Revision of the ICH guideline on detection of toxicity to reproduction for medicinal products: SWOT analysis

Accepted Manuscript Title: Revision of the ICH guideline on detection of toxicity to reproduction for medicinal products: SWOT analysis Author: Paul Barrow PII: DOI: Reference:

S0890-6238(16)30049-1 http://dx.doi.org/doi:10.1016/j.reprotox.2016.03.048 RTX 7262

To appear in:

Reproductive Toxicology

Received date: Revised date: Accepted date:

2-3-2016 29-3-2016 31-3-2016

Please cite this article as: Barrow Paul.Revision of the ICH guideline on detection of toxicity to reproduction for medicinal products: SWOT analysis.Reproductive Toxicology http://dx.doi.org/10.1016/j.reprotox.2016.03.048 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Revision of the ICH guideline on detection of toxicity to reproduction for medicinal products: SWOT analysis Paul Barrow Roche Pharmaceutical Research and Early Development, F. Hoffmann-La-Roche, Ltd., Basel, Switzerland. Corresponding author Paul Barrow Roche Pharmaceutical Research and Early Development Pharmaceutical Sciences Roche Innovation Center Basel F. Hoffmann-La Roche Ltd Grenzacherstrasse 124 CH 4070 Basel Switzerland [email protected] +41 61 687 8040

1

Highlights 

SWOT analysis of current ICH S5(2) on toxicity to reproduction for pharmaceuticals: o

Strengths: safety record, acceptance, flexibility…

o

Weaknesses: alternative tests, risk assessment, limit dose, mAbs, drugs in semen…

o

Opportunities: 3Rs, TK, testing strategies, new technologies, maternal toxicity…

o

Threats: emerging science, contradiction with other guidelines, bureaucracy, etc.

2

Abstract SWOT analysis was used to gain insights and perspectives into the revision of the ICH S5(R2) guideline on detection of toxicity to reproduction for medicinal products. The current ICH guideline was rapidly adopted worldwide and has an excellent safety record for more than 20 years. The revised guideline should aim to further improve reproductive and developmental (DART) safety testing for new drugs. Alternative methods to animal experiments should be used whenever possible. Modern technology should be used to obtain high quality data from fewer animals. Additions to the guideline should include considerations on the following: limit dose setting, maternal toxicity, biopharmaceuticals, vaccines, testing strategies by indication, developmental immunotoxicity, and male-mediated developmental toxicity. Emerging issues, such as epigenetics and the microbiome, will most likely pose challenges to DART testing in the future. It is hoped that the new guideline will be adopted even outside the ICH regions.

Keywords: reproductive and developmental safety testing; guidelines; teratogenicity; ICHS5

3

1. Introduction This article is intended to accompany an invited talk at the 44th annual meeting of the European Teratology Society, the views expressed are those of the author alone and do not represent the policies, positions or opinions of any organization, group or company. The ICH guideline on “Detection of toxicity to reproduction for medicinal products” was the first ICH safety guideline. Following its issue in 1993[1], it rapidly gained worldwide acceptance (see below for the notable exceptions of China and India). This guideline successfully harmonized the disparate requirements of the regulatory authorities in Europe, the USA and Japan. A significant reduction in experimental animal use was thus accomplished by abolishing the need to duplicate the various nonclinical reproductive toxicity studies to achieve marketing authorization for a new drug across the three regions. This remarkable achievement marked the birth of the International Conference on Harmonisation (recently renamed International Council on Harmonisation). The principles of the ICH guideline were based on a previous guideline issued by the FDA in 1966[2] in response to the thalidomide tragedy and subsequent guidelines from the EC and Japan. The initial ICH S5 guideline left open some questions on the minimum duration of treatment of males before mating and the relative value of semen analysis, mating performance and histopathology for the evaluation of testicular toxicity. Subsequent literature surveys[3] and validation studies[4, 5] concluded that histopathology of reproductive organs is the most sensitive method for detecting effects on spermatogenesis. An addendum to the guideline to incorporate these findings was issued in 2000. The ICH S5(R2) nomenclature was added in 2005, when the title of the guideline was changed to “Detection of Toxicity to Reproduction for Medicinal Products & Toxicity to Male Fertility”. The studies described in the guideline are designed to detect each of the four known manifestations of developmental toxicity, i.e.: (1) death (embryo–fetal resorption, abortion, stillbirth or post-natal mortality), (2) growth retardation (resulting in low birth weight or depressed post-natal growth), (3) malformation, and (4) functional deficit[6]. The guidelines define six phases of reproduction that need to be assessed: (A) adult fertility, (B) early embryonic development before implantation on the uterus, (C) embryonic organogenesis, (D) fetal development, (E) birth and pre-weaning development and (F) post-weaning development up to sexual maturity. A three segment strategy is proposed to cover the evaluation of all of these phases, comprised of a fertility study (usually in the rat), embryofetal development (EFD) studies in two species (usually rat and rabbit) and pre- and post-natal development (PPND) studies (usually in the rat). Various options are also proposed for combining two or more of the various rodent studies into a single experiment. More than 20 years after taking effect, ICH S5 guideline is about to undergo its first major revision[7]. This review uses a SWOT (strengths, weaknesses, opportunities and threats) analysis to identify and discuss salient points to be considered in the ICH S5 guideline revision.

2. Methods The strengths and weaknesses of the current ICH S5(R2) guideline were assessed, along with opportunities and threats pertaining to the revision process. Each item is listed by category and 4

discussed with regard to the future revised guideline (R3). Identifying the strengths of the guideline highlights the current existing elements that should not be removed or disrupted in the revision process. The identified weaknesses highlight issues that have arisen since the adoption of the guideline or where scientific thinking or technology has evolved. Opportunities represent desirable additions to the guideline, such as alignments with other ICH guidance documents. Finally, threats are issues that should be dealt with proactively where possible in case they become obstacles to the successful completion of the revision process.

3. Results 3.1. Strengths of the current guideline Safety Record. The existing procedures described in the guideline have been remarkably successful in identifying the reproductive hazards of new drugs. Since the adoption of the guideline, there have been no developmental toxicity-related tragedies with marketed drugs. Effective drug labeling and more cautious drug prescription practices have of course also contributed to this success. The study designs described in the guideline have proven to be effective in detecting reproductive and developmental hazards associated with mechanisms of action that were not yet envisaged when the guideline was devised. This, however, is not a reason to be complacent (see Weaknesses). Wide acceptance. ICH S5(R2) rapidly gained acceptance by most health authorities worldwide, including those outside of the ICH regions (with the exception of China and India, see Threats) and has been the de facto standard for more than 20 years. 3 Rs. From its adoption, ICH S5 resulted in a significant reduction in the number of experimental animals used for the regulatory developmental and reproductive toxicity (DART) testing of new drugs. By harmonizing study designs, the guideline removed the need for specific DART studies to meet the regulatory requirements of each region for a worldwide marketing submission. We must continue to strive to reduce or eliminate animal use for drug safety testing, without compromising safety. Established robust study designs. Despite their complexity, the DART study designs have proven to be practical and effective, having incorporated the best elements from the previous guidelines in the three ICH regions. The necessary equipment is readily available and the methods have become routine. Safety testing laboratories have built up large databases of reference values over the last two decades. Flexibility. The guideline avoids mandatory rules, favoring flexibility (for instance, with respect to the combination of various rodent studies). This flexibility has allowed the guideline to remain relevant even for classes of drug that had not been invented when the guideline was published. This flexibility should be retained in the revised guideline. Testicular toxicity. The research culminating in the amendment on testicular toxicity remains pertinent today and even anticipated issues that would arise years later with respect to the testing of biopharmaceuticals (see Weaknesses, below).

3.2. Weaknesses of the current guideline 5

The title. “Detection of Toxicity to Reproduction for Medicinal Products & Toxicity to Male Fertility” is long, clumsy and misleading. Male fertility is part of toxicity to reproduction and does not need to be repeated. Furthermore, female fertility is not mentioned, even though it is covered in the guideline. It is ironic that male fertility was added to the title in acknowledgment of work that concluded that the most sensitive marker of testicular toxicity can be determined in the general toxicity studies rather than in the DART studies. Hopefully, the revised guideline will have a more balanced title. No provision for alternative tests. Some alternative tests –i.e. in-vitro, ex-vivo, in-silico and nonmammalian systems– are mentioned in the current guideline “for encouragement”, but no options are provided to replace studies in live mammals. While very few new alternative tests for DART have been developed over the last 20 years, many of the previously existing tests have been extensively validated and qualified[8]. Unfortunately, a predictability of more than 80% with respect to animal studies has not been reliably reported for any of the available tests[9]. Also, the physical characteristics of drug candidates (solubility, pH, osmolality, etc.) often render alternative test systems impractical. Because of these limitations, alternative test systems cannot be expected to completely replace live animal studies for regulatory DART testing in the near future. Nonetheless, alternative systems, such as the embryonic stem cell test[10] and the zebra fish teratogenicity test[11], are used by many companies for drug candidate selection and are thus contributing to reduced animal use in pharmaceutical development. The revised guideline should give guidance on how to qualify an alternative test system and the conditions that should be met before an alternative method can be used in the place of a mammalian study. The (fortunate) paucity of human data for the majority of known teratogens, makes it very difficult to demonstrate that a new alternative test is more or as effective than an established animal study for the detection of human teratogenicity. One current school of thought considers that a prospective alternative test should be demonstrated to be predictive of the animal test that it is intended to replace. So, for instance, before a zebra fish test can be used in the place of the rabbit EFD, it would have to be shown to reliably detect the same list of known teratogens as the rabbit EFD. This principle is illogical at best, however, when applied to a human stem cell test, which logically could be expected to be more predictive for the human than an animal-based test. Alternative tests for teratogenicity in general show poor specificity (i.e. true negative rate) due to a very poor capacity to predict developmental effects arising from maternal influences (e.g. reduced perfusion of the placenta) or as the result of active metabolites produced by maternal metabolism (e.g. allyl alcohol metabolized to ethanol). On the other hand, many alternative systems show high sensitivity (true positive rate), so a positive result in an alternative system may be highly predictive of teratogenicity in in-vivo animal studies. So, while a negative result for teratogenicity in an alternative test may not be completely reassuring of lack of teratogenicity in the human, a positive teratogenicity finding in an alternative test may make further animals studies redundant. Few alternative tests, however, provide sufficient information on the exposure at which teratogenicity occurs relative to the human therapeutic exposure. The revised guideline could give guidance on situations where a positive finding in an alternative test could negate the need for in-vivo studies in one or both species.

6

No guidance on human risk assessment. The current guideline is essentially restricted to hazard assessment. The revised guideline could give guidance on how to assess the human relevance of reproductive or developmental hazards detected in regulatory safety studies. One major consideration is the acceptance (or not) of a safety margin for teratogenicity. Karnofsky’s law[12] states that any substance can be teratogenic if given to the right species at the right dose during the right stage in development. Drugs that are teratogenic at very high doses in animals may be safely used at a lower dose in the human. The revised guideline should give some indication of the expected safety margin between the no adverse effect level (NOAEL) in an EFD study and the anticipated human exposure at the maximum recommended human dose (MRHD). Similar considerations apply to other manifestations of developmental and reproductive toxicity, each of which could have a different safety margin, e.g. embryotoxicity versus reduced sperm motility. Another aspect of reproductive risk assessment, not related to the DART studies but which may necessitate contraception requirements in the drug label, is genetic toxicity. Although unlikely, congenital anomalies can be induced by genetic toxicity to the germ cells[13]. Generally, the routine genetic toxicity studies using somatic cells are also capable of detecting effects on germ cells. If a drug is considered to present a risk of genetic toxicity under the conditions of use, contraception will most likely be required for patients of reproductive potential. The risk of genetic toxicity to germ cells is greater in the male due to continuous division of spermatic stem cells, than in the female where the primary oocytes are retained in an arrested prophase. Exposure of the germ cells occurs in the gonads, so plasma and tissue concentrations (rather than semen concentration) are the appropriate measures of exposure for transgenerational genetic toxicity. An effective form of contraception (i.e. not condoms alone) should be used during treatment of the male partner and for at least three months after the drug exposure has fallen below a level of concern for genetic toxicity. This 90-day period of contraception after the last dose corresponds to the duration of a complete spermatic cycle, as defined in a recent guideline issued by the European Clinical Trial Facilitation Group[14]. The recent ICH M7 guideline [15] gives guidance on how to determine a threshold of concern for direct or indirect genetic toxicity. The rapidly dividing cells of the embryo and fetus are sensitive to genetic toxicity, so effective contraception should be used by women of child-bearing potential whilst taking a genotoxic drug. Outdated approach to high dose setting: limit dose. The current guideline states that some minimal toxicity is expected to be induced in the high dose dams. A list of factors limiting the high dose level is given. For drugs that do not induce maternal toxicity at high dose levels, a limit dose of 1000 mg/kg/day is specified, but no provision is made to limit the dose to a pre-defined multiple of the human exposure. For general toxicity studies it has become accepted practice to limit the high dose level in safety studies to that which provides a 50-fold margin above the anticipated AUC in the human at the MRHD (or 10-fold for biopharmaceuticals). ICH M3(R2)[16] causes confusion by stating that the principle of limiting the high dose level to a multiple of the human AUC is consistent with the recommendations for reproductive studies, whilst failing to revoke the requirement for maternal toxicity in ICH S5(R2). Today it is not unknown for regulatory authorities to insist on maternally toxic doses in reproductive studies, far exceeding the dose levels used in the general toxicity studies. The consensus among reproductive toxicologists is that high doses in excess of 50-times the AUC at the MHRD, do not contribute to the interpretation of the study. Indeed, many reproductive toxicologists would be in favor of a 25-times AUC limit dose, above which a positive indication of developmental 7

toxicity is generally considered to be of less concern for human risk assessment[17]. This lower margin would allow tighter intervals between dose levels, with the low dose exposure close to the therapeutic range. According to an analysis of recent FDA approvals[18], 90% of drugs showed maternal and/or fetal toxicity in rat or rabbit EFD studies at exposure levels below 25-times the human AUC, even in studies where the high dose level provided an exposure in excess of 50-times the human AUC. The revised guideline should give a clear recommendation on high dose setting. Number of dose levels. The current guideline makes provision for the use of a single dose group given 1000 mg/kg/day when no adverse effects are expected. In practice, this provision has been very rarely accepted, since dose-response data are often crucial to risk assessment. Even if no effects are seen, multiple dose groups improve the predictive power of the study. In general, three dose levels are preferred, though two are often sufficient for mAbs. No guidance on male-mediated developmental toxicity (drugs in semen). There are no documented examples to date of male-mediated developmental toxicity in either animals or humans. One possible mechanism by which such an effect could occur (see also epigenetics under Threats, below) is by transmission of the drug from a treated male via the semen to a potentially pregnant female partner during sexual intercourse. For drugs that are potent teratogens at a small fraction of the MHRD, seminal transfer could conceivably present a risk of male-mediated developmental toxicity. Many drugs are present in the semen, usually at a similar concentration to that in plasma. The highest recorded drug concentration in human semen was 11-times the plasma concentration for cydamycin[19]. Recent work has demonstrated that the principle route of exposure of the embryo following vaginal administration in semen is via the maternal circulation after absorption across the vaginal mucosa[20]. The total dose of drug transmitted to the female in the semen is at least 1000-times lower than the dose administered to the male, owing to the low volume of ejaculate (5-6 mL). The resulting maximum female plasma exposure can be calculated by dividing the total dose in the ejaculate by the maternal plasma volume. Alternatively, routine pharmacokinetic modelling techniques may be used to obtain a more accurate estimate by taking into account the kinetics of absorption from the vagina. The accuracy of the calculations may be further improved by using the human semen concentration, determined during clinical trials, in the place of male plasma concentration. Condoms, while not an effective form of contraception, are effective for the prevention of seminal transfer of drugs provided they are used throughout treatment and after the last dose until male plasma levels have fallen below a level of concern. These precautions are also warranted for vasectomized men. This risk is not pertinent for biopharmaceuticals owing to their low concentration in semen, low bioavailability following vaginal administration and poor systemic absorption. The revised guideline should give guidance on how to asses developmental risk posed by the seminal transport of a drug and the measures to be taken if a risk is identified[21]. No guidance on the testing of biopharmaceuticals. The DART testing of humanized therapeutic mAbs was not envisioned in the current guideline. The first humanized mAb was approved in 1997. For mAbs that do not exert their pharmacological action in lower species, a non-human primate (NHP), usually the cynomolgus monkey, is most often used. However, developmental toxicity studies should only be conducted in NHPs when they are the only relevant species. ICH S6(R1), issued in 8

1997, defines an ePPND study design in the NHP that is now routinely used for mAbs in the place of separate EFD and PPND studies. Homologous (or surrogate) antibodies have been used in the place of the humanized clinical mAb to allow the DART studies to be performed in rodents. ICH S6(R1), however emphasized the limited value of surrogate studies, “which are generally not useful for quantitative risk assessment”. Nonetheless, surrogate studies can be useful to help understand the developmental consequences of the intended pharmacological action of a mAb. The revised guideline should give guidance on the testing of biopharmaceuticals. No guidance on the testing of vaccines. When the current guideline was issued, there were no requirements to test vaccines for reproductive toxicity. Developmental toxicity studies (but not fertility studies) are now required prior to approval of vaccines in the USA and Europe, following issue of an FDA guidance on considerations for developmental toxicity studies for preventive and therapeutic vaccines for infectious disease indications in 2006[22, 23]. The revised guideline should align with the FDA requirements for preventative vaccines and should also give recommendations for therapeutic vaccines.

3.3. Opportunities 3 Rs. The revision of the guideline will be an opportune time to review all recommendations with a view to replacing in-vivo animal studies wherever possible, reducing the numbers of animals used and refining the study protocols to minimize animal suffering. The 3Rs principles must be applied throughout the entire guideline. Integrative testing strategies by indication. The testing requirements for DART and any resulting precautions may be more or less stringent depending on the indication(s) for which the drug will be used. The risk-benefit analysis for a drug intended to treat advanced cancer, for instance, is very different than that for a preventative vaccine. Likewise, it seems reasonable to accept fewer or smaller studies for drugs that will be used in a very limited population or under very controlled conditions (e.g. hospitalized patients in intensive care) than for indications that affect large numbers of women or men of reproductive age. For instance, when an EFD or a preliminary EFD (pEFD) study shows teratogenicity in one species, the question should be asked if further EFD studies are necessary. The answer to this question will depend on the exposure level at which the effect was seen and the perceived safety margin appropriate to the indication and conditions of use of the drug. The revised guideline should also assess how the testing requirements can be simplified in order to accelerate the marketing authorization for drugs of great medical need. Development and qualification of alternative tests. The revised guideline should pave the way for the qualification of future alternative tests as they become available, thus allowing live animal experiments to be replaced in the future. Elaborate on toxicokinetic parameters used for exposure comparison between species. The existing ICH S5(R2) guideline emphasizes the importance of toxicokinetic data in DART studies. AUC (“body burden”) is suggested as the principle metric of exposure for the purpose of dose selection. The revised guideline should give further guidance on the extrapolation of exposure data from DART studies to the human clinical situation. In particular, the influence of plasma protein binding should 9

be mentioned[24]. For very highly-bound drugs, small differences in plasma binding affinity between species may result in large differences in bioavailability. For example, a drug that is 98% bound in the human and 99% bound in the rat will have a two-fold higher bioavailability in the human for the same total exposure. For less highly-bound drugs, minor differences in plasma protein affinity have much less impact on bioavailability. Integration of new methods. Methods and techniques have evolved considerably since the issue of the guideline, e.g. fetal examination by X-ray micro computer tomography[25] or by magnetic resonance imaging[26]. Likewise, much more sophisticated methods are now available for toxicokinetic investigations. Microsampling techniques may be used to perform more analyses with low volumes of blood and hence fewer animals[27]. The revised guideline should promote the use of any modern methods that can improve the quality of the generated data and maximize the useful information obtained from each experimental animal. Flexibility should also be retained to allow the integration of future technologies as they become available. Clarify the appropriate duration of contraception requirements. Some confusion and inconsistencies in the duration of contraption requirements for female and male patients between regions could be avoided by clear guidance. If a risk of harm to the fetus has been established under the clinical conditions of use, effective contraception should be used by women of child-bearing potential during treatment with the drug. Contraception should be continued after the last dose until drug plasma levels have fallen below the level of concern (usually five half-lives). For requirements in men, see genetic toxicity and male-mediated developmental toxicity under Weaknesses, above. These issues may be more appropriately covered in a clinical guideline. Alignment with other ICH guidelines. The revised guideline should eliminate any inconsistencies regarding DART studies between the various ICH guidelines, e.g. see S6, S9 and M3 cited above. A prime example is the S11 guideline currently under development for nonclinical safety testing in support of development of pediatric medicines[7]. This opportunity should be seized to promote the integration of juvenile endpoints into the DART protocols (e.g. PPND or ePPND studies) and avoid the need for additional dedicated juvenile studies whenever possible. Alignment with other pharmaceutical legislation in ICH regions. The revised guideline should be designed to comply with the national pharmaceutical legislation in each of the ICH regions. For instance, the non-clinical studies must provide the data required for inclusion in the drug label in each country. Get agreement from non-ICH regions. Whether or not China and India finally join the ICH, it is in the interests of all parties if the revised guideline can be adopted worldwide. The ICH guidelines are currently not applicable in China and India, but studies designed and run to ICH principles are generally acceptable[28]. The EFD and fertility studies are currently required before the start of Phase 1 clinical trials in China, even if only male volunteers will be exposed. Elaborate and expand on possible study combinations. The current guideline already presents options for combining the various DART studies (see above). The possible strategies should be evaluated for use with new pharmaceutical classes (e.g. oligonucleotides). The opportunity should also be taken to evaluate which endpoints could be dropped to save animals without compromising the power of the study. For example, the EFD and PPND rodent studies could be combined without 10

increasing animal numbers provided that post-natal examinations of the pups (as in the PPND study) can be shown to be as sensitive for the detection of embryotoxicity and of teratogenicity as morphological examination of the fetuses derived by c-section (as in the EFD study). Elaborate on species selection. The preferred species for DART studies, i.e. rat, rabbit and mouse, have not changed. However, EFD studies in the minipig are now commonplace[29], while other species cited in the original guideline present severe practical difficulties, .i.e. guinea pigs, hamsters, ferrets and dogs. The revised guideline should outline the relative advantages and drawbacks of each species. Maternal toxicity. Scientific thinking regarding the influence of maternal toxicity on adverse pregnancy outcomes has evolved over recent years[30]. In addition to considerations for dose selection (see Weaknesses, above), the revised guideline could give guidance on the impact of maternal toxicity on the interpretation of study results. It is now considered that maternal toxicity cannot simply be used to explain fetal malformations. Maternal toxicity can be responsible for manifestations of reproductive toxicity, but should not be assumed to be the cause of developmental effects in a study unless the link can be demonstrated.

4. Threats Poor understanding of teratogenic mechanisms. While several mechanisms of teratogenicity have been elucidated over recent decades, the mode of action of many teratogens is still poorly if at all understood. The classic example is thalidomide, which ironically sparked the publication of the FDA guideline on reproductive toxicity and marked the start of the current era of regulatory toxicology. More than 30 theories for thalidomide embryopathy have been proposed, but the mechanism of action is still not fully understood[31]. Indeed, there is still doubt whether a ‘new thalidomide’ would be detected today. The main categories of teratogenic mechanism according to current knowledge[32] are: folate antagonism, neural crest cell disruption, endocrine (sex hormone) disruption, oxidative stress, vascular disruption and specific receptor- or enzyme-mediated teratogenesis. The final category includes many diverse teratogenic classes, including: angiotensinconverting enzyme inhibitors and angiotensin II antagonists, histone deacetylase inhibitors, cyclooxygenase-1 inhibitors, serotonin agonists or antagonists, carbonic anhydrase inhibitors and potentially any novel therapeutic class or innovative drug yet to be discovered. Fortunately, there is very little human experience with recent classes of teratogen, so it is unknown whether the hazards detected in animal safety studies actually represent a human risk under the clinical conditions of use. The revised guideline should at least be capable of detecting all known classes of reproductive hazard, without being over-predictive. Contradiction with other guidelines. The scope of the S5 revision does not allow contradiction with more recent ICH guidelines. This restricts the scope of the improvements that can be made. For example, a postponement of the requirement for the fertility study is not within the scope of the ICH S5 revision, since the timing of the DART studies relative to clinical trials is specified in ICH M3(R2). Fertility evaluations for biopharmaceuticals that can only be tested in the NHP are limited to a histopathological examination of the reproductive organs in the general toxicity studies. A dedicated fertility study is not required, mating studies not being reliable in NHPs[33]. For small molecule drugs, on the other hand, a fertility study is currently required before Phase 3, as stipulated in ICH 11

M3(R2). If the fertility study could be postponed until after Phase 3, it could possibly be combined with the PPND study, thus saving animals. The risk to patients of permanent effects on the reproductive organs would be minimal, since these are evaluated in the sub-chronic and chronic repeat dose studies. At worse, effects on libido and sexual function could remain undetected until after the Phase 3 trials. Discordance between ICH regions. One area where there is clear discordance on DART testing requirements between ICH regions is the need for EFD studies prior to the inclusion of women of child-bearing potential (WOCBP) in clinical trials, as stipulated in ICH M3(R2). In the USA, no EFD studies are required before Phase 3, provided that stringent precautions are taken to prevent pregnancy during clinical trials. In Europe and Japan, EFD studies are required before inclusion of WCBP in clinical trials, though up to 150 WOCBP may be included in studies of up to three months in duration if acceptable preliminary EFD (pEFD) studies are available. The current S5 guideline states that group sizes resulting in less than 16 litters for examination give inconclusive results, whereas the pEFD studies have a minimum of six dams per group. This contradiction is likely to persist because of the limited scope of the S6 revision (see above). Immunotoxicity. Not all possible reproductive and developmental hazards can be reliably detected using current testing methodologies. One example is developmental immunotoxicology. Several drugs have been shown to adversely influence the development and maturation of the immune system in animals. The rat appears to be a good model to detect such effects[34]. The appropriate exposure period covers gestation and the post-natal period up to sexual maturity, which is outside of the scope of the current DART study designs. A developmental immunotoxicity arm is now included in the OECD testing protocols for chemicals[35]. An equivalent dosing scenario for pharmaceuticals would be difficult to achieve within the constraints of the current organization of the ICH S5 guideline. The juvenile toxicity studies may include an assessment of developmental immunotoxicity, as will be defined in the ICH S11 guideline, but these studies do not encompass the pre-natal period of vulnerability. Transgenerational epigenetic effects. Epigenetic mechanisms of gene regulation, such as DNA methylation and chromatin modification, appear to play an important role in the fetal basis of adult disease susceptibility[36]. These factors also play a role in male infertility and have been shown to be transmissible across generations[37]. For example, paternal cyclophosphamide treatment is suspected of causing aberrant epigenetic programming in the early embryo[38]. The current package of DART studies is not designed to detect paternally-mediated effects other than reduced fertility in an exposed generation. Furthermore, it is not clear if second-generation developmental defects could be detected if they were not already manifest in the offspring of exposed females. If these hazards prove to be pertinent to hazard identification, it is hoped that biochemical assays or in-vitro tests could be developed to detect the induced changes in DNA or histones without the need for invivo studies. Microbiome. Perturbation of the microbiome (i.e. the collective genome of all the microorganisms colonizing the body) during development has been hypothesized to play a role in pediatric disease and also in disease onset later in life (e.g. Type 1 diabetes)[39]. Seeding of the microbiome in the neonate relies on transfer of microorganisms from the mother before, during and after birth. Drugs, particularly antibiotics, may have the capacity to unbalance this process resulting in a permanent 12

skewing of the microbiome, with possible adverse consequences throughout life. More research will be necessary in this area before experimental endpoints or disease models can be developed and integrated into a DART testing guideline. Scientific disagreement. For the revised guideline to reach a successful conclusion, the expert working group will have to reach agreement on a series of complex scientific issues in order to produce a technical document (Step 1). Once agreed amongst the ICH parties (Step 2a), the draft guideline will be adopted by the regulators in each region (Step 2b) and submitted for regulatory consultation and discussion (Step 3). Only after consensus has been reached will the final guideline be adopted (Step 4). The process will be finally complete after incorporation of the guideline into the legislation of each region (Step 5). In reality, complete disaccord amongst scientists in the field of DART is unlikely since the published literature in the field is fairly conclusive. When disagreements do occur, additional data may be needed to clarify the issues, as was the case regarding male reproductive toxicity in the original guideline (see above). In theory, if the parties fail to agree, the final decision can be taken by the regulators alone, but in reality the differences in points of view almost never fall along a neat line of fracture between the regulators and industry (or between regions). Lawyers and bureaucracy. The implementation of the revised guideline could easily be hindered by legal or procedural challenges, which could come from the industry, national health authorities or public advocacy groups. The pregnancy and lactation labelling rule in the USA was held up for six years before issue of the final rule in 2014[40]. Politics. Regulatory safety legislation is used as a trade barrier by governments wishing to curb imports. Nationalistic interests could easily derail the ICH process.

5. Discussion and conclusion SWOT analysis is a useful tool to gain insights and perspectives into a project, but it is not a quantitative evaluation. Furthermore, the methodology has not been rigorously applied in this paper; it is intended here mainly as a vehicle to highlight the points worthy of discussion. The number of items under each SWOT heading gives no indication of their relative importance. In reality, the strengths of the current ICH S5(R2) guideline far outweigh the weaknesses; and the opportunities outweigh the threats. The first ICH guideline issued in 1993 remains highly relevant today. By harmonizing regulatory requirements, this initiative had an immediate beneficial impact on the numbers of experimental animals used for the safety testing of new drugs. ICH S5(R2) provides a sound foundation upon which to build for the future. The revised guideline needs to be extended to cover the unique considerations posed by new categories of drugs, such as biopharmaceuticals. Improvements also need to be made to incorporate new technologies and the latest scientific thinking. It is essential to give clear directives on the requirements for DART testing, so that the pharmaceutical industry can reliably estimate the investments necessary in terms of time, costs and resources for a successful marketing submission. On the other hand, the guideline must not be so proscriptive that it encourages the industry and regulatory experts to blindly following its instructions without reflecting deeply on the best strategy to obtain full and complete information of 13

the DART profile of each new innovative medicine. Likewise, the guideline must only promote the use of animals in circumstances where the data is absolutely essential and cannot be obtained by other means. The new guideline must be written with a view to its future evolution as new knowledge, technology and methods become available.

14

References [1]

ICH Harmonised tripartite guideline: detection of toxicity to reproduction for medicinal

products & toxicity to male fertility S5(R2). http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Safety/S5/Step4/S5_R2__ Guideline.pdf (accessed February 17, 2016). [2]

FDA, Guidelines for reproduction studies for safety evaluation of drugs for human use. 1966.

[3]

Ulbrich, B.; Palmer, A. K., Detection of effects on male reproduction—a literature survey. Int.

J. Toxicol., 1995, 14, 293-327. [4]

Takayama, S.; Akaike, M.; Kawashima, K.; Takahashi, M.; Kurokawa, Y., A collaborative study

in Japan on optimal treatment period and parameters for detection of male fertility disorders induced by drugs in rats. Int. J. Toxicol., 1995, 14, 266-292. [5]

Sakai, T.; Takahashi, M.; Mitsumori, K.; Yasuhara, K.; Kawashima, K.; Mayahara, H.; Ohno, Y.,

Collaborative work to evaluate toxicity on male reproductive organs by repeated dose studies in rats: overview of the studies. The Journal of toxicological sciences, 2000, 25, 1-21. [6]

Barrow, P., Reproductive toxicity testing for pharmaceuticals under ICH. Reprod. Toxicol.,

2009, 28, 172-9. [7]

ICH Final Concept Paper. S11: Nonclinical Safety Testing in Support of Development of

Pediatric Medicines. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Safety/S11/S11_Final_Co ncept_Paper_10_November_2014.pdf (accessed August 05, 2015). [8]

Piersma, A. H., Innovations in Testing Strategies in Reproductive Toxicology. In

Teratogenicity Testing: Methods and Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 327-341. [9]

Judson, R. S.; Kavlock, R. J.; Setzer, R. W.; Cohen Hubal, E. A.; Martin, M. T.; Knudsen, T. B.;

Houck, K. A.; Thomas, R. S.; Wetmore, B. A.; Dix, D. J., Estimating toxicity-related biological pathway altering doses for high-throughput chemical risk assessment. Chem. Res. Toxicol., 2011, 24, 451-462.

15

[10]

Schulpen, S. H. W.; Piersma, A. H., The Embryonic Stem Cell Test. In Teratogenicity Testing:

Methods and Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 375-382. [11]

Brannen, K. C.; Charlap, J. H.; Lewis, E. M., Zebrafish Teratogenicity Testing. In Teratogenicity

Testing: Methods and Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 383-401. [12]

Karnofsky, D., Mechanisms of action of certain growth-inhibiting drugs In: Wilson JG.

University of Chicago Press: Chicago, 1965; p 185-213. [13]

Olshan, A. F.; Faustman, E. M., Male-mediated developmental toxicity. Reprod. Toxicol.,

1993, 7, 191-202. [14]

Clinical Trial Facilitation Group Recommendations related to contraception and pregnancy

testing in clinical trials. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Multidisciplinary/M7/M7 _Step_4.pdf (accessed January 13, 2016). [15]

ICH Assessment and control of DNA reactive (mutagenic) impurities in pharmaceuticals to

limit potential carcinogenic risk. M7. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Multidisciplinary/M7/M7 _Step_4.pdf. (accessed January 13, 2016). [16]

ICH Guideline M3(R2) on non-clinical safety studies for the conduct of human clinical trials

and marketing authorisation for pharmaceuticals. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Multidisciplinary/M3_R2/ Step4/M3_R2__Guideline.pdf (accessed February 17, 2016). [17]

FDA Guidance for Industry Reproductive and Developmental Toxicities — Integrating Study

Results to Assess Concerns http://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm07 9240.pdf (accessed February 17, 2016). [18]

Ishihara-Hattori, K.; Barrow, P., Review of embryo-fetal developmental toxicity studies

performed for recent FDA-approved pharmaceuticals. Reproductive Toxicolgy, 2016, This issue. [19]

Klemmt, L.; Scialli, A. R., The transport of chemicals in semen. Birth Defects Research Part B:

Developmental and Reproductive Toxicology, 2005, 74, 119-131.

16

[20]

Scialli, A. R.; Bailey, G.; Beyer, B. K.; Bøgh, I. B.; Breslin, W. J.; Chen, C. L.; DeLise, A. M.; Hui, J.

Y.; Moffat, G. J.; Stewart, J., Potential seminal transport of pharmaceuticals to the conceptus. Reprod. Toxicol., 2015, 58, 213-221. [21]

Banholzer, M. L.; Buergin, H.; Wandel, C.; Schmitt, G.; Gocke, E.; Peck, R.; Singer, T.;

Reynolds, T.; Mannino, M.; Deutsch, J.; Doessegger, L., Clinical trial considerations on male contraception and collection of pregnancy information from female partners. J. Transl. Med., 2012, 10, 129. [22]

FDA Guidance for Industry: Considerations for Developmental Toxicity Studies for Preventive

and Therapeutic Vaccines for Infectious Disease Indications. http://www.fda.gov/downloads/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformatio n/Guidances/Vaccines/ucm092170.pdf (accessed January 07, 2016). [23]

Barrow, P.; Allais, L., Developmental Toxicity Testing of Vaccines. In Teratogenicity Testing:

Methods and Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 81-89. [24]

ICH Harmonised tripartite guideline note for guidance on toxicokinetics: the assessment of

systemic exposure in toxicity studies S3A. http://www.ich.org/fileadmin/Public_Web_Site/ICH_Products/Guidelines/Safety/S3A/Step4/S3A_G uideline.pdf (accessed February 16, 2016). [25]

Vasquez, S. X.; Shah, N.; Hoberman, A. M., Small animal imaging and examination by micro-

CT. In Teratogenicity Testing, Springer: 2013; pp 223-231. [26]

French, J. M.; Woodhouse, N., Soft Tissue Examination of the Fetal Rat and Rabbit Head by

Magnetic Resonance Imaging. In Teratogenicity Testing: Methods and Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 255-273. [27]

Nilsson, L. B.; Ahnoff, M.; Jonsson, O., Capillary microsampling in the regulatory

environment: Validation and use of bioanalytical capillary microsampling methods. Bioanalysis, 2013, 5, 731-738. [28]

Rao, K. S.; Dong, J., Nonclinical Reproductive Toxicity Testing Requirements for Drugs,

Pesticides, and Industrial Chemicals in India and China. In Teratogenicity Testing: Methods and Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 13-30.

17

[29]

McAnulty, P. A., Teratology Studies in the Minipig. In Teratogenicity Testing: Methods and

Protocols, Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 157-167. [30]

Danielsson, B. R., Maternal Toxicity. In Teratogenicity Testing: Methods and Protocols,

Barrow, C. P., Ed. Humana Press: Totowa, NJ, 2013; pp 311-325. [31]

Vargesson, N., Thalidomide-induced teratogenesis: history and mechanisms. Birth Defects

Res C Embryo Today, 2015, 105, 140-56. [32]

van Gelder, M. M. H. J.; van Rooij, I. A. L. M.; Miller, R. K.; Zielhuis, G. A.; de Jong-van den

Berg, L. T. W.; Roeleveld, N., Teratogenic mechanisms of medical drugs. Hum. Reprod. Update, 2010, 16, 378-394. [33]

Martin, P. L.; Weinbauer, G. F., Developmental Toxicity Testing of Biopharmaceuticals in

Nonhuman Primates Previous Experience and Future Directions. Int. J. Toxicol., 2010, 29, 552-568. [34]

Smialowicz, R. J., The rat as a model in developmental immunotoxicology. Hum. Exp.

Toxicol., 2002, 21, 513-519. [35]

OECD, OECD Guidelines for the Testing of Chemicals, Section 4: Health Effects: Test No. 443:

Extended One-Generation Reproductive Toxicity Study. 2011. [36]

Dolinoy, D. C.; Weidman, J. R.; Jirtle, R. L., Epigenetic gene regulation: Linking early

developmental environment to adult disease. Reprod. Toxicol., 2007, 23, 297-307. [37]

Dada, R.; Kumar, M.; Jesudasan, R.; Fernández, J. L.; Gosálvez, J.; Agarwal, A., Epigenetics

and its role in male infertility. J. Assist. Reprod. Genet., 2012, 29, 213-223. [38]

Barton, T. S.; Robaire, B.; Hales, B. F., Epigenetic programming in the preimplantation rat

embryo is disrupted by chronic paternal cyclophosphamide exposure. Proceedings of the National Academy of Sciences, 2005, 102, 7865-7870. [39]

Dietert, R. R., The Microbiome in Early Life: Self-Completion and Microbiota Protection as

Health Priorities. Birth Defects Research Part B: Developmental and Reproductive Toxicology, 2014, 101, 333-340. [40]

FDA Pregnancy and Lactation Labeling Final Rule.

http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/Labeling/ucm093 307.htm (accessed December 17, 2015). 18