By Troy Seidle, Humane Society of the United States, Humane Society International & Humane Society Legislative Fund
Conventional animal tests for acute systemic toxicity are rooted in the 1920s concept of the median lethal dose, or LD50: "the single dose of a substance that can be expected to cause death in 50% of the animals in an experimental group" (1). Since their conception, LD50-type studies have become the most pervasive in vivo tests in most regulatory toxicology frameworks (2), and are often conducted via more than one exposure route––oral, inhalation and dermal being the most common, but also intra-venous, intra-peritoneal, intra-muscular, etc. (3). It has been postulated that the regulatory appeal of LD50-type studies may lie in their substitution of a simplistic index of toxicity in place of real-world biological complexity (4).
Despite their continued presence in international toxicological testing frameworks, the LD50 and related methodologies have been among the most widely and harshly criticised on both scientific and ethical grounds (5). It has been stated, for instance, that "[f]or the recognition of symptomatology of acute poisoning in man, and for the determination of the human lethal dose, the LD50 in animals is of very little value" (6), and "... even if the LD50 could be measured exactly and reproducibly, the knowledge of its precise numerical value would barely be of practical importance, because an extrapolation from the experimental animals to man is hardly possible" (7). It has likewise been pointed out that "[t]he test has never been formally validated. The widespread use of the test has therefore not been based on a documented good performance, but on the lack of better tests" (8).
Whilst acknowledging that the "classical" oral LD50 test was deleted from international pharmaceutical and chemical test guidelines in 1991 and 2002, respectively (9-10), and replaced with several reduction/refinement methods (11-13), and that similar methodologies are currently under development for acute systemic studies via the dermal and inhalation routes (14), the scientific criticisms cited above appear to apply equally to the reduction/refinement variants as to the original LD/LC50 tests. Thus, the inevitable question arises: Is not 21st century science capable of greater sophistication––and humanity––than this? (15)
Opportunities for immediate replacement
Pharmaceutical acute toxicity working group
Current guidelines promulgated by the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) for non-clinical safety studies (16) state that: "The single dose (acute) toxicity for a pharmaceutical should be evaluated in two mammalian species prior to the first human exposure." However, in light of the controversy associated with acute lethality studies, as well as the knowledge that as many as 92% of drugs that pass preclinical testing fail clinical trials (17), an expert group comprised of 15 international pharmaceutical companies and contract testing facilities, together with the UK National Centre for the 3Rs, was established with the mandate to evaluate the utility and necessity of acute systemic toxicity studies in the overall preclinical safety testing paradigm.
On the basis of an extensive inter-company data sharing exercise (18), the expert group determined that "the information obtained from conventional acute toxicity studies is of little or no value in the pharmaceutical development process" (19). The conclusions of the expert group were subsequently considered and endorsed by regulators and scientists from the EU, US and Japan at a workshop in November 2006. Results from quantitative surveys completed at the end of the workshop further revealed the following (18):
- "100% of respondents found data from acute toxicity studies of little or no use and only used the information in dose setting for other studies in exceptional circumstances.
- "100% of respondents agreed that they would not carry out acute toxicity testing if it were not a regulatory requirement.
- "100% of respondents agreed that acute toxicity studies were not used to identify target organs.
- "100% of respondents never use acute toxicity data to help set the starting dose in man.
- "81% of respondents thought the data obtained from acute toxicity studies was of no use to regulators or clinicians."
Opportunities for near-term reduction and refinement
Discontinuing multi-route/multi-species studies
Current regulatory frameworks for industrial and agrochemicals, pharmaceuticals, and other substances with potentially significant human exposure call for acute systemic toxicity testing by up to three different exposure routes in rodents, rabbits, dogs and/or other animal species (16). For example, current EU and US data requirements for agrochemicals prescribe "six-pack" (i.e., oral, dermal, inhalation LD/LC50 values or estimates, skin and eye irritation, and skin sensitisation) testing for all finished formulations/end-use products (20-21), whilst recent proposals to revise these regulations would see this requirement expanded to include pesticidal active ingredients as well (22-23), potentially doubling testing requirements for studies which continue to use lethality as the principal experimental endpoint.
A similar trend is evident in the industrial chemicals sector with the EU's Registration, Evaluation and Authorisation of Chemicals (REACH) regulation. Initially, REACH called for acute systemic toxicity data via a single route, and only for chemicals marketed in volumes of ≥10 metric tonnes per annum (24). However, amendments tabled by several EU member states have led to acute lethality data being required for all substances covered under REACH (i.e., 30,000 chemicals marketed in volumes ≥1 tonne per year), plus a further requirement for lethality data via a second exposure route for chemicals marketed annually in volumes ≥10 tonnes (25). According to one stakeholder group, these new EU standards prescribe "considerably more information than required by regulatory authorities to assure that chemicals are produced and used as safely as possible" (26). Indeed, REACH information requirements for low- and medium-production volume chemicals are generally more extensive than most national base-set requirements for high-volume (HPV) substances (27).
In contrast to the brazen duplication of animal studies to evaluate acute effects, there is generally much less redundancy in testing schemes as exposure scenarios become more chronic. Indeed, risk assessors often rely on data from a single toxicity study (one species, one route) as a basis for estimates of chronic risk via multiple routes of exposure by using biokinetic modelling and related techniques of route-to-route extrapolation (28). Whilst undoubtedly motivated more by considerations of time and/or cost than animal welfare (29), it nonetheless stands to reason that if risk assessors are capable of reaching sound regulatory decisions by means of route-to-route extrapolation in lieu of redundant animal testing for some exposure scenarios, this approach could likewise become the norm for all exposure scenarios.
Oral starting dose
Based on the premise that the actions of chemicals that produce toxicity do so at the cellular level, numerous in vitro cytotoxicity assays have been developed, using a variety of cell lines and endpoint measures, as potential components in an integrated testing strategy for acute systemic toxicity (30-32). The Multicentre Evaluation of In Vitro Cytotoxicity (MEIC) programme examined the accuracy of 61 such assays, alone and in various combinations, in predicting human lethal blood concentrations for 50 chemicals (8). Several combinations of in vitro assays were found to be significantly more predictive of acute lethality in humans (R2 = 0.77 to 0.83) than rat and mouse LD50 values (R2 = 0.65). In October 2000, participants at an International Workshop on In Vitro Methods for Assessing Acute Systemic Toxicity recommended that basal cytotoxicity tests be put to immediate use in establishing starting doses for acute oral toxicity studies in animals––a strategy with the potential to reduce animal use by up to 40% (33).
The following year, US validation authorities published a guidance document on the use of in vitro data to estimate starting doses in vivo (34), and federal regulatory and research agencies in the US agreed to encourage use of these methods within the scope of their programmes (35). For example, the Environmental Protection Agency (EPA) posted a notice on its web site to "encourage those participating in the HPV Challenge Program to consider using the recommended In vitro tests ... as a supplemental component in conducting any new In vivo acute oral toxicity studies under the HPV Challenge Program..." (36). However, little evidence could be found in either the EPA's HPV Information System database (37) or the open literature to demonstrate meaningful implementation of this reduction strategy.
Longer-term replacement initiatives
One of a number of 3Rs initiatives funded through the EU's Sixth Framework Programme for research and technological development (FP6), the five-year, €15.6 million ACuteTox integrated project aims to "develop and pre-validate a simple and robust in vitro testing strategy for prediction of human acute toxicity" (38). ACuteTox brings together 37 regulatory, corporate, academic and other partners from 13 countries to build on the foundation created by the MEIC programme in an effort to optimise in vitro-in vivo correlations for acute systemic toxicity. The project is divided into the following scientific work packages (39):
- Generation of high quality in vivo and in vitro databases (which currently contain LD50 values from 2,206 animal studies and human data from 2,902 case reports).
- Iterative amendment of the testing strategy, including adapting various cell lines to commercially available high-throughput robotic platforms.
- New cell systems and new endpoints, such as in vitro production of cytokines in whole human blood cultures.
- Role of kinetics (i.e., absorption, distribution and elimination), including further evaluation of in vitro and in silico models for gut and blood-brain-barrier passage.
- Role of metabolism, including further evaluation of a variety of metabolically active cell systems.
- Role of target organ toxicity, with an emphasis on neuro-, nephro- and hepato-toxicities.
- Technical optimisation of the amended test strategy.
- Pre-validation of the test strategy.
The in-lab testing portion of the ACuteTox programme is expected to be complete by the end of December 2007, after which data will be analysed the best performing assays will be selected for pre-validation, which is slated to begin by the summer of 2008 (40).
NRC 21st century vision
In June 2007, the US National Research Council (NRC) published the interim report, Toxicity Testing in the Twenty-first Century: A Vision and a Strategy, calling for a paradigm shift that would "rely less heavily on animal studies and instead focus on in vitro methods that evaluate chemicals' effects on biological processes using cells, cell lines, or cellular components, preferably of human origin. The new approach would generate more-relevant data to evaluate risks people face, expand the number of chemicals that could be scrutinized, and reduce the time, money, and animals involved in testing" (41). It is hoped that the NRC vision will inspire US agencies to increase their support for the development and use of in vitro methods as replacement measures in appropriately validated, integrated test batteries.
A number of (quantitative) structure-activity relationship [(Q)SAR], expert system, and other in silico models have been developed for the prediction of various acute toxicity values (42). For example, the TOPKAT (Accelrys Inc., Cambridge, UK) models of oral LD50 and inhalation LC50 in rats and mice have been constructed based on published experimental toxicity data for hundreds or thousands of substances. Regulatory authorities in Canada, Denmark and the US have reported extensive use of (Q)SAR models as a basis for interim regulatory classifications of existing and/or new chemicals (43-46), with the Danish EPA noting that "according to validation results the models available ... identify the substances [of concern] with a degree of accuracy of approximately 70-85 per cent, depending on the model used" (44). Further work to build, refine and validate (Q)SAR models is under way in numerous academic, corporate, governmental and other laboratories and forums (46-48), including projects emphasising the prediction of acute toxicity via the inhalation route (49).
Overcoming barriers to progress
The NRC 21st century vision report acknowledged that "current toxicity-testing practices are long established and deeply ingrained in some sectors. Thus, some resistance to the vision proposed by this committee is expected" (50). This may partially account for the underwhelming rate of progress in some parts of the globe (51-52), despite the near consensus view expressed at the October 2000 International Workshop on In Vitro Methods for Assessing Acute Systemic Toxicity that "if the commitment to conducting a formal validation study was strong enough, the scientific resources could be harnessed for this effort with facility and the in vitro tests studied proved good enough, a replacement test battery might be achieved in as short a time as 2-3 years" (33).
Whatever the ultimate reason(s) for the delay, it is simply unacceptable that in the 21st century, the "state of the art" for acute systemic toxicity testing should remain tethered to an 80-year-old, inadequately validated methodology (53) that subjects animals to extreme, unrelieved suffering to the point of moribundity or death (54). There is an urgent need for high-level commitment from member countries of the Organisation for Economic Co-operation and Development (OECD) to buy-in to the findings of ACuteTox and similar initiatives, and to participate actively and constructively in follow-up R&D, validation and peer review activities, as needed. It is also vital that OECD members take steps to establish bilateral or multilateral "mutual recognition" agreements to facilitate more rapid and efficient international regulatory acceptance and use of validated replacement, reduction and/or refinement methods without requiring a detailed review by more than one validation authority (55).
©2007 Troy Seidle