Reproductive and developmental toxicity testing: Examination of the extended one-generation reproductive toxicity study guideline.

An important aspect of safety assessment of chemicals (industrial and agricultural chemicals and pharmaceuticals) is determining their potential reproductive and developmental toxicity. A number of guidelines have outlined a series of separate reproductive and developmental toxicity studies from fertilization through adulthood and in some cases to second generation. The Extended One-Generation Reproductive Toxicity Study (EOGRTS) is the most recent and comprehensive guideline in this series. EOGRTS design makes toxicity testing progressive, comprehensive, and efficient by assessing key endpoints across multiple life-stages at relevant doses using a minimum number of animals, combining studies/evaluations and proposing tiered-testing approaches based on outcomes. EOGRTS determines toxicity during preconception, development of embryo/fetus and newborn, adolescence, and adults, with specific emphasis on the nervous, immunological, and endocrine systems, EOGRTS also assesses maternal and paternal toxicity. However, EOGRTS guideline is complex, criteria for selecting doses is unclear, and monitoring systemic dose during the course of the study for better interpretation and human relevance is not clear. This paper discusses potential simplification of EOGRTS, suggests procedures for relevant dose selection and monitors systemic dose at multiple life-stages for better interpretation of data and human relevance.


Background
In order to provide a background for discussion of the EOGRTS guidance, the readers should be aware of several other guideline studies routinely conducted, primarily in rats, to determine immediate and latent reproductive effects of chemical exposure. Assessment of toxicity to reproduction includes possible effects of chemicals on fertility, embryonic and fetal development, peri-and postnatal development, and maternal function. Traditionally, separate reproductive/developmental toxicity studies are conducted to evaluate these effects. Guidelines OECD 414 and OPPTS 870.3700 determine effects of chemicals on embryo-fetal development/death, altered growth and structural changes (ICH, 2005;OECD, 2001a;USEPA, 1998a). Effects of chemicals on maternal behavior, length of gestation, dystocia, number and sex of pups, live births, runts, presence of gross abnormalities, and abnormal behavior in pups are determined in guidelines OECD 421 (screening test) and OPPTS 870.3550 (ICH, 2005;OECD, 1995;USEPA, 2000a). General and reproductive/developmental toxicity endpoints are combined in OECD 422 (screening test) and OPPTS 870.3650 guidelines (OECD, 1996;USEPA, 2000b).
Guidelines OECD 415 and 416 determine effects of chemicals on reproduction in one-and two-generation studies, respectively (OECD, 1983(OECD, , 2001bUSEPA, 1998b). The two-generation study (OECD 416;OPPTS 870.3800) is considered the most comprehensive design to assess reproductive toxicity (Carney and Sattivari, 2013) and the effects of chemicals on the reproductive performance of the F1 parents. The two-generation study assesses effects of chemicals on reproductive parameters listed for OECD 421 in P and F1 generations as well as the presence of gross abnormalities and abnormal behavior in F1 and F2 animals. The NTP's modified one-generation study design determines effects of chemicals on animals from gestation through weaning of F2 animals (Foster, 2014); however, no formal guideline document exists. The difference between the NTP design and other approved guidelines include retention of multiple pups per litter rather than 1 pup/sex/ litter/dose group and premating treatment of males for a full 10 weeks. However, both ICH and OECD guidelines indicate that a full 10-week premating period is often not needed, especially when other general toxicity studies (e.g., existing subchronic studies) indicate a lack of toxicity to the testes or uterus.

Current guidelines and modified approach
Most of the above described individual guidelines evaluate toxicity of chemicals to only parts of the reproductive and developmental stages with the exception the two-generation reproductive toxicity study. These guideline studies have not been updated to reflect advancements in the assessment of developmental and reproductive toxicity. For example, researchers now like to combine multiple reproductive and developmental toxicity studies into a single study and determine systemic exposure during dose rangefinding or other general toxicity studies for the selection of appropriate doses (Chapman et al., 2013;Dorato et al., 2014;Marty et al., 2013;Saghir et al., 2013). Although the two-generation toxicity study is considered "the gold standard" for the assessment of reproductive toxicity, it is complex in design, high in the utilization in animals (~2600 animals for study in rats) and with debatable value of the F2 generation (Janer et al., 2007a(Janer et al., , 2007bMoore et al., 2009;Piersma et al., 2011;Rorije et al., 2011). The two-generation toxicity study is also not designed to evaluate developmental neurotoxicity (DNT) or developmental immunotoxicity (DIT) endpoints, which require standalone studies using an additional 1280 animals.

Extended one-generation reproductive toxicity study (EOGRTS)
To make the toxicity testing across life stages state-of-the-art, the International Life Sciences Institute/Health and Environmental Science Institute (ILSI/HESI) Agricultural Chemicals Safety Assessment (ACSA) Technical Committee was charged with proposing an improved testing paradigm to assess potential effects of chemicals across life stages by incorporating the current understanding of developmental and reproductive toxicity (ILSI/HESI, 2001;Cooper et al., 2006). The committee identified key toxicity profiles across life stages beyond developmental and reproductive phases, combined studies/evaluations of endpoints across multiple life stages, and proposed a tiered testing approach for flexibility based on the needs and available data. The committee considered approaches to assess the potential of chemicals to cause adverse effects on reproduction, developmental life stages, and in the elderly. The life stage toxicity was defined as the potential adverse effects of chemicals on preconception, development (embryo/fetal and newborn/pre-weaning life stages), adolescence, and adults of all ages for reproductive and developmental toxicity, any special sensitivity with respect to general toxicity and specific effects on the nervous, immunological, and endocrine systems at critical life  stages. Additionally, they emphasized using doses that are relevant to realistic human exposures while maintaining adequate power to detect toxicity utilizing a systemic dose in a minimum number of animals (see Cooper et al., 2006;Marty et al., 2013;Saghir, 2015;Saghir et al., 2012Saghir et al., , 2013. The ILSI/HESI-ACSA proposed study design (Cooper et al., 2006) became the basis for the OECD 443 EOGRTS guideline (OECD, 2012). The study starts with exposing a sufficient number of adult male and female rats (to achieve 20 litters/dose) to the test chemical for two weeks prior to mating through weaning. Both parents are then sacrificed on study day (SD) 71 and evaluated while pups are continuously dosed with the test chemical until their scheduled sacrifice after evaluation for possible toxicological effects (Fig. 1). Groups of pups are evaluated for developmental neurotoxicity and at sexual maturity for reproductive, immuno, neuro, and general toxicity, and bred, when triggered, to produce F2 litters. The trigger to generate F2 animals in EOGRTS is based on developmental landmarks (e.g., anogenital distance, nipple retention, puberty onset) in F1 animals. In addition to the enhanced interpretative value, the EOGRTS protocol also retains multiple pups per litter, similar to the NTP study design and in contrast to retaining 1 pup/ sex/dose in conventional two-generation reproductive toxicity study protocols . Therefore, it is not clear how the NTP design offers additional advantage as mentioned by Foster (2014). Feasibility/validation of EOGRTS was achieved in four studies conducted for 2,4-dichlorophenoxyacetic acid (2,4-D) methimazole, vinclozolin, and lead acetate (Fegert et al., 2012;Marty et al., 2013;Milius et al., 2010;Schneider et al., 2011;Wright et al., 2011).
Although, the EOGRTS approach provides advantage by combining evaluations, adding DNT and DIT parameters and decreasing animal use, it is not without criticism. Even though Schiffelers et al. (2015) raised concern about the acceptance of the current EOGRTS protocol in the Europe without amendments due to criticism, the European Commission has recently adopted the EOGRTS (EC, 2015). However, the Commission has left an option for European Chemicals Agency (ECHA) to request performance of the F2 generation when justified (EC, 2015). In addition to the debate on the limited added value of the second generation (Janer et al., 2007a(Janer et al., , 2007b  et al., 2011), the organization of the OECD 443 guideline is perceived to be difficult to follow. The criteria for selecting the highest dose is unclear (even though it recommends using toxicokinetic data generated in dose range-finding or other earlier studies). In addition, procedures for monitoring the systemic dose, for better interpretation and human relevance of the animal data, are not included. Saghir et al. (2013), on the other hand, offered criteria that can effectively guide dose selection and provide a direct example of the strategy for practical implementation of EOGRTS protocols. This paper examines ways to monitor systemic dose during the course of EOGRTS using core study animals and to select appropriate doses within dose-proportional range that are relevant to actual human exposure (Saghir, 2015;Saghir et al., 2012Saghir et al., , 2013.

Role of kinetics in dose selection and incorporation into EOGRTS
Safety assessment of chemicals should focus on doses in animals that are relevant to human exposure while adequate to detect toxicity. One of the ways to determine the top dose for EOGRTS is to determine systemic dose proportionality and select the top dose based on the kinetically-derived maximum dose (KMD) at or slightly above the point of departure (POD) from dose proportionality Saghir, 2015;Saghir et al., 2012Saghir et al., , 2013. The POD from dose proportionality can be determined in a dose range-finding developmental study or in other repeat-dose toxicity studies as described by Saghir et al. (2012) and Saghir (2015). An effect observed in animals at the non-proportional systemic dose may not be relevant to the assessment of actual human risk; especially when the actual human exposure is many orders of magnitude lower than those used in animal studies. Additionally, it is recommended to have some kinetic information of chemicals in the test animal species along with likely human exposure estimates for appropriate margin of exposure before the initiation of reproductive toxicity studies with collection of further kinetic information in pregnant and lactating animals and in pups (ILSI/HESI, 2001;Cooper et al., 2006;Saghir et al., 2013). An example is given in Fig. 2 where the top dose for 2,4-D EOGRTS was selected based on KMD at slightly above the POD from proportionality of the systemic dose; the dose selected was half of the maximum tolerated dose and still several orders of magnitude higher than the expected human exposure (see Marty et al., 2013;Saghir et al., 2013 for detail). Determining systemic dose during the course of a reproductive/ developmental study is also helpful in understanding the exposure at different life-stages (Fig. 3) for better interpretation of the human relevance of the results in test animals (Fegert et al., 2012;Marty et al., 2013;Saghir et al., 2013). In dietary exposure studies, the importance of adequately adjusting doses during different lifestages is emphasized in Fig. 3. Failure to adjust dietary concentrations can result in dramatically different systemic doses of test chemicals reflective of differences in bodyweight to food intake ratio, skewing the resulting risk assessment. In order to accomplish the determination of systemic dose at various developmental life stages, an approach for a single blood collection (100 ml) at reproductive/developmental landmarks during the course of the study is proposed in Fig. 4 and Tables 1e4. Cord/pup along with maternal blood may be collected from animals used in the doserange-finding (DRF) study or from dedicated groups in the main study as outlined in Fig. 4, pooled for each litter/dose group to achieve the minimum volume required for analysis. For the collection of blood from PND 4 pups, use of culled and extra animals in Cohort 3 of EOGRTS is recommended. Blood from each litter may be pooled when needed. The DRF study for EOGRT, when needed, can be designed to determine systemic dose in dams during gestation and in dams and pups at birth or even in dams and fetuses one day before parturition, if warranted, in a few designated animals in the DRF or main study. Similarly, milk can be obtained from dams during lactation from the DRF or designated main study animals (see OECD, 2012;Marty et al., 2013;Saghir et al., 2013 for detail). The collected blood (or processed plasma/serum) is analyzed for the parent chemical and/or metabolite(s) of interest.

Detailed and easy to follow steps for EOGRTS
An easily followed general outline of the EOGRTS, beginning before cohabitation through their final termination, is presented in Table 1. The outline also includes blood sampling (a single sample of 100 ml) at different life stages in order to determine changes in the potentially fluctuating systemic dose (Figs. 2 and 3), and for better understanding of the systemic exposure and associated human relevance of the outcome. Collection of one blood sample during designated life-stages is considered sufficient due to likely steadystate systemic dose with diurnal fluctuations based on the mode of dosing (e.g., dietary, through drinking water, daily oral gavage). When a single sample-based approach fails to adequately establish KMD of a chemical, collection of up to three blood samples at specific times needs to be considered as described by Saghir et al. (2006Saghir et al. ( , 2013. Timing for the collection of blood sample(s) will depend on the kinetics of the test chemical; see Saghir (2015) and Saghir et al. (2006) for detail. The current EOGRTS design and suggested modifications to assess the systemic dose at different life stages are outlined in Figs. 1 and 4, respectively. Table 2 and Figs. 1 and 4 outline the EOGRTS processes for the F1 pups from their birth through maturation including blood sampling for the assessment of a systemic dose. Reproductive systems and general toxicity assessments for the Cohort 1A and reproductive performance, when triggered, of Cohort 1B pups along with the relationship with the systemic dose of the test chemicals and/or metabolite(s) are outlined in Table 3. The triggers to mate Cohort 1B animals to generate F2 animals is based on developmental landmarks (e.g., anogenital distance, nipple retention, puberty onset) in F1 animals. Table 4 outlines procedures for the neurotoxicity evaluation of test chemicals in Cohorts 2A and 2B and immunotoxicity evaluation in Cohort 3. A list of neurobehavioral assessments are presented in Table 5.

Conclusion
The EOGRT studies are complex and require a close and committed conversation among registrants, laboratories conducting EOGRTS, and regulatory agencies. The inclusion of additional studies for immunotoxicity assessment must also be carefully considered prior to finalization of the protocol; registrants must have a clear idea of the data needed to address the critical questions for each test chemical, and that the complex approach to a large multipurpose study is warranted for both the critical Table 3 Outline of EOGRTS for F1 cohort 1 with option for F2. with the proposed modifications, or a variant of it based on the properties of the test chemicals and issues at hand, will accomplish an overall reduction in the use of resources including the number of animals used in a series of studies conducted to assess the developmental and reproductive toxicity (and possible mode-of-action studies) of test chemicals by consolidating them into one large multipurpose study. It is agreed that the benefits of consolidating developmental and reproductive toxicity studies into one large multipurpose study must be evaluated carefully in relation to the questions needing answers.