Hi Mike
There's a substantial effort in revision of the whole ISO 10993 standard series to cement biological evaluation as a risk management activity not just a menu of testing activities. The key point is that the standard requires
evaluation of a series of biological endpoints - according to the nature of the device and its evasiveness and duration of use. Evaluation does not necessarily require new testing - if there is sufficient information from existing sources, such as comprehensive knowledge of the chemical composition of the final device or testing data (or better still, clinical history) from closely similar devices.
The latest version (2018) of ISO 10993-1 specifically calls out a need to evaluate chemical composition. Again that doesn't necessarily mean you have to go and do chemical analysis, if you have other information to hand such as detailed knowledge of the formulation and processing. A surgical steel device with conventional machining processes and post process cleaning would not normally require chemical analysis for example.
I have participated in the ISO TC 194 writing these standards for some time and there's been very strong engagement in that process from FDA - as well as some other global regulators. FDA have been very supportive of minimising biological testing. But the key message from FDA has been that a premarket submission (510k or PMA) needs to address each and every endpoint called out by ISO 10993-1 for the particular device type, either by providing direct testing results or by justifying an alternative approach - such as chemical composition and risk assessment. Where manufacturers get into deep water is when they omit something entirely from the submission, or their justification for not doing direct testing is inadequate.
Sometimes in the conduct of individual reviews, reviewers can, as you say, be quite prescriptive in requesting specific testing. Occasionally that can be because of an inexperienced reviewer, but it's more often resulting from inadequate (or absent) justification from the applicant for the chosen approach to biological evaluation. If it's the former - there are formal means of escalating the issue with FDA. If the latter- more work may need to be done in response to an FDA Additional Info Request.
ISO 10993-1 requires for every device category an evaluation with regard to cytotoxicity, irritation and sensitisation.
- Cytotoxicity is inexpensive and a valuable screen - it really should be done in most cases.
- Irritation requires an animal test, but there are alternative in vitro tests under active development - so there may be some movement in the future. It is also possible to establish from literature, safe exposure thresholds -which can in some cases provide an argument for safety based on chemical composition, without animal testing.
- Sensitisation testing is the one where it's most difficult to avoid testing - as it's difficult to establish safe exposure thresholds (very low exposures can trigger sometimes potent allergic responses for some sensitisers) and there are no suitable alternate in vitro models. But again prior data on closely similar devices can sometimes help.
Arthur
------------------------------
Arthur Brandwood PhD FRAPS
Founder and Principal Consultant
Brandwood Biomedical
Sydney, Australia
Arthur@brandwoodbiomedical.com------------------------------
Original Message:
Sent: 30-Apr-2019 09:54
From: Mike Kilkelly
Subject: Animal testing and FDA 510k
Hi
The FDA have been quite prescriptive in indicating the bio-compatibility testing they expect for a 510k submission (ISO 10993 and guidance document) and in my experience animal testing for sensitization and irritation has been part of what they expect to see.
However in other territories there is a big drive to avoid testing on animals.
I would be interested in hearing from anyone on how these can be reconciled?
regards
Mike
------------------------------
Mike Kilkelly
Quality Manager
Galway
Ireland
------------------------------