Technical Report 130

3.THE STEPS I-VII OF THE ECETOC 7SI-ED

Step I: Gathering of relevant information with regard to adverse effects and endocrine modes-of-action (MoA)

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

Gather all the available information that may be relevant for the identification of:

  • The substance;
  • Adverse effects that may be indicative of endocrine disrupting properties;
  • Evidence on endocrine (or non-endocrine) MoA that may be relevant for the adverse effects described in bullet point 2;
  • Evidence for other endocrine activity.

[/vc_column_text][/vc_column][/vc_row]

Step II: Evaluation of quality, reliability, reproducibility and consistency of the individual studies

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

Use the JRC ToxR Tool to assess the reliability of the data presented in each relevant study:

  • Category 1: reliable without restrictions;
  • Category 2: reliable with restrictions;
  • Category 3: not reliable;
  • Category 4: not assignable.

Studies assigned to Category 3 (not reliable) or Category 4 (not assignable) are summarised and documented, justifying the Category assignment, taking great care to determine how much weight can justifiably be given to such low-quality data. Studies that are not interpretable should be excluded from analysis.

After the reliability assessment, the remaining studies (and other information) are placed into the appropriate level of the OECD CF.

[/vc_column_text][/vc_column][/vc_row]

Step III: Evaluation and summary of the evidence for an adverse effect

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

Use OECD GD 150 to examine OECD CF Level 4 and 5 studies for the presence or absence of adverse effects. Relevant in vivo studies using test methods or apical endpoints not mentioned in the OECD CF may also be used, if a justification for their relevance is provided:

  • Are there substance-related adverse effects which may be indicative of endocrine disrupting properties?
  • Are there substance-related adverse effects which may not be indicative of endocrine disrupting properties?

For bullet points 1 or 2, a WoE approach is applied to consider sufficiency of evidence to support a substance-induced adverse effect. First, conclusions on the strength of evidence are made for individual studies and then WoE collectively across studies, with attention given to the quality of the study, as evidence may strengthen in support of an adverse effect (or lack thereof) with higher quality data.

When evaluating the strength of evidence in support of a substance-induced effect in a single study, important considerations include dose-responsiveness (i.e. toxic effects increase in incidence magnitude and/or severity as the dose increases), statistical significance, difference from concurrent controls, and historical control range, and consistency of effects across apical endpoints within one study.

Provided several studies are available, consistency across studies and coherence amongst different lines of evidence are also key considerations (cf. Definitions):

  • Is the same adverse effect seen in more than one study of the same type?
  • Is a consistent pattern of observations emerging across endpoints within one study in support of an adverse effect? Or across similar study types within the same line of evidence (i.e. in vivo rodent assays)?
  • Is a coherent pattern of observations emerging from other lines of evidence in support of an adverse effect (e.g. in vivo mammalian studies, in vivo non-mammalian studies)?

Concluding how much to weight a given study with limitations is an important part of the WoE evaluation:

  • Low-quality studies are carefully considered, particularly if those studies conflict with results observed in comparable, higher quality studies. The limitations of low-quality studies should not be overlooked.
  • Negative or null data are also weighted.

As this framework is explicit to the identification of substances with endocrine disrupting properties as defined in WHO/IPCS (2002) and by the criteria set out in the ongoing revision of the EU plant protection products and biocidal products legislations, the following considerations are also warranted:

  • Is it plausible that any of the adverse effects are caused by an endocrine MoA?
  • If more than one adverse effect is observed, is it plausible that they are caused by different endocrine MoAs?
  • Are the adverse effects relevant to humans and/or are they relevant at the population level for wildlife?

The evidence for the presence of an adverse effect which may be indicative of endocrine disrupting properties is categorised as sufficient, insufficient or none:

Sufficient evidence: Clear evidence of a substance-related effect(s) that may be indicative of endocrine disrupting properties and that is not related to general systemic toxicity or other potentially confounding (specific) toxicities across a range of study types. Sufficient evidence may include a dose-response for an effect with statistically significant difference from concurrent controls and which is outside the historical control range; and consistently observed in high-quality studies. The evidence is strengthened by being seen in a ToxR Tool Reliability-Category 1 study. Evidence from ToxR Tool Reliability-Category 2 studies is considered to have less strength and, dependent on the nature and the extent of the limitations, may result in a consideration that evidence is insufficient.

Insufficient evidence: A single effect that may be indicative of endocrine disrupting properties, which was not seen in similar studies, or an array of different effects, which do not form a coherent/consistent pattern. Insufficient evidence includes an effect which was only observed at the highest dose tested, but which was only marginally increased over concurrent control or historical background control levels.

No evidence: No adverse effects identified that may be indicative of endocrine disrupting properties (evidence of absence or absence of evidence). Absence of evidence includes a statistically significant effect which does not form part of a dose-response curve and which is not seen in other studies which might provide evidence of this effect. If there is absence of evidence or evidence of absence of an adverse effect indicative of endocrine disrupting properties, the definition for an endocrine disruptor is not fulfilled and an endocrine disrupting property cannot be identified.

[/vc_column_text][/vc_column][/vc_row]

Step IV: Evaluation and summary of the evidence for endocrine activity (and non-endocrine activity, if available)

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

Use OECD GD 150 to examine OECD CF Level 2 and 3 assays and other reliable assays and studies providing relevant data; further include relevant assays to investigate non-endocrine activities, where available:

  • Are there indications of endocrine activity?
  • Are there indications of non-endocrine activity?

For bullet points 1 or 2, a WoE approach is applied to consider sufficiency of evidence to support substance-induced endocrine activity. First, conclusions on the strength of evidence are made for individual assays and then WoE collectively across assays, with attention given to the quality of the assay, as evidence may strengthen in support of endocrine activity (or lack thereof) with higher quality data.

When evaluating the strength of evidence in support of substance-induced activity in a single assay, important considerations include dose-responsiveness (i.e. activity increases in incidence, magnitude and/or severity as the dose increases), statistical significance, difference from concurrent controls.

Provided several assays (and studies) are available, consistency across assays (and studies) and coherence amongst different lines of evidence are also key considerations (cf. Definitions):

  • Is the same endocrine activity seen in more than one assay and/ or study of the same type?
  • Are indications consistent within similar assay and/or study types?
  • Is a consistent pattern of observations emerging across studies in support of a specific endocrine activity and the directional change in this activity (e.g. agonistic versus antagonistic).
  • Is a coherent pattern of observations emerging from other lines of evidence in support of a specific endocrine activity (e.g. in vitro assays, in vivo assays)?

Concluding how much to weight a given assay and/or study with limitations is an important part of the WoE evaluation:

  • Low-quality assays and/or studies need to be carefully considered, particularly if they conflict with results observed in comparable, higher quality assays and/or studies.
  • Negative or null data should also be weighted.
  • The limitations of in vitro models must be addressed and weighted, particularly in the presence of incoherent data (e.g. endocrine activity in in vitro assays but not in in vivo assays).
  • Are the indications consistent within similar assay and/or study types?
  • For human health: Is the endocrine activity relevant to humans?

The evidence for a substance to be endocrine active is categorised as sufficient, insufficient or none.

Sufficient evidence: Clear evidence of a substance-related effect(s) indicative of endocrine activity from Level 3 in vivo assays (or comparable assays and/or studies), which is consistent with indications from Level 2 in vitro assays (or comparable assays) of the same type of endocrine activity.

Insufficient evidence: Some endocrine activity in one in vivo assay which is not part of a consistent pattern in other in vivo assays is assessed as insufficient evidence. Incoherent evidence for endocrine activity, e.g., in vitro assays are positive in the absence of any in vivo data on endocrine activity (from either Level 3 in vivo assays or from diagnostic endpoints, etc., from Level 4 and 5 studies)) is also interpreted as insufficient evidence (i.e. in vitro assays positive – no in vivo data on endocrine activity). This scenario cannot be taken as conclusive evidence of absence of endocrine activity because endocrine activity could still be seen in vivo.

No evidence for endocrine activity: Endocrine activity was not seen in the Level 2 and 3 assays or comparable assays and studies. No evidence for endocrine activity includes activity which is restricted to in vitro assays, but which was not seen in the corresponding in vivo assay(s) or studies, which should cover both human health and ecotoxicological aspects (i.e. in vitro assays positive – comprehensive in vivo data on endocrine activity negative).

[/vc_column_text][/vc_column][/vc_row]

Step V: Integration of the evidence and evaluation of biological plausibility that adverse effect and endocrine activity are linked by specific endocrine MoA

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

The evidence from Step III and Step IV is considered together (i.e. integrated) in two phases using the following questions as a guide. If the evidence from Steps III and IV support more than one MoA, the respective data need to be evaluated separately:

Step V.A: Is there sufficient evidence from Step III assessing adverse effect(s) that may be indicative of endocrine disrupting properties and sufficient evidence from Step IV assessing endocrine activity?

  • If Yes, proceed to Step V.B.
  • If insufficient evidence in Step III and/or Step IV, proceed to Step VI: Identification of uncertainties as the scientific criteria set out in the context of the plant protection products and biocidal products Regulations (EC) No 1107/2009 and (EU) No 528/2012 have not been met.
  • If there is evidence of absence of adverse effect(s) that may be indicative of endocrine disrupting properties (Step III) or evidence of absence of endocrine activity (Step IV), the WHO/IPCS (2002) definition for endocrine disruptor is not fulfilled and an endocrine disrupting property has not been identified in accordance with the scientific criteria set out in the context of Regulations (EC) No 1107/2009 and (EU) No 528/2012.


Step V.B: The following questions about the evidence from Step III and Step IV are considered:

  • Based upon the WoE from Steps III and IV, is a specified endocrine MoA postulated to be activated / to operate (i.e. can the endocrine activity identified in Step IV plausibly explain the adverse effect(s) observed in Step III)?
  • Is there evidence for the activation of a key event(s) of the postulated endocrine MoA?
  • Is there dose/concentration-response concordance between the evidence for adverse effect(s) and the evidence for endocrine activity for the postulated endocrine MoA?
  • Are there data providing evidence on the temporality or the essentiality of the endocrine key events (i.e. is the sequence of events reversible if a key event is stopped or prevented)?
  • Are there data supporting that a non-endocrine MoA is more likely for the adverse effect(s) than an endocrine MoA?
  • Is the postulated endocrine MoA relevant to human health, and are adverse effect(s) relevant at population level for wildlife?

For each postulated MoA, the answers to these questions are used to determine whether a biologically plausible link has been established. This demands that:

  • A specified endocrine MoA is postulated to be activated.
  • There is sufficient evidence for the activation of a key event in the postulated endocrine MoA, for example the stimulation or blocking of a receptor that is associated with the MoA. This evidence is most likely to come from Level 2 in vitro
  • There is sufficient evidence of one or more in vivo downstream measurable and relevant consequences which are diagnostic of the postulated endocrine MoA. This evidence is most likely to come from Level 3 in vivo assays (or from diagnostic endpoints or other relevant parameters from Level 4 or 5 in vivo studies).
  • There is sufficient evidence of one or more in vivo downstream relevant adverse effects which would be expected from activation of the postulated endocrine MoA. This evidence is most likely to come from Level 4 or 5 in vivo
  • The dose/concentration-response in each of the three sets of evidence form a coherent pattern. Reference can be made to model substances (e.g., positive control substances) which are known to operate by the postulated endocrine MoA to estimate the relationship which may be expected amongst the three sets of evidence. Differences from the expected relationship should be explained.
  • If evidence on the temporality or the essentiality of the endocrine key events is available, the evidence does not conflict with the postulated endocrine MoA.
  • Non-endocrine MoA(s), which could also have a biologically plausible link to the adverse effect(s), are considered less likely be activated than the endocrine MoA (based upon the comparison of the available information on biological concordance, dose/concentration-response concordance or temporal association).
  • The postulated endocrine MoA is considered to be relevant to human health, or the adverse effect(s) is/are considered to be relevant at population level for wildlife.

A biologically plausible link between an adverse effect and an endocrine MoA can only be established if all these conditions can be met. Proceed to Step VII: Conclusion on endocrine disrupting properties.

If a biologically plausible link is not established, then proceed to Step VI: Identification of Uncertainties.

[/vc_column_text][/vc_column][/vc_row]

Step VI: Identification of uncertainties

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

Uncertainties are most likely to arise in the assessment of data for endocrine disrupting properties when there is insufficient evidence from Steps III or IV, or when the evidence is sufficient to go to Step V.B, but a biologically plausible link between adverse effect and endocrine MoA cannot be established.

 There are 3 major sources of uncertainty:

  • The quality of data: Poor quality data (from ToxR Tool Reliability-Category 3 (or 4) studies and assays) from the data quality evaluation in Step II.
  • The equivocality of data: Examine conclusions of insufficient evidence from Step III and Step IV to determine if data are equivocal (or absent, see next bullet point). If data are equivocal then consider potentially confounding factors in individual studies (e.g. exceeding of the Maximum Tolerated Dose (MTD) or Maximum Tolerated Concentration (MTC); presence of specific non-endocrine toxicity). If data across ToxR Tool Reliability-Category 1 and 2 studies consistently show equivocal effects, then it is unlikely that more data would reduce uncertainty. Also the inconsistency of effects observed in ToxR Tool Reliability-Category 1 and 2 studies may be a source of uncertainty and not allow a firm conclusion.
  • The lack of data: There may be evidence of endocrine activity in one set of data but no data available from one of the other necessary sets of data (e.g. evidence from Level 2 in vitro assays, but no Level 3 in vivo assays or Level 4 and 5 in vivo studies). In this situation, it would be reasonable to generate further complementary data.

[/vc_column_text][/vc_column][/vc_row]

Step VII: Conclusions on endocrine disrupting properties

[vc_row][vc_column width="1/1"][vc_column_text el_class="green-box"]

  • If a biologically plausible link is established in Step V, then an endocrine disrupting property in accordance with the WHO/IPCS (2002) definition has been identified.
  • If there is insufficient evidence for a biologically plausible link in Step V, the WHO/IPCS (2002) definition for endocrine disruptor is not fulfilled and an endocrine disrupting property has not been identified.

[/vc_column_text][/vc_column][/vc_row]