Professional Documents
Culture Documents
1
IARC Monographs 112
amine salts are readily soluble in water 1.2 Production and use
(Tomlin, 2000).
Volatility: Vapour pressure, 1.31102 mPa at 1.2.1 Production
25 C (negligible) (Tomlin, 2000). (a) Manufacturing processes
Stability: Glyphosate is stable to hydrolysis Glyphosate was first synthesized in 1950 as
in the range of pH3 to pH9, and relatively a potential pharmaceutical compound, but its
stable to photodegradation (Tomlin, 2000). herbicidal activity was not discovered until it
Glyphosate is not readily hydrolysed or was re-synthesized and tested in 1970 (Szkcs
oxidized in the field (Rueppel et al. 1977). & Darvas, 2012). The isopropylamine, sodium,
It decomposes on heating, producing toxic and ammonium salts were introduced in 1974,
fumes that include nitrogen oxides and phos- and the trimesium (trimethylsulfonium) salt was
phorus oxides (IPCS, 2005). introduced in Spain in 1989. The original patent
Reactivity: Attacks iron and galvanized steel protection expired outside the USA in 1991, and
(IPCS, 2005). within the USA in 2000. Thereafter, production
Octanol/water partition coefficient (P): log expanded to other major agrochemical manu-
P, <3.2 (pH25, 20C) (OECD method 107) facturers in the USA, Europe, Australia, and
(Tomlin, 2000). elsewhere (including large-scale production in
Henrys law: <2.1107 Pa m3 mol1 (Tomlin, China), but the leading preparation producer
2000). remained in the USA (Szkcs & Darvas, 2012).
There are two dominant families of commer-
Conversion factor: Assuming normal temper-
cial production of glyphosate, the alkyl ester
ature (25 C) and pressure (101 kPa), mg/m3
pathways, predominant in China, and the
=6.92ppm.
iminodiacetic acid pathways, with imino-
diacetic acid produced from iminodiacetonitrile
1.1.4 Technical products and impurities (produced from hydrogen cyanide), diethanol
amine, or chloroacetic acid (Dill et al., 2010; Tian
Glyphosate is formulated as an isopropyl
et al., 2012).
amine, ammonium, or sodium salt in water-
To increase the solubility of technical-grade
soluble concentrates and water-soluble gran-
glyphosate acid in water, it is formulated as its
ules. The relevant impurities in glyphosate technical
isopropylamine, monoammonium, potassium,
concentrates are formaldehyde (maximum, 1.3g/kg),
sodium, or trimesium salts. Most common
N-nitrosoglyphosate (maximum, 1mg/kg), and N-
is the isopropylamine salt, which is formu-
nitroso-N-phosphonomethylglycine (FAO, 2000).
lated as a liquid concentrate (active ingredient,
Surfactants and sulfuric and phosphoric acids
5.062%), ready-to-use liquid (active ingredient,
may be added to formulations of glyphosate, with
0.520%), pressurized liquid (active ingredient,
type and concentration differing by formulation
0.750.96%), solid (active ingredient, 7694%),
(IPCS, 1994).
or pellet/tablet (active ingredient, 6083%) (EPA,
1993a).
There are reportedly more than 750 products
containing glyphosate for sale in the USA alone
(NPIC, 2010). Formulated products contain
various non-ionic surfactants, most notably
polyethyloxylated tallowamine (POEA), to
2
Glyphosate
facilitate uptake by plants (Szkcs & Darvas, and reportedly has sufficient production capacity
2012). Formulations might contain other active to satisfy total global demand (Yin, 2011).
ingredients, such as simasine, 2,4-dichlorophen
oxyacetic acid (2,4-D), or 4-chloro-2-methyl- 1.2.2 Uses
phenoxyacetic acid (IPCS, 1996), with herbicide
resistance driving demand for new herbicide Glyphosate is a broad-spectrum, post-emergent,
formulations containing multiple active ingredi- non-selective, systemic herbicide, which effectively
ents (Freedonia, 2012). kills or suppresses all plant types, including grasses,
perennials, vines, shrubs, and trees. When applied
(b) Production volume at lower rates, glyphosate is a plant-growth regulator
and desiccant. It has agricultural and non-agricul-
Glyphosate is reported to be manufactured
tural uses throughout the world.
by at least 91 producers in 20 countries, including
53 in China, 9 in India, 5 in the USA, and others (a) Agriculture
in Australia, Canada, Cyprus, Egypt, Germany,
Guatemala, Hungary, Israel, Malaysia, Mexico, Glyphosate is effective against more than 100
Singapore, Spain, Taiwan (China), Thailand, annual broadleaf weed and grass species, and
Turkey, the United Kingdom, and Venezuela more than 60 perennial weed species (Dill et al.,
(Farm Chemicals International, 2015). Glyph 2010). Application rates are about 1.52 kg/ha
osate was registered in over 130 countries as of for pre-harvest, post-planting, and pre-emer-
2010 and is probably the most heavily used herbi- gence use; about 4.3kg/ha as a directed spray in
cide in the world, with an annual global produc- vines, orchards, pastures, forestry, and industrial
tion volume estimated at approximately 600000 weed control; and about 2 kg/ha as an aquatic
tonnes in 2008, rising to about 650000 tonnes in herbicide (Tomlin, 2000). Common application
2011, and to 720000 tonnes in 2012 (Dill et al., methods include broadcast, aerial, spot, and
2010; CCM International, 2011; Hilton, 2012; directed spray applications (EPA, 1993a).
Transparency Market Research, 2014). Due to its broad-spectrum activity, the
Production and use of glyphosate have risen use of glyphosate in agriculture was formerly
dramatically due to the expiry of patent protec- limited to post-harvest treatments and weed
tion (see above), with increased promotion of control between established rows of tree, nut,
non-till agriculture, and with the introduction and vine crops. Widespread adoption of no-till
in 1996 of genetically modified glyphosate-tol- and conservation-till practices (which require
erant crop varieties (Szkcs & Darvas, 2012). chemical weed control while reducing soil
In the USA alone, more than 80 000 tonnes of erosion and labour and fuel costs) and the intro-
glyphosate were used in 2007 (rising from less duction of transgenic crop varieties engineered
than 4000 tonnes in 1987) (EPA, 1997, 2011). to be resistant to glyphosate have transformed
This rapid growth rate was also observed in glyphosate to a post-emergent, selective herbi-
Asia, which accounted for 30% of world demand cide for use on annual crops (Duke & Powles,
for glyphosate in 2012 (Transparency Market 2009; Dill et al. 2010). Glyphosate-resistant
Research, 2014). In India, production increased transgenic varieties have been widely adopted
from 308 tonnes in 20032004, to 2100 tonnes in for the production of corn, cotton, canola, and
20072008 (Ministry of Chemicals & Fertilizers, soybean (Duke & Powles, 2009). Production
2008). China currently produces more than of such crops accounted for 45% of worldwide
40% of the global supply of glyphosate, exports demand for glyphosate in 2012 (Transparency
almost 35% of the global supply (Hilton, 2012), Market Research, 2014). However, in Europe,
3
IARC Monographs 112
where the planting of genetically modified crops during non-crop use was identified as requiring
has been largely restricted, post-harvest treat- particular attention in the short term (European
ment is still the most common application of Commission, 2002).
glyphosate (Glyphosate Task Force, 2014). Intense Nevertheless, as worldwide rates of adoption
and continuous use of glyphosate has led to the of herbicide-resistant crops and of glyphosate use
emergence of resistant weeds that may reduce its have risen in recent years (Duke & Powles, 2009),
effectiveness (Duke & Powles, 2009). restriction of glyphosate use has been enacted or
proposed in several countries, although docu-
(b) Residential use mented actions are few. In 2013, the Legislative
Glyphosate is widely used for household Assembly of El Salvador voted a ban on the use of
weed control throughout the world. In the USA, pesticides containing glyphosate (Repblica de
glyphosate was consistently ranked as the second El Salvador, 2013). Sri Lanka is reported to have
most commonly used pesticide (after 2,4-D) in instituted a partial ban based on an increasing
the home and garden market sector between number of cases of chronic kidney disease among
2001 and 2007, with an annual use of 20004000 agricultural workers, but the ban was lifted after
tonnes (EPA, 2011). 2months (ColomboPage, 2014). The reasons for
such actions have included the development of
(c) Other uses resistance among weed species, as well as health
Glyphosate was initially used to control concerns.
perennial weeds on ditch banks and roadsides No limits for occupational exposure were
and under power lines (Dill et al., 2010). It is also identified by the Working Group.
used to control invasive species in aquatic or
wetland systems (Tu et al., 2001). Approximately 1.3 Measurement and analysis
12% of total glyphosate use in the USA is in
forest management (Mance, 2012). Several methods exist for the measurement of
Glyphosate has been used in a large-scale glyphosate and its major metabolite aminomethyl
aerial herbicide-spraying programme begun phosphonic acid (AMPA) in various media,
in 2000 to reduce the production of cocaine in including air, water, urine, and serum (Table1.1).
Colombia (Lubick, 2009), and of marijuana in The methods largely involve derivatization with
Mexico and South America (Szkcs & Darvas, 9-fluorenylmethyl chloro formate (FMOC-Cl)
2012). to reach sufficient retention in chromatographic
columns (Kuang et al., 2011; Botero-Coy et al.,
(d) Regulation 2013). Chromatographic techniques that do not
Glyphosate has been registered for use in require derivatization and enzyme-linked immuno-
at least 130 countries (Dill et al., 2010). In the sorbent assays (ELISA) are under development
USA, all uses are eligible for registration on the (Sanchs et al., 2012).
basis of a finding that glyphosate does not pose
unreasonable risks or adverse effects to humans
or the environment (EPA, 1993a). A review
conducted in 2001 in connection with the regis-
tration process in the European Union reached
similar conclusions regarding animal and human
safety, although the protection of groundwater
4
Glyphosate
1.4 Occurrence and exposure farming families (Acquavella et al., 2004; Curwin
et al., 2007). These studies are summarized in
1.4.1 Exposure Table1.2.
(a) Occupational exposure (b) Community exposure
Studies related to occupational exposure Glyphosate can be found in soil, air, surface
to glyphosate have included farmers and tree water, and groundwater (EPA, 1993a). Once in
nursery workers in the USA, forestry workers in the environment, glyphosate is adsorbed to soil
Canada and Finland, and municipal weed-con- and is broken down by soil microbes to AMPA
trol workers in the United Kingdom (Centre de (Borggaard & Gimsing, 2008). In surface water,
Toxicologie du Qubec, 1988; Jauhiainen et al., glyphosate is not readily broken down by water
1991; Lavy et al., 1992; Acquavella et al., 2004; or sunlight (EPA, 1993a). Despite extensive
Johnson et al., 2005). Para-occupational expo- worldwide use, there are relatively few studies
sures to glyphosate have also been measured in
5
6
Table 1.2 Occupational and para-occupational exposure to glyphosate
7
Glyphosate
IARC Monographs 112
on the environmental occurrence of glyphosate in 9.5% of samples tested from member states
(Kolpin et al., 2006). of the European Union, and from Norway and
Iceland in 2007 (Granby & Vahl, 2001; EFSA,
(i) Air
2009). In the United Kingdom, food sampling
Very few studies of glyphosate in air were for glyphosate residues has concentrated mainly
available to the Working Group. Air and rain- on cereals, including bread and flour. Glyphosate
water samples were collected during two has been detected regularly and usually below the
growing seasons in agricultural areas in Indiana, reporting limit (Pesticide Residues Committee,
Mississippi, and Iowa, USA (Chang et al., 2011). 2007, 2008, 2009, 2010). Six out of eight samples
The frequency of glyphosate detection ranged of tofu made from Brazilian soy contained
from 60% to 100% in air and rain samples, and glyphosate, with the highest level registered
concentrations ranged from <0.01 to 9.1 ng/m3 being 1.1 mg/kg (Pesticide Residues Committee,
in air samples and from < 0.1 to 2.5 g/L in 2007).
rainwater samples. Atmospheric deposition
was measured at three sites in Alberta, Canada. (iv) Household exposure
Rainfall and particulate matter were collected In a survey of 246 California households,
as total deposition at 7-day intervals throughout 14% were found to possess at least one product
the growing season. Glyphosate deposition containing glyphosate (Guha et al., 2013).
rates ranged from <0.01 to 1.51 g/m2 per day
(v) Biological markers
(Humphries et al., 2005).
No data were available to the Working Group Glyphosate concentrations in urine were
regarding glyphosate concentrations in indoor analysed in urban populations in Europe, and
air. in a rural population living near areas sprayed
for drug eradication in Colombia (MLHB, 2013;
(ii) Water Varona et al., 2009). Glyphosate concentrations
Glyphosate in the soil can leach into ground- in Colombia were considerably higher than in
water, although the rate of leaching is believed to Europe, with means of 7.6 g/L and 0.02 g/L,
be low (Borggaard & Gimsing, 2008; Simonsen respectively (Table 1.5). In a study in Canada,
et al., 2008). It can also reach surface waters by glyphosate concentrations in serum ranged from
direct emission, atmospheric deposition, and by undetectable to 93.6 g/mL in non-pregnant
adsorption to soil particles suspended in runoff women (n=39), and were undetectable in serum
water (EPA, 1993a; Humphries et al., 2005). of pregnant women (n=30) and fetal cord serum
Table 1.3 summarizes data on concentrations (Aris & Leblanc, 2011).
of glyphosate or AMPA in surface water and
groundwater. 1.4.2 Exposure assessment
(iii) Residues in food and dietary intake Exposure assessment methods in epidemio
Glyphosate residues have been measured logical studies on glyphosate and cancer are
in cereals, fruits, and vegetables (Table 1.4). discussed in Section 2.0 of the Monograph on
Residues were detected in 0.04% of 74 305 Malathion, in the present volume.
samples of fruits, vegetables, and cereals tested
from 27 member states of the European Union,
and from Norway, and Iceland in 2007 (EFSA,
2009). In cereals, residues were detected in 50%
of samples tested in Denmark in 19981999, and
8
Table 1.3 Concentration of glyphosate and AMPA in water
9
Glyphosate
10
Table 1.4 Concentrations of glyphosate in food
Australia, 2006 Composite sample of foods 75% of samples had detectable 20 total samples from 43 McQueen et al. (2012)
consumed in 24 hours residues pregnant women
Mean, 0.08 mg/kg
Range, <0.005 to 0.5 mg/kg
Table 1.5 Concentrations of glyphosate and AMPA in urine and serum in the general population
11
12
Table 2.1 Cohort studies of cancer and exposure to glyphosate
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
study location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period/follow-
up, study-design
De Roos et al. 54315 (after exclusions, from a total Lung Ever use NR 0.9 (0.61.3) Age, smoking, AHS
(2005a) cohort of 57311) licensed pesticide Cumulative other Cancer sites
Iowa and North applicators exposure pesticides, investigated: lung,
Carolina, USA Exposure assessment method: alcohol
IARC Monographs 112
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
study location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period/follow-
up, study-design
Flower et al. 21375; children (aged <19 years) Childhood Maternal 13 0.61 Childs age at AHS
(2004) of licensed pesticide applicators in cancer use of (0.321.16) enrolment Glyphosate results relate
Iowa and North Iowa (n=17357) and North Carolina glyphosate to the Iowa participants
Carolina, USA (n=4018) (ever) only
Enrolment, Exposure assessment method: Paternal 6 0.84 [Strengths: Large cohort;
19931997; questionnaire use of (0.352.34) specific assessment of
follow-up, glyphosate glyphosate. Limitations:
19751998 (prenatal) based on self-reported
exposure; potential
exposure to multiple
pesticides; limited
power for glyphosate
exposure]
Engel et al. 30454 wives of licensed pesticide Breast Direct 82 0.9 (0.71.1) Age, race, state AHS
(2005) applicators with no history of breast exposure to [Strengths: large cohort;
Iowa and North cancer at enrolment glyphosate specific assessment of
Carolina, USA Exposure assessment method: Husbands 109 1.3 (0.81.9) glyphosate. Limitations:
Enrolment, questionnaire use of based on self-reported
19931997 glyphosate exposure; limited to
follow-up to licensed applicators;
2000 potential exposure to
multiple pesticides]
Lee et al. (2007) 56813 licensed pesticide applicators Colorectum Exposed to 225 1.2 (0.91.6) Age, smoking, AHS
Iowa and North Exposure assessment method: glyphosate state, total [Strengths: large cohort.
Carolina, USA questionnaire Colon Exposed to 151 days of any Limitations: based on
Enrolment, glyphosate pesticide self-reported exposure,
19931997; Rectum Exposed to 74 application limited to licensed
follow-up to 2002 glyphosate applicators, potential
exposure to multiple
pesticides]
13
Glyphosate
14
Table 2.1 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
study location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period/follow-
up, study-design
Andreotti et al. Cases: 93 (response rate, NR); identified Pancreas Ever 55 1.1 (0.61.7) Age, smoking, AHS
(2009) from population-based state-cancer (C25.0 exposure to diabetes [Strengths: large cohort.
Iowa and North registries. Incident cases diagnosed C25.9) glyphosate Limitations: based on
IARC Monographs 112
(De Roos et al., 2005b). [The study had limited and other factors was obtained at enrolment by
power for the analysis of multiple myeloma; there self-administered questionnaire from the women
were missing data on covariates when multiple and their husbands. A total of 309 incident cases
adjustments were done, limiting the interpreta- of cancer of the breast were identified until 2000.
tion of the findings.] A re-analysis of these data There was no difference in incidence of cancer of
conducted by Sorahan (2015) confirmed that the the breast for women who reported ever applying
excess risk of multiple myeloma was present only pesticides compared with the general popula-
in the subset with no missing information (of 22 tion. The relative risk for cancer of the breast
cases in the restricted data set). In a subsequent among women who had personally used glypho-
cross-sectional analysis of 678 male participants sate was 0.9 (95% CI, 0.71.1; 82 cases) and 1.3
from the same cohort, Landgren et al. (2009) (95% CI, 0.81.9; 109 cases) among women who
did not find an association between exposure to never used pesticides but whose husband had
glyphosate and risk of monoclonal gammopathy used glyphosate. [No information on duration of
of undetermined significance (MGUS), a prema- glyphosate use by the husband was presented.]
lignant plasma disorder that often precedes Results for glyphosate were not further stratified
multiple myeloma (odds ratio, OR, 0.5; 95% CI, by menopausal status.
0.21.0; 27 exposed cases). Lee et al. (2007) investigated the relation-
Flower et al. (2004) reported the results of the ship between exposure to agricultural pesticides
analyses of risk of childhood cancer associated and incidence of cancer of the colorectum in
with pesticide application by parents in the AHS. the AHS. A total of 56813 pesticide applicators
The analyses for glyphosate were conducted with no prior history of cancer of the colorectum
among 17357 children of Iowa pesticide appli- were included in this analysis, and 305 incident
cators from the AHS. Parents provided data cancers of the colorectum (colon, 212; rectum,
via questionnaires (19931997) and the cancer 93) were diagnosed during the study period,
follow-up (retrospectively and prospectively) 19932002. Most of the 50 pesticides studied
was done through the state cancer registries. were not associated with risk of cancer of the
Fifty incident childhood cancers were identi- colorectum, and the relative risks with expo-
fied (19751998; age, 019 years). For all the sure to glyphosate were 1.2 (95% CI, 0.91.6), 1.0
children of the pesticide applicators, risk was (95% CI, 0.71.5), and 1.6 (95% CI, 0.92.9) for
increased for all childhood cancers combined, cancers of the colorectum, colon, and rectum,
for all lymphomas combined, and for Hodgkin respectively.
lymphoma, compared with the general popula- Andreotti et al. (2009) examined associations
tion. The odds ratio for use of glyphosate and risk between the use of pesticides and cancer of the
of childhood cancer was 0.61 (95% CI, 0.321.16; pancreas using a casecontrol analysis nested
13 exposed cases) for maternal use and 0.84 (95% in the AHS. This analysis included 93 incident
CI, 0.352.34; 6 exposed cases) for paternal use. cases of cancer of the pancreas (64 applicators,
[The Working Group noted that this analysis 29 spouses) and 82503 cancer-free controls who
had limited power to study a rare disease such as completed the enrolment questionnaire. Ever-use
childhood cancer.] of 24 pesticides and intensity-weighted life-
Engel et al. (2005) reported on incidence of time days [(lifetime exposure days)(exposure
cancer of the breast among farmers wives in the intensity score)] of 13 pesticides were assessed.
AHS cohort, which included 30454 women with Risk estimates were calculated controlling for
no history of cancer of the breast before enrol- age, smoking, and diabetes. The odds ratio for
ment in 19931997. Information on pesticide use ever- versus never-exposure to glyphosate was
15
IARC Monographs 112
1.1 (95% CI, 0.61.7; 55 exposed cases), while Malathion, Section 2.0, for a detailed description
the odds ratio for the highest category of level of of this study). A population-based casecontrol
intensity-weighted lifetime days was 1.2 (95% CI, study of 173 white men with multiple myeloma
0.62.6; 19 exposed cases). and 650 controls was conducted in Iowa, USA, an
Dennis et al. (2010) reported that exposure area with a large farming population. A non-sig-
to glyphosate was not associated with cutaneous nificantly elevated risk of multiple myeloma
melanoma within the AHS. [The authors did not was seen among farmers compared with never-
report a risk estimate.] farmers. The odds ratio related to exposure to
glyphosate was 1.7 (95% CI, 0.83.6; 11 exposed
cases). [This study had limited power to assess
2.2 Casecontrol studies on non- the association between multiple myeloma and
Hodgkin lymphoma, multiple exposure to glyphosate. Multiple myeloma is
myeloma, and leukaemia now considered to be a subtype of NHL.]
De Roos et al. (2003) used pooled data from
2.2.1 Non-Hodgkin lymphoma three casecontrol studies of NHL conducted in
See Table2.2 the 1980s in Nebraska (Zahm et al., 1990), Kansas
(Hoar et al., 1986), and in Iowa and Minnesota
(a) Casecontrol studies in the midwest USA (Cantor et al., 1992) (see the Monograph on
Cantor et al. (1992) conducted a casecontrol Malathion, Section 2.0, for a detailed description
study of incident non-Hodgkin lymphoma (NHL) of these studies) to examine pesticide exposures in
among males in Iowa and Minnesota, USA (see farming as risk factors for NHL in men. The study
the Monograph on Malathion, Section 2.0, for a population included 870 cases and 2569 controls;
detailed description of this study). A total of 622 650 cases and 1933 controls were included for the
white men and 1245 population-based controls analysis of 47 pesticides controlling for potential
were interviewed in person. The association with confounding by other pesticides. Both logistic
farming occupation and specific agricultural regression and hierarchical regression (adjusted
exposures were evaluated. When compared with estimates were based on prior distributions
non-farmers, the odds ratios for NHL were 1.2 for the pesticide effects, which provides more
(95% CI, 1.01.5) for men who had ever farmed, conservative estimates than logistic regression)
and 1.1 (95% CI, 0.71.9; 26 exposed cases; adjusted were used in data analysis, and all models were
for vital status, age, state, cigarette smoking essentially adjusted for age, study site, and other
status, family history of lymphohaematopoietic pesticides. Reported use of glyphosate as well
cancer, high-risk occupations, and high-risk as several individual pesticides was associated
exposures) for ever handling glyphosate. [There with increased incidence of NHL. Based on 36
was low power to assess the risk of NHL associ- cases exposed, the odds ratios for the association
ated with exposure to glyphosate. There was no between exposure to glyphosate and NHL were
adjustment for other pesticides. These data were 2.1 (95% CI, 1.14.0) in the logistic regression
included in the pooled analysis by De Roos et al. analyses and 1.6 (95% CI, 0.92.8) in the hier-
(2003).] archical regression analysis. [The numbers of
Brown et al. (1993) reported the results of cases and controls were lower than those in the
a study to evaluate the association between pooled analysis by Waddell et al. (2001) because
multiple myeloma and agricultural risk factors only subjects with no missing data on pesticides
in the midwest USA (see the Monograph on were included. The strengths of this study when
compared with other studies are that it was large,
16
Table 2.2 Casecontrol studies of leukaemia and lymphoma and exposure to glyphosate
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
USA
Brown et al. Cases: 578 (340 living, 238 Leukaemia Any 15 0.9 (0.51.6) Age, vital status, [Strengths: large
(1990) deceased) (response rate, 86%); glyphosate state, tobacco use, population based
Iowa and cancer registry or hospital family history study in a farming
Minnesota, USA records lymphopoietic area.
19811983 Controls: 1245 (820 living, cancer, high-risk Limitations: not
425 deceased) (response rate, occupations, high controlled for
7779%); random-digit dialling risk exposures exposure to other
for those aged < 65 years and pesticides. Limited
Medicare for those aged 65 power for glyphosate
years exposure]
Exposure assessment method:
questionnaire
Cantor et al. Cases: 622 (response rate, 89.0%); NHL Ever handled 26 1.1 (0.71.9) Age, vital Data subsequentially
(1992) Iowa health registry records glyphosate status, state, pooled in De Roos
Iowa and and Minnesota hospital and smoking status, et al. (2003); white
Minnesota, USA pathology records family history men only
19801982 Controls: 1245 (response rate, lymphopoietic [Strengths: large
7679%); population-based; cancer, high-risk population-based
no cancer of the lympho- occupations, study in farming
haematopoietic system; high-risk areas.
frequency-matched to cases by exposures Limitations: not
age (5-year group), vital status, controlled for
state. Random-digit dialling exposure to other
(aged <65 years); Medicare pesticides. Limited
records (aged 65 years); state power for glyphosate
death certificate files (deceased exposure]
subjects)
Exposure assessment method:
questionnaire; in-person
interview
17
Glyphosate
18
Table 2.2 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Brown et al. Cases: 173 (response rate, 84%); Multiple Any 11 1.7 (0.83.6) Age, vital status [Strengths:
(1993) Iowa health registry myeloma glyphosate population-based
Iowa, USA Controls: 650 (response rate, study. Areas with high
19811984 78%); Random-digit dialling prevalence of farming.
IARC Monographs 112
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Lee et al. (2004a) Cases: 872 (response rate, NR); NHL Exposed to 53 1.4 (0.982.1) Age, vital status, 177 participants
Iowa, Minnesota diagnosed with NHL from 1980 glyphosate state (45 NHL cases, 132
and Nebraska, to 1986 non- controls) reported
USA Controls: 2381 (response rate, asthmatics having been told by
19801986 NR); frequency-matched Exposed to 6 1.2 (0.43.3) their doctor that they
controls glyphosate had asthma
Exposure assessment method: asthmatics
questionnaire; information on
use of pesticides and history of
asthma was based on interviews
Canada
McDuffie et al. Cases: 517 (response rate, 67.1%), NHL Exposed to 51 1.2 (0.831.74) Age, province of Cross-Canada study
(2001) from cancer registries and glyphosate residence [Strengths: large
Canada hospitals population based
19911994 Controls: 1506 (response rate, study. Limitations:
48%); random sample from Unexposed 464 1 no quantitative
health insurance and voting >0 and 2 28 1.0 (0.631.57) exposure data.
records days Exposure assessment
Exposure assessment by questionnaire.
>2 days 23 2.12 (1.23.73)
method: questionnaire, some Relatively low
administered by telephone, some participation]
by post
19
Glyphosate
20
Table 2.2 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Karunanayake Incident cases: 316 (response HL (ICDO2 Glyphosate- 38 1.14 (0.741.76) Age group, Cross Canada study
et al. (2012) rate, 68.4%); men aged 19 years; included based province of Based on the statistical
Six provinces ascertained from provincial nodular formulation residence analysis of pilot study
in Canada cancer registries, except in sclerosis Glyphosate- 38 0.99 (0.621.56) Age group, data, it was decided
IARC Monographs 112
(Quebec, Ontario, Quebec (hospital ascertainment) (M9656/3; based province of that the most efficient
Manitoba, Controls: 1506 (response rate, M9663/3; formulation residence, medical definition of pesticide
Saskatchewan, 48%); matched by age2years M9664/3; history exposure was a
Alberta, and to be comparable with the age M9665/3; cumulative exposure
British Columbia) distribution of the entire case M9666/3; 10 hours/year to
19911994 group (HL, NHL, MM, and M9667/3), any combination
STS) within each province of lymphocytic of pesticides. This
residence. Potential controls predominance discriminated (a)
(men aged 19 years) selected at (M9651/3; between incidental,
random within age constraints M9657/3; bystander, and
from the provincial health M9658/3; environmental
insurance records (Alberta, M9659/3), exposure vs more
Saskatchewan, Manitoba, mixed intensive exposure,
Quebec), computerized cellularity and (b) between cases
telephone listings (Ontario), or (M9652/3), and controls
voters lists (British Columbia) lymphocytic [Strengths: large study.
Exposure assessment method: depletion Limitations: low
questionnaire; stage 1 used (M9653/3; response rates]
a self-administered postal M9654/3),
questionnaire; and in stage 2 miscellaneous
detailed pesticide exposure (other
information was collected by M9650-M9669
telephone interview codes for HL)
Table 2.2 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Kachuri et al. Cases: 342 (response rate, 58%); Multiple Glyphosate 32 1.19 (0.761.87) Age, province of Cross-Canada study
(2013) men aged 19 years diagnosed myeloma use residence, use of a [Strengths:
Six Canadian between 1991 and 1994 were Use of 15 0.72 (0.391.32) proxy respondent, population-based
provinces (British ascertained from provincial glyphosate smoking status, casecontrol study.
Columbia, cancer registries except in (>0 and medical variables, Limitations: relatively
Alberta, Quebec, where ascertained from 2days per family history of low response rates]
Saskatchewan, hospitals year) cancer
Manitoba, Controls: 1357 (response rate, Use of 12 2.04 (0.984.23)
Ontario and 48%); men aged 19 years glyphosate
Quebec) selected randomly using (>2days per
19911994 provincial health insurance year)
records, random digit dialling,
or voters lists, frequency-
matched to cases by age
(2years) and province of
residence
Exposure assessment method:
questionnaire
Sweden
Nordstrm et al. Cases: 111 (response rate, 91%); HCL Exposed to 4 3.1 (0.812) Age Overlaps with Hardell
(1998) 121 HCL cases in men identified glyphosate et al. (2002). HCL is a
Sweden from Swedish cancer registry subtype of NHL
19871992 Controls: 400 (response rate, [Strengths:
83%); 484 (four controls/case) population-based
matched for age and county; casecontrol study.
national population registry Limitations: Limited
Exposure assessment method: power. There was no
questionnaire; considered adjustment for other
exposed if minimum exposure exposures]
of 1 working day (8 h) and an
induction period of at least
1year
21
Glyphosate
22
Table 2.2 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Hardell & Cases: 404 (192 deceased) NHL (ICD-9 Ever 4 2.3 (0.413) Not specified in Overlaps with Hardell
Eriksson (1999) (response rate, 91%); regional 200 and 202) glyphosate the multivariable et al. (2002)
Northern and cancer registries univariate analysis [Strengths:
middle Sweden Controls: 741 (response rate, Ever NR 5.8 (0.654) population-based
IARC Monographs 112
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Eriksson et al. Cases: 910 (response rate, NHL Any 29 2.02 (1.13.71) Age, sex, year of [Strengths:
(2008) 91%); incident NHL cases glyphosate enrolment population-based
Sweden. Four were enrolled from university Any 29 1.51 (0.772.94) case-control.
health service hospitals glyphosate* Limitations: limited
areas (Lund, Controls: 1016 (response rate, power for glyphosate]
Linkoping, 92%); national population * Exposure to other
Orebro and registry 10 days per 12 1.69 (0.74.07) pesticides (e.g. MPCA)
Umea) Exposure assessment method: year use controlled in the
19992002 questionnaire >10 days per 17 2.36 (1.045.37) analysis
year use
NHL 110 yrs NR 1.11 (0.245.08)
>10 yrs NR 2.26 (1.164.4)
B-cell Exposure to NR 1.87 (0.9983.51)
lymphoma glyphosate
Lymphocytic Exposure to NR 3.35 (1.427.89)
lymphoma/B- glyphosate
CLL
Diffuse Exposure to NR 1.22 (0.443.35)
large B-cell glyphosate
lymphoma
Follicular, Exposure to NR 1.89 (0.625.79)
grade IIII glyphosate
Other Exposure to NR 1.63 (0.534.96)
specified B-cell glyphosate
lymphoma
Unspecified Exposure to NR 1.47 (0.336.61)
B-cell glyphosate
lymphoma
T-cell Exposure to NR 2.29 (0.5110.4)
lymphoma glyphosate
Unspecified Exposure to NR 5.63 (1.4422)
NHL glyphosate
23
Glyphosate
24
Table 2.2 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Other studies in Europe
Orsi et al. (2009) Cases: 491 (response rate, 95.7%); NHL Any 12 1.0 (0.52.2) Age, centre, [Limitations: limited
France cases (244 NHL; 87 HL; 104 glyphosate socioeconomic power for glyphosate]
20002004 LPSs; 56 MM) were recruited exposure category (blue/
IARC Monographs 112
from main hospitals of the HL Any exposure 6 1.7 (0.65) white collar)
French cities of Brest, Caen, to glyphosate
Nantes, Lille, Toulouse and LPS Any exposure 4 0.6 (0.22.1)
Bordeaux, aged 2075 years; ALL to glyphosate
cases excluded
MM Any exposure 5 2.4 (0.87.3)
Controls: 456 (response rate,
to glyphosate
91.2%); matched on age and sex,
recruited in the same hospitals as All lymphoid Any exposure 27 1.2 (0.62.1)
the cases, mainly in orthopaedic neoplasms to glyphosate
and rheumatological
departments and residing in the NHL, diffuse Occupational 5 1.0 (0.32.7)
hospitals catchment area large cell use of
Exposure assessment method: lymphoma glyphosate
questionnaire NHL, follicular Occupational 3 1.4 (0.45.2)
lymphoma exposure to
glyphosate
LPS/CLL Occupational 2 0.4 (0.11.8)
exposure to
glyphosate
LPS/HCL Occupational 2 1.8 (0.39.3)
exposure to
glyphosate
Table 2.2 (continued)
Reference, Population size, description, Organ site Exposure Exposed Risk estimate Covariates Comments
location, exposure assessment method (ICD code) category or cases/ (95% CI) controlled
enrolment level deaths
period
Cocco et al. Cases: 2348 (response rate, 88%); B-cell Occupational 4 3.1 (0.617.1) Age, sex, EPILYMPH case-
(2013) cases were all consecutive adult lymphoma exposure to education, centre control study in six
Czech Republic, patients first diagnosed with glyphosate European countries
France, Germany, lymphoma during the study
Italy, Ireland and period, resident in the referral
Spain area of the participating centres
19982004 Controls: 2462 (response rate,
81% hospital; 52% population);
controls from Germany and
Italy were randomly selected
by sampling from the general
population and matched to cases
on sex, 5-year age-group, and
residence area. The rest of the
centres used matched hospital
controls, excluding diagnoses of
cancer, infectious diseases and
immunodeficiency diseases
Exposure assessment method:
questionnaire; support of a crop-
exposure matrix to supplement
the available information,
industrial hygienists and
occupational experts in each
participating centre reviewed the
general questionnaires and job
modules to assess exposure to
pesticides
ALL, acute lymphocytic leukaemia; B-CLL, chronic lymphocytic leukaemia; CLL, chronic lymphocytic leukaemia; HCL, hairy cell leukaemia; HL, Hodgkin lymphoma; LPS,
lymphoproliferative syndrome; MCPA, 2-methyl-4-chlorophenoxyacetic acid; MM, multiple myeloma; NHL, non-Hodgkin lymphoma; NR, not reported; ref., reference; STS, soft tissue
sarcoma
25
Glyphosate
IARC Monographs 112
population-based, and conducted in farming exposed cases) compared with those with some,
areas. Potential confounding from multiple but 2days of exposure. [The study was large,
exposures was accounted for in the analysis.] but had relatively low participation rates.]
Using the data set of the pooled popu- Kachuri et al. (2013) investigated the asso-
lation-based casecontrol studies in Iowa, ciation between lifetime use of pesticides and
Minnesota, and Nebraska, USA, Lee et al. multiple myeloma in a population-based case
(2004a) investigated whether asthma acts as an control study among men in six Canadian
effect modifier of the association between pesti- provinces between 1991 and 1994 (see the
cide exposure and NHL. The study included 872 Monograph on Malathion, Section 2.0, for a
cases diagnosed with NHL from 1980 to 1986 and detailed description of this study). Data from
2381 frequency-matched controls. Information 342 cases of multiple myeloma and 1357 controls
on use of pesticides and history of asthma was were obtained for ever-use of pesticides, number
based on interviews. A total of 177 subjects (45 of pesticides used, and days per year of pesticide
cases, 132 controls) reported having been told use. The odds ratios were adjusted for age, prov-
by their doctor that they had asthma. Subjects ince of residence, type of respondent, smoking
with a history of asthma had a non-significantly and medical history. The odds ratio for ever-use
lower risk of NHL than non-asthmatics, and of glyphosate was 1.19 (95% CI, 0.761.87; 32
there was no main effect of pesticide exposure. cases). When the analysis was conducted by level
In general, asthmatics tended to have larger odds of exposure, no association was found for light
ratios associated with exposure to pesticides users (2days per year) of glyphosate (OR, 0.72;
than non-asthmatics. There was no indication 95% CI, 0.391.32; 15 exposed cases) while the
of effect modification: the odds ratio associated odds ratio in heavier users (>2days per year) was
with glyphosate use was 1.4 (95% CI, 0.982.1; 2.04 (95% CI, 0.984.23; 12 exposed cases). [The
53 exposed cases) among non-asthmatics and 1.2 study had relatively low response rates. Multiple
(95% CI, 0.43.3; 6 exposed cases) for asthmatics, myeloma is now considered a subtype of NHL.]
when compared with non-asthmatic non-ex-
posed farmers). [This analysis overlapped with (c) Casecontrol studies in Sweden
that of De Roos et al. (2003).] Nordstrm et al. (1998) conducted a popu-
lation casecontrol study in Sweden on hairy
(b) The cross-Canada casecontrol study cell leukaemia (considered to be a subgroup
McDuffie et al. (2001) studied the associa- of NHL). The study included 121 cases in men
tions between exposure to specific pesticides and and 484 controls matched for age and sex. An
NHL in a multicentre population-based study age-adjusted odds ratio of 3.1 (95% CI, 0.812;
with 517 cases and 1506 controls among men of 4 exposed cases) was observed for exposure to
six Canadian provinces (see the Monograph on glyphosate. [This study had limited power to
Malathion, Section 2.0, for a detailed descrip- detect an effect, and there was no adjustment for
tion of this study). Odds ratios of 1.26 (95% other exposures.]
CI, 0.871.80; 51 exposed cases; adjusted for Hardell & Eriksson (1999) reported the
age and province) and 1.20 (95% CI, 0.831.74, results of a population-based casecontrol study
adjusted for age, province, high-risk exposures) on the incidence of NHL in men associated with
were observed for exposure to glyphosate. In an pesticide exposure in four northern counties in
analysis by frequency of exposure to glyphosate, Sweden. Exposure data was collected by ques-
participants with > 2 days of exposure per year tionnaire (also supplemented by telephone inter-
had an odds ratio of 2.12 (95% CI, 1.203.73, 23 views) from 404 cases (192 deceased) and 741
26
Glyphosate
controls (matched by age, sex, county, and vital or with an odds ratio of > 1.50 and at least 10
status). Increased risks of NHL were found for exposed subjects (2,4,5-T and/or 2,4-D; mercu-
subjects exposed to herbicides and fungicides. rial seed dressing, arsenic, creosote, tar), age,
The odds ratio for ever-use of glyphosate was 2.3 sex, year of diagnosis or enrolment. The odds
(95% CI, 0.413; 4 exposed cases) in a univariate ratio for exposure to glyphosate was 2.02 (95%
analysis, and 5.8 (95% CI, 0.654) in a multivar- CI, 1.103.71) in a univariate analysis, and 1.51
iable analysis. [The exposure frequency was low (95% CI, 0.772.94) in a multivariable analysis.
for glyphosate, and the study had limited power When exposure for more than 10 days per year
to detect an effect. The variables included in the was considered, the odds ratio was 2.36 (95% CI,
multivariate analysis were not specified. This 1.045.37). With a latency period of >10 years,
study may have overlapped partially with those the odds ratio was 2.26 (95% CI, 1.164.40).
of Hardell et al. (2002).] The associations with exposure to glyphosate
Hardell et al. (2002) conducted a pooled anal- were reported also for lymphoma subtypes, and
ysis of two casecontrol studies, one on NHL elevated odds ratios were reported for most of the
(already reported in Hardell & Eriksson, 1999) cancer forms, including B-cell lymphoma (OR,
and another on hairy cell leukaemia, a subtype 1.87; 95% CI, 0.9983.51) and the subcategory of
of NHL (already reported by Nordstrm et al., small lymphocytic lymphoma/chronic lympho-
1998). The pooled analysis of NHL and hairy cytic leukaemia (OR, 3.35; 95% CI, 1.427.89;
cell leukaemia was based on 515 cases and 1141 [not adjusted for other pesticides]). [This was a
controls. Increased risk was found for exposure large study; there was possible confounding from
to glyphosate (OR, 3.04; 95% CI, 1.088.52; 8 use of other pesticides including MCPA, but this
exposed cases) in the univariate analysis, but the was considered in the analysis.]
odds ratio decreased to 1.85 (95% CI, 0.556.20)
when study, study area, and vital status were (d) Other casecontrol studies in Europe
considered in a multivariate analysis. [The expo- Orsi et al. (2009) reported the results of a
sure frequency was low for glyphosate and the hospital-based casecontrol study conducted in
study had limited power. This study partially six centres in France between 2000 and 2004.
overlapped with those of Hardell & Eriksson Incident cases with a diagnosis of lymphoid
(1999) and Nordstrm et al. (1998).] neoplasm aged 2075 years and controls of the
Eriksson et al. (2008) reported the results of same age and sex as the cases were recruited in
a population based casecontrol study of expo- the same hospital, mainly in the orthopaedic and
sure to pesticides as a risk factor for NHL. Men rheumatological departments during the same
and women aged 1874 years living in Sweden period. [The Working Group noted that the age
were included from 1 December 1999 to 30 of case eligibility was given in the publication as
April 2002. Incident cases of NHL were enrolled 2075 years in the materials and methods section,
from university hospitals in Lund, Linkping, but as 1875 years in the abstract.] Exposures
rebro, and Ume. Controls (matched by age to pesticides were evaluated through specific
and sex) were selected from the national popu- interviews and case-by-case expert reviews. The
lation registry. Exposure to different agents was analyses included 491 cases (244 cases of NHL,
assessed by questionnaire. In total, 910 (91%) 87 cases of Hodgkin lymphoma), 104 of lymph-
cases and 1016 (92%) controls participated. oproliferative syndrome, and 56 cases of multiple
Multivariable models included agents with myeloma), and 456 age- and sex-matched controls.
statistically significant increased odds ratios Positive associations between some subtypes
(MCPA, 2-methyl-4-chlorophenoxyacetic acid), and occupational exposure to several pesticides
27
IARC Monographs 112
were noted. The odds ratios associated with any and the odds ratio for exposure to glyphosate
exposure to glyphosate were 1.2 (95% CI, 0.62.1; and B-cell lymphoma was 3.1 (95% CI, 0.617.1;
27 exposed cases) for all lymphoid neoplasms 4 exposed cases and 2 exposed controls). [The
combined, 1.0 (95% CI, 0.52.2; 12 exposed study had a very limited power to assess the
cases) for NHL, 0.6 (95% CI, 0.22.1; 4 exposed effects of glyphosate on risk of NHL.]
cases) for lymphoproliferative syndrome, 2.4
(95% CI, 0.87.3) for multiple myeloma, and 1.7 2.2.2 Other haematopoietic cancers
(95% CI, 0.65.0; 6 exposed cases) for Hodgkin
lymphoma, after adjusting for age, centre, and Orsi et al. (2009) also reported results for
socioeconomic category (blue/white collar). Hodgkin lymphoma (see Section 2.2.1).
Cocco et al. (2013) reported the results of a Karunanayake et al. (2012) conducted a case
pooled analysis of casecontrol studies conducted control study of Hodgkin lymphoma among
in six European countries in 19982004 white men, aged 19 years or older, in six regions of
(EPILYMPH, Czech Republic, France, Germany, Canada (see the Malathion Monograph, Section
Ireland, Italy, and Spain) to investigate the role of 2.0, for a detailed description of this study). The
occupational exposure to specific groups of chem- analysis included 316 cases and 1506 age-matched
icals in the etiology of lymphoma overall, B-cell (2years) controls. Based on 38 cases exposed
lymphoma, and its most prevalent subtypes. A to glyphosate, the odds ratios were 1.14 (95% CI,
total of 2348 incident cases of lymphoma and 0.741.76) adjusted for age and province, and 0.99
2462 controls were recruited. Controls from (95% CI, 0.621.56) when additionally adjusted
Germany and Italy were randomly selected by for medical history variables.
sampling from the general population, while the Brown et al. (1990) evaluated exposure
rest of the centres used matched hospital controls. to carcinogens in an agricultural setting and
Overall, the participation rate was 88% for cases, the relationship with leukaemia in a popula-
81% for hospital controls, and 52% for population tion-based casecontrol interview study in Iowa
controls. An occupational history was collected and Minnesota, USA, including 578 white men
with farm work-specific questions on type of with leukaemia and 1245 controls. The exposure
crop, farm size, pests being treated, type and assessment was done with a personal interview
schedule of pesticide use. In each study centre, of the living subjects or the next-of-kin. Farmers
industrial hygienists and occupational experts had a higher risk of all leukaemias compared
assessed exposure to specific groups of pesti- with non-farmers, and associations were found
cides and individual compounds with the aid of for exposure to specific animal insecticides,
agronomists. [Therefore any exposure misclas- including the organophosphates crotoxyphos,
sification would be non-differential.] Analyses dichlorvos, famphur, pyrethrins, and methoxy-
were conducted for lymphoma and the most chlor. The odds ratio for glyphosate was 0.9 (95%
prevalent lymphoma subtypes adjusting for age, CI, 0.51.6; 15 exposed cases; adjusted for vital
sex, education, and centre. Lymphoma overall, status, age, state, tobacco use, family history of
and B-cell lymphoma were not associated with lymphopoietic cancer, high-risk occupations,
any class of the investigated pesticides, while and high-risk exposures). [This was a large study
the risk of chronic lymphocytic leukaemia was in an agricultural setting, but had limited power
elevated among those ever exposed to inorganic for studying the effects of glyphosate use.]
and organic pesticides. Only for a few individual
agrochemicals was there a sizeable number of
study subjects to conduct a meaningful analysis,
28
Glyphosate
2.3 Casecontrol studies on other Carren et al. (2005) evaluated the effects of
cancer sites rural exposures to pesticides on risk of glioma
among women aged 1880 years who were
2.3.1 Cancer of the oesophagus and stomach nonmetropolitan residents of Iowa, Michigan,
Minnesota, and Wisconsin in the Upper Midwest
Lee et al. (2004b) evaluated the risk of adeno- Health Study, USA. A total of 341 cases of glioma
carcinomas of the oesophagus and stomach and 528 controls were enrolled. A personal inter-
associated with farming and agricultural pesti- view was carried out for exposure assessment. The
cide use. The population-based casecontrol response rates were 90% and 72%, respectively.
study was conducted in eastern Nebraska, USA. After adjusting for age, age group, education, and
Subjects of both sexes diagnosed with adenocar- farm residence, no association with glioma was
cinoma of the stomach (n=170) or oesophagus observed for exposure to several pesticide classes
(n=137) between 1988 and 1993 were enrolled. or individual pesticides. There was a reduced
Controls (n=502) were randomly selected from risk for glyphosate (OR, 0.7; 95% CI, 0.41.3; 18
the population registry of the same geographical exposed cases). These results were not affected by
area. The response rates were 79% for cancer of the the exclusion of proxy respondents (43% of cases,
stomach, 88% for cancer of the oesophagus, and 2% of controls).
83% for controls. Adjusted odds ratios were esti- Lee et al. (2005) evaluated the association
mated for use of individual and chemical classes between farming and agricultural pesticide use
of insecticides and herbicides, with non-farmers and risk of adult glioma in a population-based
as the reference category. No association was casecontrol study in eastern Nebraska, USA.
found with farming or ever-use of insecticides Cases of glioma were in men and women (n=251)
or herbicides, or with individual pesticides. For and were compared with population controls
ever-use of glyphosate, the odds ratio was 0.8 from a previous study (n = 498). A telephone
(95% CI, 0.41.4; 12 exposed cases) for cancer of interview was conducted for 89% of the cases
the stomach, and 0.7 (95% CI, 0.31.4; 12 exposed and 83% of the controls. Adjusted odds ratios
cases) for oesophageal cancer. [The study was for farming and for use of individual and chem-
conducted in a farming area, but the power to ical classes of insecticides and herbicides were
detect an effect of glyphosate use was limited.] calculated using non-farmers as the reference
category. Among men, ever living or working
2.3.2 Cancer of the brain on a farm and duration of farming were associ-
Ruder et al. (2004) conducted a casecontrol ated with significantly increased risks of glioma,
study on glioma among nonmetropolitan but the positive findings were limited to proxy
residents of Iowa, Michigan, Minnesota, and respondents. Among women, there were no posi-
Wisconsin in the Upper Midwest Health Study, tive associations with farming activities among
USA. The study included 457 cases of glioma self or proxy respondents. Some specific pesti-
and 648 population-based controls, all adult cide families and individual pesticides were asso-
men. Exposure assessment was done with inter- ciated with significantly increased risks among
views of the subject or the relatives. The response male farmers, but most of the positive associa-
rates were 93% and 70% for cases and controls, tions were limited to proxy respondents. There
respectively. No association were found with any was a non-significant excess risk with glyphosate
of the pesticides assessed, including glyphosate. use for the overall group (OR, 1.5; 95% CI, 0.73.1;
[Glyphosate use was assessed, but specific results 17 exposed cases), but there was inconsistency
were not presented.] between observations for self-respondents (OR,
29
IARC Monographs 112
0.4; 95% CI, 0.11.6) and observations for proxy 2.4. Meta-analyses
respondents (OR, 3.1; 95% CI, 1.28.2). [The
study had limited power to detect an effect of Schinasi & Leon (2014) conducted a system-
glyphosate use, and the inconsistencies for self atic review and meta-analysis of NHL and occu-
and proxy respondents made the results difficult pational exposure to agricultural pesticides,
to interpret.] including glyphosate. The meta-analysis for
glyphosate included six studies (McDuffie et al.,
2.3.3 Soft tissue sarcoma 2001; Hardell et al., 2002; De Roos et al., 2003;
2005a; Eriksson et al., 2008; Orsi et al., 2009) and
Pahwa et al. (2011) reported the results of yielded a meta risk-ratio of 1.5 (95% CI, 1.12.0).
the soft tissue sarcoma component of the cross- [The Working Group noted that the most fully
Canada study in relation to specific pesticides, adjusted risk estimates from the articles by
including 357 cases of soft tissue sarcoma and Hardell et al. (2002) and Eriksson et al. (2008)
1506 population controls from 19911994. The were not used in this analysis. After considering
fully adjusted odds ratio for glyphosate use was the adjusted estimates of the two Swedish studies
0.90 (95% CI, 0.581.40). in the meta-analysis, the Working Group esti-
mated a meta risk-ratio of 1.3 (95% CI, 1.031.65),
2.3.4 Cancer of the prostate I2=0%, P for heterogeneity 0.589.]
Band et al. (2011) report results of a case
control study including 1516 patients with cancer
of the prostate (ascertained by the cancer registry 3. Cancer in Experimental Animals
of British Columbia, Canada, for 198390) and
4994 age-matched controls with cancers at all 3.1 Mouse
other cancer sites excluding lung and unknown
primary site. Agricultural exposures were See Table3.1
assessed by job-exposure matrix. A total of 60
cases were exposed to glyphosate (adjusted OR, 3.1.1 Dietary administration
1.36; 95% CI, 0.832.25). Groups of 50 male and 50 female CD-1 mice
[age not reported] were given diets containing
2.3.5 Childhood cancer glyphosate (purity, 99.7%) at a concentration of
Parental exposure to pesticides, including 0, 1000, 5000, or 30000ppm, ad libitum, for 24
glyphosate, was assessed in a population-based months. There was no treatment-related effect on
casecontrol study of childhood leukaemia in body weight in male and female mice at the lowest
Costa Rica (Monge et al., 2007). However, associ- or intermediate dose. There was a consistent
ations of childhood cancer with glyphosate were decrease in body weight in the male and female
reported only for an other pesticides category mice at the highest dose compared with controls.
that also included paraquat, chlorothalonil, and Survival in all dose groups was similar to that of
other chemicals. [Because glyphosate was not controls. There was a positive trend (P = 0.016,
specifically assessed, this study was not evalu- trend test; see EPA, 1985b) in the incidence of
ated by the Working Group.] renal tubule adenoma in dosed male mice: 0/49,
0/49, 1/50 (2%), 3/50 (6%). [The Working Group
noted that renal tubule adenoma is a rare tumour
in CD-1 mice.] No data on tumours of the kidney
30
Table 3.1 Studies of carcinogenicity with glyphosate in mice
Species, strain (sex) Dosing regimen, For each target organ: incidence Significance Comments
Duration Animals/group at start (%) and/or multiplicity of tumours
Reference
Mouse, CD-1 (M,F) Diet containing glyphosate (technical Males P for trend=0.016; No information was provided on
24 mo grade; purity, 99.7%) at concentrations of Renal tubule adenoma: 0/49, 0/49, see Comments renal tubule adenomas in female
EPA (1985a, b, 1986, 0, 1000, 5000, or 30000 ppm, ad libitum, 1/50 (2%), 3/50 (6%) mice, or on statistical analyses of
1991a) for 24 mo Females tumour data
50 M and 50 F/group [age, NR] No data provided on the kidney EPA recommended that additional
renal sections be cut and evaluated
from all control and treated male
mice. The pathology report for
Report from the PWG of the EPA
these additional sections (EPA,
(1986):
1985b) showed the same incidence
Males of renal tubule adenomas as
Renal tubule adenoma: 1/49 (2%), [NS] originally reported, with no
0/49, 0/50, 1/50 (2%) significant difference in incidence
Renal tubule carcinoma: 0/49, 0/49, [P=0.037; Cochran when comparing control and
1/50 (2%), 2/50 (4%) Armitage trend test] treated groups; however, the test
Renal tubule adenoma or carcinoma [P=0.034; Cochran for linear trend in proportions
(combined): 1/49 (2%), 0/49, 1/50 Armitage trend test] resulted in P=0.016
(2%), 3/50 (6%) EPA (1986) convened a PWG and
requested additional pathological
and statistical information on
kidney tumours observed in male
mice treated with glyphosate
Mouse, CD-1 (M,F) Diet containing glyphosate (purity, Males
104 wk 98.6%) at doses of 0, 100, 300, 1000 mg/kg Haemangiosarcoma: 0/50, 0/50, [P<0.001; Cochran
JMPR (2006) bw, ad libitum, for 104 wk 0/50, 4/50 (8%) Armitage trend test]
50 M and 50 F/group [age, NR] Histiocytic sarcoma in the NS
lymphoreticular/haemopoietic
tissue: 0/50, 2/50 (4%), 0/50, 2/50
(4%)
Females
Haemangiosarcoma: 0/50, 2/50 NS
(4%), 0/50, 1/50 (2%)
Histiocytic sarcoma in the NS
lymphoreticular/haemopoietic
tissue: 0/50, 3/50 (6%), 3/50 (6%),
1/50 (2%)
31
Glyphosate
32
Table 3.1 (continued)
Species, strain (sex) Dosing regimen, For each target organ: incidence Significance Comments
Duration Animals/group at start (%) and/or multiplicity of tumours
Reference
Mouse, Swiss (M) Initiationpromotion study Skin tumours [called papillomas Short duration of treatment, no
32 wk Skin application of glyphosate-based by the authors, following gross solvent controls, and lack of any
George et al. (2010) formulation (glyphosate, 41%; POEA, examination only] histopathological evaluation
~15%) (referred to as glyphosate) Age at start, NR (mice weighed
dissolved in 50% ethanol; DMBA 1215 g bw)
IARC Monographs 112
were provided for female mice. No other tumour morphological continuum in the development
sites were identified (EPA, 1985a). Subsequent to and progression of renal neoplasia. Thus larger
its initial report (EPA, 1985a), the United States masses may exhibit greater heterogeneity in histo-
Environmental Protection Agency (EPA) recom- logical growth pattern, and cytologically more
mended that additional renal sections be cut and pleomorphism and atypia than smaller lesions
evaluated from all male mice in the control and (Eustis et al., 1994). Of note, a renal tumour
treated groups. The pathology report for these confirmed by the PWG after re-evaluation of the
additional sections (EPA, 1985b) indicated the original slides (EPA, 1986), had not been seen in
same incidence of renal tubule adenoma as orig- the re-sectioned kidney slides (EPA, 1985b). This
inally reported, with no significant increase in may be related to the growth of tumour that
incidence between the control group and treated in contrast to tumours in other organs is not
groups by pairwise comparison. However, as spherical but elliptical because of the potential
already reported above, the test for linear trend expansion in tubules. In addition, the concept
in proportions resulted in a significance of of tubular expansion without compression of
P=0.016. The EPA (1986) also requested that a adjacent parenchyma may be at the basis of the
pathology working group (PWG) be convened discrepancy between the first (EPA, 1985a, b) and
to evaluate the tumours of the kidney observed second evaluation (EPA, 1986).]
in male mice treated with glyphosate, including In another study reported to the Joint FAO/
the additional renal sections. In this second eval- WHO Meeting on Pesticide Residues (JMPR),
uation, the PWG reported that the incidence of groups of 50 male and 50 female CD-1 mice [age
adenoma of the renal tubule was 1/49 (2%), 0/49, at start not reported] were given diets containing
0/50, 1/50 (2%) [not statistically significant]; the glyphosate (purity, 98.6%) at a concentration
incidence of carcinoma of the renal tubule was that was adjusted weekly for the first 13 weeks
0/49, 0/49, 1/50 (2%), 2/50 (4%) [P=0.037, trend and every 4 weeks thereafter to give doses of 0,
test for carcinoma]; and the incidence of adenoma 100, 300, or 1000 mg/kg bw, ad libitum, for 104
or carcinoma (combined) of the renal tubule was weeks (JMPR, 2006). There was no treatment-re-
1/49 (2%), 0/49, 1/50 (2%), 3/50 (6%) [P=0.034, lated effect on body weight or survival in any
trend test for combined]. [The Working Group of the dosed groups. There was an increase in
considered that this second evaluation indicated the incidence of haemangiosarcoma in males
a significant increase in the incidence of rare 0/50, 0/50, 0/50, 4/50 (8%) [P<0.001, Cochran
tumours, with a dose-related trend, which could Armitage trend test], and in females 0/50, 2/50
be attributed to glyphosate. Chandra & Frith (4%), 0/50, 1/50 (2%) [not statistically significant],
(1994) reported that only 1 out of 725 [0.14%] and an increase in the incidence of histiocytic
CD-1 male mice in their historical database had sarcoma in the lymphoreticular/haemopoietic
developed renal cell tumours (one carcinoma).] tissue in males 0/50, 2/50 (4%), 0/50, 2/50 (4%),
[The Working Group noted the differences and in females 0/50, 3/50 (6%), 3/50 (6%), 1/50
in histopathological diagnosis between pathol- (2%) [not statistically significant for males or
ogists. Proliferative lesions of the renal tubules females]. [The Working Group considered that
are typically categorized according to published this study was adequately reported.]
criteria as hyperplasia, adenoma, or carcinoma.
The difference is not trivial, because focal hyper-
plasia, a potentially preneoplastic lesion, should
be carefully differentiated from the regenerative
changes of the tubular epithelium. There is a
33
IARC Monographs 112
34
Glyphosate
mice [age at start not reported] were given diets evaluate this study because of the limited exper-
containing glyphosate (purity, 9496%) at a imental data provided in the review article and
concentration of 0, 1600, 8000, or 40 000 ppm supplemental information.]
for 18 months. The increase in the incidence of
bronchiolo-alveolar adenoma and carcinoma,
and of lymphoma, was reported to be not statis-
3.2 Rat
tically significant in males and females receiving See Table3.2
glyphosate. [The Working Group was unable to
evaluate this study because of the limited exper- 3.2.1 Drinking-water
imental data provided in the review article and
supplemental information.] Groups of 10 male and 10 female Sprague-
In the second study (identified as Study 13, Dawley rats (age, 5 weeks) were given drinking-
2001), groups of 50 male and 50 female Swiss water containing a glyphosate-based formulation
albino mice [age at start not reported] were at a dose of 0 (control), 1.1108% (5.0105 mg/L),
given diets containing glyphosate (purity, >95%) 0.09% (400mg/L) or 0.5% (2.25103 mg/L), ad
at a concentration of 0 (control), 100, 1000, or libitum, for 24 months (Sralini et al., 2014). [The
10000ppm for 18 months. The authors reported study reported is a life-long toxicology study on
a statistically significant increase in the incidence a glyphosate-based formulation and on geneti-
of malignant lymphoma (not otherwise specified, cally modified NK603 maize, which the authors
NOS) in males at the highest dose: 10/50 (20%), stated was designed as a full study of long-term
15/50 (30%), 16/50 (32%), 19/50 (38%; P < 0.05; toxicity and not a study of carcinogenicity. No
pairwise test); and in females at the highest dose: information was provided on the identity or
18/50 (36%), 20/50 (40%), 19/50 (38%), 25/50 concentration of other chemicals contained in
(50%; P < 0.05; pairwise test). [The Working this formulation.] Survival was similar in treated
Group was unable to evaluate this study because and control rats. [No data on body weight were
of the limited experimental data provided in the provided.] In female rats, there was an almost
review article and supplemental information.] twofold increase in the incidence of tumours
In the third study (identified as Study 14, of the mammary gland (mainly fibroadenoma
2009a), groups of 51 male and 51 female CD-1 and adenocarcinoma) in animals exposed to
mice [age at start not reported] were given diets the glyphosate-based formulation only versus
containing glyphosate (purity, 94.697.6%) at a control animals: control, 5/10 (50%); lowest dose,
concentration of 0, 500, 1500, or 5000ppm for 9/10 (90%); intermediate dose, 10/10 (100%)
18 months. Incidences for bronchiolo-alveolar [P < 0.05; Fisher exact test]; highest dose, 9/10
adenoma and carcinoma, malignant lymphoma (90%). [The Working Group concluded that this
(NOS), and hepatocellular adenoma and carci- study conducted on a glyphosate-based formu-
noma in males, and for bronchiolo-alveolar lation was inadequate for evaluation because
adenoma and carcinoma, malignant lymphoma the number of animals per group was small, the
(NOS) and pituitary adenoma in females, were histopathological description of tumours was
included in the article. In males, the authors poor, and incidences of tumours for individual
reported that there was a significant positive trend animals were not provided.]
[statistical test not specified] in the incidence of In another study with drinking-water,
bronchiolo-alveolar carcinoma (5/51, 5/51, 7/51, Chruscielska et al. (2000) gave groups of 55
11/51) and of malignant lymphoma (0/51, 1/51, male and 55 female Wistar rats (age, 67 weeks)
2/51, 5/51). [The Working Group was unable to drinking-water containing an ammonium salt
35
IARC Monographs 112
36
Table 3.2 Studies of carcinogenicity with glyphosate in rats
Species, strain (sex) Dosing regimen, For each target organ: Significance Comments
Duration Animals/group at start incidence (%) and/or
Reference multiplicity of tumours
Rat, Sprague-Dawley Drinking-water containing a glyphosate- Males Data are from an in-depth life-long toxicology
(M,F) based formulation at a concentration No significant increase in NS study on a glyphosate-based formulation and
24 mo of 0 (control), 1.1108% (glyphosate, tumour incidence observed in NK603 genetically modified maize; authors
Sralini et al. (2014) 5.0105 mg/L), 0.09% (glyphosate, any of the treated groups stated that the study was designed as a full
400 mg/L) or 0.5% (glyphosate, Females chronic toxicity and not a carcinogenicity study.
2.25103 mg/L), ad libitum, for 24 mo No information provided on the identity or
Mammary tumours *[P<0.05]
10 M and 10 F/group (age, 5 wk) concentration of other chemicals contained in
(mainly fibroadenomas and
this formulation
adenocarcinomas): 5/10
Histopathology poorly described and tumour
(50%), 9/10 (90%), 10/10
incidences for individual animals not discussed
(100%)*, 9/10 (90%)
in detail. Small number of animals per group
Pituitary lesions [NS] [The Working Group concluded this was an
(hypertrophy, hyperplasia, inadequate study for the evaluation of glyphosate
and adenoma): 6/10 (60%), carcinogenicity]
8/10 (80%), 7/10 (70%), 7/10
(70%)
Rat, Wistar (M,F) Drinking-water containing ammonium No significant increase in NS Limited information on dosing regimen,
24 mo salt of glyphosate (13.85% solution) tumour incidence observed in histopathological examination methods, and
Chruscielska et al. [purity of glyphosate, NR] was used to any of the treated groups tumour incidences
(2000) make aqueous solutions of 0, 300, 900,
and 2700 mg/L
[Details on dosing regimen, NR]
55 M and 55 F/group (age, 67 wk)
Rat, Wistar- Diet containing glyphosate (purity, No significant increase in NS Short duration of exposure
Alpk:APfSD (M,F) 95.6%) at concentrations of 0, 2000, tumour incidence observed in
1yr 8000, or 20000 ppm, ad libitum, for 1yr any groups of treated animals
JMPR (2006) 24 M and 24 F/group [age, NR]
Rat, Sprague-Dawley Diet containing glyphosate (purity, No significant increase in NS
(M,F) 98.798.9%) at doses of 0, 10, 100, 300, or tumour incidence observed in
104 wk 1000 mg/kg bw, ad libitum, for 104 wk any groups of treated animals
JMPR (2006) 50 M and 50 F/group [age, NR]
Rat, Wistar- Diet containing glyphosate (purity, No significant increase in NS
Alpk:APfSD (M,F) 97.6%) at concentrations of 0, 2000, tumour incidence observed in
24 mo 6000, or 20000 ppm, ad libitum, for 2yr any groups of treated animals
JMPR (2006) 52 M and 52 F/group [age, NR]
37
Glyphosate
38
Table 3.2 (continued)
Species, strain (sex) Dosing regimen, For each target organ: Significance Comments
Duration Animals/group at start incidence (%) and/or
Reference multiplicity of tumours
Rat Sprague-Dawley Diet containing glyphosate (technical Males Historical control range for pancreatic islet cell
(M,F) grade; purity, 96.5%) at concentrations of Pancreas (islet cell): Adenoma, adenoma reported in males at this laboratory,
24 mo 0, 2000, 8000, or 20000ppm, ad libitum, Adenoma: 1/58 (2%), 8/57 * P0.05 1.88.5%
EPA (1991a, b, c, d) for 24 mo (14%)*, 5/60 (8%), 7/59 (12%) (Fisher exact EPA (1991a) performed additional analyses using
60 M and 60 F/group (age, 8 wk) test with the CochranArmitage trend test and Fisher
Carcinoma: 1/58 (2%), 0/57,
IARC Monographs 112
10 rats/group killed after 12 mo Bonferroni exact test, and excluding animals that died or
0/60, 0/59
inequality); were killed before wk 5455:
Adenoma or carcinoma Males
see
(combined): 2/58 (3%), 8/57 Pancreas (islet cell):
comments
(14%), 5/60 (8%), 7/59 (12%) Adenoma: 1/43 (2%), 8/45 (18%; P=0.018), 5/49
Liver: Adenoma, (10%), 7/48 (15%; P=0.042)
Hepatocellular adenoma: 2/60 P for trend Carcinoma: 1/43 (2%), 0/45 (0%), 0/49 (0%), 0/48
(3%), 2/60 (3%), 3/60 (6%), = 0.016; see (0%)
7/60 (12%) comments Adenoma or carcinoma (combined): 2/43 (5%),
Hepatocellular carcinoma: 8/45 (18%), 5/49 (10%), 7/48 (15%)
3/60 (5%), 2/60 (3%), 1/60 [There was no statistically significant positive
(2%), 2/60 (3%) trend in the incidence of pancreatic tumours,
Females and no apparent progression to carcinoma]
Liver:
Pancreas (islet cell): NS
Hepatocellular adenoma: 2/44 (5%; P for trend =
Adenoma: 5/60 (8%), 1/60
0.016), 2/45 (4%), 3/49 (6%), 7/48 (15%)
(2%), 4/60 (7%), 0/59
Hepatocellular carcinoma: 3/44 (7%); 2/45 (4%),
Carcinoma: 0/60, 0/60, 0/60,
1/49 (2%), 2/48 (4%)
0/59
Hepatocellular adenoma or carcinoma
Adenoma or carcinoma
(combined): 5/44 (11%), 4/45 (9%), 4/49 (8%),
(combined): 5/60 (8%), 1/60
9/48 (19%)
(2%), 4/60 (7%), 0/59
[There was no apparent progression to
Thyroid: Adenoma, carcinoma]
C-cell adenoma: 2/60 (3%), P for trend Females
2/60 (3%), 6/60 (10%), 6/60 = 0.031; see Thyroid:
(10%) comments C-cell adenoma: 2/57 (4%; P for trend = 0.031),
C-cell carcinoma: 0/60, 0/60, 2/60 (3%), 6/59 (10%), 6/55 (11%)
1/60, 0/60 C-cell carcinoma: 0/57, 0/60, 1/59 (2%), 0/55
C-cell adenoma or carcinoma (combined): 2/57
(4%), 2/60 (3%), 7/59 (12%), 6/55 (11%)
[There was no apparent progression to
carcinoma]
Table 3.2 (continued)
Species, strain (sex) Dosing regimen, For each target organ: Significance Comments
Duration Animals/group at start incidence (%) and/or
Reference multiplicity of tumours
Rat Sprague-Dawley Diet containing glyphosate (purity, Males
(M,F) 98.7%) at concentrations of 0 ppm, Pancreas (islet cell): Adenoma, [There was no statistically significant positive
Lifetime (up to 26 30 ppm (3 mg/kg bw per day), 100 Adenoma: 0/50 (0%), 5/49* *[P<0.05; trend in the incidence of pancreatic tumours,
mo) ppm (10 mg/kg bw per day), 300 ppm (10%), 2/50 (4%), 2/50 (4%) Fisher exact and no apparent progression to carcinoma]
EPA (1991a, b, c, d) (31 mg/kg bw per day), ad libitum, up to test]
26 mo Carcinoma: 0/50 (0%), 0/49
50 M and 50 F/group [age, NR] (0%), 0/50 (0%), 1/50 (2%)
Adenoma or carcinoma
(combined): 0/50 (0%), 5/49
(10%), 2/50 (4%), 3/50 (6%)
Females
Pancreas (islet cell): NS
Adenoma: 2/50 (4%), 1/50
(2%), 1/50 (2%), 0/50 (0%)
Carcinoma: 0/50 (0%), 1/50
(2%), 1/50 (2%), 1/50 (2%)
Adenoma or carcinoma
(combined): 2/50 (10%), 2/50
(2%), 2/50 (74%), 1/50 (2%)
bw, body weight; d, day; F, female; M, male; mo, month; NR, not reported; NS, not significant; wk, week; yr, year
39
Glyphosate
IARC Monographs 112
males (P = 0.016) and of thyroid follicular cell not included in Table3.2 and Section 5.3, because
adenoma in females (P = 0.031). [The Working the information provided in the review article
Group noted that there was no apparent progres- and its supplement was insufficient (e.g. infor-
sion to carcinoma for either tumour type.] mation lacking on statistical methods, choice of
The EPA (1991a, b, c, d) provided information doses, body-weight gain, survival data, details on
on another long-term study in which groups of histopathological examination and/or stability
50 male and 50 female Sprague-Dawley rats [age of dosed feed mixture).]
at start not reported] were given diets containing In one study (identified as Study 4, 1996),
glyphosate (purity, 98.7%) at a concentration of groups of 50 male and 50 female Wistar rats [age
0, 30 (3 mg/kg bw per day), 100 (10 mg/kg bw at start not reported] were given diets containing
per day), or 300 ppm (31 mg/kg bw per day), ad glyphosate (purity, 96%) at a concentration
libitum, for life (up to 26 months). No informa- of 0, 100, 1000, or 10 000 ppm, ad libitum, for
tion was provided on body weight or survival of 24 months. It was reported that hepatocellular
the study animals. An increase in the incidence adenomas and hepatocellular carcinomas were
of pancreatic islet cell adenoma was reported found at non-statistically significant incidences
in males at the lowest dose: controls, 0/50 (0%); in both males and females. There was no signifi-
lowest dose, 5/49 (10%) [P < 0.05; Fisher exact cant increase in tumour incidence in the treated
test]; intermediate dose, 2/50 (4%); highest dose, groups. [The Working Group was unable to
2/50 (4%). [The Working Group noted that there evaluate this study because of the limited exper-
was no statistically significant positive dose-re- imental data provided in the review article and
lated trend in the incidence of these tumours, supplemental information.]
and no apparent progression to carcinoma.] In one study in Sprague-Dawley rats (iden-
tified as Study 5, 1997), groups of 50 male and
3.2.3 Review articles 50 female rats [age at start not reported] were
given diets containing glyphosate technical acid
Greim et al. (2015) have published a review [purity not reported] at a concentration of 0, 3000,
article containing information on nine long- 15000, or 25000ppm, ad libitum, for 24 months.
term bioassay feeding studies in rats. Of these There was no significant increase in tumour inci-
studies, two had been submitted for review to dence in the treated groups. [The Working Group
the EPA (1991a, b, c, d), two to the JMPR (JMPR, was unable to evaluate this study because of the
2006), and one had been published in the openly limited experimental data provided in the review
available scientific literature (Chruscielska article and supplemental information.]
et al., 2000); these studies are discussed earlier In a second study in Sprague Dawley rats
in Section 3.2. The review article reported on an (identified as Study 6, 1997b), groups of 50
additional four long-term bioassay studies in rats males and 50 females [age at start not reported]
that had not been previously published, but had were given diets containing glyphosate (purity,
been submitted to various organizations for regis- 94.697.6%) at a concentration of 0, 3000, 10000,
tration purposes. The review article provided a or 30 000 ppm, ad libitum, for 24 months.
brief summary of each study and referred to an Non-significant increases in tumour incidences
online data supplement containing the original compared with controls were noted for skin
data on tumour incidence from study reports. keratoacanthoma in males at the highest dose,
The four additional long-term bioassay studies and for fibroadenoma of the mammary gland
in rats are summarized below. [The Working in females at the lowest and intermediate doses.
Group did not evaluate these studies, which are [The Working Group was unable to evaluate this
40
Glyphosate
study because of the limited experimental data microbial metabolism in humans (Motojyuku
provided in the review article and supplemental et al., 2008) and rodents (Brewster et al., 1991).
information.]
In another study in male and female Wistar 4.1.2 Absorption
rats (identified as Study 8, 2009b), groups of
51 male and 51 female rats [age at start not (a) Humans
reported] were fed diets containing glyphosate Data on the absorption of glyphosate via
(purity, 95.7%) at a concentration of 0, 1500, intake of food and water in humans were not
5000, or 15000 ppm, ad libitum, for 24 months. available to the Working Group. Inhalation of
The highest dose was progressively increased glyphosate is considered to be a minor route
to reach 24000ppm by week 40. A non-signif- of exposure in humans, because glyphosate is
icant increase in tumour incidence was noted usually formulated as an isopropylamine salt
for adenocarcinoma of the mammary gland in with a very low vapour pressure (Tomlin, 2000).
females at the highest dose (6/51) compared with In the Farm Family Exposure Study, 60% of
controls (2/51). [The Working Group was unable farmers had detectable levels of glyphosate in
to evaluate this study because of the limited 24-hour composite urine samples taken on the
experimental data provided in the review article day they had applied a glyphosate-based formu-
and supplemental information. The Working lation (Acquavella et al., 2004). Farmers who
Group noted that tumours of the mammary did not use rubber gloves had higher urinary
gland had been observed in other studies in rats concentrations of glyphosate than those who did
reviewed for the present Monograph.] use gloves [indicating that dermal absorption is
a relevant route of exposure]. In a separate study,
detectable levels of glyphosate were found in
4. Mechanistic and Other urine samples from farm families and non-farm
Relevant Data families (Curwin et al., 2007).
In accidental and deliberate intoxication cases
involving ingestion of glyphosate-based formu-
4.1 Toxicokinetic data lations, glyphosate was readily detectable in the
4.1.1 Introduction blood (Zouaoui et al., 2013). After deliberate
or accidental ingestion, one glyphosate-based
The herbicidal activity of glyphosate is attrib- formulation was found to be more lethal to
uted to interference with the production of essen- humans than another (Srensen & Gregersen,
tial aromatic amino acids (EPA, 1993b). In plants, 1999). [Greater lethality was attributed to the
glyphosate competitively inhibits the activity presence of trimethylsulfonium counterion,
of enolpyruvylshikimate phosphate synthase, which might facilitate greater absorption after
an enzyme that is not present in mammalian oral exposure.]
cells. Glyphosate is degraded by soil microbes Small amounts of glyphosate can be absorbed
to aminomethylphosphonic acid (AMPA) (see after dermal exposures in humans in vitro.
Fig.4.1), a metabolite that can accumulate in the For example, when an aqueous solution of 1%
environment. In mammals, glyphosate is not glyphosate was applied in an in-vitro human
metabolized efficiently, and is mainly excreted skin model, only 1.4% of the applied dose was
unchanged into the urine; however, it has been absorbed through the skin. Glyphosate is typi-
suggested that glyphosate can undergo gut cally formulated as an isopropylamine salt, and
is dissolved in a water-based vehicle, while the
41
IARC Monographs 112
stratum corneum is a lipid-rich tissue (Wester In a study reviewed by the EPA, Sprague-
et al., 1991). In-vitro studies using human skin Dawley rats were given an oral dose of glyphosate
showed that percutaneous absorption of a (10 mg/kg bw); 30% and 36% of the administered
glyphosate-based formulation was no more than dose was absorbed in males and females, respec-
2% of the administered dose over a concentration tively (EPA, 1993b). At a dose that was ~10-fold
range of 0.5154 g/cm2 and a topical volume higher (1000 mg/kg bw), oral absorption of
range of 0.0140.14 mL/cm2. In addition, very glyphosate by the rats was slightly reduced.
little glyphosate ( 0.05% of the administered In a 14-day feeding study in Wistar rats given
dose) was sequestered in the stratum corneum glyphosate at dietary concentrations of up to 100
after dermal application (Wester et al., 1991). ppm, only ~15% of the administered dose was
In the human Caco-2 cell line, an in-vitro found to be absorbed (JMPR, 2006). In New
model of intestinal enterocytes, glyphosate Zealand White rabbits or lactating goats given
(>10mg/mL) was shown to significantly disrupt glyphosate as single oral doses (69 mg/kg bw),
barrier properties, leading to an increase in para- a large percentage of the administered dose was
cellular permeability (transport of substances recovered in the faeces [suggesting very poor
that pass through the intercellular space between gastrointestinal absorption of glyphosate in
the cells) (Vasiluk et al., 2005). these animal models] (JMPR, 2006).
In monkeys given glyphosate by dermal appli-
(b) Experimental systems cation, percutaneous absorption was estimated
Three studies have been conducted to inves- to be between 1% and 2% of the administered
tigate the absorption of a single oral dose of dose (Wester et al., 1991). Most of the adminis-
glyphosate in rats (Brewster et al., 1991; Chan & tered dose was removed by surface washes of the
Mahler, 1992; EPA, 1993b). exposed skin.
In male Sprague-Dawley rats given
[14C]-labelled glyphosate (10 mg/kg bw), the 4.1.3 Distribution
majority of the radiolabel was associated with
(a) Humans
the gastrointestinal contents and small intestinal
tissue 2 hours after administration (Brewster No data in humans on the distribution of
et al., 1991). Approximately 3540% of the admin- glyphosate in systemic tissues other than blood
istered dose was found to be absorbed from the were available to the Working Group. In cases
gastrointestinal tract. Urinary and faecal routes of accidental or deliberate intoxication involving
of elimination were equally important. [The ingestion of glyphosate-based formulations,
Working Group concluded that glyphosate is glyphosate was measured in blood. Mean blood
incompletely absorbed from the gastrointestinal concentrations of glyphosate were 61mg/L and
tract after oral exposure in rats.] 4146mg/L in mild-to-moderate cases of intoxi-
In a study by the United States National cation and in fatal cases, respectively (Zouaoui
Toxicology Programme (NTP) in Fisher 344 rats, et al., 2013).
30% of the administered oral dose (5.6 mg/kg bw) One report, using optical spectroscopy and
was absorbed, as determined by urinary excre- molecular modelling, indicated that glyphosate
tion data (Chan & Mahler, 1992). This finding could bind to human serum albumin, mainly
was in accordance with the previously described by hydrogen bonding; however, the fraction of
study of oral exposure in rats (Brewster et al., glyphosate that might bind to serum proteins
1991). in blood was not actually measured (Yue et al.,
2008).
42
Glyphosate
(b) Experimental systems in the blood by day 6 (JMPR, 2006). The tissue
In Sprague-Dawley rats given a single oral concentrations of glyphosate had the following
dose of glyphosate (100 mg/kg bw), glypho- rank order: kidneys > spleen > fat > liver.
sate concentrations in plasma reached peak Tissue levels declined rapidly after cessation of
levels, then declined slowly from day 1 to day 5 exposure to glyphosate. A second study in rats
(Bernal et al., 2010). The plasma data appeared given glyphosate (10 mg/kg bw per day, 14 days)
to fit a one-compartment model with an elim- followed by a single oral dose of [14C]-glyphosate
ination rate constant of kel = 0.021 hour-1. [The (at 10 mg/kg bw) showed that repeated dosing
Working Group estimated the elimination half- did not alter the tissue distribution of glyphosate
life of glyphosate to be 33 hours.] Tissue levels of (JMPR, 2006).
glyphosate were not determined in this study. In In rhesus monkeys, tissues harvested 7days
a study by Brewster et al. (1991), the tissue levels after dermal exposures to [14C]-labelled glypho-
of glyphosate at 2, 6.3, 28, 96, and 168 hours in sate did not contain radiolabel at detectable levels
Sprague-Dawley rats given a single oral dose (Wester et al., 1991).
(10 mg/kg bw) declined rapidly. Tissues with the
greatest amounts of detectable radiolabel (>1% of 4.1.4 Metabolism and modulation of
the administered dose) were the small intestine, metabolic enzymes
colon, kidney, and bone. Peak levels were reached (a) Metabolism
in small intestine tissue and blood by 2 hours,
while peak levels in other tissues occurred at Glyphosate is degraded in the environ-
6.3 hours after dosing. After 7 days, the total ment by soil microbes, primarily to AMPA
body burden of [14C]-labelled residues was ~1% of and carbon dioxide (Fig. 4.1; Jacob et al.,
the administered dose, and was primarily asso- 1988). A minor pathway for the degradation of
ciated with the bone (~1 ppm). In every tissue glyphosate in bacteria (Pseudomonas sp. strain
examined after administration of [14C]-labelled LBr) is via conversion to glycine (Jacob et al.,
glyphosate, essentially 100% of the radiolabel 1988). In a case of deliberate poisoning with a
that was present in the tissue was unmetabolized glyphosate-based formulation, small amounts
parent glyphosate. Thus, essentially 100% of the of AMPA (15.1 g/mL) were detectable in the
body burden was parent compound, with no blood (Motojyuku et al., 2008) [suggesting that
significant persistence of glyphosate after 7days this pathway might also operate in humans]. In
(Brewster et al., 1991). In a 14-day feeding study rats given a single high oral dose of glyphosate
in Wistar rats given diets containing glyphosate (100 mg/kg bw), small amounts of AMPA were
at 100ppm, glyphosate reached steady-state levels detected in the plasma (Bernal et al., 2010). In
43
IARC Monographs 112
44
Glyphosate
1993b). By either route, most (98%) of the admin- provided in the supplement was insufficient
istered dose was excreted as unchanged parent regarding topics such as details of statistical
compound. AMPA was the only metabolite found methods, choice of the highest dose tested, and
in the urine (0.20.3% of the administered dose) verification of the target tissue exposure. The
and faeces (0.20.4% of the administered dose). Working Group determined that the informa-
[The large amount of glyphosate excreted in the tion in the supplement to Kier & Kirkland (2013)
faeces is consistent with its poor oral absorption.] did not meet the criteria for data inclusion as laid
Less than 0.3% of the administered dose was out in the Preamble to the IARC Monographs,
expired as carbon dioxide. being neither reports that have been published
In rhesus monkeys given glyphosate as or accepted for publication in the openly avail-
an intravenous dose (9 or 93 g), > 95% of the able scientific literature nor data from govern-
administered dose was excreted in the urine mental reports that are publicly available (IARC,
(Wester et al., 1991). Nearly all the administered 2006). The review article and supplement were
dose was eliminated within 24 hours. In contrast, not considered further in the evaluation.]
in rhesus monkeys given glyphosate by dermal
application (5400 g/20 cm2), only 2.2% of the (a) Humans
administered dose was excreted in the urine (i) Studies in exposed humans
within 7days (Wester et al., 1991). See Table4.1
Overall, systemically absorbed glyphosate In exposed individuals (n = 24) living in
is not metabolized efficiently, and is mainly northern Ecuador in areas sprayed with a glypho-
excreted unchanged into the urine. sate-based formulation, a statistically significant
increase in DNA damage (DNA strand breaks)
4.2 Mechanisms of carcinogenesis was observed in blood cells collected 2 weeks to
2months after spraying (Paz-y-Mio et al., 2007).
4.2.1 Genetic and related effects The same authors studied blood cells from indi-
Glyphosate has been studied for genotoxic viduals (n=92) in 10 communities in Ecuadors
potential in a wide variety of assays. Studies northern border, who were sampled 2years after
carried out in exposed humans, in human cells the last aerial spraying with a herbicide mix
in vitro, in other mammals in vivo and in vitro, containing glyphosate, and showed that their
and in non-mammalian systems in vivo and in karyotypes were normal compared with those of
vitro, respectively, are summarized in Table4.1, a control group (Paz-y-Mio et al., 2011).
Table 4.2, Table 4.3, Table 4.4, and Table 4.5. Bolognesi et al. (2009) studied community
[A review article by Kier & Kirkland (2013) residents (137 women of reproductive age and
summarized the results of published articles their 137 spouses) from five regions in Colombia.
and unpublished reports of studies pertaining In three regions with exposures to glypho-
to the genotoxicity of glyphosate and glypho- sate-based formulations from aerial spraying,
sate formulations. A supplement to this report blood samples were taken from the same indi-
contained information on 66 unpublished regu- viduals at three time-points (before spraying
latory studies. The conclusions and data tables (baseline), 5 days after spraying and 4 months
for each individual study were included in the after spraying) to determine the frequency of
supplement; however, the primary study reports micronucleus formation in lymphocytes. The
from which these data were extracted were not baseline frequency of binucleated cells with
available to the Working Group. The information micronuclei was significantly higher in subjects
45
IARC Monographs 112
from the three regions where there had been in micronucleus formation in human lympho-
aerial spraying with glyphosate-formulations cytes at levels estimated to correspond to occupa-
and in a fourth region with pesticide exposure tional and residential exposure (Mladinic et al.,
(but not through aerial spraying), compared 2009a). Sister-chromatid exchange was induced
with a reference region (without use of pesti- by glyphosate (Bolognesi et al., 1997), and by
cide). The frequency of micronucleus formation a glyphosate-based formulation (Vigfusson &
in peripheral blood lymphocytes was signifi- Vyse, 1980; Bolognesi et al., 1997) in human
cantly increased, compared with baseline levels lymphocytes exposed in vitro.
in the same individuals, after aerial spraying
with glyphosate-based formulations in each of (b) Experimental systems
the three regions (see Table4.1; Bolognesi et al., (i) Non-human mammals in vivo
2009). Immediately after spraying, subjects who See Table4.3
reported direct contact with the glyphosate-based The ability of glyphosate or a glypho-
spray showed a higher frequency of binucleated sate-based formulation to induce DNA adducts
cells with micronuclei. However, the increase in was studied in mice given a single intraperito-
frequency of micronucleus formation observed neal dose. Glyphosate induced DNA adducts
immediately after spraying was not consistent (8-hydroxy deoxyguanosine) in the liver, but not
with the rates of application used in the regions, in the kidney, while a glyphosate-based formula-
and there was no association between self-re- tion caused a slight increase in DNA adducts in
ported direct contact with pesticide sprays and the kidney, but not in the liver (Bolognesi et al.,
frequency of binucleated cells with micronuclei. 1997). Peluso et al. (1998) showed that a glypho-
In subjects from one but not other regions, the sate-based formulation (glyphosate, 30.4%), but
frequency of binucleated cells with micronu- not glyphosate alone, caused DNA adducts (as
clei was significantly decreased 4 months after detected by 32P-DNA post-labelling) in mouse
spraying, compared with immediately after liver and kidney. Glyphosate and a glypho-
spraying. sate-based formulation produced DNA strand
(ii) Human cells in vitro breaks in the liver and kidney after a single intra-
See Table4.2 peritoneal dose (Bolognesi et al., 1997).
Glyphosate induced DNA strand breaks (as In mice given a single dose of glyphosate by
measured by the comet assay) in liver Hep-2 cells gavage, no genotoxic effect was observed by the
(Maas et al., 2009a), lymphocytes (Mladinic dominant lethal test (EPA, 1980a).
et al., 2009b; Alvarez-Moya et al., 2014), GM38 After a single intraperitoneal dose, no
fibroblasts, the HT1080 fibrosarcoma cell line chromosomal aberrations were observed in the
(Monroy et al., 2005), and the TR146 buccal bone marrow of rats treated with glyphosate (Li
carcinoma line (Koller et al., 2012). DNA strand & Long 1988), while chromosomal aberrations
breaks were induced by AMPA in Hep-2 cells were increased in the bone marrow of mice given
(Maas et al., 2009b), and by a glyphosate-based a glyphosate-based formulation (glyphosate
formulation in the TR146 buccal carcinoma cell isopropylamine salt, ~41%) (Prasad et al., 2009).
line (Koller et al., 2012). A single oral dose of a glyphosate-based formu-
In human lymphocytes, AMPA (Maas et al., lation did not cause chromosomal aberrations in
2009b), but not glyphosate (Maas et al., 2009a), mice (Dimitrov et al., 2006).
produced chromosomal aberrations. Glyphosate In mice treated by intraperitoneal injec-
did not induce a concentration-related increase tion, a single dose of glyphosate did not cause
46
Table 4.1 Genetic and related effects of glyphosate in exposed humans
Tissue Cell type End-point Test Description of exposure and controls Responsea/ Comments Reference
(if specified) significance
Blood NR DNA damage DNA strand 24 exposed individuals in northern + P<0.001 Paz-y-Mio et al.
breaks, comet Ecuador; areas sprayed with glyphosate- (2007)
assay based formulation (sampling 2 weeks to
2months after spraying); control group
was 21 non-exposed individuals
Blood NR Chromosomal Chromosomal 92 individuals in 10 communities, 182 karyotypes were Paz-y-Mio et al.
damage aberrations northern border of Ecuador; sampling considered normal (2011)
2years after last aerial spraying with [Smoking status, NR]
herbicide mix containing glyphosate);
control group was 90 healthy individuals
from several provinces without
background of smoking or exposure to
genotoxic substances (hydrocarbons,
X-rays, or pesticides)
Blood Lymphocytes Chromosomal Micronucleus 55 community residents, Nario, + [P<0.001] P values for after Bolognesi et al.
damage formation Colombia; area with aerial glyphosate- spraying vs before (2009)
based formulation spraying for coca and spraying in the same
poppy eradication (glyphosate was tank- individuals
mixed with an adjuvant)
Blood Lymphocytes Chromosomal Micronucleus 53 community residents, Putumayo, + [P=0.01] P values for after Bolognesi et al.
damage formation Colombia; area with aerial glyphosate- spraying vs before (2009)
based formulation spraying for coca and spraying in the same
poppy eradication (glyphosate was tank- individuals
mixed with an adjuvant)
Blood Lymphocytes Chromosomal Micronucleus 27 community residents, Valle del Cauca, + [P<0.001] P values for after Bolognesi et al.
damage formation Colombia; area where glyphosate-based spraying vs before (2009)
formulation was applied through aerial spraying in the same
spraying for sugar-cane maturation individuals
(glyphosate was applied without
adjuvant)
a +, positive; , negative
47
Glyphosate
IARC Monographs 112
48
Table 4.2 Genetic and related effects of glyphosate, AMPA, and glyphosate-based formulations in human cells in vitro
49
Glyphosate
50
Table 4.2 (continued)
Liver Hep-2 DNA damage DNA strand breaks, + NT 4.5 mM P<0.05 at 4.5 mM; Maas et al. (2009b)
comet assay [500 g/mL] P<0.01 at up to
7.5mM
Doseresponse
relationship (r 0.90;
P<0.05)
Lymphocytes Chromosomal Chromosomal + NT 1.8 mM P<0.05 Maas et al. (2009b)
damage aberrations [200 g/mL]
Glyphosate-based formulations
Liver HepG2 DNA damage DNA strand breaks, (+) NT 5 ppm Glyphosate, 400 g/L Gasnier et al. (2009)
comet assay Dose-dependent
increase; greatest
increase at 10 ppm
Statistical analysis, NR
Buccal carcinoma DNA damage DNA strand breaks, + NT 20 g/mL Glyphosate acid, Koller et al. (2012)
TR146 SCGE assay 450g/L
Dose-dependent
increase (P0.05)
Lymphocytes Chromosomal Sister-chromatid + NT 250 g/mL P<0.001 Vigfusson & Vyse
damage exchange No growth at 25mg/ (1980)
mL
Lymphocytes Chromosomal Sister-chromatid + NT 100 g/mL Glyphosate, 30.4% Bolognesi et al.
damage exchange P<0.05 (1997)
a +, positive; -, negative; (+) or (-) positive/negative in a study with limited quality
AMPA, aminomethyl phosphonic acid; HID, highest ineffective dose; hOGG1, human 8-hydroxyguanosine DNA-glycosylase; LED, lowest effective dose; NR, not reported; NT, not
tested; S9, 9000 g supernatant; SCGE, single cell gel electrophoresis; vs, versus
Glyphosate
51
52
Table 4.3 Genetic and related effects of glyphosate, AMPA, and glyphosate-based formulations in non-human mammals in vivo
Species, strain Tissue End-point Test Results Dose (LED or Route, duration, Comments Reference
(sex) HID) dosing regimen
Glyphosate
Mouse, Swiss Liver DNA damage DNA adducts, + 300 mg/kg bw i.p.; 1; sampled Single dose tested only Bolognesi et al.
CD1 8-OHdG by after 8 and 24 h P<0.05 after 24 h (1997)
(M) LC/UV
Mouse, Swiss Kidney DNA damage DNA adducts, 300 mg/kg bw i.p.; 1; sampled Single dose tested only Bolognesi et al.
CD1 8-OHdG by after 8 and 24 h (1997)
IARC Monographs 112
(M) LC/UV
Mouse, Swiss Kidney DNA damage DNA adducts, 270 mg/kg bw i.p.; 1; sampled Glyphosate Peluso et al. (1998)
CD1 32P-DNA post after 24 h isopropylammonium salt
(M, F) labelling
Mouse, Swiss Liver DNA damage DNA adducts, 270 mg/kg bw i.p.; 1; sampled Glyphosate Peluso et al. (1998)
CD1 32P-DNA post after 24 h isopropylammonium salt
(M, F) labelling
Mouse, Swiss Liver DNA damage DNA strand + 300 mg/kg bw i.p.; 1; sampled Single dose tested only Bolognesi et al.
CD1 breaks, alkaline after 4 and 24 h P<0.05 after 4 h (1997)
(M) elution assay
Mouse, Swiss Kidney DNA damage DNA strand + 300 mg/kg bw i.p.; 1; sampled Single dose tested only Bolognesi et al.
CD1 breaks, alkaline after 4 and 24 h P<0.05 after 4 h (1997)
(M) elution assay
Mouse, CD-1 Uterus Mutation Dominant 2000 mg/kg bw Oral gavage; 1 Proportion of early EPA (1980)
(M) after lethal test resorptions evaluated after
mating mating of non-treated
females with glyphosate-
treated male mice
Rat, Sprague- Bone Chromosomal Chromosomal 1000 mg/kg bw i.p.; 1; sampled Single dose tested only Li & Long (1988)
Dawley marrow damage aberrations after 6, 12 and 24 h
(M, F)
Mouse, NMRI- Bone Chromosomal Micronucleus 200 mg/kg bw i.p.; 1; sampled Glyphosate Rank et al. (1993)
bom marrow damage formation after 24 and 48 h isopropylamine salt
(M, F) (PCE)
Mouse, Swiss Bone Chromosomal Micronucleus + 300 mg/kg bw i.p.; 2150 mg/ Single dose tested only Bolognesi et al.
CD1 marrow damage formation kg bw with 24 h P<0.05 after 24 h (1997)
(M) (PCE) interval; sampled
6 or 24 h after the
last injection
Table 4.3 (continued)
Species, strain Tissue End-point Test Results Dose (LED or Route, duration, Comments Reference
(sex) HID) dosing regimen
Mouse, Balb C Bone Chromosomal Micronucleus + 400 mg/kg bw i.p.; one injection P<0.01 at the highest dose Maas et al.
(M, F) marrow damage formation per 24 h, 2 200, (400 mg/kg bw) (2009a)
(PCE) sampled 24 h after
the last injection
AMPA
Mouse, Balb C Bone Chromosomal Micronucleus + 200 mg/kg bw i.p.; one injection P<0.01 at the lowest dose Maas et al.
(M, F) marrow damage formation per 24 h, 2 100, (200 mg/kg bw) (2009b)
(PCE) sampled 24 h after
the last injection
Glyphosate-based formulations
Mouse, Swiss Liver DNA damage DNA adducts, ~300 mg/kg bw i.p.; 1, sampled Glyphosate, 30.4% Bolognesi et al.
CD1 8-OHdG by after 8 and 24 h Single dose tested only (1997)
(M) LC/UV
Mouse, Swiss Kidney DNA damage DNA adducts, + ~300 mg/kg bw i.p.; 1, sampled Glyphosate, 30.4% Bolognesi et al.
CD1 8-OHdG by after 8 and 24 h Single dose tested only (1997)
(M) LC/UV P<0.05
Mouse, Swiss Kidney DNA damage DNA adducts, + 400 mg/kg bw i.p.; 1; sampled Glyphosate Peluso et al. (1998)
CD1 32P-DNA post after 24 h isopropylammonium salt,
(M, F) labelling 30.4%
Mouse, Swiss Liver DNA damage DNA adducts, + 400 mg/kg bw i.p.; 1; sampled Glyphosate Peluso et al. (1998)
CD1 32P-DNA post after 24 h isopropylammonium salt,
(M, F) labelling 30.4%
Mouse, Swiss Liver DNA damage DNA strand + ~300 mg/kg bw i.p.; 1; sampled Glyphosate, 30.4% Bolognesi et al.
CD1 breaks, alkaline after 4 and 24 h Single dose tested only (1997)
(M) elution assay P<0.05 only after 4 h
Mouse, Swiss Kidney DNA damage DNA strand + ~300 mg/kg bw i.p.; 1; sampled Glyphosate, 30.4% Bolognesi et al.
CD1 breaks, alkaline after 4and 24h Single dose tested only (1997)
(M) elution assay P<0.05 only after 4 h
Mouse, C57BL Bone Chromosomal Chromosomal 1080 mg/kg bw p.o. in distilled Single dose tested only Dimitrov et al.
(M) marrow damage aberrations water; 1; (2006)
(PCE) sampled after 6,
24, 48, 72, 96 and
120h
53
Glyphosate
54
Table 4.3 (continued)
Species, strain Tissue End-point Test Results Dose (LED or Route, duration, Comments Reference
(sex) HID) dosing regimen
Mouse, Swiss Bone Chromosomal Chromosomal + 25 mg/kg bw i.p.; 1; sampled Glyphosate Prasad et al. (2009)
albino marrow damage aberrations after 24, 48 and isopropylamine salt, >41%
(M) 72 h The percentage of aberrant
cells was increased vs
control in a dose- and
time-dependent manner
IARC Monographs 112
(P<0.05)
Mouse, NMRI- Bone Chromosomal Micronucleus 200 mg/kg bw i.p.; 1; sampled Glyphosate Rank et al. (1993)
bom marrow damage formation after 24 h isopropylammonium salt,
(M, F) (PCE) 480g/L
The percentage of PCE
decreased
Mouse, Swiss Bone Chromosomal Micronucleus 200 mg/kg bw i.p.; 2within Glyphosate Grisolia (2002)
(M, F) marrow damage formation 24 h interval and isopropylammonium salt,
(PCE) sampled 24 h after 480g/L
the last injection
Mouse, Swiss Bone Chromosomal Micronucleus + 25 mg/kg bw i.p.; 1; sampled Glyphosate Prasad et al. (2009)
albino marrow damage formation after 24, 48 and isopropylamine salt, >41%
(M) (PCE) 72 h Significant induction of
micronuclei vs control at
both doses and all times
(P<0.05)
Mouse, Swiss Bone Chromosomal Micronucleus + 450 mg/kg bw i.p.; 2225 mg/kg Glyphosate, 30.4% Bolognesi et al.
CD1 marrow damage formation with 24 h interval; Single dose tested only (1997)
(M) (PCE) sampled 6 or 24 P<0.05 after 6 h and 24 h
h after the last
injection
Mouse, C57BL Bone Chromosomal Micronucleus 1080 mg/kg bw p.o. in distilled Single dose tested only Dimitrov et al.
(M) marrow damage formation water; 1; (2006)
sampled after 24,
48, 72, 96 and
120 h
a +, positive; -, negative; (+) or (-) positive/negative in a study with limited quality
bw, body weight; F, female; h, hour; HID, highest effective dose; i.p., intraperitoneal; LC, liquid chromatography; LED, lowest effective dose; M, male; PCE, polychromatic erythrocytes;
p.o., oral; 8-OHdG, 8-hydroxydeoxyguanosine; UV, ultraviolet
Table 4.4 Genetic and related effects of glyphosate, AMPA, and glyphosate-based formulations in non-human mammalian
cells in vitro
AMPA, aminomethyl phosphonic acid; HIC, highest ineffective concentration; Hprt, hypoxanthine guanine phosphoribosyl transferase gene; LEC, lowest effective concentration; NT,
not tested
55
Glyphosate
56
Table 4.5 Genetic and related effects of glyphosate, AMPA, and glyphosate-based formulations in non-mammalian systems
in vivo
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Glyphosate
Fish Prochilodus lineatus DNA damage DNA strand + 0.48 mg/L Time of exposure 6, 24, and Moreno et al. (2014)
(sbalo), erythrocytes breaks, comet 96h
and gill cells assay For erythrocytes, P=0.01
after 6h, and P=0.014 after
IARC Monographs 112
96 h; no significant increase
after 24h
For gill cells, P=0.02 only
after 6 h at 2.4 mg/L
Fish Anguilla anguilla L. DNA damage DNA strand + 0.0179 mg/L Time of exposure 1 and Guilherme et al.
(European eel), blood breaks, comet 3days (2012b)
cells assay P<0.05
Fish Danio rerio DNA damage DNA strand + 10 mg/L After 96 h, DNA Lopes et al. (2014)
(zebrafish), sperm breaks, integrity was 78.33.5%,
acridine significantly reduced from
orange control (94.70.9%) and
method 5 mg/L (92.61.9%),
(P<0.05)
Fish Oreochromis DNA damage DNA strand + 7 M Time of exposure, 10 days Alvarez-Moya et al.
niloticus (Nile breaks, comet [1.2mg/L] P<0.001 with (2014)
tilapia) branchial assay concentrations 7 M
erythrocytes
Oyster Oyster spermatozoa DNA damage DNA strand 0.005 mg/L Time of exposure, 1h Akcha et al. (2012)
breaks, comet
assay
Insect Drosophila standard Mutation SMART + 1 mM Purity, 96% Kaya et al. (2000)
cross [0.169mg/L] Increased frequency of
small single spots (1 mM)
and total spots (2mM)
P = 0.05
Insect Drosophila Mutation SMART 10 mM Purity, 96% Kaya et al. (2000)
melanogaster, high [1.69mg/L]
bioactivation cross
Table 4.5 (continued)
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Plant systems Tradescantia clone DNA damage DNA strand + 0.0007 mM Glyphosate isopropylamine Alvarez-Moya et al.
4430 (spiderworts), breaks, comet [0.12g/mL] salt (2011)
staminal hair nuclei assay P<0.01 for directly
exposed nuclei (dose-
dependent increase) and
plants
Plant systems Allium cepa (onion) Chromosomal Chromosomal + 3% Single dose tested only Frescura et al. (2013)
damage aberrations Partial but significant
reversal with distilled water
Plant systems Allium cepa (onion) Chromosomal Chromosomal 2.88 g/mL Glyphosate isopropylamine Rank et al. (1993)
damage aberrations
Plant systems Trigonella foenum- Chromosomal Chromosomal + 0.2% P<0.001; positive dose Siddiqui et al. (2012)
graecum L. damage aberrations response relationship
(fenugreek)
Plant systems Vicia faba (bean) Chromosomal Micronucleus 1400 ppm Tested with two types of De Marco et al.
damage formation (1400g/g of soil, but not without soil (1992)
soil)
AMPA
Fish Anguilla anguilla L. DNA damage DNA strand + 0.0118 mg/L Time of exposure, 1 and Guilherme et al.
(European eel) breaks, comet 3days (2014b)
assay P<0.05 after 1day of
exposure
Fish Anguilla anguilla L. Chromosomal Other (ENA) + 0.0236 mg/L P<0.05 only at highest Guilherme et al.
(European eel) damage dose after 3day exposure (2014b)
(not after 1 day)
Glyphosate-based formulations
Fish Anguilla anguilla L. DNA damage DNA strand + 0.058 mg/L P<0.05 Guilherme et al.
(European eel), blood breaks, comet Positive doseresponse (2010)
cells assay relationship
Fish Anguilla anguilla L. DNA damage DNA strand + 0.058 mg/L Glyphosate-based Guilherme et al.
(European eel), blood breaks, formulation, 30.8% (2012b)
cells comet assay Time of exposure, 1 and
improved with 3days
the DNA- With FPG, P<0.05; with
lesion-specific comet assay alone, P< 0.05
FPG and Endo at 116 g/L
III
57
Glyphosate
58
Table 4.5 (continued)
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Fish Anguilla anguilla L. DNA damage DNA strand + 0.116 mg/L Single dose tested only Guilherme et al.
(European eel), blood breaks, Time of exposure, 3days; (2014a)
cells comet assay recovery from non-specific
improved with DNA damage, but not
the DNA- oxidative DNA damage, 14
lesion-specific days after exposure
IARC Monographs 112
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Fish Corydoras paleatus DNA damage DNA strand + 0.0067 mg/L Glyphosate, 48% de Castilhos Ghisi &
(blue leopard breaks, comet (corresponding to Cestari (2013)
corydoras, mottled assay 3.20g/L)
corydoras and Single dose tested only, for
peppered catfish), 3, 6, and 9days
blood and hepatic P<0.01, in blood and in
cells liver cells
Fish Cyprinus carpio DNA damage DNA strand + 2 mg/L (10% Glyphosate, equivalent to Gholami-Seyedkolaei
Linnaeus (carp), breaks, comet LC50, 96h) 360 g/L et al. (2013)
erythrocytes assay Single dose tested only, for
16 days
P<0.01
Fish Carassius auratus DNA damage DNA strand + 5 ppm Glyphosate equivalent to Cava & Knen
(goldfish), breaks, comet 360 g/L (2007)
erythrocytes assay Time of exposure, 2, 4 and
6days
After 48 h: P<0.05
(5mg/L) and P<0.001 (10
and 15 mg/L)
Fish Prochilodus lineatus Chromosomal Micronucleus 10 mg/L Single dose tested only, for Cavalcante et al.
(sbalo) erythrocytes damage formation 6, 24, and 96h (2008)
Nuclear abnormalities
(lobed nuclei, segmented
nuclei and kidney-shaped
nuclei)
Fish Corydoras paleatus Chromosomal Micronucleus 0.0067 mg/L Glyphosate, 48% de Castilhos Ghisi &
(blue leopard damage formation (corresponding to Cestari (2013)
corydoras, mottled 3.20g/L)
corydoras and Single dose tested only, for
peppered catfish), 3, 6 and 9days
blood and hepatic
cells
59
Glyphosate
60
Table 4.5 (continued)
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Fish Tilapia rendalli Chromosomal Micronucleus + 42mg/kg bw Glyphosate, 480 g/L Grisolia (2002)
(redbreast tilapia) damage formation Increased frequency of
blood erythrocytes micronucleus formation
vs control (P<0.05) in
blood samples collected 4
days after a single intra-
IARC Monographs 112
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Fish Prochilodus lineatus Chromosomal Chromosomal 10 mg/L Single dose tested only, for Cavalcante et al.
(sbalo) erythrocytes damage aberrations 6, 24, and 96h (2008)
Nuclear abnormalities
(lobed nuclei, segmented
nuclei and kidney-shaped
nuclei)
Fish Anguilla anguilla Chromosomal Other (ENA) + 0.058 mg/L Time of exposure, 1 and Guilherme et al.
L. (European eel), damage 3days (2010)
peripheral mature Chromosomal breakage
erythrocytes and/or chromosomal
segregational abnormalities
after 3days of exposure,
P<0.05
Caiman Caiman latirostris DNA damage DNA strand + 0.500 mg/egg Glyphosate, 66.2% Poletta et al. (2009)
(broad-snouted breaks, comet In-ovo exposure; blood
caiman), erythrocytes assay sampling at the time of
hatching
P<0.05 in both
experiments (501000g/
egg in experiment 1; 500
1750g/egg in experiment 2)
Caiman Caiman latirostris DNA damage DNA strand 19 800 mg/L Glyphosate, 66.2% Poletta et al. (2011)
(broad-snouted breaks, comet Single dose tested only; in-
caiman), erythrocytes assay ovo exposure
First spraying exposure
at the beginning of
incubation period, a second
exposure on day 35, then
incubation until hatching
Caiman Caiman latirostris Chromosomal Micronucleus + 0.500 mg/egg Glyphosate, 66.2% Poletta et al. (2009)
(broad-snouted damage fomation In-ovo exposure; blood
caiman), erythrocytes sampling at the time of
hatching
P<0.05 in both
experiments (501000g/
egg in experiment 1; 500
1750g/egg in experiment 2)
61
Glyphosate
62
Table 4.5 (continued)
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Caiman Caiman latirostris Chromosomal Micronucleus + 19.8g/L Glyphosate, 66.2% Poletta et al. (2011)
(broad-snouted damage fomation One dose tested; in-ovo
caiman), erythrocytes exposure
First spraying exposure
at the beginning of
incubation period, a second
IARC Monographs 112
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Clam Corbicula fluminea DNA damage DNA strand 10 mg/L Time of exposure, 96 h dos Santos &
(Asian clam) breaks, comet Significant increase when Martinez (2014)
haemocytes assay atrazine (2 or 10mg/L)
was added to glyphosate
(P<0.05)
No increase after exposure
to atrazine or glyphosate
separately
Mussels Utterbackia imbecillis DNA damage DNA strand 5 mg/L Glyphosate, 18% Conners & Black
(Bivalvia: Unionidae) breaks, comet Doses tested: 2.5 and (2004)
glochidia mussels assay 5mg/L for 24h
(larvae) NOEC, 10.04mg/L
Worm Earthworm, Eisenia DNA damage DNA strand 240 g a.e./cm2 Monoammonium salt, Piola et al. (2013)
andrei, coelomocytes breaks, comet 85.4%, a.e.
assay Epidermic exposure during
72 h (on filter paper)
Worm Earthworm, Eisenia DNA damage DNA strand + 15 g a.e./cm2 Monoammonium salt, Piola et al. (2013)
andrei, coelomocytes breaks, comet 72%, a.e.
assay Epidermic exposure during
72h (on filter paper)
P<0.001
Worm Earthworm, DNA damage DNA strand 251.50g/cm2 Active ingredient, 36% Muangphra et al.
Pheretima peguana, breaks, comet (w/v) (2014)
coelomocytes assay Epidermic exposure 48 h on
filter paper; LC50, 251.50g/
cm2
Worm Earthworm, Chromosomal Micronucleus + 251.50 g/cm2 Active ingredient, 36% Muangphra et al.
Pheretima peguana, damage formation (w/v) (2014)
coelomocytes Exposure, 48h on filter
paper; LC50, 251.50g/cm2
filter paper
P<0.05, for total
micro-, bi-, and trinuclei
frequencies at 0.25g/cm2;
when analysed separately,
micro- and trinuclei
frequencies significantly
differed from controls only
63
Glyphosate
at the LC50
64
Table 4.5 (continued)
Phylogenetic Species, strain, tissue End-point Test Resultsa Dose Comments Reference
class (LED or HID)
Insect Drosophila Mutation Sex-linked + 1ppm Single dose tested only Kale et al. (1995)
melanogaster recessive P<0.001
lethal
mutations
Plant systems Allium cepa (onion) Chromosomal Chromosomal + 1.44g/mL Glyphosate-based Rank et al. (1993)
damage aberrations formulation, 480g/L
IARC Monographs 112
a.e., acid equivalent; AMPA, aminomethyl phosphonic acid; bw, body weight; ENA, erythrocytic nuclear abnormalities; Endo III, endonuclease III; FPG, formamidopyrimidine
glycosylase; h, hour; HID, highest ineffective dose; LC50, median lethal dose; LED, lowest effective dose; NOEC, no-observed effect concentration; p.o., oral; SMART, somatic mutation
and recombination test
Table 4.6 Genetic and related effects of glyphosate and glyphosate-based formulations on non-mammalian systems in vitro
65
Glyphosate
66
Table 4.6 (continued)
Additionally, although all four glyphosate-based between a nitrogen atom of glyphosate and
formulations dramatically reduced the transcrip- the active site of the purified equine aromatase
tion of ER and ER in ERE-transfected HepG2 enzyme.
cells, glyphosate alone had no significant effect. In the mouse MA-10 Leydig cell tumour cell
Glyphosate and all four formulations reduced line, a glyphosate-based formulation (glypho-
androgen-receptor transcription in the breast sate, 180 mg/L) markedly reduced [(Bu)2]
cancer cell line MDA-MB453-kb2, which has a cAMP-stimulated progesterone production
high level of androgen receptor, with the formu- (Walsh et al., 2000). The inhibition was dose-de-
lations showing greater activity than glyphosate pendent, and occurred in the absence of toxicity
alone. or parallel reductions in total protein synthesis.
In a human placental cell line derived from In companion studies, the formulation also
choriocarcinoma (JEG3 cells), 18 hours of disrupted steroidogenic acute regulatory protein
exposure to a glyphosate-based formulation expression, which is critical for steroid hormone
(IC50 = 0.04%) decreased aromatase activity synthesis. Glyphosate alone did not affect steroi-
(Richard et al., 2005). Glyphosate alone was dogenesis at any dose tested up to 100 g/L.
without effect. The concentrations used did not Forgacs et al. (2012) found that glyphosate (300 M)
affect cell viability. had no effect on testosterone production in a novel
Glyphosate, at non-overtly toxic concen- murine Leydig cell line (BLTK1). Glyphosate did
trations, decreased aromatase activity in fresh not modulate the effect of recombinant human
human placental microsomes and transformed chorionic gonadotropin, which served as the
human embryonic kidney cells (293) transfected positive control for testosterone production.
with human aromatase cDNA (Benachour
(iii) Non-mammalian experimental systems
et al., 2007). A glyphosate-based formulation, at
non-overtly toxic concentrations, had the same Gonadal tissue levels of testosterone, 17-estra-
effect. The formulation was more active at equiv- diol and total microsomal protein were signifi-
alent doses than glyphosate alone. cantly reduced in adult snails (Biomphalaria
In human androgen receptor and ER and alexandrina) exposed for 3 weeks to a glypho-
ER reporter gene assays using the Chinese sate-based formulation (glyphosate, 48%) at
hamster ovary cell line (CHO-K1), glypho- the LC10 (10% lethal concentration) (Omran
sate had neither agonist nor antagonist activity & Salama, 2013). These effects persisted after a
(Kojima et al., 2004, 2010). 2-week recovery period, although the impact
on 17-estradiol was reduced in the recovery
(ii) Non-human mammalian experimental animals. The formulation also induced marked
systems degenerative changes in the ovotestis, including
absence of almost all the gametogenesis stages.
In vivo
CYP450 1B1, measured by enzyme-linked
No data were available to the Working Group. immunosorbent assay (ELISA), was substantially
In vitro increased in the treated snails, including after the
Benachour et al. (2007) and Richard et al. recovery period.
(2005) reported that glyphosate and a glypho- Glyphosate (0.11 mg/L for 7 days) did not
sate-based formulation inhibited aromatase increase plasma vittelogenin levels in juvenile
activity in microsomes derived from equine rainbow trout (Xie et al., 2005).
testis. Richard et al. (2005) reported an absorb-
ance spectrum consistent with an interaction
67
IARC Monographs 112
(b) Other pathways gene assay was increased by the formulation, and
(i) Humans a retinoic acid antagonist blocked the effect. This
indicated a possible significant modulation of
Studies in exposed humans retinoic acid activity by glyphosate.
No data were available to the Working Group.
4.2.3 Oxidative stress, inflammation, and
Human cells in vitro
immunosuppression
Glyphosate did not exhibit agonist activity in
an assay for a human pregnane X receptor (PXR) (a) Oxidative stress
reporter gene in a CHO-K1 cell line (Kojima (i) Humans
et al., 2010).
Studies in exposed humans
(ii) Non-human mammalian experimental
systems No data were available to the Working Group.
Human cells in vitro
In vivo
Several studies examined the effects of
In rats, glyphosate (300 mg/kg bw, 5days per glyphosate on oxidative stress parameters in the
week, for 2 weeks) had no effect on the formation human keratinocyte cell line HaCaT. Gehin et al.
of peroxisomes, or the activity of hepatic carni- (2005) found that a glyphosate-based formu-
tine acetyltransferase and catalase, and did not lation was cytotoxic to HaCaT cells, but that
cause hypolipidaemia, suggesting that glyphosate addition of antioxidants reduced cytotoxicity.
does not have peroxisome proliferator-activated Elie-Caille et al. (2010) showed that incubation
receptor activity (Vainio et al., 1983). of HaCaT cells with glyphosate at 21 mM (the
In vitro half maximal inhibitory concentration for cyto-
Glyphosate was not an agonist for mouse toxicity, IC50) for 18 hours increased production
peroxisome proliferator-activated receptors of hydrogen peroxide (H2O2) as shown by dichlo-
PPAR or PPAR in reporter gene assays using rodihydrofluorescein diacetate assay. Similarly,
CV-1 monkey kidney cells in vitro (Kojima et al., George & Shukla (2013) exposed HaCaT cells
2010). Glyphosate was also not an agonist for the to a glyphosate-based formulation (glyphosate,
aryl hydrocarbon receptor in mouse hepatoma 41%; concentration, up to 0.1 mM) and evalu-
Hepa1c1c7 cells stably transfected with a reporter ated oxidative stress using the dichlorodihydro
plasmid containing copies of dioxin-responsive fluorescein diacetate assay. The formulation
element (Takeuchi et al., 2008). (0.1 mM) increased maximum oxidant levels
by approximately 90% compared with vehicle,
(iii) Non-mammalian experimental systems an effect similar to that of H2O2 (100 mM).
As a follow-up to experiments in which Pre-treatment of the cells with the antioxi-
injection of glyphosate, or incubation with a dant N-acetylcysteine abrogated generation of
glyphosate-based formulation (glyphosate, oxidants by both the formulation and by H2O2.
48%), caused chick and frog (Xenopus laevis) N-Acetylcysteine also inhibited cell proliferation
cephalic and neural crest terata characteristic of induced by the glyphosate-based formulation
retinoic acid signalling dysfunction, Paganelli (0.1 mM). [The Working Group noted the recog-
et al., (2010) measured retinoic acid activity in nized limitations of using dichlorodihydrofluo-
tadpoles exposed to a glyphosate-based formu- rescein diacetate as a marker of oxidative stress
lation. Retinoic activity measured by a reporter (Bonini et al., 2006; Kalyanaraman et al., 2012),
68
Glyphosate
and that the studies that reported this end-point (ii) Non-human mammalian experimental
as the sole evidence for oxidative stress should systems
thus be interpreted with caution.] Most of the studies of oxidative stress and
Chaufan et al. (2014) evaluated the effects glyphosate were conducted in rats and mice, and
of glyphosate, AMPA (the main metabolite of examined a range of exposure durations, doses,
glyphosate), and a glyphosate-based formulation preparations (glyphosate and glyphosate-based
on oxidative stress in HepG2 cells. The formula- formulations), administration routes and tissues.
tion, but not glyphosate or AMPA, had adverse In addition, various end-points were evaluated
effects. Specifically, the formulation increased to determine whether oxidative stress is induced
levels of reactive oxygen species, nitrotyrosine by exposure to glyphosate. Specifically, it was
formation, superoxide dismutase activity, and found that glyphosate induces production of free
glutathione, but did not have an effect on cata- radicals and oxidative stress in mouse and rat
lase or glutathione-S-transferase activities. tissues through alteration of antioxidant enzyme
Coalova et al. (2014) exposed Hep2 cells to a activity, depletion of glutathione, and increases
glyphosate-based formulation (glyphosate as in lipid peroxidation. Increases in biomarkers of
isopropylamine salt, 48%) at the LC20 (concen- oxidative stress upon exposure to glyphosate in
tration not otherwise specified) and evaluated vivo have been observed in blood plasma (Astiz
various parameters of oxidative stress. Exposure et al., 2009b), liver (Bolognesi et al., 1997; Astiz
to the formulation for 24 hours increased catalase et al., 2009b), skin (George et al., 2010), kidney
activity and glutathione levels, but did not have (Bolognesi et al., 1997; Astiz et al., 2009b), and
an effect on superoxide dismutase or glutathione- brain (Astiz et al., 2009b). Several studies demon-
S-transferase activity. strated similar effects with a glyphosate-based
Using blood samples from non-smoking formulation in the liver (Bolognesi et al., 1997;
male donors, Mladinic et al. (2009b) examined Cavuolu et al., 2011; Jasper et al., 2012), kidney
the effects of in-vitro exposure to glyphosate on (Bolognesi et al., 1997; Cavuolu et al., 2011)
oxidative DNA damage in primary lymphocyte and brain (Cattani et al., 2014), or with a pesti-
cultures and on lipid peroxidation in plasma. Both cide mixture containing glyphosate in the testes
parameters were significantly elevated at glypho- (Astiz et al., 2013). Pre-treatment with antioxi-
sate concentrations of 580 g/mL (~3.4 mM), dants has been shown to mitigate the induction
but not at lower concentrations. Kwiatkowska of oxidative stress by a glyphosate-based formu-
et al. (2014) examined the effects of glyphosate, lation (Cavuolu et al., 2011) and by a pesticide
its metabolite AMPA, and N-methylglyphosate mixture containing glyphosate (Astiz et al., 2013).
(among other related compounds) in human DNA damage associated with oxidative stress
erythrocytes isolated from healthy donors. The after exposure to glyphosate (e.g. as reported in
erythrocytes were exposed at concentrations Bolognesi et al., 1997) is reviewed in Section 4.2.1.
of 0.015 mM for 1, 4, or 24 hours before flow
cytometric measurement of the production of (iii) Non-mammalian experimental systems
reactive oxygen species with dihydrorhodamine Positive associations between exposure to
123. Production of reactive oxygen species was glyphosate and oxidative stress were reported in
increased by glyphosate ( 0.25 mM), AMPA various tissues in aquatic organisms (reviewed in
(0.25 mM), and N-methylglyphosate (0.5 mM). Slaninova et al., 2009). Glyphosate and various
glyphosate-based formulations have been tested
in various fish species for effects on a plethora
of end-points (e.g. lipid peroxidation, DNA
69
IARC Monographs 112
70
Glyphosate
71
IARC Monographs 112
Several studies evaluated the impact of glypho- of cell viability, that was similar to that shown
sate or glyphosate-based formulations on apop- by one formulation. The calculated 50% lethal
totic cell death in the HepG2 human hepatoma concentration in JEG3 cells for mitochondrial
cell line. Glyphosate-based formulations induced succinate dehydrogenase activity was greater for
apoptosis in HepG2 cells, while glyphosate alone three formulations, although the values appeared
was generally without effect or showed effects inconsistent with the concentrationresponse
at considerably higher concentrations (Gasnier curves.]
et al., 2009, 2010; Mesnage et al., 2013; Chaufan et In HUVEC primary neonate umbilical cord
al., 2014; Coalova et al., 2014). For example, 23.5% vein cells, and 293 embryonic kidney and JEG3
of the nuclei of HepG2 cells exposed to a glypho- placental cell lines, Benachour & Sralini (2009)
sate-based formulation showed condensed and found that glyphosate at relatively high concen-
fragmented chromatin (P < 0.01), and caspases trations induced apoptosis, as indicated by
3 and 7 were significantly activated, both effects induction of caspases 3 and 7, and DNA staining
being indicative of apoptosis (Chaufan et al., and microscopy. At comparable or lower concen-
2014). Caspases were unaffected by glyphosate trations, four glyphosate-based formulations all
or AMPA alone. Glyphosate and AMPA did caused primarily necrotic cell death. The umbil-
not affect cell viability at concentrations up to ical cord HUVEC cells were the most sensitive
1000 mg/L, a concentration that increased rather (by about 100-fold) to the apoptotic effects of
than decreased cell viability after 48 and 72 glyphosate.
hours of incubation. In contrast, cells exposed to Heu et al. (2012) evaluated apoptosis in
glyphosate-based formulation at lower concen- immortalized human keratinocytes (HaCaT)
trations were not viable. Similarly, Coalova et al. exposed to glyphosate (570 mM). Based on
(2014) reported that a glyphosate-based formu- annexin V, propidium iodide and mitochondrial
lation (glyphosate, 48%) induced apoptotic cell staining, exposures leading to 15% cytotoxicity
death in HepG2 cells. Apoptosis was indicated gave evidence of early apoptosis, while increases
by activation of caspases 3 and 7, and the signif- in late apoptosis and necrosis were observed at
icant fraction (17.7%) of nuclei with condensed higher levels of cytotoxicity.
and fragmented chromatin (P<0.001).
In studies with glyphosate and nine different (b) Non-human mammalian experimental
glyphosate-based formulations in three cell lines, systems
glyphosate alone did not increase the activity (i) In vivo
of adenylate kinase (Mesnage et al., 2013). The In male Wistar rats, glyphosate (10 mg/kg
activity of caspases 3 and 7 was significantly bw, injected intraperitoneally three times per
increased by glyphosate in HepG2 and embry- week for 5 weeks) reduced, but not significantly,
onic kidney HEK293 cells, and elevated (although the inner mitochondrial membrane integrity of
not significantly) about 1.8 times above control the substantia nigra and cerebral cortex (Astiz
levels in placental choriocarcinoma JEG-3 cells. et al. 2009a). Caspase 3 activity was unaltered in
Two formulations containing an ethoxylated these tissues. Mitochondrial cardiolipin content
adjuvant induced adenylate kinase activity to a was significantly reduced, particularly in the
greater extent than caspase activity. All formu- substantia nigra, where calpain activity was
lations were reported to be more cytotoxic than substantially higher. Glyphosate induced DNA
glyphosate. [In concentrationresponse curves, fragmentation in the brain and liver.
glyphosate showed an effect on mitochondrial
succinate dehydrogenase activity, a measure
72
Glyphosate
73
IARC Monographs 112
have been reported in many countries, including In the study by the NTP, cytoplasmic alteration
Australia (Stella & Ryan, 2004), Denmark was observed in the parotid and submandibular
(Mortensen et al., 2000), India (Mahendrakar salivary glands of rats (Chan & Mahler, 1992).
et al., 2014), Japan (Motojyuku et al., 2008), In a study of carcinogenicity in mice given
Republic of Korea (Park et al., 2013), New Zealand glyphosate at doses of 150, 1500, or 4500 mg/kg
(Temple & Smith, 1992), Sri Lanka (Roberts et al., bw per day, liver hypertrophy and necrosis were
2010), Taiwan, China (Chen et al., 2009), and observed in males at the highest dose (EPA, 1983).
Thailand (Sribanditmongkol et al., 2012). Other effects in males at the highest dose included
Glyphosate demonstrated no potential for increased testes weight, interstitial nephritis, and
photo-irritation or photo-sensitization in 346 decreased body weight. In females at the highest
volunteers exposed dermally on normal or dose, ovary weights were increased, proximal
abraded skin (Hayes & Laws, 1991). On the other tubule epithelial basophilia and hypertrophy was
hand, Mariager et al. (2013) reported severe burns observed, and body weights were decreased. In
after prolonged accidental dermal exposure to a the study by the NTP, cytoplasmic alteration was
glyphosate-based formulation. observed in the parotid salivary glands in mice
(Chan & Mahler, 1992).
4.5.2 Experimental systems
Developmental and reproductive toxicity
Glyphosate was tested in nine regulatory
In a study of developmental toxicity in
submissions included in the Toxicity Reference
rats given glyphosate at a dose of 300, 1000, or
Database (ToxRefDB) and reviewed by the EPA
3500 mg/kg bw per day, reduced implantation
(EPA, 2015). Specifically, study design, treatment
rates and fewer live fetuses were observed in dams
group, and treatment-related effect information
at the highest dose (EPA, 1980b). In fetuses at the
were captured for four long-term studies and/or
highest dose, unossified sternebra were observed
carcinogenicity studies, one short-term study, two
and fetal weight was reduced.
multigeneration studies of reproductivity, and two
studies of developmental toxicity. The NTP also
tested glyphosate in a 13-week study in rats and
mice (Chan & Mahler, 1992).
5. Summary of Data Reported
In a long-term combined study of toxicity
and carcinogenicity in rats given glyphosate 5.1 Exposure data
at nominal doses of 100, 400, and 1000 mg/kg
Glyphosate is a broad-spectrum herbicide
bw per day, inflammation was observed in the
that is effective at killing or suppressing all
stomach mucosa of females at the intermediate
plant types, including grasses, perennials, and
and highest doses (EPA, 1990, 1991b). In males
woody plants. The herbicidal activity of glypho-
at the highest dose, liver weight, cataracts and
sate was discovered in 1970 and since then its
lens degeneration in the eyes, and urine specific
use has increased to a point where it is now the
gravity were increased, while body weight, body-
most heavily used herbicide in the world, with
weight gain, and urinary pH were decreased.
an annual global production volume in 2012 of
Pancreatic acinar cell atrophy was observed in
more than 700 000 tonnes used in more than
males at the highest dose. Pancreatic inflamma-
750 different products. Changes in farming prac-
tion was also observed in male rats at the highest
tice and the development of genetically modi-
dose in a short-term study (nominal doses of 50,
fied crops that are resistant to glyphosate have
250, and 1000 mg/kg bw per day) (EPA, 1987).
contributed to the increase in use.
74
Glyphosate
There is little information available on occu- studies reported excesses for NHL associated
pational or community exposure to glyphosate. with exposure to glyphosate, after adjustment
Glyphosate can be found in soil, air, surface for other pesticides (reported odds ratio were 2.1
water and groundwater, as well as in food. It (95% CI, 1.14.0); 1.85 (95% CI, 0.556.2); and
has been detected in air during agricultural 1.51 (95% CI, 0.772.94). Subtype-specific anal-
herbicide-spraying operations. Glyphosate was yses in a Swedish casecontrol study indicated
detected in urine in two studies of farmers in positive associations for total NHL, as well as all
the USA, in urban populations in Europe, and in subtypes, but this association was statistically
a rural population living near areas sprayed for significant only for the subgroup of lymphocytic
drug eradication in Columbia. However, urinary lymphoma/chronic lymphocytic leukaemia (OR,
concentrations were mostly below the limit of 3.35; 95% CI, 1.427.89). An elevated risk (OR,
detection in several earlier studies of forestry 3.1; 95% CI, 0.617.1) was also found for B-cell
workers who sprayed glyphosate. Exposure of lymphoma in an European study based on few
the general population occurs mainly through cases. One hospital-based casecontrol study
diet. from France did not find an association between
exposure to glyphosate and NHL (OR, 1.0; 95%
CI, 0.52.2) based on few exposed cases.
5.2 Human carcinogenicity data A roughly twofold excess of multiple myeloma,
In its evaluation of the epidemiological a subtype of NHL, was reported in three studies:
studies reporting on cancer risks associated with only among the highest category of glyphosate
exposure to glyphosate, the Working Group use (>2days/year) in the large Canadian case
identified seven reports from the Agricultural control study, in a casecontrol study from Iowa,
Health Study (AHS) cohort and several reports USA, and in a French casecontrol study (all not
from casecontrol studies. The AHS cohort, the statistically significant). These three studies did
pooled analyses of the casecontrol studies in not adjust for the effect of other pesticides. In the
the midwest USA, and the cross-Canada study AHS, there was no association with NHL (OR,
were considered key investigations because of 1.1; 0.71.9). For multiple myeloma, relative risk
their relatively large size. Reports from two was 1.1 (95% CI, 0.52.4) when adjusted for age
or more independent studies were available only; but was 2.6 (95% CI, 0.79.4) when adjusted
for non-Hodgkin lymphoma (NHL), multiple for multiple confounders. No excess in leukaemia
myeloma, Hodgkin lymphoma, glioma, and was observed in a casecontrol study in Iowa and
prostate. For the other cancer sites, results from Minnesota, USA, or in the AHS.
only one study were available for evaluation. In summary, casecontrol studies in the USA,
Canada, and Sweden reported increased risks
5.2.1 NHL and other haematopoietic cancers for NHL associated with exposure to glyphosate.
The increased risk persisted in the studies that
Two large casecontrol studies of NHL from adjusted for exposure to other pesticides. The
Canada and the USA, and two casecontrol AHS cohort did not show an excess of NHL. The
studies from Sweden reported statistically signif- Working Group noted that there were excesses
icant increased risks of NHL in association with reported for multiple myeloma in three studies;
exposure to glyphosate. For the study in Canada, however, they did not weight this evidence as
the association was seen among those with more strongly as that of NHL because of the possibility
than 2 days/year of exposure, but no adjustment that chance could not be excluded; none of the
for other pesticides was done. The other three
75
IARC Monographs 112
risk estimates were statistically significant nor in the incidences of hepatocellular adenoma in
were they adjusted for other pesticide exposures. males and of thyroid C-cell adenoma in females.
Two studies (one in Sprague-Dawley rats, one
5.2.2. Other cancer sites in Wistar rats) found no significant increase in
tumour incidence at any site. One study in Wistar
No association of glyphosate with cancer rats was inadequate for the evaluation because of
of the brain in adults was found in the Upper the short duration of exposure.
Midwest Health casecontrol study. No associa- In the study in Wistar rats given drinking-water
tions in single casecontrol studies were found for containing glyphosate, there was no significant
cancers of the oesophagus and stomach, prostate, increase in tumour incidence.
and soft-tissue sarcoma. For all other cancer sites A glyphosate-based formulation was found
(lung, oral cavity, colorectal, pancreas, kidney, to be a skin-tumour promoter in the initiation
bladder, breast, prostate, melanoma) investigated promotion study in male Swiss mice. The study of
in the large AHS, no association with exposure to a glyphosate-based formulation in drinking-water
glyphosate was found. in Sprague-Dawley rats was inadequate for the
evaluation because of the small number of animals
5.3 Animal carcinogenicity data per group, and the limited information provided
on tumour histopathology and incidence in indi-
Glyphosate was tested for carcinogenicity in vidual animals. These studies of a chemical mixture
male and female mice by dietary administration containing glyphosate were considered inadequate
in two studies, and in male and female rats by to evaluate the carcinogenicity of glyphosate alone.
dietary administration in five studies and in
drinking-water in one study. A glyphosate-based
formulation was also tested in drinking-water in 5.4. Other relevant data
one study in male and female rats, and by skin Direct data on absorption of glyphosate in
application in one initiationpromotion study in humans were not available to the Working Group.
male mice. Glyphosate was detected in the urine of agricul-
There was a positive trend in the incidence tural workers in several studies, and in the blood
of renal tubule carcinoma and of renal tubule of poisoning cases, indicative of absorption.
adenoma or carcinoma (combined) in males in Some evidence for absorption through human
one feeding study in CD-1 mice. Renal tubule skin (~2%) was reported in studies in vitro.
carcinoma is a rare tumour in this strain of mice. The minor role of dermal absorption was also
No significant increase in tumour incidence was shown in a study in non-human primate model
seen in female mice in this study. In the second in vivo. However, no study examined the rates
feeding study, there was a significant positive of absorption in humans. In rodents, several
trend in the incidence of haemangiosarcoma studies showed up to 40% absorption after oral
in male CD-1 mice. No significant increase in administration of a single or repeated dose.
tumour incidence was seen in female mice in Glyphosate was measured in human blood.
this study. No data on parenchymal tissue distribution
For the five feeding studies in rats, two for glyphosate in humans were available to the
studies in the Sprague-Dawley strain showed a Working Group. In rats given glyphosate by oral
significant increase in the incidence of pancre- administration, concentrations in tissues had
atic islet cell adenoma in males one of these two the following rank order: kidneys >spleen >fat
studies also showed a significant positive trend > liver. Repeated administration had no effect
76
Glyphosate
77
IARC Monographs 112
increased the confidence of the Working Group Overall, the mechanistic data provide strong
in the overall database. evidence for genotoxicity and oxidative stress.
There is weak evidence that glyphosate There is evidence that these effects can operate
or glyphosate-based formulations induce in humans.
receptor-mediated effects. In multiple experi-
ments, glyphosate-based formulations affected
aromatase activity; glyphosate was active in a few 6. Evaluation
of these studies. Some activity in other nuclear
receptor-mediated pathways has been observed 6.1 Cancer in humans
for glyphosate or glyphosate-based formula-
tions. In one series of experiments, glyphosate There is limited evidence in humans for the
was not found to be a ligand to several receptors carcinogenicity of glyphosate. A positive asso-
and related proteins (aryl hydrocarbon receptor, ciation has been observed for non-Hodgkin
peroxisome proliferator-activated receptors, lymphoma.
pregnane X receptor).
There is weak evidence that glyphosate may
affect cell proliferation or death. Several studies
6.2 Cancer in experimental animals
in human and rodent cell lines have reported There is sufficient evidence in experimental
cytotoxicity and cell death, the latter attributed to animals for the carcinogenicity of glyphosate.
the apoptosis pathway. Studies that examined the
effects of glyphosate alone or a glyphosate-based
formulation found that glyphosate alone had no 6.3 Overall evaluation
effect, or a weaker effect than the formulation. Glyphosate is probably carcinogenic to
There is weak evidence that glyphosate may humans (Group 2A).
affect the immune system, both the humoral and
cellular response, upon long-term treatment in
rodents. Several studies in fish, with glyphosate 6.4 Rationale
or its formulations, also reported immunosup-
In making this overall evaluation, the
pressive effects.
Working Group noted that the mechanistic and
With regard to the other key characteristics of
other relevant data support the classification of
human carcinogens (IARC, 2014), the Working
glyphosate in Group 2A.
Group considered that the data were too few for
In addition to limited evidence for the carcino-
an evaluation to be made.
genicity of glyphosate in humans and sufficient
Severe or fatal human poisoning cases have
evidence for the carcinogenicity of glyphosate in
been documented worldwide. In rodents, organ
experimental animals, there is strong evidence
and systemic toxicity from exposures to glypho-
that glyphosate can operate through two key
sate are demonstrated by liver-weight effects and
characteristics of known human carcinogens,
necrosis in animals at high doses. Additionally,
and that these can be operative in humans.
effects on the pancreas, testes, kidney and ovaries,
Specifically:
as well as reduced implantations and unossified
sternebra were seen at similar doses. There is strong evidence that exposure to
No data on cancer-related susceptibility after glyphosate or glyphosate-based formulations
exposure to glyphosate were available to the is genotoxic based on studies in humans in
Working Group. vitro and studies in experimental animals.
78
Glyphosate
79
IARC Monographs 112
80
Glyphosate
administered in dosed feed to F344/N rats and B6C3F1 Cocco P, Satta G, Dubois S, Pili C, Pilleri M, Zucca M
mice. Toxic Rep Ser, 16:158. PMID:12209170 et al. (2013). Lymphoma risk and occupational expo-
Chandra M, Frith CH (1994). Spontaneous renal lesions in sure to pesticides: results of the Epilymph study. Occup
CD-1 and B6C3F1 mice. Exp Toxicol Pathol, 46(3):189 Environ Med, 70(2):918. doi:10.1136/oemed-2012-
98. doi:10.1016/S0940-2993(11)80080-1 PMID:8000238 100845 PMID:23117219
Chang FC, Simcik MF, Capel PD (2011). Occurrence ColomboPage News Desk (2014). Sri Lanka lifts ban on
and fate of the herbicide glyphosate and its degradate sale of glyphosate. ColomboPage, Sri Lanka Internet
aminomethylphosphonic acid in the atmosphere. Newspaper [online newspaper]. 13 May, 12:13 am Sri
Environ Toxicol Chem, 30(3):54855. doi:10.1002/ Lanka time. Available from: http://www.colombopage.
etc.431 PMID:21128261 com /a rch ive _14 A /May13_1399920230CH.php,
Chaufan G, Coalova I, Ros de Molina MC (2014). accessed June 2015.
Glyphosate commercial formulation causes cyto- Conners DE, Black MC (2004). Evaluation of lethality
toxicity, oxidative effects, and apoptosis on human and genotoxicity in the freshwater mussel Utterbackia
cells: differences with its active ingredient. Int J imbecillis (Bivalvia: Unionidae) exposed singly and
Toxicol, 33(1):2938. doi:10.1177/1091581813517906 in combination to chemicals used in lawn care. Arch
PMID:24434723 Environ Contam Toxicol, 46(3):36271. doi:10.1007/
Chen L, Xie M, Bi Y, Wang G, Deng S, Liu Y (2012). s00244-003-3003-z PMID:15195808
The combined effects of UV-B radiation and herbi- Costa MJ, Monteiro DA, Oliveira-Neto AL, Rantin FT,
cides on photosynthesis, antioxidant enzymes and Kalinin AL (2008). Oxidative stress biomarkers and
DNA damage in two bloom-forming cyanobacteria. heart function in bullfrog tadpoles exposed to Roundup
Ecotoxicol Environ Saf, 80:22430. doi:10.1016/j. Original. Ecotoxicology, 17(3):15363. doi:10.1007/
ecoenv.2012.03.007 PMID:22464588 s10646-007-0178-5 PMID:17987383
Chen M-X, Cao Z-Y, Jiang Y, Zhu Z-W (2013). Direct Culbreth ME, Harrill JA, Freudenrich TM, Mundy WR,
determination of glyphosate and its major metabolite, Shafer TJ (2012). Comparison of chemical-induced
aminomethylphosphonic acid, in fruits and vegeta- changes in proliferation and apoptosis in human
bles by mixed-mode hydrophilic interaction/weak and mouse neuroprogenitor cells. Neurotoxicology,
anion-exchange liquid chromatography coupled with 33(6):1499510. doi:10.1016/j.neuro.2012.05.012
electrospray tandem mass spectrometry. J Chromatogr PMID:22634143
A, 1272:909. doi:10.1016/j.chroma.2012.11.069 Curwin BD, Hein MJ, Sanderson WT, Nishioka MG,
PMID:23261284 Reynolds SJ, Ward EM etal. (2005). Pesticide contami-
Chen YJ, Wu ML, Deng JF, Yang CC (2009). The epidemi- nation inside farm and nonfarm homes. J Occup Environ
ology of glyphosate-surfactant herbicide poisoning in Hyg, 2(7):35767. doi:10.1080/15459620591001606
Taiwan, 19862007: a poison center study. Clin Toxicol PMID:16020099
(Phila), 47(7):6707. doi:10.1080/15563650903140399 Curwin BD, Hein MJ, Sanderson WT, Striley C, Heederik
PMID:19640238 D, Kromhout H et al. (2007). Urinary pesticide
Chruscielska K, Brzezinski J, Kita K, Kalhorn D, Kita I, concentrations among children, mothers and fathers
Graffstein B et al. (2000). Glyphosate - Evaluation of living in farm and non-farm households in Iowa. Ann
chronic activity and possible far-reaching effects. Part Occup Hyg, 51(1):5365. doi:10.1093/annhyg/mel062
1. Studies on chronic toxicity. Pestycydy (Warsaw), PMID:16984946
34:1120. de Castilhos Ghisi N, Cestari MM (2013). Genotoxic
Clair E, Mesnage R, Travert C, Sralini G (2012). A glypho- effects of the herbicide Roundup() in the fish Corydoras
sate-based herbicide induces necrosis and apoptosis in paleatus (Jenyns 1842) after short-term, environ-
mature rat testicular cells in vitro, and testosterone mentally low concentration exposure. Environ Monit
decrease at lower levels. Toxicol In Vitro, 26(2):26979. Assess, 185(4):32017. doi:10.1007/s10661-012-2783-x
doi:10.1016/j.tiv.2011.12.009 PMID:22200534 PMID:22821326
Clements C, Ralph S, Petras M (1997). Genotoxicity De Marco A, De Simone C, Raglione M, Testa A, Trinca
of select herbicides in Rana catesbeiana tadpoles S (1992). Importance of the type of soil for the induc-
using the alkaline single-cell gel DNA elec- tion of micronuclei and the growth of primary roots of
trophoresis (comet) assay. Environ Mol Vicia faba treated with the herbicides atrazine, glypho-
Mutagen, 29(3):27788. doi:10.1002/(SICI)1098- sate and maleic hydrazide. Mutat Res, 279(1):913.
2 2 8 0 (19 9 7 ) 2 9 : 3 < 2 7 7: : A I D - E M 8 >3 . 0 . C O ; 2 - 9 doi:10.1016/0165-1218(92)90260-7 PMID:1374535
PMID:9142171 de Menezes CC, da Fonseca MB, Loro VL, Santi A, Cattaneo
Coalova I, Ros de Molina MC, Chaufan G (2014). R, Clasen B etal. (2011). Roundup effects on oxidative
Influence of the spray adjuvant on the toxicity effects of stress parameters and recovery pattern of Rhamdia
a glyphosate formulation. Toxicol In Vitro, 28(7):1306 quelen. Arch Environ Contam Toxicol, 60(4):66571.
11. doi:10.1016/j.tiv.2014.06.014 PMID:24999230 doi:10.1007/s00244-010-9574-6 PMID:20680259
81
IARC Monographs 112
De Roos AJ, Blair A, Rusiecki JA, Hoppin JA, Svec M, Elie-Caille C, Heu C, Guyon C, Nicod L (2010).
Dosemeci M et al. (2005a). Cancer incidence among Morphological damages of a glyphosate-treated human
glyphosate-exposed pesticide applicators in the keratinocyte cell line revealed by a micro- to nanoscale
Agricultural Health Study. Environ Health Perspect, microscopic investigation. Cell Biol Toxicol, 26(4):3319.
113(1):4954. doi:10.1289/ehp.7340 PMID:15626647 doi:10.1007/s10565-009-9146-6 PMID:20043237
De Roos AJ, Svec MA, Blair A, Rusiecki JA, Dosemeci M, Engel LS, Hill DA, Hoppin JA, Lubin JH, Lynch CF,
Alavanja MC et al. (2005b). Glyphosate results revis- Pierce J et al. (2005). Pesticide use and breast cancer
ited: De Roos et al. respond. Environ Health Perspect, risk among farmers wives in the Agricultural Health
113(6):A3667. doi:10.1289/ehp.113-a366 Study. Am J Epidemiol, 161(2):12135. doi:10.1093/aje/
De Roos AJ, Zahm SH, Cantor KP, Weisenburger DD, kwi022 PMID:15632262
Holmes FF, Burmeister LF et al. (2003). Integrative EPA (1980a). Glyphosate; Submission of rat teratology,
assessment of multiple pesticides as risk factors rabbit teratology, dominant lethal mutagenicity
for non-Hodgkins lymphoma among men. Occup assay in mice. Washington (DC): United States
Environ Med, 60(9):E11 doi:10.1136/oem.60.9.e11 Environmental Protection Agency, Office of Toxic
PMID:12937207 substances. Available from: http://www.epa.gov/
De Souza Filho J, Sousa CC, Da Silva CC, De Sabia- pesticides/chemicalsearch/chemical/foia/cleared-
Morais SM, Grisolia CK (2013). Mutagenicity and reviews/reviews/103601/103601-090.pdf, accessed 10
genotoxicity in gill erythrocyte cells of Poecilia reticu- March 2015.
lata exposed to a glyphosate formulation. Bull Environ EPA (1980b). Review of Rodwell DE, Tasker EJ, Blair AM, et
Contam Toxicol, 91(5):5837. doi:10.1007/s00128-013- al. (1980). Teratology study in rats: IRDC No. 401054.
1103-7 PMID:24042842 MRID 00046362. Washington (DC): United States
Dennis LK, Lynch CF, Sandler DP, Alavanja MC (2010). Environmental Protection Agency. Available from:
Pesticide use and cutaneous melanoma in pesticide http://www.epa.gov/ncct/toxrefdb/, and from http://
applicators in the Agricultural Health Study. Environ www.epa.gov/pesticides/chemicalsearch/chemical/
Health Perspect, 118(6):8127. doi:10.1289/ehp.0901518 foia/cleared-reviews/reviews/103601/103601-090.pdf,
PMID:20164001 accessed 10 March 2015.
Dill GM, Sammons RD, Feng PCC, Kohn F, Kretzmer EPA (1983). Review of Knezevich A, Hogan G (1983).
K, Mehrsheikh A etal. (2010). Chapter 1: Glyphosate: A chronic feeding study of glyphosate (Roundup
discovery, development, applications, and properties. Technical) in mice: Project No. 772061: Bdn-77- 420.
In: Nandula VK editor. Glyphosate resistance in crops Final Report. MRID 00130406. Washington (DC):
and weeds: history, development, and management. United States Environmental Protection Agency.
Hoboken (NJ): Wiley; pp. 133. Available from: http://www.epa.gov/ncct/toxrefdb/,
Dimitrov BD, Gadeva PG, Benova DK, Bineva MV (2006). accessed 10 March 2015.
Comparative genotoxicity of the herbicides Roundup, EPA (1985a). Glyphosate; EPA Reg.#: 524308; Mouse
Stomp and Reglone in plant and mammalian test oncogenicity study. Document No. 004370. Washington
systems. Mutagenesis, 21(6):37582. doi:10.1093/ (DC): Office of Pesticides and Toxic Substances, United
mutage/gel044 PMID:16998229 States Environmental Protection Agency. Available
dos Santos KC, Martinez CB (2014). Genotoxic and from: http://www.epa.gov/pesticides/chemicalsearch/
biochemical effects of atrazine and Roundup(), alone chemical/foia/cleared-reviews/reviews/103601/103601-
and in combination, on the Asian clam Corbicula 183.pdf, accessed 10 March 2015.
fluminea. Ecotoxicol Environ Saf, 100:714. doi:10.1016/j. EPA (1985b). EPA Reg.#: 524308; Roundup; glypho-
ecoenv.2013.11.014 PMID:24433785 sate; pathology report on additional kidney sections.
Duke SO, Powles SB (2009). Glyphosate-resistant crops Document No. 004855. Washington (DC): Office
and weeds. Now and in the future. AgBioForum, of Pesticides and Toxic Substances, United States
12(3&4):34657. Environmental Protection Agency. Available from:
EFSA (2009). 2007 Annual Report on Pesticide Residues http://w w w.epa.gov/pesticides/chemica lsearch/
according to Article 32 of Regulation (EC) No 396/2005. chemical/foia/cleared-reviews/reviews/103601/103601-
Parma: European Food Safety Authority. Available 206.pdf, accessed 10 March 2015.
from: http://www.efsa.europa.eu/en/efsajournal/ EPA (1986). Glyphosate; EPA Registration No. 524308;
pub/305r.htm, accessed 1 November 2014. Roundup; additional histopathological evaluations
el-Gendy KS, Aly NM, el-Sebae AH (1998). Effects of of kidneys in the chronic feeding study of glypho-
edifenphos and glyphosate on the immune response sate in mice. Document No. 005590. Washington
and protein biosynthesis of bolti fish (Tilapia (DC): Office of Pesticides and Toxic Substances,
nilotica). J Environ Sci Health B, 33(2):13549. United States Environmental Protection Agency.
doi:10.1080/03601239809373135 PMID:9536512 Available from: http://www.epa.gov/pesticides/
chemica lsearch/chemica l/foia/cleared-rev iews/
82
Glyphosate
83
IARC Monographs 112
84
Glyphosate
85
IARC Monographs 112
86
Glyphosate
glyphosate. Arch Environ Contam Toxicol, 22(1):613. Lushchak OV, Kubrak OI, Storey JM, Storey KB, Lushchak
doi:10.1007/BF00213295 PMID:1554254 VI (2009). Low toxic herbicide Roundup induces mild
Lee EA, Strahan AP, Thurman EM (2001). Methods oxidative stress in goldfish tissues. Chemosphere,
of analysis by the U.S. Geological Survey Organic 76(7):9327. doi:10.1016/j.chemosphere.2009.04.045
Geochemistry Research Group determination of PMID:19450865
glyphosate, aminomethylphophonic acid, and glufos- Mahendrakar K, Venkategowda PM, Rao SM, Mutkule
inate in water using online solid-phase extraction DP (2014). Glyphosate surfactant herbicide poisoning
and high-performance liquid chromatography/mass and management. Indian J Crit Care Med, 18(5):328
spectrometry. Open-File Report 01454. Lawrence 30. doi:10.4103/0972-5229.132508 PMID:24914265
(KS): United States Geological Survey. Available from: Malatesta M, Perdoni F, Santin G, Battistelli S, Muller
http://ks.water.usgs.gov/pubs/reports/ofr.01-454.pdf, S, Biggiogera M (2008). Hepatoma tissue culture
accessed 28 July 2015. (HTC) cells as a model for investigating the effects of
Lee WJ, Cantor KP, Berzofsky JA, Zahm SH, Blair A low concentrations of herbicide on cell structure and
(2004a). Non-Hodgkins lymphoma among asthmatics function. Toxicol In Vitro, 22(8):185360. doi:10.1016/j.
exposed to pesticides. Int J Cancer, 111(2):298302. tiv.2008.09.006 PMID:18835430
doi:10.1002/ijc.20273 PMID:15197786 Maas F, Peralta L, Raviolo J, Garca Ovando H, Weyers A,
Lee WJ, Colt JS, Heineman EF, McComb R, Weisenburger Ugnia L etal. (2009b). Genotoxicity of AMPA, the envi-
DD, Lijinsky W et al. (2005). Agricultural pesticide ronmental metabolite of glyphosate, assessed by the
use and risk of glioma in Nebraska, United States. Comet assay and cytogenetic tests. Ecotoxicol Environ
Occup Environ Med, 62(11):78692. doi:10.1136/ Saf, 72(3):8347. doi:10.1016/j.ecoenv.2008.09.019
oem.2005.020230 PMID:16234405 PMID:19013644
Lee WJ, Lijinsky W, Heineman EF, Markin RS, Maas F, Peralta L, Raviolo J, Ovando HG, Weyers A,
Weisenburger DD, Ward MH (2004b). Agricultural Ugnia L et al. (2009a). Genotoxicity of glyphosate
pesticide use and adenocarcinomas of the stomach assessed by the comet assay and cytogenetic tests.
and oesophagus. Occup Environ Med, 61(9):7439. Environ Toxicol Pharmacol, 28(1):3741. doi:10.1016/j.
doi:10.1136/oem.2003.011858 PMID:15317914 etap.2009.02.001 PMID:21783980
Lee WJ, Sandler DP, Blair A, Samanic C, Cross AJ, Alavanja Mance D 3rd (2012). The great glyphosate debate. Northern
MC (2007). Pesticide use and colorectal cancer risk in Woodlands [online magazine]. 8 March. Available
the Agricultural Health Study. Int J Cancer, 121(2):339 from: http://northernwoodlands.org/articles/article/
46. doi:10.1002/ijc.22635 PMID:17390374 the-great-glyphosate-debate, accessed 28 July 2015.
Li AP, Long TJ (1988). An evaluation of the genotoxic poten- Mariager TP, Madsen PV, Ebbehj NE, Schmidt B, Juhl A
tial of glyphosate. Fundam Appl Toxicol, 10(3):53746. (2013). Severe adverse effects related to dermal expo-
doi:10.1016/0272-0590(88)90300-4 PMID:3286348 sure to a glyphosate-surfactant herbicide. Clin Toxicol
Li Q, Lambrechts MJ, Zhang Q, Liu S, Ge D, Yin R etal. (Phila), 51(2):1113. doi:10.3109/15563650.2013.763951
(2013). Glyphosate and AMPA inhibit cancer cell growth PMID:23360343
through inhibiting intracellular glycine synthesis. Marques A, Guilherme S, Gaivo I, Santos MA, Pacheco
Drug Des Dev Ther, 7:63543. PMID:23983455 M (2014). Progression of DNA damage induced by a
Lioi MB, Scarf MR, Santoro A, Barbieri R, Zeni O, Di glyphosate-based herbicide in fish (Anguilla anguilla)
Berardino D etal. (1998). Genotoxicity and oxidative upon exposure and post-exposure periodsinsights
stress induced by pesticide exposure in bovine lympho- into the mechanisms of genotoxicity and DNA repair.
cyte cultures in vitro. Mutat Res, 403(12):1320. Comp Biochem Physiol C Toxicol Pharmacol, 166:126
doi:10.1016/S0027-5107(98)00010-4 PMID:9726001 33. doi:10.1016/j.cbpc.2014.07.009 PMID:25110831
Lopes FM, Varela Junior AS, Corcini CD, da Silva AC, Marques A, Guilherme S, Gaivo I, Santos MA, Pacheco
Guazzelli VG, Tavares G etal. (2014). Effect of glypho- M (2015). Erratum to: Progression of DNA damage
sate on the sperm quality of zebrafish Danio rerio. Aquat induced by a glyphosate-based herbicide in fish
Toxicol, 155:3226. doi:10.1016/j.aquatox.2014.07.006 (Anguilla anguilla) upon exposure and post-exposure
PMID:25089920 periods - Insights into the mechanisms of genotoxicity
Lubick N (2009). Environmental impact of cocaine and DNA repair [Comp. Biochem. Physiol. C 166 (2014)
strategy assessed [News]Nature, Published online 12 126133]. Comp Biochem Physiol C Toxicol Pharmacol,
November, doi doi:10.1038/news.2009.1080 168C:1 doi:10.1016/j.cbpc.2014.10.008 PMID:25521452
Lueken A, Juhl-Strauss U, Krieger G, Witte I (2004). Martini CN, Gabrielli M, Vila MC (2012). A commercial
Synergistic DNA damage by oxidative stress (induced formulation of glyphosate inhibits proliferation and
by H2O2) and nongenotoxic environmental chemi- differentiation to adipocytes and induces apoptosis
cals in human fibroblasts. Toxicol Lett, 147(1):3543. in 3T3L1 fibroblasts. Toxicol In Vitro, 26(6):100713.
doi:10.1016/j.toxlet.2003.10.020 PMID:14700526 doi:10.1016/j.tiv.2012.04.017 PMID:22546541
87
IARC Monographs 112
McDuffie HH, Pahwa P, McLaughlin JR, Spinelli JJ, Ecotoxicol Environ Saf, 74(4):75460. doi:10.1016/j.
Fincham S, Dosman JA et al. (2001). Non-Hodgkins ecoenv.2010.10.037 PMID:21126764
lymphoma and specific pesticide exposures in Monge P, Wesseling C, Guardado J, Lundberg I, Ahlbom
men: cross-Canada study of pesticides and health. A, Cantor KP etal. (2007). Parental occupational expo-
Cancer Epidemiol Biomarkers Prev, 10(11):115563. sure to pesticides and the risk of childhood leukemia in
PMID:11700263 Costa Rica. Scand J Work Environ Health, 33(4):293
McQueen H, Callan AC, Hinwood AL (2012). Estimating 303. doi:10.5271/sjweh.1146 PMID:17717622
maternal and prenatal exposure to glyphosate in Monroy CM, Corts AC, Sicard DM, de Restrepo HG
the community setting. Int J Hyg Environ Health, (2005). [Cytotoxicity and genotoxicity of human cells
215(6):5706. doi:10.1016/j.ijheh.2011.12.002 exposed in vitro to glyphosate] Biomedica, 25(3):335
PMID:22261298 45. doi:10.7705/biomedica.v25i3.1358 PMID:16276681
Mesnage R, Bernay B, Sralini GE (2013). Ethoxylated adju- Moreno NC, Sofia SH, Martinez CB (2014). Genotoxic
vants of glyphosate-based herbicides are active princi- effects of the herbicide Roundup Transorb and its active
ples of human cell toxicity. Toxicology, 313(23):1228. ingredient glyphosate on the fish Prochilodus lineatus.
doi:10.1016/j.tox.2012.09.006 PMID:23000283 Environ Toxicol Pharmacol, 37(1):44854. doi:10.1016/j.
Meza-Joya FL, Ramrez-Pinilla MP, Fuentes-Lorenzo JL etap.2013.12.012 PMID:24448465
(2013). Toxic, cytotoxic, and genotoxic effects of a glypho- Mortensen OS, Srensen FW, Gregersen M, Jensen K
sate formulation (RoundupSL-Cosmoflux411F) in the (2000). [Poisonings with the herbicides glyphosate
direct-developing frog Eleutherodactylus johnstonei. and glyphosate-trimesium] [in Danish] Ugeskr Laeger,
Environ Mol Mutagen, 54(5):36273. doi:10.1002/ 162(35):46569. PMID:10986892
em.21775 PMID:23625742 Motojyuku M, Saito T, Akieda K, Otsuka H, Yamamoto
Ministry of Chemicals & Fertilizers (2008). Performance I, Inokuchi S (2008). Determination of glyphosate,
of chemical & petrochemical industry at a glance (2001 glyphosate metabolites, and glufosinate in human
2007). New Delhi: Monitoring and Evaluation Division, serum by gas chromatography-mass spectrometry.
Department of Chemicals and Petrochemicals, J Chromatogr B Analyt Technol Biomed Life Sci,
Government of India. Available from: http://chemicals. 875(2):50914. doi:10.1016/j.jchromb.2008.10.003
nic.in/stat0107.pdf, accessed February 2015. PMID:18945648
Mladinic M, Berend S, Vrdoljak AL, Kopjar N, Radic B, Muangphra P, Kwankua W, Gooneratne R (2014).
Zeljezic D (2009b). Evaluation of genome damage and Genotoxic effects of glyphosate or paraquat on earth-
its relation to oxidative stress induced by glyphosate worm coelomocytes. Environ Toxicol, 29(6):61220.
in human lymphocytes in vitro. Environ Mol Mutagen, doi:10.1002/tox.21787 PMID:22644885
50(9):8007. doi:10.1002/em.20495 PMID:19402152 Nakashima K, Yoshimura T, Mori H, Kawaguchi M,
Mladinic M, Perkovic P, Zeljezic D (2009a). Adachi S, Nakao T etal. (2002). [Effects of pesticides
Characterization of chromatin instabilities induced on cytokine production by human peripheral blood
by glyphosate, terbuthylazine and carbofuran using mononuclear cellsfenitrothion and glyphosate]
cytome FISH assay. Toxicol Lett, 189(2):1307. Chudoku Kenkyu, 15(2):15965. PMID:12108020
doi:10.1016/j.toxlet.2009.05.012 PMID:19477249 NCBI (2015). Glyphosate. Compound summary for CID
MLHB (2013). Determination of glyphosate residues in 3496. PubChem Open Chemistry Database. Bethesda
human urine samples from 18 European countries. (MD): National Center for Biotechnology Information,
Bremen: Medical Laboratory of Bremen. Available United States National Library of Medicine. Available
from: https://www.foeeurope.org/sites/default/files/ from: http://pubchem.ncbi.nlm.nih.gov/summary/
glyphosate_studyresults_june12.pdf, accessed 24 summary.cgi?cid=3496, accessed 5 March 2015.
November 2014. Nedelkoska TV, Low GKC (2004). High-performance
Modesto KA, Martinez CB (2010a). Effects of Roundup liquid chromatographic determination of glyphosate in
Transorb on fish: hematology, antioxidant defenses water and plant material after pre-column derivatisa-
and acetylcholinesterase activity. Chemosphere, tion with 9-fluorenylmethyl chloroformate. Anal Chim
81(6):7817. doi:10.1016/j.chemosphere.2010.07.005 Acta, 511(1):14553. doi:10.1016/j.aca.2004.01.027
PMID:20684975 NIH (2015). Questionnaires and study data. Agricultural
Modesto KA, Martinez CB (2010b). Roundup causes Health Study. National Institutes of Health. Available
oxidative stress in liver and inhibits acetylcholinest- from: http://aghealth.nih.gov/collaboration/
erase in muscle and brain of the fish Prochilodus questionnaires.html, accessed 12 June 2015.
lineatus. Chemosphere, 78(3):2949. doi:10.1016/j. Nordstrm M, Hardell L, Magnuson A, Hagberg H,
chemosphere.2009.10.047 PMID:19910015 Rask-Andersen A (1998). Occupational exposures,
Mohamed AH (2011). Sublethal toxicity of Roundup to animal exposure and smoking as risk factors for hairy
immunological and molecular aspects of Biomphalaria cell leukaemia evaluated in a case-control study. Br
alexandrina to Schistosoma mansoni infection.
88
Glyphosate
89
IARC Monographs 112
and developmental alterations observed in Caiman Rumack BH (2015). Emergency medical treatment.
latirostris exposed in ovo to pesticide formulations Glyphosate isopropylamine salt. POISINDEX(R)
and mixtures in an experiment simulating environ- Information System. CCIS Volume 164, edition expires
mental exposure. Ecotoxicol Environ Saf, 74(4):8529. May, 2015. Available from: http://toxnet.nlm.nih.gov/
doi:10.1016/j.ecoenv.2010.12.005 PMID:21185601 cgi-bin/sis/search2/f?./temp/~M2Dk5e:2.
Poletta GL, Larriera A, Kleinsorge E, Mudry MD Sanchs J, Kantiani L, Llorca M, Rubio F, Ginebreda
(2009). Genotoxicity of the herbicide formulation A, Fraile J et al. (2012). Determination of glypho-
Roundup (glyphosate) in broad-snouted caiman sate in groundwater samples using an ultrasensitive
(Caiman latirostris) evidenced by the Comet assay immunoassay and confirmation by on-line solid-
and the Micronucleus test. Mutat Res, 672(2):95102. phase extraction followed by liquid chromatography
doi:10.1016/j.mrgentox.2008.10.007 PMID:19022394 coupled to tandem mass spectrometry. Anal Bioanal
Prasad S, Srivastava S, Singh M, Shukla Y (2009). Chem, 402(7):233545. doi:10.1007/s00216-011-5541-y
Clastogenic effects of glyphosate in bone marrow PMID:22101424
cells of swiss albino mice. J Toxicol, 2009:308985 Schinasi L, Leon ME (2014). Non-Hodgkin lymphoma
doi:10.1155/2009/308985 PMID:20107585 and occupational exposure to agricultural pesticide
Rank J, Jensen AG, Skov B, Pedersen LH, Jensen K (1993). chemical groups and active ingredients: a systematic
Genotoxicity testing of the herbicide Roundup and its review and meta-analysis. Int J Environ Res Public
active ingredient glyphosate isopropylamine using the Health, 11(4):4449527. doi:10.3390/ijerph110404449
mouse bone marrow micronucleus test, Salmonella PMID:24762670
mutagenicity test, and Allium anaphase-telophase Sralini GE, Clair E, Mesnage R, Gress S, Defarge N,
test. Mutat Res, 300(1):2936. doi:10.1016/0165- Manuela Malatesta M et al. (2014). Republished
1218(93)90136-2 PMID:7683765 study:long-term toxicity of a Roundup herbicide
Repblica de El Salvador (2013). Asamblea Legislativa and a Roundup-tolerant genetically modified maize.
aprueba reformas que prohben pesticidas que daan Environmental Sciences Europe, 26(1):114. doi:10.1186/
la salud, 5 September 2013. Available from: http:// s12302-014-0014-5
www.asamblea.gob.sv/noticias/archivo-de-noticias/ Siddiqui S, Meghvansi MK, Khan SS (2012). Glyphosate,
asamblea-legislativa-aprueba-reformas-que-prohiben- alachor and maleic hydrazide have genotoxic effect on
pesticidas-que-danan-la-salud, accessed 28 April 2015. Trigonella foenum-graecum L. Bull Environ Contam
[Spanish] Toxicol, 88(5):65965. doi:10.1007/s00128-012-0570-6
Richard S, Moslemi S, Sipahutar H, Benachour N, Seralini PMID:22392005
GE (2005). Differential effects of glyphosate and Simonsen L, Fomsgaard IS, Svensmark B, Spliid NH
Roundup on human placental cells and aromatase. (2008). Fate and availability of glyphosate and AMPA
Environ Health Perspect, 113(6):71620. doi:10.1289/ in agricultural soil. J Environ Sci Health B, 43(5):365
ehp.7728 PMID:15929894 75. doi:10.1080/03601230802062000 PMID:18576216
Roberts DM, Buckley NA, Mohamed F, Eddleston M, Sinhorin VD, Sinhorin AP, Teixeira JM, Milski KM,
Goldstein DA, Mehrsheikh A etal. (2010). A prospec- Hansen PC, Moreira PS et al. (2014). Effects of the
tive observational study of the clinical toxicology of acute exposition to glyphosate-based herbicide on
glyphosate-containing herbicides in adults with acute oxidative stress parameters and antioxidant responses
self-poisoning. Clin Toxicol (Phila), 48(2):12936. in a hybrid Amazon fish surubim (Pseudoplatystoma
doi:10.3109/15563650903476491 PMID:20136481 sp). Ecotoxicol Environ Saf, 106:1817. doi:10.1016/j.
Roustan A, Aye M, De Meo M, Di Giorgio C (2014). ecoenv.2014.04.040 PMID:24840881
Genotoxicity of mixtures of glyphosate and atra- Sivikov K, Dianovsk J (2006). Cytogenetic effect of
zine and their environmental transformation prod- technical glyphosate on cultivated bovine peripheral
ucts before and after photoactivation. Chemosphere, lymphocytes Int J Hyg Environ Health, 209(1):1520.
108:93100. doi:10.1016/j.chemosphere.2014.02.079 doi:10.1016/j.ijheh.2005.07.005 PMID:16373198
PMID:24875917 Slaninova A, Smutna M, Modra H, Svobodova Z (2009).
Ruder AM, Waters MA, Butler MA, Carren T, Calvert GM, A review: oxidative stress in fish induced by pesti-
Davis-King KE etal. ; Brain Cancer Collaborative Study cides. Neuro Endocrinol Lett, 30:Suppl 1: 212.
Group(2004). Gliomas and farm pesticide exposure in PMID:20027135
men: the Upper Midwest Health Study. Arch Environ Solomon KR, Anadn A, Carrasquilla G, Cerdeira AL,
Health, 59(12):6507. doi:10.1080/00039890409602949 Marshall J, Sanin LH (2007). Coca and poppy eradi-
PMID:16789473 cation in Colombia: environmental and human health
Rueppel ML, Brightwell BB, Schaefer J, Marvel JT (1977). assessment of aerially applied glyphosate. Rev Environ
Metabolism and degradation of glyphosphate in soil Contam Toxicol, 190:43125. doi:10.1007/978-0-387-
and water. J Agric Food Chem, 25(3):51728. doi:10.1021/ 36903-7_2 PMID:17432331
jf60211a018 PMID:858844
90
Glyphosate
91
IARC Monographs 112
Environ Health Perspect, 108(8):76976. doi:10.1289/ Zouaoui K, Dulaurent S, Gaulier JM, Moesch C, Lachtre
ehp.00108769 PMID:10964798 G (2013). Determination of glyphosate and AMPA
Wang G, Deng S, Li C, Liu Y, Chen L, Hu C (2012). Damage in blood and urine from humans: about 13 cases of
to DNA caused by UV-B radiation in the desert cyano- acute intoxication. Forensic Sci Int, 226(13):e205.
bacterium Scytonema javanicum and the effects of doi:10.1016/j.forsciint.2012.12.010 PMID:23291146
exogenous chemicals on the process. Chemosphere,
88(4):4137. doi:10.1016/j.chemosphere.2012.02.056
PMID:22436589
Wester RC, Melendres J, Sarason R, McMaster J, Maibach
HI (1991). Glyphosate skin binding, absorption,
residual tissue distribution, and skin decontamination.
Fundam Appl Toxicol, 16(4):72532. doi:10.1016/0272-
0590(91)90158-Z PMID:1884912
Xie L, Thrippleton K, Irwin MA, Siemering GS, Mekebri A,
Crane D etal. (2005). Evaluation of estrogenic activities
of aquatic herbicides and surfactants using an rainbow
trout vitellogenin assay. Toxicol Sci, 87(2):3918.
doi:10.1093/toxsci/kfi249 PMID:16049272
Yadav SS, Giri S, Singha U, Boro F, Giri A (2013). Toxic
and genotoxic effects of Roundup on tadpoles of the
Indian skittering frog (Euflictis cyanophlyctis) in
the presence and absence of predator stress. Aquat
Toxicol, 132133:18. doi:10.1016/j.aquatox.2013.01.016
PMID:23454306
Yin G (2011). Glyphosate: There is no substitute. Farm
Chemicals International. 3 March 2011. Willoughby
(OH): Meister Media Worldwide. Available from:
http://www.farmchemicalsinternational.com/crop-
inputs/herbicides/glyphosate-there-is-no-substitute/,
accessed June 2015.
Yoshioka N, Asano M, Kuse A, Mitsuhashi T, Nagasaki
Y, Ueno Y (2011). Rapid determination of glyphosate,
glufosinate, bialaphos, and their major metabolites in
serum by liquid chromatography-tandem mass spec-
trometry using hydrophilic interaction chromatog-
raphy. J Chromatogr A, 1218(23):367580. doi:10.1016/j.
chroma.2011.04.021 PMID:21530973
Yue Y, Zhang Y, Zhou L, Qin J, Chen X (2008). In vitro
study on the binding of herbicide glyphosate to human
serum albumin by optical spectroscopy and molec-
ular modeling. J Photochem Photobiol B, 90(1):2632.
doi:10.1016/j.jphotobiol.2007.10.003 PMID:18035550
Zahm SH, Weisenburger DD, Babbitt PA, Saal RC, Vaught
JB, Cantor KP et al. (1990). A case-control study of
non-Hodgkins lymphoma and the herbicide 2,4-dichlo-
rophenoxyacetic acid (2,4-D) in eastern Nebraska.
Epidemiology, 1(5):34956. doi:10.1097/00001648-
199009000-00004 PMID:2078610
Zhao W, Yu H, Zhang J, Shu L (2013). [Effects of glypho-
sate on apoptosis and expressions of androgen-binding
protein and vimentin mRNA in mouse Sertoli cells]
Nan Fang Yi Ke Da Xue Xue Bao, 33(11):170913.
PMID:24273285
92