Читать книгу Biopolymers for Biomedical and Biotechnological Applications - Группа авторов - Страница 14

It is important to start the design of a chemistry testing strategy with the end goal in mind. In the case of chemistry for biocompatibility, the end goal is to be able to screen for unexpected contaminants with enough sensitivity and with enough accuracy that toxicological conclusions can be made based on the data produced (Figure 1.3). Determining the proper sensitivity can be a matter of debate but should be low enough so that any chemicals that are present – but not reported because they are below the sensitivity – are known to not be toxicologically concerning. In other words, a threshold of toxicological concern (TTC) is needed.

Figure 1.3 Important aspects for setting up a chemical characterization study.

The TTC concept was developed to define an acceptable intake for any unstudied/understudied chemical that, if below the TTC, would pose a negligible risk of carcinogenicity, systemic toxicity, and reproductive toxicity. The concept was developed for chemicals present in the human diet and is accepted by the US Food and Drug Administration (FDA), International Conference on Harmonization (ICH), and the European Medicines Agency (EMA) for the evaluation of impurities in pharmaceuticals. It has also been used for assessing contaminants in consumer products and environmental contaminants. The methods upon which the TTC is based are generally considered very conservative since they involve data for the most sensitive species and most sensitive site induction (several “worst‐case” assumptions). The TTC concept provides an estimate of safe exposures values for any compound not on the TTC exclusion list (i.e. metals, nitrosamines, and polycyclic aromatic hydrocarbons). The most conservative TTC value has been set at 1.5 μg/d and is assigned for greater than 10 years to a lifetime of exposure. A TTC of 120 μg/d has been proposed for genotoxic exposures limited to one month or less [12]. Exceeding the TTC is not necessarily associated with an increased risk given the conservative assumptions employed in the derivation of the TTC value [13,17]. When adequate evidence exists that a constituent is non‐carcinogenic, a non‐carcinogenic TTC value may be used to address the constituent (e.g. Cramer classification) [18,19].

The TTC concept for medical devices was formalized in ISO 21726 published in February 2019. This brief international standard outlines the appropriate strategy for using the Cramer class and TTC. When adequate toxicological data is not available in the literature, the Cramer classification should be used for non‐cancer effects; for cancer‐based effects, the ICH M7 TTC values should be used based on the contact duration of the device. Cramer classification stratifies compounds into three groups (I, II, and III, with III being the highest risk); the acceptable daily exposures are 1800 μg/d for class I, 540 μg/d for class II, and 90 μg/d for class III compounds. The TTC values from ISO 21726 for carcinogenic endpoints depend on contact duration and are shown in Table 1.3.

Table 1.3 Recommended TTC values from ISO 21726.

Medical device contact category	Limited (<24 h)	Prolonged (24 h to 30 d)	Long terma (>30 d)
Duration of body contact	≤1 mo	>1–12 mo	>1–10 yr	>10 yr to lifetime
TTC for any one compound (μg/d)	120	20	10	1.5b

^a Considered permanent according to ISO 10993‐1.

^b This value incorporates a 10⁻⁵ cancer risk for a 60 kg adult.

In addition to the sensitivity, the breadth of the analysis is critical. ISO 10993‐12, ISO 10993‐17, and ISO 10993‐18 provide guidance on the sample preparation and scope of analysis to give the required breadth. The device should be extracted in multiple solvents covering a range of polarities to be representative of the range of matrices that are found in the body. Extraction conditions should be selected to appropriately exaggerate the amount of chemicals found. For example, extraction of the device at 50 °C for 72 hours is prescribed by ISO 10993‐12 and is the most commonly used extraction condition. Typical extraction solvents are purified water, isopropyl alcohol, and hexane. Following extraction, the extracts must be analyzed for volatile organic compounds (VOCs), semi‐volatile organic compounds (SVOCs), non‐volatile organic compounds (NVOCs), and metals using a suite of techniques that are both qualitative and quantitative; these are almost always chromatography with mass spectroscopy (MS) for organic compounds and inductively coupled plasma for metals.

VOCs are typically analyzed for only in aqueous extracts, as semipolar and nonpolar solvents are often VOCs themselves. Two main techniques are available for VOCs: headspace gas chromatography with mass spectroscopy (HS‐GC/MS) and purge and trap GC/MS. HS‐GC/MS measures the volatiles present in the gas above a water sample in a closed vial; the vial might be slightly heated to encourage volatiles to enter the gas phase above the liquid. The gas is directed through a gas chromatograph, which separates molecules in the gaseous mixture by polarity. Different molecular polarities are retained in the instrument for different amounts of time; how long a molecule remains in the instrument is referred to as the retention time. After separation, the molecules are identified using mass spectroscopy. Briefly, mass spectroscopy works by fragmenting molecules into electrically charged pieces and then measuring the weight of those pieces very precisely. With knowledge of both the retention time and mass fragmentation patterns, VOCs can almost always be positively identified by comparison with large public or commercial databases. Purge and trap measurements differ from headspace only in the way compounds are sampled; first volatile organics are purged from the water by bubbling inert gas through the liquid and trapped in an adsorbent tube. VOCs are released from the tube into the GC/MS for analysis as with HS‐GC/MS.

SVOC measurement methods provide the single broadest source of information regarding the content of extracts and are amenable to both aqueous and nonaqueous extraction matrices. The term SVOC is ill defined in the medical device community but generally is considered to be those compounds most well suited for analysis by direct injection GC/MS. The distinction of this definition is important, as there are many molecules amenable to direct injection GC/MS that are considered to be NVOCs by every other definition. The methods used for SVOCs by GC/MS are mostly characterized by the details of their sample preparation and rigor of data analysis; instrumental details of the GC/MS remain largely harmonized. Water extracts are prepared for analysis by first doing a solvent exchange to a solvent compatible with GC/MS. Typically this is accomplished by repeatedly shaking the extract with methylene chloride under acidic, neutral, and basic conditions. The methylene chloride can then be concentrated and directly injected into the instrument. Organic solvents do not need a solvent exchange and are typically concentrated and then directly injected.

NVOCs not amenable for analysis by GC/MS are most clearly those compounds that have such a high molecular weight or polarity that they are not capable of vaporization without decomposition. For these compounds, liquid chromatography with mass spectroscopy (LC/MS) must be used. Unlike GC/MS analyses, which have more or less standardized instrument parameters, LC methods are highly variable. Because of this variability, large public databases are of limited utility, and effective interpretation of data relies much more on the level of expertise of the analyst and internal experience of the analyzing lab. LC techniques coupled with advanced mass spectroscopy tools providing high‐resolution accurate mass (HRAM) such as quantitative time of flight (qTOF) or Orbitrap can be a significant advantage, as these more sensitive methods can greatly narrow down the number of possible compounds in the identification process.

One of the key variables in chemical analysis for toxicological risk assessment and biocompatibility is the degree of certainty in the identification and quantification of compounds. Quality of identification can range from a fully automated comparison to a public database, without peer review of the results to fully confident identification. Fully automated identification can lead to scenarios where compounds with very low match scores are reported as compounds for which they are almost certainly not. On the other end of the identification spectrum is a fully validated identification where the compound in question has been injected using a standard on the same instrument and under the same conditions and under expert review. Of course, in practice, results can be a mix. It is not possible to inject standards for every compound that might occur from a biomaterial. With respect to quantification, results can vary based on the amount of evidence that is present to support the accuracy and precision of the presented results. On one end of the spectrum, results can be fully validated with calibration curves and precision and accuracy measurements. On the other end, results may be estimates based only on the concentration of an internal standard. Because patient safety may hinge on the result, often toxicologists want something more than a blind estimate of concentration of the compound is on the edge of being considered safe.

Chemistry results must be evaluated and assessed through the lens of toxicology to understand the possible systemic risks associated with the findings and the route of exposure of the device per ISO 10993‐17. This assessment should complement the results of traditional biocompatibility tests performed on biopolymeric device materials.