Summary of public comments received on the Draft Bioactivity-Exposure Ratio (BER) Science Approach Document (SciAD)

The draft Bioactivity-Exposure Ratio (BER) Science Approach Document (SciAD) published under the Chemicals Management Plan (CMP) was published in March 2021, followed by a 60-day public comment period (Canada 2021). Three stakeholders submitted comments on the document. One stakeholder has indicated their submission is confidential. Therefore, the identity of one of the stakeholders is not specified; meanwhile, their comments on the draft SciAD are masked to protect confidential business information, accordingly.

The other 2 stakeholders are as follows:

• Forest Products Association of Canada
• Technical Subcommittee of the Canadian Environmental Protection Act, 1999 (CEPA) Industry Coordinating Group (ICG)

Summarized public comments and responses from the Government of Canada are organized by topic.

Overarching

Comment summary 1: Submitter acknowledges that this SciAD represents a significant step forward in the integration of New Approach Methodologies (NAMs) to support future prioritization and risk screening activities under the CMP.

The submitter believes the SciAD establishes a logical construct that will facilitate regulatory decision-making, leveraging non-animal and next-generation toxicology tools in order to inform risk screening and prioritization, and, in some cases, become a significant line of evidence to support ‘non-toxic’ decisions.

Subject matter experts are available should further input be needed.

Response 1: Thank you for the comments.

Comment summary 2: Submitter acknowledges that this SciAD is a significant step towards the integration on NAMs to support future prioritization and risk screening activities under the CMP.

The submitter is supportive of this risk-based initiative as a progressive and modernized approach for risk screening and prioritization.

Response 2: Thank you for the comments.

Comment summary 3: Stakeholder supports the integration of NAMs into the chemical safety assessment process and case studies as outlined in the SciAD allow for the methods to be challenged and improved.

Response 3:  Thank you for the comments.

Methodology

Comment summary 4: The submitter acknowledges the utility of the presented case study as an early proof of concept and that other parallel evaluations of approaches for other exposure routes are also needed. Moreover, submitter suggests expanding case study chemical list to cover more chemical space and specific toxicities to further validate the approach.

Submitter concurs with the stated uncertainties and challenges in the SciAD of accounting for metabolism of substances within the body when using in vitro based assays. Submitter encourages Health Canada to support development of NAM approaches that take into account potential metabolism. Moreover, in the interim, the submitter encourages Health Canada to consider other lines of evidence for metabolism including read-across and predictive modelling.

Submitter suggests further research and development within the approach to account for intracellular dose/concentration for chemicals with poor solubility profiles and volatiles.

Submitter supports the need and department plan to delineate a separate approach to screen for potential genotoxic substances.

Submitter also agrees with the statement that the presented approach will continue to evolve as new sources of information become available and as further research needs are addressed.

Response 4: Expanding the chemical space for analysis is challenging, as each chemical requires in vitro data, toxicokinetics data, and existing in vivo studies for comparison. The approach presented here builds upon an analysis of a much broader chemical space (N=448) (Paul-Friedman et al., 2020). The findings regarding the conservativeness of the point of departure (POD; in this case, POD_Bioactivity) were similar to those for the SciAD. This approach has been shown to work well for a large chemical space covering various toxicities. However, application to mixtures, substances with unknown or variable composition, complex reaction products and biological materials (UVCBs), and natural products will remain challenging. This challenge is not unique to in vitro testing and is also encountered with more traditional animal models.

Metabolism is identified as an uncertainty in the approach document. Health Canada recognizes that there are efforts internationally to retrofit high-throughput screening assays and better account for metabolism. As these new techniques are further developed, the BER approach presented here will be adjusted where necessary.

The impact of in vitro distribution was explored for the latest update of approach presented in the SciAD. Specifically, various mass balance models were examined to predict cell/tissue concentrations rather than using nominal concentration for the in vitro-based POD. Ongoing efforts are necessary to collect the parameters needed for nominal concentrations estimates. These adjustments will be incorporated into the approach in the future as the science evolves.

Comment summary 5: Multiple publications have not demonstrated a correlation between ToxCast responses and in vitro PODs. Thus, using ToxCast for prioritization is less helpful as it does not distinguish high hazard chemicals from low hazard chemicals. The approach outlined in the SciAD only demonstrates that the in vitro-based PODs are lower than in vivo PODs.

Response 5: The purpose of the approach presented is to develop a protective POD based on an in vitro threshold concentration where biological activity was observed and use this value as the initial basis for prioritization and screening. The approach is not intended to be predictive of individual hazards or the respective doses at which such hazards are certain to occur in humans (or animals). By demonstrating that the POD_Bioactivity is protective in comparison to in vivo-based PODs in most cases, the approach remains useful for its intended purpose of prioritization and screening.

Since the publication of the original SciAD, further analysis has been conducted as part of the Accelerating the Pace of Chemical Risk Assessment (APCRA) international working group looking specifically at the concordance and level of protection between in vitro-based bioactivity approaches and non-clinical/clinical toxicity data for pharmaceuticals (Weitekamp et al., 2025). In the study, authors evaluated the quantitative and qualitative concordance of lowest observed adverse effect levels (LOAELs) and adverse endpoints between in vivo (rat and mouse) and in vitro (ToxCast coupled with in vitro to in vivo extrapolation (IVIVE)) models of human health and human clinical trials of pharmaceuticals. One of the key findings suggests that rodent-based LOAELs were generally higher than the human LOAEL values (that is, not health protective) but, when combined with typical composite uncertainty factors (i.e., 100-1000), were protective of >97% of drugs evaluated. By contrast, in vitro bioactivity POD values were lower than the human LOAEL values (that is, health protective) and would require lower composite uncertainty factors to achieve the same level of human health protection. POD_Bioactivity defined by using the 5th percentile from a distribution of ToxCast assays and following a similar IVIVE approach used in the SciAD (that is, selection of the 0.95 (sensitive individual) quantiles from a Monte Carlo simulation of interindividual variability in toxicokinetic parameters) coupled with an uncertainty factor of 10 resulted in protection for >95% of drugs studied (Weitekamp et al., 2025). The study comparisons were conducted using LOAEL values. It is generally the practice of risk assessments conducted under CEPA to compare to the no observed adverse effect level (NOAEL) values or, where a NOAEL is not available, to compare to a LOAEL with an additional uncertainty factor of 10. Thus, a composite uncertainty factor of 100 applied to the POD_Bioactivity could provide a similar level of protection compared to the standard approaches used in human health risk assessment.

A better POD_Bioactivity correlated with observed in vivo PODs would increase the utility of the approach and may more effectively stratify high hazard chemicals from low hazard chemicals. Refining the methods for establishing POD_Bioactivity to better correlate with in vivo PODs is an active area of development for both Health Canada and other regulators participating in the APCRA group. Refinements under consideration in the latest update include:

• modifying the filtering rules of ToxCast assays included in the AC₅₀ distribution
• accounting for in vitro disposition rather than using nominal concentration; and
• using the latest IVIVE methods

Future considerations are ongoing to evaluate responses from other high-throughput screening technologies (for example., transcriptomics) when deriving the in vitro bioactive concentration. As these refinements are developed, the approach taken by Health Canada will be updated to reflect these changes.

Comment summary 6: Submitter proposed that any future work to develop an HTS prioritization strategy for genotoxicity also include an in silico comparison. In silico methods may provide optimal utility when considering the need to generate new data.

Response 6: Thank you for the comment. In silico predictions for genotoxicity are indeed used as part of the strategy to prioritize substances for further work on the Domestic Substances List (DSL). A recently published SciAD entitled Chemical Screening and Prioritization: Health Canada’s Automated Workflow for Prioritization (HAWPr) outlines the approach.

Comment summary 7: Submitter suggested that the case studies using this approach (that is, the SciAD and Friedman et al., 2020) should be verified as sufficient to cover the chemical space (structural and physicochemical properties) of the DSL prior to application.

Response 7: A chemical space analysis between the DSL and the applicable case studies where the approach was developed was not presented in the SciAD. However, conducting such an analysis would reduce uncertainty when applying the approach and will be explored further.

Applications in risk assessment and prioritisation

Comment summary 8: Submitter is supportive of the premise for BER thresholds outlined in the SciAD but suggests that they should be flexible based on the context (that is, the profile of activity in vitro assays as well as evidence available from other sources of hazard data). Submitter suggests the outlined BER thresholds should be flexible and a starting point and not absolute. Submitter provided an analogy to what is currently used in assessment based on a margin of exposure (MOE). Conventionally, an acceptable MOE for systemic toxicity is 100 but may be lower and still adequately protective when accounting for other factors (for example, allometric scaling).

The submitter suggested more manual spot-checking in addition to the in vitro assay filtering criteria outlined in the approach to ensure the removal of activities that do not have a clear dose-response.

The submitter suggested integration of available ToxCast data tailored to specific biological mechanisms, so-called adverse outcome pathways (AOPs), for which traditional in vitro PODs are available, and to compare harmonized biological endpoints for prioritization.

The submitter suggested that the BER approach may be a useful tool when conducting alternative assessments (that is, regrettable substitution) which will be a focus of the CMP moving forward.

Response 8: As expressed in the SciAD, the approach will continue to evolve as experience is gained and the underlying science matures to address the uncertainties. The outlined BER thresholds are starting points, and it is envisioned that some flexibility will be incorporated into the approach during the application phase depending on the substance-specific context and the available data beyond in vitro bioactivity. The BER is envisioned to be analogous to the MOE approach, where flexibility is currently exercised in risk assessment and prioritization.

The assay filtering criteria presented in the SciAD were intended to provide a workflow applicable to a broad chemical space, with minimal manual intervention required for the initial comparison and confidence-building exercise. This is also aligned with the broad-based retrospective analysis developed by Health Canada in collaboration with international partners. However, it is acknowledged that, during the application phase, on a substance-by-substance basis, more explicit interrogation of the concentration–response curves from the individual assays that comprise the distribution of AC₅₀ values, upon which the POD_Bioactivity is based, is warranted.

The value of mapping the ToxCast data to specific AOPs is recognized, as this would allow for a refined and targeted BER applicable to a specific biological endpoint. This type of mapping is, in practice, difficult given the paucity of established AOPs at this given time. Moreover, the SciAD is intended to provide a screening-based approach that is independent of potential hazards that may be observed in vitro at higher doses. The approach makes use of a minimal bioactivity threshold that is protective (rather than predictive) against systemic toxicity.

Comment summary 9: The government will use the approach to identify low concern substances or substances that require further assessment action.

Stakeholder recommends that this approach, as proposed, is used only for priority-setting purposes, and not to determine whether a substance is toxic or non-toxic.

Stakeholder recommends clear communication regarding substances with a BER of less than 1000 in that this threshold does not specifically identify an exposure or risk level of concern but only that a substance has been identified for a more refined assessment.

Response 9: Section 7 outlines the applications of the approach under the CMP. Substances with a BER of less than 1000 may be considered a priority for further action, but not necessarily as “toxic” under section 64(c) of CEPA without additional data. For certain screening-level risk assessments, in the absence of in vivo data or when other indicators of potential hazard are limited, a BER greater than 1000 may be used as a line of evidence to support a conclusion of “not toxic” under section 64(c) of CEPA. These are generally low-exposure substances, and the BER, combined with other non-test method approaches such as read-across and computational modelling, may be used for rapid screening and a conclusion of not toxic. This approach is considered consistent with other “Type 2” assessment approaches that form part of the CMP risk assessment toolbox.

There is support for clear communication regarding the risks associated with substances that have a BER of less than 1000 as the application of this approach moves forward. In the SciAD, it is stated that the proposed approach, using in vitro assays to derive a POD_Bioactivity (and subsequently a BER), does not determine a level at which adverse health effects would occur. Rather, it utilizes perturbations observed in in vitro assays, which cover a broad biological range of possible biochemical and cellular targets that may form the basis of events in an adverse outcome pathway, but are not, on their own, indicative of an adverse health effect. This distinction will be further emphasized during the implementation of the approach.

References

Canada. 2020. Science approach document - Bioactivity exposure ratio: Application in priority setting and risk assessment.

Paul Friedman K, Gagne M, Loo H, Karamertzanis P, Netzeva T, Sobanski T, Franzosa J, Richard A, Lougee R, Gissi A, et al. 2019. Utility of in vitro bioactivity as a lower bound estimate of in vivo adverse effect levels and in risk-based prioritization. Tox Sci 173(1):202-225.

Weitekamp CA, Paul-Friedman K, Harrill AH, Auerbach S, Bandele O, Barton-Maclaren TS, Fitzpatrick S, Mezencev R, Santillo M, Simanainen U, Smith D, Whelan M, Thomas RS. Quantitative and qualitative concordance between clinical and nonclinical toxicity data, Tox Sci 206(2):253–272.