Close Menu

ANSES report in contrast to global scientific consensus on BPA

ANSES report in contrast to the most recent global scientific consensus on Bisphenol A
Comment of the members of the Polycarbonate/Bisphenol A group (PC/BPA) and Epoxy Resin
Committee (ERC) on the four French Food Safety Agency´s (ANSES) reports on Bisphenol A (BPA)
After a more detailed review of the new risk assessment report on BPA issued by ANSES on 9.4.2013,
the PC/BPA and ERC members would like to add the following additional comments, expressing
concern at the approach taken and conclusions presented by ANSES.
Not a weight-of-evidence approach
ANSES applies “an innovative mechanism enabling to conduct an extremely detailed risk assessment
for the first time taking into account all routes of exposure”. It is not the first time that such an overall
exposure assessment is done, on the contrary: it is official practice since more than a decade in
European Chemicals Safety Policy. The ANSES approach deviates from the current method applied in
regulatory processes by taking a selective approach to define the safety threshold. The fact that
ANSES takes a new approach also means that comparisons are impossible. The results and
conclusions drawn remain standing out alone, without an opportunity to validate or put them into
In its approach, ANSES defines four “effects” of key importance to them, and selects four studies (one
per effect) to base its conclusions on, without providing a clear rationale based on transparent criteria
for why exactly these four studies were selected. In doing so, ANSES obviously gives strong weight to
these four selected studies, while at the same time basically blinding out a lot of other relevant
comprehensive studies which reach other conclusions, including solid studies which show no effect.
This is not a weight-of-evidence approach. Most worryingly, those four selected studies had already
been evaluated and discredited for being unsuitable for regulatory purposes by other regulatory
Highly selective methodology
ANSES introduces the concept of an „internal toxicological value” to take account of the presumed
amount of all „active“ BPA that may enter the body through oral, dermal or inhalative exposure. While
this is indeed a reasonable approach, the focus on the four selected studies leads to setting an
extremely low adverse effect level to start with, which is then used to put calculated internal exposure
into perspective. In addition, ANSES systematically uses and adds up sometimes irrealistic and
arbitrary worst case assumptions (e.g. always using the highest value, not the median) and applies
unrealistic calculation methods and figures for internal exposure (e.g. tripling the additionally applied
safety factor (100 -> 300), grossly overestimating dermal bioavailability to be 100%, using similar oral
bioavailability in rodents and humans although bioavailability is reported to be lower in humans than
rats) resulting in a maximal increase of the final figures. Their main area of concern relates to a single
study in rats when exposed to a daily dose of BPA at levels over 600 times higher than actual human
exposure. It would appear therefore that the employed methodology might lead to an unbalanced
Avenue E. Van Nieuwenhuyse 4/3 – B -1160 Brussels – Belgium
Tel (32-2) 675 32 97 – Fax (32-2) 675 39 35
Four studies of moderate or negligible concern
In addition to noting the ANSES approach of selecting flawed studies and applying unrealistic
calculation methods, it is interesting to read the ANSES expert team´s own assessment on the
consequences of these studies´ results. The confidence level of the expert panel associated with the
results of the Moral et al study re potential risk for human health was not united: the majority of their
expert group members judged it as “moderate”, some experts regard it as “limited”. Concerning the
other three studies/types of effects, the panel classifies the confidence level associated with the
results re potential risk for human health as “negligible”.
The ANSES risk assessment on BPA appears to be based on a set of multiple, worst case estimates,
which, through the employment of an unclear and seemingly selective approach, could create a
conclusion of somewhat limited reliability. Being aware of this assessment, and taking into account the
comprehensive amount of existing studies and assessments on BPA, it appears highly surprising how
ANSES draws the conclusion as presented in its report.
EU risk assessment and other global safety bodies see no need for additional measures
A well established and reliable Chemicals Safety system is in place in Europe, under which BPA was
investigated as extensive as hardly any other substance in the market. The Risk Assessment covers
all exposure routes of a chemical to humans and to the environment. For BPA, all exposure routes
(oral, dermal, inhalation) were investigated and assessed by European experts. On that basis, the
European Risk assessment for BPA (2003, and updated 2008) concluded that BPA poses no risk to
consumers and can be safely used in its current applications. A most recent update from FDA of
March 2013 reconfirmed the safety of BPA: “FDA’s current assessment is that BPA is safe at the very
low levels that occur in some foods. This assessment is based on review by FDA scientists of
hundreds of studies including the latest findings from new studies initiated by the agency.”
“We still cannot follow the ANSES interpretation of the existing scientific evidence. It seems to
disregard the accepted norms of European risk assessment and is in contrast to the EFSA opinion and
the statements made by other regulatory authorities worldwide”, says Jasmin Bird of the PC/BPAgroup. Consumers and other country authorities and governments alike should wait for the results of
the currently running FDA-studies on BPA and for the comprehensive EFSA re-assessment of BPA
which has been announced for publication in November 2013.
For more information:
Jasmin Bird
PC/BPA-Group PlasticsEurope
Tel: +32 2 676 17 38
Fax: +32 2 675 39 35
Avenue E. Van Nieuwenhuyse 4/3 – B -1160 Brussels – Belgium
Tel (32-2) 675 32 97 – Fax (32-2) 675 39 35
Background information on the chosen methodology and the four selected studies
Examples of the ANSES methodology
ANSES applies an additional worst case assessment factor to characterise the
“toxicological value” without giving the reasoning for doing so, or giving the criteria for
the actual factor selected.
In 2006, EFSA considered the database for BPA so strong that an additional uncertainty factor
was not warranted any more. The safety factor of 100 applied by EFSA is tripled by ANSES to
use safety factor 300, resulting in an even lower level BPA regarded to be of concern.
ANSES employs grossly over-estimated bioavailability factors to calculate the internal
dose of BPA after exposure.
o For dermal exposure, ANSES takes 100% of assumed dermal BPA exposure to be
entering through the skin. This is completely unrealistic. Comprehensive studies like
the recent Swiss authority study on dermal penetration confirmed a penetration of
max. 10 %, i.e. a 10 fold lower factor.
o As calculation basis for oral bioavailability ANSES uses 3 % of ingested BPA to
enter the body in rats and humans, based a study in rats. However, ANSES does not
use the more recent results from valid research data indicating substantially lower
bioavailability in monkey and humans compared to rats.
ANSES generally calculates with worst case 95th percentiles instead of using the
median values, which is common practice.
Reasons for dismissing studies for regulatory purposes generally include methodological flaws,
missing reliability of data, too small test groups, inconsistencies or unavailabilities of raw data,
unvalidated test procedures etc.
Moral et al, 2008, mammary gland
The study used by ANSES to justify a potential effect on pregnant women and unborn children
had been evaluated by EFSA in its 2010 opinion, by the US FDA in its 2008 interim
assessment, as well as by the Japanese AIST in 2011. They concluded: “However, this study
is severely weakened deficiencies in methodology (lack of control for estrogens, the control for
litter effects was not documented and the authors rely on data not shown). All of the studies
are limited by the nature of the findings (these studies do not demonstrate findings that are
clearly relatable to adverse findings in humans and do not demonstrate progression to
tumors).” Claims that link BPA to e.g. cancer are not supported by robust research studies
that have thoroughly investigated this question.
Rubin et al, 2001, female reproductive system (“low dose study”)
The study was assessed by EFSA 2006. It did not lead to a change of the EFSA opinion on
the safety of BPA, the current TDI was confirmed: “The Panel considered that low-dose effects
of BPA in rodents have not been demonstrated in a robust and reproducible way, such that
they could be used as pivotal studies for risk assessment.” (page 4)
Miyawaki et al, 2007, metabolism and obesity (“low dose study”)
The study was considered by EFSA 2010. It did not lead to a change of the EFSA opinion on
the safety of BPA, the current TDI was confirmed: “Because of the lack of a common clearly
defined mode of action of BPA at low doses, the toxicological relevance of the BPA effects
described cannot be evaluated and the results cannot be taken into consideration for
derivation of a TDI. While low dose effects of BPA are reported for some biochemical changes
the Panel is not aware of any clearly reproducible adverse effect expressed specifically at low
BPA doses only.” (page 5)
Avenue E. Van Nieuwenhuyse 4/3 – B -1160 Brussels – Belgium
Tel (32-2) 675 32 97 – Fax (32-2) 675 39 35
Xu et al 2010a, brain and behaviour
This study will be considered by EFSA in their upcoming opinion. It was already considered by
WHO 2010: “There is considerable uncertainty regarding the validity and relevance of these
observations. While it would be premature to conclude that these evaluations provide a
realistic estimate of the human health risk, given the uncertainties, these findings should drive
the direction of future research with the objective of reducing this uncertainty.”