Thank you all for the great interest in
Bob Przygoda
's presentation and for participating in the discussions on the topic “Clinical relevance of extractions for in vitro and in vivo studies”!
As
Bob Przygoda
did not have the time to answer all your questions during the discussion he kindly provided some comments and clarifications to his presentation, as well answers to the questions, that we would like to share with you.
?Comments and clarifications, slide #5
This figure shows the range of exposures that a medical device investigator could encounter.?? The center line shows exposures ranging from 10-9 mg/kg to 10,000 mg/kg.?
The next line below shows the range of NOEL for chronically toxic chemicals. The dose range is from 0.1 mg/kg to approximately 5000, mg/kg.? The dose that would elicit a response would be expected to be slightly greater than the NOELs.?
The line below the NOEL is the range for chronic TI.? The range was based on the assumption that a100x UF would be used.? The 3 vertical lines indicate, where Cramer Class TTC would fall along this chronic TI line.? Note that none of the Cramer Class limits would be detected in a chronic toxicity test.
?The line below chronic TI line is TD50 values (from Gold).? There are 2 lines, one for Cohort of Concern and the other, for all other mutagenic chemicals.?? Excluding the cohort of concern, the TD50 values appeared to be slightly higher than the chronic NOEL.?? It would probably be consistent with the chronic least effective dose.?
The cohort of concern represent less than1% of the TD50 values and these values are significantly lower than the non-cohort of concern mutagens.?? The Cohort of Concern TD50 values do not appear to be evenly distributed along this line.
The bottom line is the range of Cancer Risk Specific Doses for mutagens excluding the cohort of concern and another line for the cohort of concern.? The vertical line is the equivalent TTC dose.? Note: None of the Cancer Risk Specific Doses would be detected using in vivo a carcinogenicity test.
The Top line on this side is the range of concentrations? that could induce a positive response.? This information is from in vivo micronucleus data for 79 IARC carcinogens.? The data is the Least Positive Dose that could induce micronuclei response.?? The vertical line at each end of this line denotes the equivalent percent extracted from the device compared to does in mg/kg.? Note that the lower limit of detection for the micronucleus is not much different then the TD50 values for non-COC carcinogens.? The in vivo micronucleus test ?would not increase your ability to detect a mutagen at concentrations equivalent to cancer risk specific dose (i.e. TI).
?The take home message is biological tests are not sufficiently sensitive to ensure toxicants are not present at concentrations that would increase risk to the patient.
Comments and clarifications, slide #7
?The figure compares the percent extracted from a device with the cumulative frequency? of least positive response? in 3 genetic toxicity tests.?
The y axis is the cumulative frequency of detection using the least positive dose. The x axis is the percent extracted from a device.? For instance, if the device 2 grams and 0.2 gram was extracted, the percent extracted was 10%.? It was assumed that 100% of a mutagenic chemical would be extracted from the device.? A 10% extraction would be equivalent to 20mg/ml (200mg device / 1 ml divided 10) . The analysis assumed that if a chemical was detected at 10 ug/mL , the amount needed in a non-polar extract to obtain the 10 ug/mL concentration would be 100x greater, for polar extracts, the amount needed would be 10x greater and for culture media the amount need would be 1x.
The vertical lines indicates:
- 0.1% or greater where a CMR requires labeling of the device?
- The AET for 0.2 gram device assuming 1 device is used extracted in 1 mL and there UF is 1.? In most cases the UF is 3 – 5 depending on the method of analysis.
- The AET for a 2 gram device using the same assumptions used for a 0.2 gram device.
?Extraction using culture media (polar extraction) performed best for in vitro chromosomal aberration test.?? However, when a nonpolar extract is used, the chrom abs test was not as sensitive in detecting mutagens and was similar to in vivo micronucleus in its sensitivity to detect mutagens.
The Ames test when extracted either with a polar or nonpolar extract, detected mutagens with a sensitivity in between the polar and nonpolar extracts for the chromosomal aberration test.
?When the chemical identity is unknown, the TI is set for the TTC (1.5 ug/day).
?For this analysis the AET instead of the TTC was used since the units are the same as cell culture concentrations.??
?To ensure that that unknown mutagen was not present at concentrations above the TTC (AET), the detection curves in this graph should be to the left of the AET.? However, less than 10% of the mutagens could be detected at the AET in any genetic toxicity test.
?Note:? At concentrations at or below 0.01 percent, over 50 % of the mutagens would not be detected.?
?A negative genetic toxicity test indicates that a mutagen if present, is below the detection limit of the test.? It does not mean that the mutagen is not present? or is below the Tolerable Intake (TI).? In other words, a mutagen could be present at a concentration above the TI but below the detection limit of the test.?
?Reminder, the genetic toxicity, chronic toxicity, and carcinogenicity tests were designed for hazard identification of a relatively pure chemical.? These tests were not designed for mixtures.?
Answers to questions asked during the discussion
- Please add a note on extraction if multiple comments of varying thickness are present.Consideration should be given to which component represents the worst-case patient exposure.?? Also can all the components be extracted using the same extraction ratio. For instance using 6 cm2/mL may not feasible since the device would not be complete covered by the extract solvent.? If the thinnest component represents the largest potential for patient exposure, 6 cm2/mL would be selected.
- For 0.1% CMR wouldn’t we have to dissolve the device to determine the weight percentage?I agree that a device would have to be dissolve (or the formulation of the material indicate it is present at 0.1% or greater) to determine if a CMR is present at or above 0.1%.? The point being made in the presentation was that a mutagenic carcinogen could be extracted at 0.1% (based on non-volatile residue? divided by the total wight of the device) and it may not be detected if only genetic toxicity tests are performed.? This would be a significant miss.
- Does anyone know the origins of 0.2g/mL, 3 cm2/mL and 6 cm2/mL extraction ratios?As stated in the presentation, the extraction ratio can be traced back to USP in 1960s.? It was for biological testing of containers.? I had asked USP to see if they could explain why these ratios were chosen.? The archives were checked, and there was no documentation about the origin of the extraction ratios.
- Using saline for extraction will lead to large dilution factor needed prior to ICP/OES and ICP/MS analysis. You may then miss your needed sensitivity for some metals, as the reposting limit will be raised by the dilution factor.In the presentation, the use of saline was referring to biological tests and not to chemical analysis.? I agree that using saline to extract a device can make the ICP analysis difficult.?? I am under the impression that it can be performed and maintain the necessary sensitivity. ??However, this required additional processing of the test article.
- On slide 4, Bob states that because device extracts usually include compound mixtures the concentrations may not be optimal.? It’s not clear to me why this may be so.? Is it the case that some extracted compounds could decrease solubility of others?? This doesn’t seem consistent with my understanding of dissolution processes, though I could be wrong.According to OECD test guidelines, relatively pure chemicals are used to identify potential toxicity hazards.? In addition, the concentration of the chemical must meet specific requirements to ensure that there is no hazard.? For instance, the chemical must be tested at 2 g/kg for in vivo tests and 2 mg/mL for in vitro tests, etc.? If toxicity is observed, the test must include concentrations demonstrating a dose response. If the chemical is not soluble at the required concentrations, at least one dose most be above the solubility of the chemical. Often, when testing a medical device extract, the amount and identity of the chemicals in the extract are unknown.? Therefore, you do not know if you met the top dose criteria for each individual chemical. Many of the OECD guidelines have discouraging statements about testing mixtures or state that mixtures should only tested when required by regulation
- On slide 13, Bob questions the utility of extracts when the “amount extracted is 0.1% or less”.? Is the numerator the NVR in mass units?? What’s the denominator?The numerator is the mass of the NVR and the denominator is the mass of the device that was extracted. I would note that if 0.1% of the device weight is extracted, the total amount of chemicals in the extract is 0.2mg/mL (200ug/mL).?For the Ames test the highest possible concentration that can be achieved is 20 ug/plate (0.1 mL of 200 ug/mL = 20 ug / plate) Less that half of the 109 IARC carcinogens used in the presentation could be detected at 20 ug/mL.?For the non-polar extracts for the in vitro chromosomal aberration test and for both non-polar and polar extracts of the in vivo micronucleus test, less than 25 % of the mutagen can be detected at 0.1%.To determine the potential hazard from a single chemical, the concentration must meet the following criteria: The highest concentration was 2 mg/kg (in vivo tests), 2 mg/mL (in vitro tests) or meet the toxicity limits for the test.Unless every chemical in a mixture meets this criteria, a hazard cannot be excluded based on a negative test.
- ?Bob, do you think that regulators should stop accepting Ames +MLA as the primary dataset for addressing genotoxicity?I do not think using the Ames plus MLA as the primary dataset for addressing genetic toxicity is sufficient.? Additional data is needed.? As previously stated, chemical characterization or chemical analysis is needed to supplement the biological data for this end-point.
- Do you still need to perform chemical characterization study for a device if you have all the applicable?studies in place including systemic, genotoxicity and others? In general, if we find any unknowns expected we are still ask to perform genotox studies..The point of the presentation was that biological tests without additional characterization are not conclusive. The hazard may still be present below the level of detection.? Characterization does not have to be a chemical analysis of the extract, if could be an analysis of the formulations of the materials used to construct the device including any manufacturer aides that may leave a residue on the device. I would note that ISO 10993-1 requires a chemical characterization prior to performing any biological tests.If unknowns are reported from a chemical analysis, I usually ask for the lab to provide additional information on the unknowns (for example, can you provide the potential structure of the unknown and QSAR can be used to determine the appropriate TTC).?? If a lab consistently reports unknowns above the AET, or if an unknown is in milligram amounts,? I would look for another lab.? Most good chemical analysis labs should be able to give sufficient chemical information to perform a risk assessment.? Many labs understand what is needed to perform a risk assessment.?
- Chemical analysis appear to be better for assessing known mutagens/carcinogens, while the biological testing is better than nothing when unknown genotoxic carcinogens are present.If unknowns are reported from a chemical analysis, I usually ask for the lab to provide additional information on the unknowns (for example, can you provide the potential structure of the unknown and QSAR can be used to determine the appropriate TTC).???If it is determined that there is a potential for genetic toxicity and if the amount is greater that the TTC, biological testing may be appropriate if the potential genotoxicant can be concentrated to meet the requirements for hazard identification. Usually eliminating the source of the genotoxicant is a better option.I usually prefer to perform both chemical analysis of the extract and genetic toxicity tests.?? There are problems noted in the identification chemicals in an extract and the quality of the work is inconsistent between labs and overtime in the same lab.? The quality of the chemical analysis is dependent on the expertise of the person doing the chemical analysis. Sometimes a change in personnel in a lab, results in a change in the quality of the analysis.?When a sufficient amount of material is extracted from a device, genetic toxicity tests can be helpful in eliminating any positive synergistic genotoxic response.? Also, some reviewers may not except chemical analysis alone to determine if there is a genetic toxicity hazard and will insist on biological tests.
We hope that this gave you more knowledge about the clinical relevance of extractions for in vivo and in vitro studies and we look forward to hear more about this topic from
Bob Przygoda
in the near future, please stay tuned for more information.
We also hope to see you at the next Biocompatibility Coffee break on February 21 and at our conference Biocompatibility Insights 2024 in Copenhagen, Denmark. Training day 15 October, Conference days 16-17 October.
