, 2004) that includes
~ 4000 samples from 2529 wells.

Both reports provided a list of districts with As levels
above 50 µg/l. Therefore, district data from both reports
were integrated and dichotomously categorized as high or
low level on the basis of a cutoff of 50 µg/l. GBC rates for
each state were calculated on the basis of averaged values
of their corresponding districts. With regard to potential
confounders, most regions of India are primarily made up
of a homogenous Indian-Asian population. Furthermore,
district-level data were scarce for obesity prevalence and
hence demographics and other potentially confounding
factors were not integrated into the model.
Statistical methods
Mean values of As concentrations were calculated for
each country, county, or district from the Global, USGS
[United States Geological Survey (USGC), 2001], and
Taiwan Environmental Protection Agency data sets (Taiwan
Environmental Protection Agency, 2010). The US data were
modeled using a multivariable linear regression adjusting for
age, sex, race, and obesity. Global, Taiwan, and India data
were modeled using a simple linear regression adjusting only
for age and sex. All models were tested with both unstratified
and stratified for sex and tested for effect modification using
interaction terms with As. Because of non-normal distributions
across all data sets, nonparametric statistical tests were used for
the significance of correlation. All statistical analyses were
performed using SAS Statistical Software 9.4 (SAS Institute,
Cary, North Carolina, USA). The US county maps were
developed using R Studio Software (version 3.4.2) and cor-
responding choroplethr package (Lamstein, 2019).
Results
Worldwide
First, we explored whether countries with high levels of As in
groundwater also displayed a high incidence of GBC, by
generating two lists of the top 25 countries for both As con-
centrations in groundwater and GBC incidence in females
(Supplementary Table 1, Supplemental digital content 1,
http://links.lww.com/EJCP/A239). As illustrated by Fig. 1a, a
substantial overlap was observed, with 16 (64%) countries
ranking within top 25 countries for both As the level and GBC
incidence.
To statistically test this potential global correlation, we
extended our global analysis to 52 countries with avail-
able data (Fig. 1b).
United states
Available data for GBC analysis involved 629/3142 (20%)
counties in 13 states. In females, 114/629 counties had
GBC rates greater than 2.0 (18%) per 100 000/year
(Fig. 2a). The western states including California, Utah,
and New Mexico showed higher rates of GBC compared
with other states. In males, however, only 47/629 coun-
ties had GBC rates that were greater than 2.0 (7%) (map
not shown). Data available for analysis of As levels are
included in 1374/3142 (44%) counties. Relatively higher
Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Arsenic in gallbladder cancer Ganesan et al. 3
As levels are concentrated in the western and north-
eastern regions of the USA (Fig. 2b). In addition,
797/1374 (58%) counties showed As measurements of up
to 1 µg/l, whereas 7% had an As level of at least 10 µg/l. A
total of 254/3142 (8%) counties had available data for
both As and GBC data and were included in the final
analysis.
After stratifying for sex, no correlation was detected in females
(Spearman = 0.051; P = 0.42), but a modest correlation was
observed in males (Spearman = 0.14; P = 0.026; Fig. 2c).
Taiwan
GBC rates were all lower than 2.0 per 100 000 but 7/25 (28%)
regions showed a GBC rate of 0.9 per 100 000 or greater in
females (Fig. 3a). Data for As were available in 22/25 (88%)
counties. Analogously to the distribution of GBC, Taiwan
City, and Kaohsiung City in the southwest and Yilan County
in the mid-north showed the highest As levels (Fig. 3b).
Overall, 12 (55%) counties displayed As levels above 10 µg/l.
After stratifying for sex (Fig. 3c), positive correlations
remained significant in females (Spearman = 0.57, P = 0.005),
whereas a trend was noted in males (Spearman = 0.38,
P = 0.082).
India
Data available for analysis included 16/29 (55%) states. Rates
of GBC in females and males were greater than 2.0 per
100 000/year in 10 and three states, respectively (63 and 19%).
The states with the highest GBC rates in both males and
females appeared in northeastern states such as Assam, West
Bengal, and Madya Pradesh (Fig. 4a). As data were available
for each of the 25 corresponding districts, of which 11
(44%) showed levels above 50 µg/l (Fig. 4b). Similar to the
GBC distribution, northeast regions were characterized by
high levels of As contamination. In addition, a heatmap
showing visual correlation is provided in Supplementary
Table 2 (Supplemental digital content 1, http://links.lww.com/
EJCP/A239).
GBC rates in both males and females also positively
correlated with As levels (females: R2 = 0.23, P = 0.006;
males: R2 = 0.26, P = 0.009; Fig. 4c).
Discussion
Evidence on risk factors in GBC is limited and it is, therefore,
paramount to investigate the potential role that several
environmental carcinogens may play in the development of
this disease. Herein, we identified significant associations
between As exposure and GBC incidence rates in women
worldwide, in India and in Taiwan. Data for men also showed
a significant association in India and a modest correlation in
the USA.
As previously mentioned, the incidence of GBC mark-
edly varies across different regions/countries. Although
these variations might result from previously described
factors such as age, sex, or gallstones prevalence, one may
 4 European Journal of Cancer Prevention 2019, Vol 00 No 00

Fig. 1

Distribution of arsenic (As) level in groundwater and gallbladder cancer incidence worldwide. (a) A world map showing the distribution of the top 25
countries with the highest concentrations of arsenic in groundwater (As, light gray) and the top 25 countries for gallbladder cancer (GBC) incidence
in females (dark gray); countries in medium grey are among top 25 countries for both As levels in groundwater and GBC incidence in females. (b)
Sex-stratified scatterplot of As concentration in groundwater and GBC incidence.

Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Arsenic in gallbladder cancer Ganesan et al. 5

Fig. 2

Distribution of arsenic level in groundwater and gallbladder cancer incidence in the USA. (a) Distribution of arsenic (As) concentration in groundwater
in the US counties. (b) Distribution of gallbladder cancer (GBC) incidence in females, in the US counties. (c) Sex-stratified scatterplot of GBC and As
levels.

reasonably suspect that the latter causes cannot entirely
explain such differences. Our results provide support to our
hypothesis and this may partially explain the high GBC
incidence in North India, Chile, or Bangladesh. While we
could explore and confirm our hypothesis for North India, the
quality of data in Chile and Bangladesh did not permit ade-
quate investigations. With regard to North India, data from a
very recent study also identified a correlation between soil As
levels and GBC, which support our hypothesis and findings
(Madhawi et al., 2018).

Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 6 European Journal of Cancer Prevention 2019, Vol 00 No 00

Fig. 3

Distribution of arsenic levels in groundwater and gallbladder cancer incidence in Taiwan. (a) Distribution of arsenic (As) concentration in groundwater
in Taiwanese counties. (b) Distribution of gallbladder cancer (GBC) incidence in females, in Taiwanese counties. (c) Sex-stratified scatterplot of GBC
and As levels.

The correlation analysis for the global and Taiwanese data sets
indicates a stronger correlation between As exposure and
GBC rates in females than in males. Results of our GBC data
remains consistent with several studies showing a two-fold
increased incidence of GBC in females compared with males
(Lazcano-Ponce et al., 2001; Randi et al., 2006). This has been
partially attributed to sex hormones and differences in lifestyle
behaviors (Randi et al., 2006; Shukla et al., 2008; Saranga
Bharathi et al., 2015). The higher incidence in females could
also be partially explained by a combination of two reasons: (i)
As consumption might increase in women given they are
more likely to spend more time at home, thus having less
opportunities to drink other beverages/filtered water that are
often provided at workplaces; (ii) previous studies also sug-
gested an influence of sex hormones on As metabolism
(Huang et al., 2017).
The relatively low correlation in the US suggests a dose-
dependency. While correlations were showed in both
Taiwan and India where As concentrations in drinking
water reached levels of at least 10 and 50 µg/l, respec-
tively, no correlation with GBC incidence in females was
detected in the USA were 92.8% of counties showed As
levels of less than 10 µg/l. Several studies determined
that public drinking water exceeded 5 µg/l (Frey and
Edwards, 1997) in only 3–6% of water distribution sys-
tems. Hence, the impact we may expect from As-
contaminated water would only be observable in people
drinking water from unregulated wells, which only served
14% of the US population (Maupin et al., 2014). Of
interest, western and northeastern states showed
increased GBC rates and the former were also associated
with elevated As levels in groundwater.
Of note, although one may argue that our analyses
detected modest correlation factors at first glance, it must
be interpreted in the context of an ecological study using
population-wide data and not a basic research project

Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 Arsenic in gallbladder cancer Ganesan et al. 7

Fig. 4

Distribution of arsenic levels in groundwater and gallbladder cancer incidence in India. (a) Distribution of arsenic (As) concentration in groundwater in
Indian states. (b) Distribution of gallbladder cancer (GBC) incidence in females, in Indian states. (c) A boxplot of GBC rates stratified by gender and
dichotomized for low and high levels of arsenic in groundwater.

exploring physiological mechanisms or a clinical study
investigating biomarkers, for example. In this setting, our
results showed rather substantial correlation factors.
If environmental As exposure is truly associated with
increased risk to develop GBC, one may reasonably
expect a similar correlation with occupational As expo-
sure. The literature on the topic was reviewed to further
study this question. Smelter workers, chemical manu-
facturing industry workers, and metal recycling workers
are some of the occupational cohorts that have been
exposed to As and studied for cancer risk (Rosenberg
et al., 1980; Landrigan, 1981; Xi et al., 2011). No results on
GBC risk were reported in these studies. In a small
cohort of Italian pesticide workers with exposure to As,
mortality from GBC was increased (Giordano et al., 2006).
These limited findings stress the need to explore the
potential impact of As occupational exposure on GBC
incidence.
To explain the possible association between As exposure
and GBC risk, we developed a hypothesis-driven model
(Fig. 5). Inorganic As is metabolized by a series of
reduction and oxidative methylation reactions catalyzed
by the trivalent As-specific methyltransferase, AS3MT.
AS3MT is highly evolutionarily conserved, and its
expression is enriched in the livers of vertebrate species
(Thomas et al., 2004; Tseng, 2007; Bambino and Chu,
2017). The expression levels and catalytic activity of
AS3MT contributed to methylation capacity and sus-
ceptibility to toxicity in human populations. Human
population studies showed that a higher ratio of urinary
MMA/DMA is associated with increased risk of As toxi-
city (Chen et al., 2013; Das et al., 2016). Recent studies
revealed that mouse strains with reduced capacity for As
methylation were more susceptible to liver injury (Wu
et al., 2017). Data from multiple animal systems revealed
associations between exposure to inorganic As and

Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
 8 European Journal of Cancer Prevention 2019, Vol 00 No 00
Fig. 5
Conceptual model of As metabolism. A hypothesis-driven model of the
hepatic metabolism of Arsenic (black). While the more toxic trivalent and
monomethylated metabolites (white) are excreted through the bile and
tend to accumulate in the gallbladder, pentavalent and more completely
methylated metabolites (gray) are preferentially excreted through the
urine. Illustration by Jill K. Gregory, printed with permission from Mount
Sinai Health System.
hepatobiliary toxicity (Byron et al., 1967), gallbladder
toxicity (Cockell and Bettger, 1993; Pedlar et al., 2002;
Tokar et al., 2012), and GBC (Tokar et al., 2012). Dietary
exposure to inorganic As led to the accumulation of As in
gastrointestinal organs, skin, and scales of lake whitefish
(Pedlar and Klaverkamp, 2002). In several fish species
(Cockell and Bettger, 1993), dietary exposure to inor-
ganic As caused inflammation and fibrosis of the gall-
bladder, with macroscopic lesions detected after chronic
exposure to the highest dose tested (Pedlar et al., 2002).
Mammalian experiments showed similar results. Rats
exposed to inorganic As for 2 years developed chronic
inflammation of the extrahepatic bile ducts (Byron et al.,
1967). Lifelong exposure to inorganic As caused a dose-
dependent increase in GBC incidence in male mice
(Tokar et al., 2012). However, these animal studies must
be cautiously extrapolated to human. Indeed, GBC stu-
dies are likely to be affected by species-specific differ-
ences in the gastrointestinal absorption of arsenicals and
by differences in As metabolism (Csanaky and Gregus,
2002). For example, rat hemoglobin readily binds to
methylated As metabolites (Aposhian, 1997) and rats
excrete MMAIII only through bile. Rates of As
Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
methylation varies between species, mouse strains (Wu
et al., 2017), and with AS3MT haplotype in humans
(Engstrom et al., 2011, 2013). Taken together, these data
suggest possible mechanisms underlying an association
between environmental exposure to inorganic As and the
development of GBC. We assume that the preferential
accumulation of the more toxic trivalent and mono-
methylated metabolites (Styblo et al., 2000) in the bile
compared with the preferential secretion of pentavalent
and more completely methylated metabolites through
the urine (Csanaky and Gregus, 2002) may explain an
association between As exposure and risk of GBC
(Fig. 5). As metabolites may either be excreted in urine
or in bile. In humans, it is also plausible that genetic
factors could control As metabolism (Engstrom et al.,
2011), leading to patients with preferential As excretion
either in urine or in bile. Hence, patients with pre-
ferential As metabolites excretion in urine would be at
higher risk of urogenital cancers while the ones with
increased excretion of As metabolites in bile would pre-
sent an increased risk to develop biliary malignancies.
Noteworthy, studies have demonstrated a correlation
between bladder cancer (de la Rosa et al., 2017; Lin et al.,
2018).
Several limitations need to be addressed. Although we
tested our hypothesis through comprehensive and thor-
ough analyses using large data sets, the study is limited
by its nature. While an ecological study provides many
useful insights into the plausibility of an association
between an exposure and outcome, inferences made
about correlations found in populations may not neces-
sarily hold true for individuals. Therefore, one must
exercise caution in extrapolating these relationships to
individual exposures. Another limitation was missing
data. When reviewing preliminary data worldwide, few
countries had regional or county-level data for both GBC
and As in groundwater. Given the limitations of surveil-
lance systems combined with the fact that GBC is a rare
disease, missing data was unavoidable in this study.
Moreover, because of limited data, our models did not
integrate all potential confounders such as obesity, eth-
nicity, lifestyle behaviors, socioeconomic status, gall-
stones as well as other heavy metals that may also
contaminate groundwater with high As concentration.
Also, GBC is a difficult diagnosis primarily relying on
sophisticated imaging techniques (computed tomography
and MRI). As a consequence, one can easily hypothesize
that GBC incidence is at risk to be underestimated,
particularly in areas where access to these technologies
remains limited. GBC data are vulnerable to mis-
classification for these reasons. Although we improved
our data sets by using histology confirmed data, the data
can only be as good as the surveillance system set up to
collect it. This supports our need for subject-level studies
to confirm this association.

Despite these limitations, this study provides supports to
the hypothesis that high-level exposure to As in drinking
water may be associated with GBC risk and paves the
way for individual-based studies designed to further
investigate this association.
Conclusion
The present findings provide the proof-of-concept and
pave the way for future observational and molecular
studies designed to evaluate and confirm the effect of As
exposure on GBC development. Insights into this rela-
tionship will further support the need to not only
appropriately regulate As the level in groundwater
worldwide but also to identify other sources of con-
tamination, including occupational exposure.
Acknowledgements
The authors thank Jill Gregory for the design of the
figures and the Taiwan Cancer Registry Center for
courteously providing county-level GBC data.
K.B. is supported by T32 HD049311-09 (NICHD) and
P30ES023515 (NIEHS). I.L. is supported by a grant from
the Swiss National Science Foundation, from Foundation
Roberto a nd Gianna Gonella and Foundation SICPA.
Authors’ contributions: Study concept and design: P.B.,
I.L. Acquisition of data: N.G., K.B. Analysis and inter-
pretation of data: N.G., K.B., P.B., I.L. Drafting of the
manuscript: N.G., K.B., I.L. Critical revision of the
manuscript for important intellectual content: N.G., K.B.,
P.B., I.L.
Conflicts of interest
P.B. has consulted with the Arsenic Science Task Force
on matters unrelated to the current study. For the
remaining authors, there are no conflicts of interest.
References
Aposhian HV (1997). Enzymatic methylation of arsenic species and other new
approaches to arsenic toxicity. Annu Rev Pharmacol Toxicol 37:397–419.
Are C, Ahmad H, Ravipati A, Croo D, Clarey D, Smith L, et al. (2017). Global
epidemiological trends and variations in the burden of gallbladder cancer.
J Surg Oncol 115:580–590.
Argos M, Kalra T, Rathouz PJ, Chen Y, Pierce B, Parvez F, et al. (2010). Arsenic
exposure from drinking water, and all-cause and chronic-disease mortalities in
Bangladesh (HEALS): a prospective cohort study. Lancet 376:252–258.
Bambino K, Chu J (2017). Zebrafish in toxicology and environmental health. Curr
Top Dev Biol 124:331–367.
Bray F, Colombet M, Mery L, Pineros M, Znaor A, Zanetti R, et al. (2017). Cancer
incidence in five continents, Vol XI (electronic version). Lyon: International
Agency for Research on Cancer.
Byron WR, Bierbower GW, Brouwer JB, Hansen WH (1967). Pathologic chan-
ges in rats and dogs from two-year feeding of sodium arsenite or sodium
arsenate. Toxicol Appl Pharmacol 10:132–147.
Centers for Disease Control and Prevention (CDC), Division of Diabetes
Translation (2016). All the United States, Obesity Prevalence, by county,
2004–2013. Available at: https://www.cdc.gov/diabetes/data/countydata/
countydataindicators.html. [Accessed 12 February 2018].
Central Ground Water Board (CGWB) (2015). Groundwater quality in shallow
aquifers of India. Available at: http://cgwb.gov.in/WQ/gw_quality_in_
shallow_aquifers.pdf. [Accessed 8 February 2018].
Chen Y, Wu F, Liu M, Parvez F, Slavkovich V, Eunus M, et al. (2013). A pro-
spective study of arsenic exposure, arsenic methylation capacity, and risk
Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Arsenic in gallbladder cancer Ganesan et al. 9
of cardiovascular disease in Bangladesh. Environ Health Perspect
121:832–838.
Chhabra D, Oda K, Jagannath P, Utsunomiya H, Takekoshi S, Nimura Y (2012).
Chronic heavy metal exposure and gallbladder cancer risk in India, a com-
parative study with Japan. Asian Pac J Cancer Prev 13:187–190.
Cockell KA, Bettger WJ (1993). Investigations of the gallbladder pathology
associated with dietary exposure to disodium arsenate heptahydrate in juve-
nile rainbow trout (Oncorhynchus mykiss). Toxicology 77:233–248.
Csanaky I, Gregus Z (2002). Species variations in the biliary and urinary excretion
of arsenate, arsenite and their metabolites. Comp Biochem Physiol Toxicol
Pharmacol 131:355–365.
Curado MP, Edwards B, Shin HR, Ferlay J, Heanue M, Boyle P, et al. (2007).
Cancer incidence in five continents, Vol IX. Lyon: IARC Scientific Publications
No. 160.
Das N, Giri A, Chakraborty S, Bhattacharjee P (2016). Association of single
nucleotide polymorphism with arsenic-induced skin lesions and genetic
damage in exposed population of West Bengal, India. Mutat Res 809:50–56.
de la Rosa R, Steinmaus C, Akers NK, Conde L, Ferreccio C, Kalman D, et al.
(2017). Associations between arsenic (+ 3 oxidation state) methyltransferase
(AS3MT) and N-6 adenine-specific DNA methyltransferase 1 (N6AMT1)
polymorphisms, arsenic metabolism, and cancer risk in a chilean population.
Environ Mol Mutagen 58:411–422.
Engstrom K, Vahter M, Mlakar SJ, Concha G, Nermell B, Raqib R, et al. (2011).
Polymorphisms in arsenic(+ III oxidation state) methyltransferase (AS3MT)
predict gene expression of AS3MT as well as arsenic metabolism. Environ
Health Perspect 119:182–188.
Engstrom KS, Hossain MB, Lauss M, Ahmed S, Raqib R, Vahter M, et al. (2013).
Efficient arsenic metabolism: the AS3MT haplotype is associated with DNA
methylation and expression of multiple genes around AS3MT. PLoS One 8:
e53732.
Ferlay J, Soerjomataram I, Ervik M, Dikshit R, Eser S, Mathers C, et al. (2013).
GLOBOCAN 2012 v10, cancer incidence and mortality worldwide. Lyon:
IARC CancerBase No. 11.
Forman D, Bray F, Brewster DH, Gombe Mbalawa C, Kohler B, Piñeros M, et al.
(2013). Cancer incidence in five continents. Lyon: IARC.
Frey MM, Edwards MA (1997). Surveying arsenic occurance. J Am Water Works
Assoc 89:105–117.
Giordano F, Dell’Orco V, Giannandrea F, Lauria L, Valente P, Figa-Talamanca I
(2006). Mortality in a cohort of pesticide applicators in an urban setting: sixty
years of follow-up. Int J Immunopathol Pharmacol 19 (Suppl):61–65.
Hopenhayn-Rich C, Biggs ML, Smith AH (1998). Lung and kidney cancer mor-
tality associated with arsenic in drinking water in Cordoba, Argentina. Int J
Epidemiol 27:561–569.
Huang MC, Douillet C, Su M, Zhou K, Wu T, Chen W, et al. (2017). Metabolomic
profiles of arsenic (+ 3 oxidation state) methyltransferase knockout mice:
effect of sex and arsenic exposure. Arch Toxicol 91:189–202.
Hundal R, Shaffer EA (2014). Gallbladder cancer: epidemiology and outcome.
Clin Epidemiol 6:99–109.
Jaffee EM, Dang CV, Agus DB, Alexander BM, Anderson KC, Ashworth A, et al.
(2017). Future cancer research priorities in the USA: a Lancet Oncology
Commission. Lancet Oncol 18:e653–e706.
Jaruvongvanich V, Yang JD, Peeraphatdit T, Roberts LR (2019). The incidence
rates and survival of gallbladder cancer in the USA. Eur J Cancer Prev
28:1–9.
Lamstein A (2019). Choroplethr. Lamstein Consulting. Available at: https://ari-
lamstein.com/open-source/. [Accessed 25 April 2019].
Landrigan PJ (1981). Arsenic: state of the art. Am J Ind Med 2:5–14.
Landrigan PJ, Fuller R, Acosta NJR, Adeyi O, Arnold R, Basu NN, et al. (2018).
The Lancet Commission on pollution and health. Lancet 391:462–512.
Larsson SC, Wolk A (2007). Obesity and the risk of gallbladder cancer: a meta-
analysis. Br J Cancer 96:1457–1461.
Lazcano-Ponce EC, Miquel JF, Munoz N, Herrero R, Ferrecio C, Wistuba II, et al.
(2001). Epidemiology and molecular pathology of gallbladder cancer. CA
Cancer J Clin 51:349–364.
Levy AD, Murakata LA, Rohrmann CA Jr (2001). Gallbladder carcinoma:
radiologic-pathologic correlation. Radiographics 21:295–314.
Lin YC, Chen WJ, Huang CY, Shiue HS, Su CT, Ao PL, et al. (2018).
Polymorphisms of arsenic (+ 3 oxidation state) methyltransferase and arsenic
methylation capacity affect the risk of bladder cancer. Toxicol Sci
164:328–338.
Madhawi R, Pandey A, Raj S, Mandal M, Devi S, Sinha PK, et al. (2018).
Geographical pattern of carcinoma gallbladder in Bihar and its association
with river Ganges and arsenic levels: Retrospective individual consecutive
patient data from Regional Cancer Centre. South Asian J Cancer 7:167–170.
 10 European Journal of Cancer Prevention 2019, Vol 00 No 00
Maupin MA, Kenny JF, Hutson SS, Lovelace JK, Barber NL, Linsey KS (2014).
Estimated use of water in the United States in 2010. Reston, VA: US
Geologican Survey.
Misra S, Chaturvedi A, Misra NC, Sharma ID (2003). Carcinoma of the gall-
bladder. Lancet Oncol 4:167–176.
National Cancer Registry Programme (NRCP) (2013). Population Based Cancer
Report, 2012-2014. Available at: http://www.ncdirindia.org/All_Reports/
PBCR_REPORT_2012_2014/index.htm. [Accessed 2 February 2018].
Pedlar RM, Klaverkamp JF (2002). Accumulation and distribution of dietary
arsenic in lake whitefish (Coregonus clupeaformis). Aquat Toxicol
57:153–166.
Pedlar RM, Ptashynski MD, Evans R, Klaverkamp JF (2002). Toxicological effects
of dietary arsenic exposure in lake whitefish (Coregonus clupeaformis). Aquat
Toxicol 57:167–189.
Ponomarenko O, La Porte PF, Singh SP, Langan G, Fleming DEB, Spallholz JE,
et al. (2017). Selenium-mediated arsenic excretion in mammals: a
synchrotron-based study of whole-body distribution and tissue-specific
chemistry. Metallomics 9:1585–1595.
Randi G, Franceschi S, La Vecchia C (2006). Gallbladder cancer worldwide:
geographical distribution and risk factors. Int J Cancer 118:1591–1602.
Ravenscroft P, Brammer H, Richards K (2009). Arsenic pollution: a global
synthesis. Chichester: Wiley-Blackwell.
Rosenberg MJ, Landrigan PJ, Crowley S (1980). Low-level arsenic exposure in
wood processing plants. Am J Ind Med 1:99–107.
Saranga Bharathi R, Singh R, Gupta R, Verma GR, Kalra N, Kiran K, et al. (2015).
Female sex hormone receptors in gallbladder cancer. J Gastrointest Cancer
46:143–148.
Sawyers CL, Abate-Shen C, Anderson KC, Barker A, Baselga J, Berger NA, et al.
(2013). AACR cancer progress report 2013. Clin Cancer Res 19 (Suppl):
S4–S98.
Shukla VK, Chauhan VS, Mishra RN, Basu S (2008). Lifestyle, reproductive fac-
tors and risk of gallbladder cancer. Singapore Med J 49:912–915.
Singh AK (2004). Arsenic contamination in groundwater of North Eastern India
National seminar on hydrology. Roorkee: National Institute of Hydrology.
Copyright r 2019 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Spain D (1984). Taiwan: country profile. Int Demogr 3:4–8.
Straif K, Benbrahim-Tallaa L, Baan R, Grosse Y, Secretan B, El Ghissassi F, et al.
(2009). A review of human carcinogens: Part C: metals, arsenic, dusts,
and fibres. Lancet Oncol 10:453–454.
Styblo M, Del Razo LM, Vega L, Germolec DR, LeCluyse EL, Hamilton GA, et al.
(2000). Comparative toxicity of trivalent and pentavalent inorganic and
methylated arsenicals in rat and human cells. Arch Toxicol 74:289–299.
Surveillance Epidemiology and End Results (SEER) Program (2018). SEER is an
authoritative source for cancer statistics in the United States. Available at:
http://www.seer.cancer.gov. [Accessed: 13 November 2017].
Taiwan Environmental Protection Agency (2010). Environmental water quality
information, 1993–2007 Taipei: Environmental Protection Agency, R.O.C.
(Taiwan). Available at: https://wq.epa.gov.tw/Code/?Languages=en.
[Accessed 8 February 2018].
Thomas DJ, Waters SB, Styblo M (2004). Elucidating the pathway for arsenic
methylation. Toxicol Appl Pharmacol 198:319–326.
Tokar EJ, Diwan BA, Waalkes MP (2012). Renal, hepatic, pulmonary and adrenal
tumors induced by prenatal inorganic arsenic followed by dimethylarsinic acid
in adulthood in CD1 mice. Toxicol Lett 209:179–185.
Tseng CH (2007). Arsenic methylation, urinary arsenic metabolites and human
diseases: current perspective. J Environ Sci Health C Environ Carcinog
Ecotoxicol Rev 25:1–22.
United States Geological Survey (USGS) (2001). Map of Arsenic concentrations
in groundwater of the United States. Available at: https://water.usgs.gov/GIS/
browse/arsenic_map.png. [Accessed 19 February 2018].
Wistuba II, Gazdar AF (2004). Gallbladder cancer: lessons from a rare tumour.
Nat Rev Cancer 4:695–706.
Wu R, Wu X, Wang H, Fang X, Li Y, Gao L, et al. (2017). Strain differences in
arsenic-induced oxidative lesion via arsenic biomethylation between C57BL/
6J and 129X1/SvJ mice. Sci Rep 7:44424.
Xi S, Zheng Q, Zhang Q, Sun G (2011). Metabolic profile and assessment of
occupational arsenic exposure in copper- and steel-smelting workers
in China. Int Arch Occup Environ Health 84:347–353.