Sandi Schaible and Peter Steen, WuXi AppTec Medical Device Testing11.09.22
Many medical device manufacturers look at extractables/leachables (E/L) testing as little more than a box to check as they work toward their regulatory submission. The reality, however, is chemical characterization is an interactive process between the manufacturer, lab testing partner, and consulting toxicologist. Moreover, a thorough, thoughtful approach to E/L testing is becoming increasingly important as regulatory trends drive complete chemical identification without reporting unknown compounds.
In this column, E/L experts address six common myths associated with extractables testing and tactics manufacturers can apply to stay a step ahead during the chemical characterization process.
Myth 1: Extractables testing will reveal all potential leachables. Ideally, extractables profiling reveals all potential leachables above the analytical evaluation threshold (AET). However, factors such as the sample complexity (e.g., level and number of compounds) can suppress or mask compounds present at a low level.
Additionally, a single instrument or combination of instruments that can ensure the detection of all potential compounds does not exist. Thus, there are four methods used most widely when conducting extractables testing. They include:
Myth 2: Extractables testing is the last step in the chemical characterization process. Following ISO10993-1: 2018 Biological evaluation of medical devices - Part 1: Evaluation and testing within a risk management process, producing physical and/or chemical information by chemical characterization is a primary approach for evaluating medical device biocompatibility in order to manage biological risk. To evaluate risk, it is important to understand what chemicals are extracted, at what level they are extracted, and the potential exposure (based on contact type, duration, intended population). Toxicologists use data gathered by the manufacturer and generated during extractables testing to form an assessment of the safety of the device.
For example, two devices may be found to have the same chemical at the same concentration, only the first device has contact with intact skin for less than 24 hours, and the second device is implanted in the body for more than 30 days. The latter device’s usage means that a patient will be exposed to that chemical continuously for an extended period of time, and that distinction affects whether the device is determined to have an acceptable risk or not.
ISO10993-18: 2020 Biological evaluation of medical devices - Part 18: Chemical characterization of materials states that if the risk assessment of the extractables data does not conclude that the device has acceptable risk, more clinically relevant leachables/simulated-use extractables testing is recommended (intended use studies). Furthermore, identifying certain hazardous or potentially hazardous compounds can necessitate targeted studies to improve quantitative accuracy. These studies may be suggested by a toxicologist or requested by a regulatory body.
Medical device manufacturers also play a role in the process; detailed compositional information for a medical device is crucial for analytical chemists to derive complete and accurate chemical characterization. Without this, there is an increased risk of a potential hazard misidentification, especially with devices constructed using novel materials the testing laboratory may not have previously encountered. Compounds observed in such cases are less likely to be present in commercially available and in-house chemical databases, making their identifications more challenging.
Myth 3: Extractables testing reveals what manufacturers already know about their device. The sheer number of reportable compounds in an extractables study can be quite shocking for medical device manufacturers. Extractables identify chemical compounds that are expected, as well as those that are unexpected—e.g., manufacturing residuals or impurities. Such results can cause significant delays in the toxicological risk assessment and regulatory review processes.
Extractables testing may progress without issue and not warrant further chemistry studies, but the possibility of additional testing cannot be ruled out. Managing timeline expectations is critical. Laboratory testing partners may request additional devices for testing, or biological testing may be necessary to mitigate risks and address biocompatibility endpoints.
Targeted testing plays a crucial role in this process in tandem with broader extractables profiling, similar to how a general practice physician may refer a patient to a specialist. A general practice physician (extractables screen) may be able to diagnose a broader range of conditions but may not be an expert on any of them. On the other hand, a specialist (targeted study) might be so focused on one area of disease that he/she may not even be aware of competing diagnoses.
Extractables profiling produces identifications with varying confidence levels and semi-quantitative results generated using surrogate standards. Conversely, targeted testing utilizes authentic standards of the exact compound(s), leading to more accurate results and identification confirmation.
Myth 4: Manufacturers can anticipate and avoid all questions regarding their submission. Medical device manufacturers cannot necessarily assess the quality of an extractables report on how many requests for additional information (AI) they receive. Regulators have different scientific experience and expectations; ultimately, it is on medical device manufacturers to satisfy those AI requests.
Common AI requests from regulatory bodies include:
Myth 5: Extractables testing produces highly accurate quantitative results. Screening methods are optimized for a wide range of potential samples, and accuracy can depend on numerous factors, including but not limited to:
While concurrent data collection from non-mass spectrometry-based detectors such as Diode Array (DAD) or Charged Aerosol Detectors (CAD) can help improve quantitation in some circumstances, this approach has drawbacks.
A loss of mass spectrometry signal (due to split flow in the instrument) can necessitate a higher concentration factor to meet the AET requirement. These detectors (DAD, CAD) are generally less sensitive than mass spectrometry detectors and cannot resolve co-eluting compounds.
Targeted testing is valuable when highly accurate quantitative results are required, and its application can be based on data collected during initial extractables testing.
Myth 6: Extractables testing will inevitably result in compounds identified as “unknown.” Simply put, unknowns are unacceptable. Laboratory testing partners should be capable of complete chemical characterization with no unknowns, and that expectation should be part of the initial vetting process. The option of calling something an unknown creates an easy off-ramp for analysts when they're struggling to identify a compound not in a database. It is difficult to overemphasize the importance of working with a testing laboratory that has both the expertise and culture required to produce the highest quality identifications.
With adequate mass spectrometry data, a thorough understanding of the device and a commitment to take the time, at least partial identification of compounds should be possible using spectral interpretation skills and spectral library searching. Commercially available spectral libraries are limited; therefore, lab testing partners will often turn to their own proprietary databases to assist with identification. Complete chemical characterization also requires collaboration between manufacturers and laboratory partners. Analytical chemists can better apply scientific judgment to distinguish between chemical compounds if they know a device’s materials. Working with a testing partner that sets a high bar in compound identification is a must.
Sandi Schaible is the senior director of analytical chemistry and regulatory toxicology at WuXi AppTec Medical Device Testing, specializing in extractables and leachables studies. She is a U.S. delegate and international delegate for ISO 10993 part 18 in chemical characterization, and also a U.S. delegate for ISO 10993 part 13 and the particulates committee (TIR42).
Peter Steen received his bachelor of science degree in chemistry and master of science degree in civil engineering from the University of Minnesota. His expertise includes chromatographic method development and validation, raw materials testing, medical device testing, and mass spectrometry. Since 2014 he’s worked in extractables/leachables testing at WuXi AppTec Medical Device Testing and is a subject matter expert in gas chromatography-mass spectrometry and liquid chromatography-mass spectrometry analyses.
In this column, E/L experts address six common myths associated with extractables testing and tactics manufacturers can apply to stay a step ahead during the chemical characterization process.
Myth 1: Extractables testing will reveal all potential leachables. Ideally, extractables profiling reveals all potential leachables above the analytical evaluation threshold (AET). However, factors such as the sample complexity (e.g., level and number of compounds) can suppress or mask compounds present at a low level.
Additionally, a single instrument or combination of instruments that can ensure the detection of all potential compounds does not exist. Thus, there are four methods used most widely when conducting extractables testing. They include:
- Headspace gas chromatography-mass spectrometry (HSGC-MS) to screen for volatile organic compounds
- Direct inject GC-MS to screen for semi-volatile compounds
- Liquid chromatography-mass spectrometry to screen for non-volatile compounds
- Inductively coupled plasma mass spectrometry for elemental impurities
Myth 2: Extractables testing is the last step in the chemical characterization process. Following ISO10993-1: 2018 Biological evaluation of medical devices - Part 1: Evaluation and testing within a risk management process, producing physical and/or chemical information by chemical characterization is a primary approach for evaluating medical device biocompatibility in order to manage biological risk. To evaluate risk, it is important to understand what chemicals are extracted, at what level they are extracted, and the potential exposure (based on contact type, duration, intended population). Toxicologists use data gathered by the manufacturer and generated during extractables testing to form an assessment of the safety of the device.
For example, two devices may be found to have the same chemical at the same concentration, only the first device has contact with intact skin for less than 24 hours, and the second device is implanted in the body for more than 30 days. The latter device’s usage means that a patient will be exposed to that chemical continuously for an extended period of time, and that distinction affects whether the device is determined to have an acceptable risk or not.
ISO10993-18: 2020 Biological evaluation of medical devices - Part 18: Chemical characterization of materials states that if the risk assessment of the extractables data does not conclude that the device has acceptable risk, more clinically relevant leachables/simulated-use extractables testing is recommended (intended use studies). Furthermore, identifying certain hazardous or potentially hazardous compounds can necessitate targeted studies to improve quantitative accuracy. These studies may be suggested by a toxicologist or requested by a regulatory body.
Medical device manufacturers also play a role in the process; detailed compositional information for a medical device is crucial for analytical chemists to derive complete and accurate chemical characterization. Without this, there is an increased risk of a potential hazard misidentification, especially with devices constructed using novel materials the testing laboratory may not have previously encountered. Compounds observed in such cases are less likely to be present in commercially available and in-house chemical databases, making their identifications more challenging.
Myth 3: Extractables testing reveals what manufacturers already know about their device. The sheer number of reportable compounds in an extractables study can be quite shocking for medical device manufacturers. Extractables identify chemical compounds that are expected, as well as those that are unexpected—e.g., manufacturing residuals or impurities. Such results can cause significant delays in the toxicological risk assessment and regulatory review processes.
Extractables testing may progress without issue and not warrant further chemistry studies, but the possibility of additional testing cannot be ruled out. Managing timeline expectations is critical. Laboratory testing partners may request additional devices for testing, or biological testing may be necessary to mitigate risks and address biocompatibility endpoints.
Targeted testing plays a crucial role in this process in tandem with broader extractables profiling, similar to how a general practice physician may refer a patient to a specialist. A general practice physician (extractables screen) may be able to diagnose a broader range of conditions but may not be an expert on any of them. On the other hand, a specialist (targeted study) might be so focused on one area of disease that he/she may not even be aware of competing diagnoses.
Extractables profiling produces identifications with varying confidence levels and semi-quantitative results generated using surrogate standards. Conversely, targeted testing utilizes authentic standards of the exact compound(s), leading to more accurate results and identification confirmation.
Myth 4: Manufacturers can anticipate and avoid all questions regarding their submission. Medical device manufacturers cannot necessarily assess the quality of an extractables report on how many requests for additional information (AI) they receive. Regulators have different scientific experience and expectations; ultimately, it is on medical device manufacturers to satisfy those AI requests.
Common AI requests from regulatory bodies include:
- Additional instrument method details
- Overall study design, including extraction solvents and conditions
- Recovery results associated with sample manipulation
- Surrogate standard justification and quantitation approach
- Process for compound identification
Myth 5: Extractables testing produces highly accurate quantitative results. Screening methods are optimized for a wide range of potential samples, and accuracy can depend on numerous factors, including but not limited to:
- Number and concentration of compounds (i.e., sample complexity)
- How similar the surrogate standards respond relative to the compounds they are used to quantitate
- Sample manipulation that could result in loss or degradation of compounds
While concurrent data collection from non-mass spectrometry-based detectors such as Diode Array (DAD) or Charged Aerosol Detectors (CAD) can help improve quantitation in some circumstances, this approach has drawbacks.
A loss of mass spectrometry signal (due to split flow in the instrument) can necessitate a higher concentration factor to meet the AET requirement. These detectors (DAD, CAD) are generally less sensitive than mass spectrometry detectors and cannot resolve co-eluting compounds.
Targeted testing is valuable when highly accurate quantitative results are required, and its application can be based on data collected during initial extractables testing.
Myth 6: Extractables testing will inevitably result in compounds identified as “unknown.” Simply put, unknowns are unacceptable. Laboratory testing partners should be capable of complete chemical characterization with no unknowns, and that expectation should be part of the initial vetting process. The option of calling something an unknown creates an easy off-ramp for analysts when they're struggling to identify a compound not in a database. It is difficult to overemphasize the importance of working with a testing laboratory that has both the expertise and culture required to produce the highest quality identifications.
With adequate mass spectrometry data, a thorough understanding of the device and a commitment to take the time, at least partial identification of compounds should be possible using spectral interpretation skills and spectral library searching. Commercially available spectral libraries are limited; therefore, lab testing partners will often turn to their own proprietary databases to assist with identification. Complete chemical characterization also requires collaboration between manufacturers and laboratory partners. Analytical chemists can better apply scientific judgment to distinguish between chemical compounds if they know a device’s materials. Working with a testing partner that sets a high bar in compound identification is a must.
A Final Word
Extractables profiling is more than just a perfunctory step in the medical device preclinical testing process; it is an essential first step that may lead to further testing. As regulators increasingly demand more thorough and precise extractables profiling, consulting with a laboratory testing partner that can offer complete chemical characterization with no unknowns and guide manufacturers through the submission process is critical.Sandi Schaible is the senior director of analytical chemistry and regulatory toxicology at WuXi AppTec Medical Device Testing, specializing in extractables and leachables studies. She is a U.S. delegate and international delegate for ISO 10993 part 18 in chemical characterization, and also a U.S. delegate for ISO 10993 part 13 and the particulates committee (TIR42).
Peter Steen received his bachelor of science degree in chemistry and master of science degree in civil engineering from the University of Minnesota. His expertise includes chromatographic method development and validation, raw materials testing, medical device testing, and mass spectrometry. Since 2014 he’s worked in extractables/leachables testing at WuXi AppTec Medical Device Testing and is a subject matter expert in gas chromatography-mass spectrometry and liquid chromatography-mass spectrometry analyses.
