Research and analysis

Review of Annex VIII, Section 8.7 (Reproductive toxicity testing)

Published 29 June 2023

Applies to England, Scotland and Wales

Executive summary and conclusions

Under Article 138(9) of UK REACH (Registration, Evaluation, Authorisation and restrictions of CHemicals), the Secretary of State must review Section 8.7 of Annex VIII on reproductive toxicity testing within 18 months of the UK leaving the EU. This report has been completed in compliance with that duty.

The aim was to review the standard methodologies specified for assessing the reproductive toxicity of substances that are manufactured or imported in quantities of 10 to 100 tonnes a year. It currently provides that 2 screening tests using rats, validated by the Organisation for Economic Co-operation and Development and known as OECD 421 and 422, should be used in this assessment. Annex VIII also provides for circumstances where these tests need not be followed.

The review set out to determine if there is an alternative for reproductive toxicity testing that can be mandated as the standard test method for all substances, which is the purpose of the entries in annexes VII to X. The aim was not to consider the availability of alternatives which might be used on a case-by-case basis and adapted and applied to the circumstances of individual substances. The aim was also not to assess how individual registrants have used adaptations to avoid screening tests.

To inform this report we reviewed the equivalent report by the European Commission which also investigated the requirements for reproductive toxicity. We also reviewed the information presented by the Health and Safety Executive (HSE) in their 2022 consolidated report which covered both the operation of UK REACH and the use of non-animal test methods and testing strategies under UK REACH.

Only a relatively small number of full registration dossiers had been submitted during the first year of the operation of UK REACH. However, the HSE report indicates that the dossiers that have been submitted for substances registered between 10 to 100 tonnes (those which are within the scope of this review) have used existing data in order to meet the information requirements for reproductive toxicity, rather than commissioning new animal studies.

We also sought the advice of the Hazardous Substances Advisory Committee (HSAC) and the comments made by HSAC members are included in this report. Members recognised that a single non-animal alternative was not currently available for assessing reproductive toxicity. However, they also considered ways in which more use of alternatives might be made for this endpoint, as well as how to promote new approach methodologies (NAMs) more generally. NAMs are tests, studies or other methodologies that do not involve in vivo tests (tests on live animals). These approaches are developing rapidly.

Conclusions

Most of the evidence gathered signals that it is not possible at this time to make amendments to the standard information requirements for reproductive toxicity in Section 8.7 of Annex VIII. It also indicates that it is unlikely that validated NAMs suitable for this purpose will be available in the immediate future. That is also our conclusion. However, the appropriate use of NAMs on a case-by-case basis to satisfy the information requirements for substances is possible and should be encouraged. Any use of NAMs instead of the standard methodologies for assessing reproductive toxicity would still need to fulfil the regulatory need to ensure a high level of protection of human health.

More broadly, developments and uses of science including application of NAMs are ongoing and should be kept under review.

Alongside this report the department is publishing a report under REACH Article 117(4) on the experience gained from the operation of UK REACH. That report includes information on the resources made available for the development and evaluation of alternative test methods. The government will also be publishing a Chemicals Strategy which will support ongoing work across government to facilitate the regulatory acceptance of NAMs.

In memory of Helen Pontier whose dedicated work contributed so much to the review and this report.

1. Introduction

1.1. Article 138

This report has been completed in compliance with the duty placed on the Secretary of State under Article 138(9) of REACH.

Article 138(9) states that within 18 months of the end of the EU Exit transition period the Secretary of State shall review the testing requirements of Section 8.7 of REACH Annex VIII “in accordance with the objective of promoting non-animal testing and the replacement, reduction or refinement of animal testing required under this regulation”. Replacement, reduction and refinement are collectively known as the “3 Rs”.

This timescale inevitably means that UK REACH was in operation for only a relatively short amount of time when the review was carried out.

1.2. Annex VIII, Section 8.7

Annexes VII to X of REACH set out the standard information requirements for assessing the hazards of substances to be registered under REACH following a series of physicochemical, toxicological and ecotoxicological endpoints. Each annex addresses a higher tonnage band with the information requirements building up in a cumulative manner. Annex VIII concerns substances manufactured or imported by a registrant in quantities of 10 to 100 tonnes a year.

Annex VIII, Section 8.7 concerns reproductive toxicity, that is both fertility and developmental effects of substances. This is the first tonnage band for which reproductive toxicity is included as a required endpoint.

Column 1 of Annex VIII sets the standard information required for all substances manufactured or imported in quantities of 10 tonnes or more. Column 2 lists specific rules according to which the required standard information may be omitted, replaced by other information, provided at a different stage or adapted in another way. The specific rules for adaptation are in addition to general provisions for adaptation, such as those in Annex XI of REACH.

The standard information requirements for reproductive toxicity at this tonnage are OECD TG 421 (reproductive and developmental test) or OECD TG 422 (combined repeated dose toxicity study with the reproduction/developmental toxicity). They are methods validated by the Organisation for Economic Co-operation and Development for evaluating effects of a test chemical on male and female reproductive performance in rats. The studies involve 3 test groups and a control group for each sex, with more than 10 animals per group. Males are dosed differently to females. Although described as screening tests the purpose is not to unmask potential effects on reproduction. They are apical studies, meaning that there is an observable outcome in a whole organism, such as a clinical sign or pathologic state, that is indicative of a disease state that can result from exposure to a toxicant.

2. Aims and objectives

The aim is to review the standard information requirements for reproductive toxicity testing for substances manufactured or imported in quantities of 10 to 100 tonnes a year. It aims to determine if there is an alternative for reproductive toxicity testing that can be used as a standard test method for all substances. The aim was not to consider the availability of alternatives which might be used on a case-by-case basis and adapted and applied to the circumstances of individual substances. The aim was also not to assess how individual registrants have used adaptations to avoid screening tests.

3. Method

Reproductive toxicity is a highly technical and specialist area, so our approach was to ask experts and provide supporting information from published documents.

We consulted the Health and Safety Executive as the UK REACH Agency and the Hazardous Substances Advisory Committee (HSAC).

The documents we used were:

• the HSE’s consolidated report in 2022 concerning the operation of UK REACH and the implementation and use of non-animal test methods and testing strategies
• the report by the European Commission under the equivalent duty in EU REACH

4. Results overview

The results come from 2 types of evidence: published documents from the EU and the UK REACH Agency, and opinions from HSAC members. The focus of these sources is not always the same and so their opinions and conclusions differ. In reading the evidence, it will be necessary to have regard to the aims of this review.

5. Commission report

The European Commission’s report concluded that amendments to Annex VIII, Section 8.7 were not appropriate, as follows:

The conclusion was that despite vast research efforts undertaken in the EU and internationally, and the development of some alternative approaches that model individual parts of the reproductive cycle, there are currently no alternative test methods or other non-animal approaches available that could fully replace the required screening test for reproductive/developmental toxicity in REACH Annex VIII, section 8.7.1. However, the mechanistic understanding of reproductive toxicity has increased in recent years and a range of alternative methods and approaches has been developed based on this knowledge. Such approaches can be, and are being, used on a case-by-case basis as elements in the assessment of a chemical’s potential to cause reproductive toxicity effects, e.g. in the context of read-across or weight of evidence assessments. Therefore, a revision of this section with the aim to introduce alternative test methods or approaches as the standard information requirements is currently not warranted.

6. Article 117 review

In line with the requirements of Article 117(3), HSE reported on the status of implementation and use of non-animal test methods and the testing strategies used to generate information on intrinsic properties and for risk assessment as Part 4 of their 2021 consolidated report.

Minimisation of tests on vertebrates and the avoidance of duplicate tests were key considerations in the first year of UK REACH operation. In 2021, most potential registrants were covered by the transitional registration arrangements in UK REACH which meant that they did not immediately need to submit the standard information requirements for toxicity and ecotoxicity.

Registrants of the small number of novel substances (that is substances for which an EU REACH registration did not exist) relied primarily on non-animal test methods, where these were available, or existing data to meet their information requirements.

With regard to reproductive toxicity HSE reports:

Registrants at Annex 8 are required to submit a screening test, combined for reproductive and developmental toxicity (OECD 421 or 422). This applies if there is no evidence from structurally related substances, QSAR estimates, or in vitro methods that the substance might be a developmental toxicant.

Adaptations are possible in specific circumstances, for example, where a pre-natal developmental toxicity study or a two-generation/extended one-generation reproductive toxicity study is available. The latter tests are required at Annexes 9 and 10 (a reproduction study is triggered at Annex 9 and a standard requirement at Annex 10).

The three higher-tonnage dossiers each contained information from two developmental toxicity studies in two species (rats and rabbits). Those reported in two dossiers had been conducted in the 1980s, whilst those reported in the third had been conducted for other regulatory purposes (active substance in plant protection products).

An existing two-generation reproduction study was reported in each higher-tonnage dossier, two of which were conducted in the 1990s and one of which was performed to meet the requirements of other legislation (active substance in plant protection products). Each of these registrants omitted the reproduction/developmental toxicity screening studies, since higher-tier studies to investigate reproductive toxicity were available.

Therefore, in all cases registrants met the information requirements for reproductive toxicity with existing data.

7. Hazardous Substances Advisory Committee (HSAC)

The department approached HSAC and invited their expert input on:

• whether there were currently alternatives which could fully replace the current standard information requirements for reproductive and developmental toxicity
• evidence gaps for reproductive toxicity testing and assessment
• the availability and applicability of new approach methodologies such as QSARs (quantitative structure-activity relationships) and read-across

Several members of the committee provided input on an individual basis and these are brought together in the following discussion.

The immediate conclusion was that no single new approach methodology (NAM) is going to be comparable or replace a full in-life reproduction study in rats as defined in OECD methods 421 and 422. Nevertheless, committee members did consider ways that progress could be made through the more efficient use of the existing standard tests or structured use of NAMs in the assessment of reproductive toxicity. They also considered ways to reinforce the use of non-animal alternatives more generally.

Use of NAMs for assessing reproductive toxicity

The current studies OECD 421 and OECD 422 are not screening tests, contrary to their titles. OECD 422 is likely to be a better study for assessment if additional endpoints are included. It can be argued that the benefits of OECD 421 are appearing to have an increasingly minor role. Although OECD 421 uses fewer rats than OECD 422, if further additional studies need to be conducted after an OECD 421, then that seems to take the risk versus benefit assessment for doing the 422 towards a more ethical conclusion. This could be taken up when developing testing strategies, for example a cost benefit analysis could be conducted to find the number of unnecessary animals with each approach and compare those studies with both OECD 421 and 422 against those with OECD 421 or 422 alone. The conclusion would then be used to decide on the better testing approach.

A greater emphasis on understanding the causation and mechanisms of action could help future risk assessments beyond the current apical in vivo studies. Rat experiments provide a measure of dose and effect but the context of the reasons for toxicity comes from other studies. Using a combination of alternatives could provide more high quality information on the mechanisms at play than the status quo, and therefore provide a better science-led approach to risk assessment of chemicals.

Until there is sufficient data available upon which to base learning and understanding it will be difficult to move away from the current animal test paradigm for reproductive toxicity. Although regulators are not yet in a place to make their decisions without the in vivo data there might be opportunities to provide them with the tools and authority to make these decisions drawing on their scientific knowledge and the evidence presented. This suggests having NAM data available alongside in vivo data for a period of time while the experts learn what works.

It was also considered that regulatory expectations could change to encourage better approaches, weight of evidence and scientific insight as viable alternatives. Increased traction in applying the alternative 3Rs approaches could be secured if registrants are directly asked and required to provide such approaches. Using regulatory tools to secure such data would be a demonstration of leadership. After a period of time when alternative data for this endpoint is presented, there could be a review of the approach with an evidence-based assessment of data produced in parallel to data from standard methodologies.

Broader approaches towards NAMs

A similar approach to increasing use of, and familiarity with, non-animal studies was also proposed beyond the reproductive toxicity endpoint. Such data could be actively invited in addition to information generated by standard methodologies by use of existing or new regulatory tools in REACH rather than relying on the current permissive language. It was acknowledged that formally requesting these data in addition would clearly have a cost to both registrant in generating the data and the regulator in reviewing it. However, it was felt this could overcome the block to future potential and acceptance of non-animal methods and other NAMs, which can result from the desire of regulators to understand the results of such studies in the context of their current data and protection goals.

A large international study is in progress, Precision Tox, that is seeking genomic links between species to predict toxic responses to chemicals.

It is funded by the EU, coordinated by the University of Birmingham and involves scientists from around the world. It started in January 2021 and will conclude in February 2026. Precision Tox aims to transform regulatory toxicology by simultaneously deploying five biomedical model species and human cell lines to uncover molecular toxicity pathways shared across the animal kingdom, enabling chemicals to be classified according to their precise effects on human health.

There are many other NAMs initiatives, which demonstrate the determination to replace, reduce or refine animal testing to protect human health and the environment. It is a rapidly developing area, with potential to lead to changes, and should be kept under review.