The adverse effects of a mixture of phthalates fed to laboratory rats were equivalent to those predicted by adding up the doses of individual chemicals, according to scientists at the US Environmental Protection Agency.1 The findings have clear implications for authorising chemicals under the EU REACH regime, which is based on controlling risks by setting tolerable daily intakes (TDIs) for individual substances.
Phthalates are a group of chemicals used in cosmetic products and in flexible plastics. Considerable evidence from animal tests has demonstrated that they have endocrine disrupting effects, notably on male sexual development.
As a result, the use of certain phthalates in toys and childcare items has been banned in Europe since 2005 (ENDS Report 367, p 48 
). Similar restrictions have been discussed for the use of phthalates in medical devices and food contact items (ENDS Report 396, p 50 ), but so far none have been implemented.
Male sexual development depends on testicular testosterone production in the growing foetus of both humans and rodents.
Certain phthalates are anti-androgenic in that they suppress testosterone production. Testicular testosterone levels in male foetuses can be used as a measure of their effects.
In the study in Earl Gray’s laboratory at the USEPA, pregnant rats were given phthalates individually or in combination. Six phthalates were used - butyl benzyl phthalate (BBP), dibutyl phthalate (DBP), diethylhexyl phthalate (DEHP), diisobutyl phthalate (DiBP), diethyl phthalate (DEP) and dipentyl phthalate (DPP).
BBP, DBP, DEHP and DiBP were equally anti-androgenic, while DPP was three times more potent. DEP produced no effects.
When given as a mixture, the total anti-androgenicity was equal to the sum of the parts. For instance, a mixed dose of 300 milligram per kilogram body weight per day BBP and the same amount of DBP was equivalent to 600 mg/kg bw/day DEHP.
In combination with previous work, this study effectively puts the existence of additive effects "beyond dispute", according to Professor Richard Sharpe of the Medical Research Council’s Human Reproductive Sciences Unit in Edinburgh.
However, the evidence for effects in humans is much more thin on the ground. The landmark study on human effects is still a paper published in 2005 (ENDS Report 365, pp 3-4 ). This compared the ano-genital distance of baby boys with their mothers’ phthalate exposure. The distance, normally twice as long in boys as in girls, is an alternative measure of anti-androgenic effects.
A statistically significant relationship between phthalate exposure and reduced ano-genital distance was observed, but only small numbers of mother-baby pairs were monitored. At the time, Professor Sharpe told ENDS: "I expect to see a confirmation (or otherwise) of this result within a year."
These results have not appeared, to the frustration of those seeking a regulatory response. Gwynne Lyons of CHEMTrust criticised current regulatory practices: "Regulating endocrine disruptors on the basis of individual thresholds will not prevent serious health effects. If good science and health protection are to win out, then the EU REACH regulation must be interpreted with this in mind."
In a commentary accompanying Dr Gray’s research, Professor Sharpe concurs. "Risk assessment is largely undertaken on a chemical-by-chemical basis, with little or no account taken of other concurrent exposures, a practice that clearly has to change."
Regulators are considering concurrent exposures, but are uncertain how to proceed.
Three phthalates - DBP, DEHP and BBP - are under consideration on the REACH candidate list of "substances of very high concern". The so-called Annex XV report on DBP - required under REACH to argue the case for inclusion on the list - says a cumulative safety threshold "may be more appropriate" for assessing the exposure and health risks for simultaneous exposure to phthalates.
But in the Annex XV report for DEHP, its Swedish authors note the impact of co-exposure to other phthalates "was not allowed to be assessed" in a risk assessment conducted under the existing substances regulations - REACH’s predecessor. An agreed methodology for assessing co-exposure is also lacking, they point out.
This is a major obstacle to overcome. In order to calculate a cumulative threshold, chemicals with a common mode of action must be grouped together. But "we need to broaden our definition of common action", says Professor Alan Boobis, a member of the government’s expert Committee on Toxicity. Professor Boobis’ own work has showed additive effects of various oestrogenic chemicals.
Simply lumping together all phthalates, which have a common chemical structure, may not be enough. Other unrelated chemicals can also have anti-androgenic effects - and they might or might not have additive effects when combined with phthalates.
An alternative to regulating cumulative exposures to groups of chemicals would be to increase the margin of safety used when extrapolating from animal data on individual compounds to arrive at a human TDI. At present a margin of 100 times is used, made up of a factor of ten for interspecies differences and another factor of ten for sensitivity differences between individuals.
Ms Lyons suggests that an additional factor of 100 would be needed to account for additive effects from the wide range of endocrine disruptors to which humans are exposed.
A United States National Research Council committee has been evaluating cumulative risk assessments for phthalates since last December, and is due to publish a final report in the autumn.