Proposed new European increase in ADI for glyphosate based on flimsy science?

The European Food Safety Agency has proposed an increase of 60% for permitted exposure levels for the European population to glyphosate. After releasing the Conclusion on glyphosate, the EFSA advised in an explanatory promotional document:

'The acceptable operator exposure level (AOEL) has also been set at 0.1 mg/kg body weight per day and an acceptable daily intake (ADI) for consumers has been set in line with the ARfD at 0.5 mg/kg body weight per day.'

The study that is proposed to establish the new European ADI is a developmental or teratology study where animals are dosed throughout pregnancy and then the offspring inspected for health related effects. The study cites 50 mg/kg bw/d as the level that is the NOAEL - no observable acceptable dose level. The level that the foetus is healthy.

However study data on Pp 627-634 indicate there were effects at the same level.

Note: it is industry standard to NOT dose with the chemical in the first 7 days - effectively the first trimester. Thus the test period ran from from gestation days 7-19. Animals were then euthanised on day 29 of gestation. 

  • The laboratory for the Syngenta study is undeclared
  • The author for the Syngenta study is undeclared
  • The guidelines are outdated
  • The studies are 19 and 24 years old and unpublished.
  • EFSA has ignored the IARC findings and scientific literature that demonstrates toxicity at much lower levels of population exposure.

Health based effects dismissed:

  • Syngenta study: 10/18 pregnant rabbits at the 50 mg/kg bw/day had scours.
  • At all days noted, animals at 50 mg/kg bw/day had weight significantly lower than the control
  • Syngenta study: At 50mg/kg bw/d - 27.8% of the litters had 'anomalous' foetuses (Anomalous foetus: a foetus with one or more congenital defects.)
  • Cheminova study: The previous European assessment declared 'since a treatment related impact on fetal viability at this dosage is considered possible, the mid dose level of 150 mg/kg bw/day is assumed to represent a more reliable NOEL for fetotoxicity'.

Farmers and herbicide contractor/applicators should be concerned -

all the studies used to derive the Accepted Operator Exposure Level are 1990's private, unpublished studies. (See pp 108-9  of the Final Addendum - link below). The Rapporteur Member State (RMS), Germany, who compiled the Addendum, ignored published literature that demonstrate that current low exposure levels are dangerous. In deriving the AOEL the RMS only considered old studies at much higher dosage levels.

Unfortunately because the all studies used to derive the AOEL and ADI are private and not available for public scrutiny, further evaluation by public domain scientists are not possible.

 

Data located in Background Documents.  Click on Peer Review Report and revised Renewal Assessment Report and select Final Addendum (PDF).

Final addendum to the  Renewal Assessment Report - public version -   
Risk assessment provided by the rapporteur Member State Germany  and co-rapporteur Member State Slovakia for the active substance  
GLYPHOSATE  
according to the procedure for the renewal of the inclusion of a second group of active substances in Annex I to Council Directive 91/414/EEC laid down in Commission Regulation (EU) No. 1141/2010

In rabbits, the overall NOAEL for maternal toxicity was 50 mg/kg bw/day, based on mortality and gastrointestinal disturbances from 100 mg/kg bw/day onwards. The lowest NOAEL for developmental effects was also established at 50 mg/kg bw/d, based on post-implantation loss at 200 mg/kg bw/day. Other developmental foetal effects comprised lower foetal weight and a delay in ossification. 

2.6.12 Toxicological end point for assessment of risk following long-term dietary exposure - ADI     p.105 .   (Page 105 does not give this information - rather it gives 'Overview of glyphosate residues and its metabolites following hydroponic treatment')

 Volume 1, Level 1. Page 37.  The proposed ADI and AOEL for glyphosate are both based on the lowest NOAEL for maternal and developmental toxicity in the rabbit at 50 mg/kg bw/day since this was the most sensitive animal model. Applying the safety factor of 100, the resulting ADI is 0.5 mg/kg bw.  The ADI for glyphosate is applicable also to AMPA if health evaluation of residues is needed

For setting the AOEL, correction for 20 % oral absorption must be made. Accordingly, an AOEL of 0.1 mg/kg bw/day was established. 

TWO STUDIES WERE USED TO ARRIVE AT THE NEW ADI - 

[1] 1996 Syngenta unpublished study ASB201211499 (Study details below)

[2] 1991 Cheminova unpublished study  TOX9552391:  Believe this study to be, Brooker et al., 1991.

Details found:  Glyphosate – Volume 1, Level 1  revised 29 January 2015; 31 March 2015. Pp 77-82

[1] Table 2.6-12: Developmental toxicity studies in rabbits reviewed for previous EU evaluation that are still to be used for risk assessment 

Reference;  Study identification; Batch,  purity; Owner: 1991; TOX9552391 206-Jak-25-1, 98.6%; Cheminova

Study type, strain, duration, route: Developmental,NZW rabbit, d 719 p.c., gavage

Dose levels: 0, 50, 150, 450 mg/kg bw

NOAEL: Maternal: 50 mg/kg bw/d, dev.: 150 mg/kg bw/d

LOAEL: Maternal: 150 mg/kg bw/d, dev.: 450 mg/kg bw/d

Targets/Main effects: Maternal: GI-tract, food & bw gain ↓: dev.: late embryonic death, post implantation loss, cardiac malformations* 

Outcome of re-evaluation: Study acceptable, previous evaluation partly confirmed

[2] Table 2.6-14: New developmental toxicity studies in rabbits reviewed for EU evaluation 2012

Study reference, identification, batch, purity, owner: 1996; ASB201211499; H95D161A, 95.3%; Nufarm .

Study type, strain, duration, route: Developmental, NZW rabbit, d 719 p.c., gavage.

Dose levels: 0, 50, 200, 400 mg/kg bw/d . NOAEL: Maternal & dev.: 50 mg/kg bw/d ; LOAEL: Maternal& dev.: 200 mg/kg bw/d;

Target/Main effects: Maternal: bw gain ↓,  Dev.: postimplantation loss ;

Outcome of RMS Evaluation: Study acceptable

 

Page 81: In conclusion, the previously known and newly submitted studies have clearly shown a particular vulnerability of pregnant females of this species. The overall NOAEL for maternal toxicity was 50 mg/kg bw/d as found by   (1996, ASB2012-11499) and by  (1991, TOX9552391). It is further supported by the (supplementary) study of  (1980, TOX2552390) who had established 75 mg/kg bw/day and is well between the NOAEL of 20 mg/kg bw/day and the LOAEL in the (supplementary) study by  (1993, TOX9551106). However, in the latter study, mortality occurred already at the LOAEL of 100 mg/kg bw/day. This finding might suggest a rather steep dose response curve and provided the lowest dose at which mortality was noted in any study and species following administration of glyphosate. In other developmental studies in rabbits, maternal deaths were seen at 175 mg/kg bw/day and above ( ., 1980, TOX2552390; ., 1991, TOX9552391;  1995, ASB2012-11498). Thus, the (pregnant) rabbit proved more sensitive than the other species which were employed in toxicological testing of glyphosate. Surprisingly, there were other studies in which the does tolerated similar or even higher amounts of glyphosate (300 to 500 mg/kg bw/day) much better, i.e., without mortality at least ( , 1989, TOX9551960;  1996, ASB2012-11499;  1996, TOX2000-2002). Apart from mortality in some studies, maternal toxicity was characterised by gastrointestinal signs, lower body weight (gains) and reduced food consumption and, occasionally, abortions. The lowest LOAEL was found in the study by  (1993, TOX9551106) whereas the respective values in the other experiments ranged from 150 to 500 mg/kg bw/day.  


The lowest NOAEL for developmental effects was 50 mg/kg bw/day, based post-implantation loss at 200 mg/kg bw/day in the study by  (1996, ASB2012-11499). Due to dose spacing, the NOAELs in other studies were higher ( 1980, TOX2552390; ., 1991, TOX9552391;  1996, TOX2000-2002) but consistently below 200 mg/kg bw/day. Beside post-implanation losses and late embryonic death ( 1989, TOX9551960;  1991, TOX9552391), developmental findings at higher dose levels included a lower foetal weight and delayed ossification.

P 107: The ADI of 0.5 mg/kg bw and the approach taken for its setting as proposed by the RMS were confirmed on the Pesticides Peer Review 125 expert meeting (25 – 27 February 2015)

Details of the two studies that provides the NOAEL of 50 mg/kg bw/day justifying the increased European ADI from .3 mg/kg to .5 mg/kg

[1] 1996 Syngenta unpublished study ASB201211499

Page 627. 

Reference: IIA, 5.6.11/02 Report:  (1996) Glyphosate technical: Oral gavage teratology study in the rabbit   
Data owner: Nufarm SPL project no.: 434/020 Date: 1996-07-04 GLP: yes not published ASB2012-11499

Guidelines: OECD 414 (1981), JMAFF 59 NohSan 4200 (1985), US-EPA 83-3 (1984) Deviations: None  Acceptability: See RMS comment  
Dates of experimental work: 1995-10-13 - 1995-12-12  


Materials and methods  
Test material:  Identification: Glyphosate technical Description: White powder Lot/Batch #: H95D161A Purity: 95.3 % Stability of test compound: not reported Vehicle and/ or positive control: 1 % carboxymethyl cellulose Test animals: Species: Rabbit Strain: New Zealand White  Source: XXXXXXX  Age: 17 - 19 weeks Sex: Females (time-mated) Weight at dosing: 2.2 - 4.1 kg Acclimation period: At least 4 days Diet/Food: SQC Standard Rabbit Diet (SDS Ltd., Witham, Essex, UK), ad libitum Water: Tap water, ad libitum Housing: Individually in stainless steel cages with grid floor Environmental conditions: Temperature: 20 ± 3 °C Humidity: 50 ± 20 % Air changes: 15/hour 12 hours light/dark cycle

p.628

Study design and methods:  In life dates: 1995-10-13 - 1995-12-12  
Animal assignment and treatment in the preliminary study: Twenty-four time-mated females were supplied. Sexually mature, virgin females were paired with stud males. The day of copulation was designated Day 0 of gestation. The females were delivered to Safepharm Laboratories Ltd. at or before Day 3 of gestation and were allocated randomised to treatment groups. Groups of 6 mated New Zealand white female rabbits received 0, 50, 200 or 400 mg/kg bw/day test substance in 1 % carboxymethyl cellulose by gavage (5 mL/kg bw) from gestation Day 7-19. The dose levels were chosen based on results of a preliminary dose finding study with 6 female nulliparous rabbits, where administration of 500 or 1000 mg/kg bw resulted in toxicity signs (scours, fluid filled caecum, stomach ulceration, body weight loss, reduced food consumption). Based on these findings dose levels of ≥ 500 mg/kg bw were considered to be too high for a prolonged study.  


Animal assignment and treatment in the main study: Seventy-two time-mated females were supplied as described for the preliminary study (see above). Groups of 18 mated New Zealand white female rabbits received 0, 50, 200 or 400 mg/kg bw/day test substance in 1 % carboxymethyl cellulose by gavage (5 mL/kg bw) from gestation Day 7-19.  
Dose formulation and analysis For each dose level, the test material was suspended daily in 1% carboxymethyl cellulose by weighing the required amount into a glass jar and adding vehicle to make the appropriate final volume. Homogeneity was assured by mixing the formulation with a homogeniser. The concentration, stability and homogeneity of the test material were analysed. The formulation was stable for at least 1 h.  
Clinical observations A check for clinical signs of toxicity, ill-health or behavioural changes was made once daily during the pre- and post-dosing periods and twice daily (before and after dosing) during the dosing period.  


Body weight Individual body weights were recorded on Day 3, 7, 10, 13, 16, 19, 22, 25 and 29 of gestation.   


Food consumption Food consumption of females was recorded on Days 3 to 7, Days 7 to 10, Days 10-13, Days 13-16, Days 16-19, Days 19-22, Days 22-25 and Days 25-29 of gestation.  
Sacrifice and pathology Females were euthanatized by an i.v. injection of an overdose of sodium pentobarbitone into the auricular vein on Day 29 of gestation, examined for macroscopic abnormalities and subjected to caesarean sectioning. The ovaries and uteri were removed, weighed and then examined for the number of corpora lutea and for the number and position of implants and dead or live foetuses. Resorptions and foetal deaths were classified into implantation sites, placental remnants, and macerated foetuses according to the difference in developmental stage at which deaths had occurred. After examination of the ovaries and conceptuses, each female was necropsied. 

Developmental parameters The foetuses were killed by intrathoracic injection of sodium pentobarbitone. All foetuses were dissected and examined for visceral abnormalities macroscopically. The heads of alternate foetuses were removed and identified using an indelible marker and placed in Bouin‘s fixative. After a minimum of 14 days, the heads were transferred to 90 % industrial methylated spirits (IMS) in distilled water and examined for visceral anomalies under a low power binocular microscope. All foetuses were identified using colour coded wires and placed in 70 % IMS in distilled water. The foetuses were eviscerated, processed and the skeletons stained with alizarin red.The foetuses were examined for skeletal development and anomalies.   


Statistics in the main study Female bodyweight change (relative to Day 7 of gestation) and food consumption were analysed statistically by one-way analysis of variance with the Bonferroni multiple comparison test followed by pair wise analysis of control values against treated group values using Students ‘t‘ test where appropriate.  All foetal parameters, skeletal development, group incidence of specific visceral and skeletal anomalies were analysed statistically by Kruskall-Wallis non parametric analysis of variance followed by pair wise analysis of control values against treated values using the MannWhitney U - test where appropriate.  


Results and discussion

Analysis of dose formulations The test substance was detected at the levels of 81-102% of the target concentrations in each dosing solution.  


Food consumption In the preliminary study, significantly reduced food consumption was observed while administering in the high dose level of 400 mg/kg/day (Days 7 to19 of gestation).  This observation was confirmed in the main study. At the high dose level, there was a reduction in food consumption during the dosing period compared to controls (Days 10 to 13, p < 0.05; Days 13 to 19, p < 0.01). No other significant changes were observed in the remaining groups during the main study.  


Mortality In the preliminary study, two does were killed in extremis in the high dose group, one had aborted foetuses and the other was bleeding from the vagina. No mortalities occurred at any dose up to 400 mg/kg/day in the preliminary study.    


In the main study, two rabbits were found dead or moribund at the high dose level. One female was found dead prior to dosing on Day 19 of treatment. One female was killed in extremis on Day 20 of treatment. Clinical observations noted at this time included hunched posture, lethargy, ptosis, hypothermia and blood on the litter tray. At the intermediate dose level, one female was found dead after dosing on Day 16 of treatment. Necropsy findings of reddened lungs, a fluid filled thorax and test material in thoracic cavity are consistent with mal-dosing. At the low dose level, no mortalities occurred. One female was found dead two minutes after dosing in the control group. Necropsy findings of blood in thorax, inflated appearance of lungs and a large area of congestion on the right caudal lobe are consistent with mal-dosing.

P.630

Clinical Observations

In both the preliminary and the main study, the clinical signs were in general the same. There was a toxicologically significant increase in the incidence of clinical observations, particularly scours, reduced faecal output and diarrhoea at the high dose level (400 mg/kg bw/day). Observations of lethargy, ptosis, hunched posture, hypothermia and blood on tray were noted for one animal of the main study killed in extremis.  At 200 mg/kg bw/day, vaginal bleeding and blood on tray were noted for one animal of the main study. Scours were also noted in animals at 200 and 50 mg/kg bw/day as well as in the control group, but the incidence and duration were not as severe as at the high dose level (see Table B.6.6-28). No other treatment-related observations were evident. Thus, for the findings observed at doses below 400 mg/kg bw/day, a clear dose-response could not be established.

 
Table B.6.6-28: Observed clinical signs during the dosing period

Background document on the conclusion on the peer review of the risk assessment of glyphosate . Addendum.

Background document on the conclusion on the peer review of the risk assessment of glyphosate . Addendum.

 Body weight

In the preliminary, study a toxicologically significant decrease in body weight gain from Day 13 to 19 post coitum was evident at the high and intermediate dose levels. Likewise a reduction in group mean body weight gain from Days 9 to 29 post coitum was observed in the high dose level group during the main study. The difference in group mean bodyweight change compared to controls was statistically significant (P<0.05 to 0.01) from Days 13 to 29 post coitum. Also in the intermediate dose level group a slight reduction (although not statistically significant) in group mean body weight gain from Day 9 to Day 29 post coitum was noted. In the low dose level group body weight gain was comparable to controls throughout the study period (see Table B.6.6-29).  

P.631
Table B.6.6-29: Mean body weight gain during gestation

Background document on the conclusion on the peer review of the risk assessment of glyphosate . Addendum.

Background document on the conclusion on the peer review of the risk assessment of glyphosate . Addendum.

Pathology Necropsy

The macroscopic necropsy findings of the two does of the high-level dose group that died or were killed in extremis included fluid filled large intestines, haemorrhage, ulceration and sloughing of the stomach, duodenum congested and colon, rectum and appendix gas distended. These findings indicate that the test material may affect the gastrointestinal tract. The animal killed in extremis at this level also had both uterine horns containing blood and dead foetuses in the uterus. This may be a result of maternal toxicity. All other necropsy findings were not treatment-related.  


Observations on the ovary and uterus

No treatment related effects were evident in both the preliminary and the main study.  


In the control, low, intermediate and high dose level groups 14, 18, 16, and 16 females, respectively, survived to termination of the main study and were proven to be pregnant. The number and distribution of females that were not pregnant indicate that there were no treatment-related effects on pregnancy rates. Litter size at caesarean necropsy was comparable in all treatment groups.  


Developmental parameters

Number and viability of foetuses The litter size at caesarean section was comparable in all treatment groups. In the high dose level group, there were slight, but not statistically significant, increases in late foetal deaths and post implantation loss, mainly due to one animal that had nine late deaths, resulting in a post implantation loss of 69.2 %. This was therefore considered not to be a treatment-related effect. At 200 mg/kg bw/day, there were statistically significant increases (p<0.05) in total foetal deaths and post implantation loss. These increases were caused by a slight, but not statistically significant, rise in early foetal deaths. As at this dose level, there was no rise in late foetal deaths, as seen at the high level; the effect on early foetal deaths was considered not to be treatment-related.   

Foetal body weights  

No statistically significant differences were noted in the mean foetal body weights between the control group and the treated groups. Mean total litter weights were comparable in all treatment groups.   


External, visceral and skeletal examination

At the high dose level, there was one litter with one foetus with major malformations. This foetus was found to have spina bifida and clubbed and malrotated hind limbs. At the intermediate dose level, two foetuses of two different litters had major malformations. One foetus had retinal infolding and a haemorrhage in the retinal layer, the other acephaly, small kinked tail, bilateral forelimb flexure, interrupted aorta and an intraventricular septal defect. At skeletal examination, this foetus was found to have multiple rib and vertebral column abnormalities. At the low dose level, three foetuses of two different litters had major abnormalities. In one litter, one foetus had forked ribs with a displaced vertebral centrum. In another litter, one foetus had a small eye with retinal infolding and aphakia. A second foetus from this litter had nostrils close together, and a thin nasal septum not attached at posterior pole near the front of the nasal passages. In the control group, there were two foetuses from two different litters with major abnormalities. One foetus had gastroschisis and the other foetus had an extra vertebral arch resulting in scoliosis. These findings were considered to be within the range of normal variation for this species. There were no treatment-related effects on the degree of skeletal development. 


Table B.6.6-30: Incidence of foetal malformations and variations in rabbits treated with glyphosate acid
Foetal findings  Dose level (mg/kg bw/day) 0 50 200 400

Background document on the conclusion on the peer review of the risk assessment of glyphosate . Addendum.

Background document on the conclusion on the peer review of the risk assessment of glyphosate . Addendum.

Conclusion by the Notifiers

The oral administration of glyphosate technical to pregnant rabbits by gavage from gestation Day 7-19 resulted maternal toxicity at 400 mg/kg bw/day. There were no treatment-related effects on pregnancy or foetuses at any dose level. Therefore the ‘No Observed Adverse Effect Level’ (NOAEL) was considered to be 200 mg/kg bw/day for maternal toxicity. The ‘No Observed Adverse Effect Level’ (NOAEL) for developmental toxicity was considered to be 400 mg/kg bw/day.  


Comment by RMS:  

The study is considered acceptable. The NOAEL of 200 mg/kg bw/d for maternal toxicity is not supported. The NOAEL is considered to be 50 mg/kg bw/d due to slight reduction in body weight gain at 200 mg/kg bw/d.  The NOAEL of 400 mg/kg bw/d for developmental toxicity is neither supported. The NOAEL is considered to be 50 mg/kg bw/d due to significantly increased post-implantation loss at 200 mg/kg bw/d. The statement, that this increase was caused by a slight rise in early foetal deaths and not in late foetal deaths, as seen at the high dose level and therefore considered not to be treatment-related, cannot be followed, because there is no information given regarding the mechanism behind this foetal deaths. Due to some reporting deficiencies, it remains unclear, whether the heart was part of visceral examination.  


Comment by GTF on the first draft of the RAR (July 2013):

1. The GTF believes the developmental NOAEL in  is 400 mg/kg/day, with no evidence to test substance related dose-response, citing the published expert review: Kimmel et al., 2013, (ASB2013-3462) Evaluation of developmental toxicity studies of glyphosate with attention to cardiovascular development, Crit Rev Toxicol, 2013; 43(2): 79– 95.  


2. Regarding  (1996, ASB2012-11499) the RMS notes that “it remains unclear whether the heart was part of visceral examination”. The study sponsor contacted the contract laboratory and the response is noted….The contract laboratory Head of reprotox at the contract notes Reprotox “The examination of the heart is conducted on the fresh fetus for rabbit developmental toxicity studies. This is mentioned in our standard operating procedure“. This is reflected the draft RAR Volume 3, B.6.6.11, noting “All foetuses were dissected and examined for visceral abnormalities macroscopically”………..  


RMS comment (August 2013): 1. As already stated, the NOAEL is considered to be 50 mg/kg bw/d due to significantly increased post-implantation loss at 200 mg/kg bw/d. The statement, that this increase was caused by a slight rise in early foetal deaths and not in late foetal deaths, as seen at the high dose level and therefore considered not to be treatment-related, cannot be followed, because there is no information given regarding the mechanism behind this foetal deaths (please refer to the additionally inserted table below. Considering the individual animal data: at low dose level of 50 mg/kg bw/d 4/18 pregnant animals revealed post implantation losses, which was comparable to control animals (4/14 animals). Whereas at the mid dose of 200 mg/kg bw/d 10/15 animals were affected and at 400 mg/kg bw/d 9/15 animals. Currently, RMS does not see the need to change the assessment.  [Remark: The published expert review by Kimmel et al. (2013, ASB2013-3462) was already available and considered when preparing the first draft of the RAR, please refer to B.6.6.12- Published data] 

[2] 1991 Cheminova unpublished study  TOX9552391:  Believe this study to be, Brooker et al., 1991.

By the: European Commission.  Health & Consumer Protection Directorate-General

The legislation is here: It includes glyphosate in Annex I of Directive 91/414/EEC – Overall conclusion in the context of Directive 91/414/EEC is that it may be expected that plant protection products containing glyphosate will fulfill the safety requirements laid down in Article 5(1)(a) and (b) of Directive 91/414/EEC.

The details for the previous EC assessment of glyphosate final decisions as to glyphosate toxicity are in  Document No. 6511/VI/99-final.

Study data was located in: APPENDIX II – END POINTS AND RELATED INFORMATION GLYPHOSATE/GLYPHOSATE TRIMESIUM:

Brooker, A.J., Brennan, C., John, D.M., Anderson, A. & Dawe, I.S. (1991a): The effect of glyphosate on pregnancy of the rabbit (incorporates preliminary investigations). Unpublished.  “GLP like”. Report No. CHV 45 & 39 & 40/901303, dated 14 October 1991, from Huntingdon Research Centre Ltd., Huntingdon, England. Submitted as part of a joint dossier by Monsanto and Cheminova.

Glyphosate given by oral gavage daily to 16-20 mated NZW female rabbits at dose levels of 0, 50, 150 and 450 mg/kg bw/day from days 7-19 of pregnancy.  Results:  One mortality observed in high dose group following abortion and body weight loss. Gastro-intestinal signs of toxicity and inappetence were observed at doses of 450 and 150mg/kg bw/day.  Mean food consumption and body weight gain were reduced during the treatment period at these dose levels. Thus the lowest dose of 50mg/kg bw/day was considered NOEL for maternal toxicity. At day 29, at terminal sacrifice, 18, 12, 15 and 13 litters were available for examination in the control, low, mid and high dose groups, respectively.  There was a significant increase in embryonic deaths in treated groups compared to the controls.  Accordingly the percentage of post implantation loss was elevated.  Accordingly, a comparison with historical control data from the performing laboratory revealed that the incidence in the study group was atypically low.  In addition, a clear dose-related pattern was not demonstrated.  On the other hand, note that an increase in the occurrence of late embryonic deaths at the top dose level was also studied in a further study below.

Conclusion: Note the lowest dose of 50 mg/kg bw/day was considered NOEL for maternal toxicity by study authors.  It was concluded that no convincing indications of teratogenicity were obtained in this study up to the highest dose of 450 mg//kg bw/day.  There was some concern about the more frequent occurrence of foetuses with heart malformations like the interventricular septal defect in the high dose group, however, the incidence was still in the range of historical background data.  On the other hand, anomalies of the heart have been described in other rabbit teratogenicity studies with glyphosate too.  NOEL for fetal effects at the top dose level of 450 mg/kg bw/day.  However, since a treatment related impact on fetal viability at this dosage is considered possible, the mid dose level of 150 mg/kg bw/day is assumed to represent a more reliable NOEL for fetotoxicity.