Review:OECD External Review September, 2015 - Aop:23

From AOP-Wiki
Jump to: navigation, search

AOP Information


  • Associated wiki page: Aop:23


External Review

The review was performed in September 2015 by a team of 5 reviewers:

- Jana Blahová (Czech Republic) - ZhiChao Dang (NL) - Alex Odermatt (Switzerland) - Ioanna Katsiadaki (UK) - R. Balasubramanian (ICAPO)




Charge question 1:Check if the AOP incorporates the critical scientific literature and if the scientific content of the AOP reflects the current scientific knowledge on this specific topic

Responses

Reviewer 1:

The current AOP is well written with current scientific knowledge. It would be important to have limitation and weight of evidence captured under each key events to understand the regulatory standpoint. The wide applicability of the current AOP to evaluate different taxonomic groups and chemicals would add more clarity.

Response: Weight of evidence and major uncertainties or inconsistencies are covered on the KER pages. The taxonomic domain of applicability of the AOP is described on p.35 of the pdf snapshot.

Reviewer 2:

Reviewer 3:

The AOP is well described and included the essential literature, which reflects the current understanding in reproductive impairments resulted from the exposure to AR agonists in small model fish. To increase understanding and application of this AOP in the regulatory field, following points need to be clarified and further elucidated. Two chemicals, 17β-trenbolone and spironolactone, were listed. There are, however, fish data on other AR agonists e.g. 17-methyltestosterone, androstenedione, diazinon, and methyldihydrotestosterone, which should be included too. Explanation is need whether or not these chemicals produce similar effects as described in the AOP. Response: The reviewer raises an interesting point here. There are indeed a number of studies with 17-methyltestosterone (MT). While in vitro MT would be classified as an androgen, in vivo it can be aromatized to an estrogen. As a result, for this compound effects observed may not be consistent with the present AOP. Additional notes regarding this potential limitation of the AOP relative to domain of chemical applicability were added to the "uncertainties and inconsistencies" section of the AOP evaluation. This caveat regarding applicability to non-aromatizable estrogens only was already noted in the "Applicability of the AOP" section of the AOP page. The description for Event:25 also notes that "structures subject to aromatization may behave in vivo as estrogens despite exhibiting potent androgen receptor agonism in vitro."


Information has been filled in the sections of some, but not all of KEs. In some KEs, fathead minnow and medaka have been listed. But information on the other fish like zebrafish, sheepshead minnow etc has not yet been included. Some KEs may be applicable to vertebrates including fish. It may be important to include some results of e.g. AR and Aromatase knockout mice, and transgenic fish if possible. Information on species similarity and difference in ARs has not yet been included in the AOP. Such information may be of use for the extrapolation among species. The current description may be only applied to the genomic action. Similar to ERs, both genomic and non-genomic mechanisms are applied to ARs too. It is suggested that this issue should be mentioned too. Different names for the same assays, e.g. radioimmunoassay and RIA; enzyme immunoassay and ELISA, have been used in KEs. It is important to consistently use the same expressions for these methods in the whole AOP. In the methods described, no current OECD activities on ARTA were included. In fact, several ARTA assays have been validated and will be developed into the test guideline. This development should be included in this section. For the in vivo fish tests, secondary sex characteristics in fathead minnow and western mosquitofish were described. How about medaka? Information on medaka should be included too. Fecundity is an important apical endpoint for fish full/partial life cycle tests, and for TG240 (MEOGT). In addition to TG229, it is important to add these long term toxicity tests. Interrenal, instead of adrenal, is used in fish. Should the term interrenal be included in the AOP?

Reviewer 4:

In general the AOP incorporates relevant current knowledge, although some aspects are restricted to one publication only; for example the process of vitellogenesis is solely references as the Tyler and Sumpter paper although it was Wallace and Selman (1981) that first (and in more detail) described the cellular and dynamic aspects of oocyte growth in teleosts. Also I think, that other patterns of spawning, more common amongst fish should be presented as a reference at least. Fish are a highly diverse group encompassing more than 30,000 species of which very few share the spawning pattern observed in continuous spawners such as the fathead minnow and other regulatory species. As far as the androgen axis goes, the literature doesn't include traits such as the anal fin elongation in the medaka and the spiggin regulation in the stickleback, both of which can provide very insightful information on xenoandrogens. Importantly the AOP doesn't touch at all in the number and type of receptors present in different species; this is very important at binding to the receptor is key for the MIE and fish with only one type of AR are compromised by nature in their ability to detect a wide range of xenobiotics. The weakest by far link in this AOP is the effect of androgen agonists on gonadotrophins; if we assume there is a negative feedback then the AOP stands strong all the way from the MIE to the KE, the KER and even the population trajectory besides some knowledge gaps there. However, this is not the case; many (if not most) fish species employ different reproductive strategies to the FHM or the medaka and as such they have developed feedback mechanisms that suit them best. During puberty for example many species have a positive feedback mechanism (see review by Trudeau, 1997; Antonopoulou et al, 1999 and many many more!!). I wonder since the AOP relies on this reduction of endogenous androgens leading to reduction of endogenous oestrogens to materialise all downstream KE why there is practically no literature cited here other than a single paper on eels that if I am correct is not relevant as it is an in vitro study. In fact this reference (Huang et al, 1997) is missing from the list but is cited on page 35. In terms of taxonomic applicability, none of this is relevant to invertebrates as they do not use vertebrate steroids to regulate their reproduction. I suggest we clearly state this rather than say ...not necessarily relevant to invertebrates (i.e. page 3) it's definitely not relevant to them. A minor point, I am not sure the term lutenising hormone is used in fish; instead we recognize two FSH; trivial perhaps but worth considering

Reviewer 5:

Response from AOP author:

To reviewer 1:

To reviewer 2:

To reviewer 3:

To reviewer 4:

To reviewer 5:



Charge question 2:Verify the weight of evidence judgement/scoring provided by AOP developers for KEs, KERs and the overall AOP

Responses

Reviewer 1:

The weight of evidence summary in page 30 of this AOP lists that the WoE for androgen receptor, agonism as 'weak', while the other events are moderate to strong. This is may need further detailing to address the regulatory requirements and ways by which the limitation could be addressed.

Reviewer 2:

Reviewer 3:

According to the description of the KE of Gonadotropins, circulating concentrations, Reduction, it seems that there is a problem to measure LH and FSH in small model fish. If so, this may have a conflict with the notion that KEs must be measurable. In addition, KEs have been numbered in the concordance table, in which the KE of Gonadotropin reduction was not included. Explanation is needed why this KE is not included. Over the relationship: Androgen receptor, Agonism Directly Leads to Gonadotropins, Circulating Concentrations, Reduction, the term “directly leads to” seems confusing because the direct target of AR agonists may not be pituitary and a decrease in gonadotropin levels results from a negative feedback mechanism in which neurotransmitters are involved. This relationship can be understood as indirect process too.

In the section of Empirical Support for Linkage, results of DEHP/MEHP in mammals have been included. Did they have similar mechanisms to those of fadrozole, prochloraz in fish?

In the section of Empirical support for linkage, some lines of evidence were based on an increase in 17β-estradiol concentration resulting in an enhancement in VTG in male fish. Such evidence is different from the title of reduction and female of AOPs.

A weight of evidence table is needed in the annex.

The table of the KEs was not presented according to the orders of KEs, which may cause some confusing. It is suggested that the table of KEs should be organized according to the order of KEs.

Reviewer 4:

As expanded above, once we accept that an AR agonist is not aromatisable (species differ enormously on their ability to aromatase or not specific androgens) and has a negative feedback for LH, then the weight of evidence is moderate to strong and the AOP stands. However, my concerns are entirely on how relevant is this for ecologically important species that ultimately we are trying to protect via this exercise. If we are interested in the adverse outcome for a laboratory strain of FHM, then by all means the evidence is there in most relationships including the adversity. However, I do not believe this is how most species regulate their reproduction and more information should be made available to regulators on this critical assumptions.

Reviewer 5:

Response from AOP author:

To reviewer 1:

To reviewer 2:

To reviewer 3:

To reviewer 4:

To reviewer 5:



Charge question 3:What would be the regulatory applicability of this AOP in your opinion?

Responses

Reviewer 1:

Globally across different geographies identification and evaluation of EDCs are recommended by various regulatory agencies. This AOP covers the existing and additional events of evaluation. The set of regulatory specification/ data requirement for the AOP could help broader usage for different chemicals types

Reviewer 2:

Reviewer 3:

Identification of Endocrine Disrupting Chemicals (EDCs) is needed under several pieces of European Union (EU) legislation, including the Regulation on industrial chemicals (Registration, Evaluation, Authorization and restriction of Chemicals, EC 1907/2006, REACH), the Plant Protection Products Regulation (EC 1107/2009, PPPR), and the Biocides Products Regulation (528/2012, BPR). Currently, the regulatory identification of EDCs is mainly based on the general consensus on the WHO definition, which consists of three essential elements, i.e. chemical-induced adverse effects (adversity), chemical specific endocrine modes/mechanisms of action (MOAs) and the causal relationship (causality) between adverse effects and endocrine MOAs. AOPs cover all essential elements for identification of EDCs and show the complex biology of adversity and MOAs. These will help regulators understand the complexity of identification of EDCs. Besides, current regulatory tests focus on EATS pathways. In contrast, AOPs include not only EATS pathways but also other pathways, e.g. PPARs, RXR, that are essential to development, growth, and reproduction. Within each AOP, different targets at molecular, cellular, organ/tissue and individual levels could be identified and the adverse outcome would be predicated. Such information would be of help for prioritizing chemicals, for grouping chemicals and for developing an integrated testing strategy. It is important to indicate that the AOP needs extensive amount of data which might be possible for a few chemicals but will not be possible for a majority of chemicals. Current data requirements under REACH, PPPR, BPR, etc. do not cover all key events of the AOP.

Reviewer 4:

This is restricted clearly to female fish only as adversity is linked to reduced oestrogen synthesis (via reduced androgen synthesis); it is also limited to fully reproductive mature fish (not fish entering puberty or juvenile fish) and importantly is limited to fish that once they reach sexual maturity they spawn constantly. The latter is a reproductive strategy employed by fish that tend to occupy tropical areas (around the equator). Unfortunately most fish species have different reproductive strategies (annual life cycle) hence the level of gonadotropin expression (and consequently steroid production) is regulated by photoperiodic and temperature changes throughout the year. Even if a negative feedback mechanism operates in all of these species and in all life stages (which is certainly not the case) we still need to establish what is the relative strength of the AR agonist induced negative feedback to the environment-induced stimulation of gonadotropins! This link has never been studied and is critical if we really mean to protect wildlife.

Reviewer 5:

Response from AOP author:

To reviewer 1:

To reviewer 2:

To reviewer 3:

To reviewer 4:

To reviewer 5:



Charge question 4:Overall Assessment of the AOP - Would you recommend this AOP to be submitted to the Working group of the National Coordinators for the Test Guidelines Programme (WNT) and the Task Force on Hazard Assessment (TFHA) for endorsement?

Responses

Reviewer 1:

YES, recommend this to be submitted to the WNT. This will set the right direction for regulatory application sooner. Additionally, the strengths and weakness of this AOP should be clearly stated for continuous usage and updation

Reviewer 2:

Reviewer 3:

Yes. It is important to get the official stamp for publishing the AOP. To increase the regulatory applications of the AOP, following points are suggested: The current AOP focuses only on female fish. As the majority of test guidelines include both males and females, it is important to include certain information/statement over male fish so that regulators can get an overall picture. It is also important to indicate that the major androgen in male teleost fish is 11-ketotestosterone. In addition, it is important to specify the uncertainties or inconsistencies that are related to chemicals. This would be of great help for non-experts and risk assessors to understand confounding factors.

Reviewer 4:

Yes as long as the limitations based on feedback mechanisms and reproductive strategies are made even clearer than currently are. This is relevant ONLY for small repeat-spawning fish species but this is mentioned only once in the title throughout the AOP. Assuming this is presented as a warning on the first paragraph, then my recommendation would be to submit to WNT.

Reviewer 5:

Response from AOP author:

To reviewer 1:

To reviewer 2:

To reviewer 3:

To reviewer 4:

To reviewer 5:



Other comments

Reviewer 1:

Reviewer 2:

Reviewer 3:

Reviewer 4:

Reviewer 5: