Chemical Characterisation – no simple matter

Author: James Morrison

Understanding Chemistry is now mandatory

The most recent edition of ISO 10993-1 in 2018 added to the evaluation table (Table A.1) the interesting item “Physical and/or chemical information.” This is interesting for several reasons, not the least of which is that it appears in the list of: “Endpoints for Biological evaluation.” Clearly, it isn’t a biological endpoint, but it appears to be of sufficient importance or significance that it is required (or at least to be “addressed”) for every type of device requiring biological evaluation, regardless of contact type or duration.

Updated standards on chemical and physical assessment

In January 2020, the new edition of ISO 10993-18 on Chemical Characterisation was published offering a very comprehensive (if sometimes intimidating) array of guidance on means and methods for the task.

Then in March 2020, the new edition of ISO/TS 10993-19 on Physiochemical, morphological, and topographical characterisation was published, almost unnoticed. As a Technical Specification, Part 19 is informative only (i.e. contains no formal requirements) and therefore carries less weight than a standard. It is really aimed at the cases where the physical nature of the material (e.g. surface characteristics, porosity) have a significant effect on biological response and so is less frequently used.

Recently, ISO/TC 194/WG 11 Allowable Limits (the ISO group responsible) met again (electronically, in these uncertain times) to continue the revision process for ISO 10993-17. The revision will probably have a new title: Toxicological risk assessment of medical device chemical constituents. The process is now in the “Committee Draft” stage.

Chemistry vs Biology – the drivers of change

But before we get to that interesting event, it’s instructive to review exactly what is going on with biocompatibility, and why all these things are happening.

Biocompatibility is both a physical and a chemical issue. As always with biocompatibility: “Context is King!” The physical aspects seem to be more pertinent to implant devices and those contacting circulating blood. The chemical aspects are more universally applicable. Physical properties are often overlooked. Chemical issues are often ill-understood. This combination, along with the extensive historical precedent, has led us to excessive dependence on animal testing.

Since 1992 we have had the guidance of ISO 10993-2 Animal Welfare Requirements. The drive to replace, reduce and refine animal testing is a big factor in the field of biocompatibility. And these days the emphasis is strongly on replacement. This is obvious in the cosmetics industry, where animal testing has essentially been outlawed. Reduction in animal testing is continuing in biocompatibility, for example, the new ISO 10993-23 Tests for Irritation due to be published early 2021 will include a validated in vitro (cell culture) irritation assay as an alternative to the current animal tests.

In addition to the drive to reduce animal testing, it’s also the case that in some ways, chemical analysis can be scientifically more informative than biological testing. As modern chemical techniques become more and more sensitive, it’s becoming apparent that the detailed information available from chemical analysis can be more sensitive than the biological assays on which biocompatibility has relied for decades. (See more discussion here) This is particularly the case in the analysis of genotoxicity – where chemical analysis can quantify the presence of genotoxic substances. If the chemical analysis rules out the presence of genotoxins, there should be no need for biological testing. Conversely, chemical analysis can sometimes detect the presence of genotoxins which are missed in biological assays, such as Ames testing.

Given the impetus to minimise animal testing and the new opportunities presented by modern chemical analysis, there is an increased interest in (if not dependence on) chemical characterisation. The characterisation data provides the input to Toxicological Risk Assessment of the medical device which can be used to address a number of the (strictly) biological endpoints of Table A.1. Specifically, it is now widely accepted that this approach is valid for Systemic Toxicity (acute, sub-acute, sub-chronic & chronic), Genotoxicity and Carcinogenicity. These are all systemic endpoints driven by availability (exposure dose) of toxic substances leaching from the device and distributed systemically. Local endpoints are not suitable for evaluation in this way.

Interestingly, Material Mediated Pyrogenicity, which is a systemic effect, is not considered to be adequately addressed in this way. In the absence of a validated in vitro method, its evaluation remains an in vivo test (USP, EU or JP rabbit pyrogen test, as per ISO 10993-11 Annex G.)

Finally, don’t underestimate the importance of having methods that are validated. Equally don’t underestimate the complexity and effort involved in such validations. Method validation is a big-ticket item for many ISO Working Groups.

Expect Reviewers to be Conservative

The very broad and informative reach of the revised Part 18 provides excellent guidance, without being too prescriptive. This is helpful and appropriate as a wide range of materials used in medical devices, and the wide range of clinical uses demand a flexible approach where scientific judgment is used to determine what is important for a specific medical device and the most appropriate approach to the assessment of chemical composition. Unfortunately, the same lack of detailed prescription and reliance on expert judgement can lead to divergent interpretations and expectations between manufacturers, laboratories and regulatory reviewers. Thus, manufacturers can be greatly surprised (and disturbed) by a reviewer’s request for additional information. Particularly when the manufacturer believes they are compliant with the standard, but the reviewer takes a different (usually more conservative) view.

The conservative technical expectations of reviewers are normal – it’s their job, and we all have to deal with it. It isn’t going away soon.

The following are some strategies to help you navigate your way through chemical characterisation:

• Talk to the regulator – as always, a pre-submission before you do anything which involves spending a substantial amount of money is a very good investment. Be careful to succinctly describe the device and what you intend to do and ask just a few carefully considered questions. And of course, be prepared to follow whatever advice the regulator gives you!
• Collaborate internally – I’m aware that biocompatibility is sometimes the domain of a manufacturer’s Regulatory Affairs department. But in the case of chemical characterisation, everyone is going to have to work together, applying their expertise:
  • Clinical Affairs – how and by whom is the device used.
  • Design Engineers – what parts of the device need to be evaluated.
  • Materials Scientists – (Shameless plug on my part) what is in that material in the first place.
  • Manufacturing Engineers – what might contaminate the device during the various production processes, including residual sterilant.

• Collaborate externally – Answer as many questions as possible by information “gathering” rather than “generation.” Engage with your supply chain, analytical chemistry labs, and toxicologist to set up whatever testing you do to be most effective and profitable. (no one-size-fits-all solution)
• Explain and defend every technical consideration in a very clear fashion. “Scientific rationale shall be presented and justified” is the essential mantra.

Part 17 Toxicological Risk Assessment Revision Underway

Working Group 11 recently reviewed approximately 400 comments (possibly unprecedented) on CD 10993-17.

In these unusual times, this was dealt with by a global web meeting extended over 2 weeks, approximately 3 to 4 hours each day. (Somewhat burdensome for those delegates attending in the middle of their night. Such are the wages of “expertise!”)

Half of the comments were from the USA. The rest were from a distributed range of participating countries. Clearly not all 400 comments were individually addressed. There were many duplicate ideas, themes and suggestions, allowing for a reduced number of discussion points, whilst still doing justice to the original comments.

The overall thrust of the comments was technical accuracy and usability of the new revision. It looks like the new version will be much bigger than the current one. It will describe a lot of new applications of toxicology to the field of medical devices, where there has been historically much less interest/use (than say agriculture.) It will help toxicologists and non-toxicologists alike to prepare Toxicological Risk Assessments that are thoroughly scientific and significant in regulatory and safety terms.

And, like chemical characterisation, I’m sure that Toxicological Risk Assessment of medical devices requires and benefits from cross-functional collaboration, good science, and effective communication.

We tailor our services to the requirements of your business and can help you implement plans that are best suited to your projects.