When the International Organization for Standardization (ISO) revised 10993-18 in 2020, the preceding standard had been in place for 15 years, and an evolution was overdue. ISO 10993-18:2020 established a need for medical device manufacturers to investigate their products’ chemical constituents more rigorously. 

ISO 10993-18:2020 provides guidance on chemical characterization information, either collected or generated, for medical device evaluations. This information provides an estimate of patient exposure and is the basis for a toxicological risk assessment (TRA). Chemical characterization can be a paper-based review of all materials or a targeted study focusing on a single compound of concern, but more often necessitates data generation through extractables/leachables (E/L) testing. For many medical devices, exhaustive and exaggerated extractions are recommended by ISO 10993-18:2020 and are expected by regulators.  

Prolonged and long-term medical devices are routinely subjected to exhaustive and exaggerated extractions, which also increases the number of solvents and analytical time points. Exaggerated and exhaustive extractions can yield hundreds, sometimes thousands, of compounds for toxicologists to assess. Just based on the sheer number of chemicals, the TRA process takes considerably more time and increases the likelihood of an “equivocal” result (i.e., a finding from the TRA for which toxicologists cannot deem the level of risk acceptable given the available information). 

Historically, the total amount of each chemical in the exhaustive extraction study is conservatively assumed to be released daily for a worst-case exposure estimate. This results in an extreme overestimation of exposure, and an overestimation of risk. Understanding the release kinetics can provide a more realistic estimate of worst-case daily exposure. 

As mentioned in ISO 10993-18:2020, information on the release kinetics of extracted chemicals can be helpful to support the TRA. But there is no guidance for conducting studies to evaluate release kinetics. Exhaustive extractions often lead to additional chemical characterization to evaluate release kinetics (e.g., through simulated-use extractions) and mitigate toxicological risk, so it helps to conduct these studies concurrently or to budget longer timelines for medical device evaluation. 

The bottom line is device manufacturers should consider materials characterization a journey. Information is introduced and considered (often multiple times) before the program is complete. Proactively planning precious resources while expecting a multi-step journey will help avoid unanticipated delays.

Waiting for the ISO 10993-17 Update

The current version of 10993-17 (published 20 years ago) provides information on establishing allowable limits for chemicals but does not address exposure assessment or toxicological risk characterization. The revised version, currently in development, will help reduce the burden of evaluating large extractable datasets with tools and guidance. The standard will also provide a framework for: 

• Screening data using a toxicological screening limit (TSL)
• Applying threshold of toxicological concern (TTC)
• Establishing estimated exposure dose based on assumed or actual release kinetics data
• Deriving tolerable intake (TI) and tolerable contact levels (TCL)
• Calculating margins of safety (MOS)
• Evaluating the MOS with toxicological risk acceptance criteria 

When an exposure dose is below the TI, TCL or TTC, it results in a MOS of greater than one. When the MOS is greater than one, the toxicological risk is considered tolerable or acceptable. A MOS value below 1 requires additional risk analysis, risk evaluation or risk control. The qualifications and support of an experienced toxicologist are pivotal in determining acceptability of toxicological risk and the overall biological risk of a final finished medical device.

Defining Release Kinetics

Release kinetics is the quantity of a constituent released from a medical device as a function of time. For example, a patient with a long-term implant will be exposed to leachable chemicals at different levels over time. For most chemical constituents, it is likely that the patient will first be exposed to a greater concentration of a substance short-term, and then the exposure will taper off over time.

Understanding and calculating concentrations of estimated exposure to chemical constituents over different periods of time will be an important consideration in the revised ISO 10993-17. Similarly, exposure during different durations will be compared to the appropriate TI or TCL for that duration, and MOS values may be different for acute/subacute versus subchronic or chronic exposure durations. This can inform the risk assessor about whether there are identified short-term and/or long-term risks, and risk mitigation strategies can be tailored to address the relevant risks.

Toxicologists will use worst-case daily exposure estimates based on available chemical characterization information. This information can include a total quantity of each chemical constituent based on compositional information or based on an exhaustive or exaggerated extraction study. When an exhaustive extraction study includes analysis of multiple iterations, it can provide an indication of release kinetics. Otherwise, the estimated exposure can be calculated based on actual or assumed release kinetics.

For the risk assessor, the value of release kinetics data—actual or assumed—is threefold: (1) It achieves a more clinically relevant estimate of exposure; (2) It offers a better understanding of the short-term and long-term toxicological risks; (3) It informs the next steps, which may include additional chemical information (i.e., actual release kinetics data, targeted chemical analysis), biological testing and possibly risk control.

Designing Release Kinetics Studies

Release kinetics data can be obtained experimentally through a simulated-use study or a leachables study. Alternatively, a qualified or validated release model can be used if supporting chemical and material data are available. However, leachables studies are ideal because they provide the most relevant chemical characterization data and measure the substances actually released from the medical device under clinical use conditions.

Designing these studies may be straightforward for devices that contact patients indirectly through saline or other intended solutions used clinically (like drugs or biologics in the case of combination products), utilizing the clinically relevant matrix, time and temperature. However, for medical devices in which the device contacts the body directly or indirectly, and chemicals are released into biological matrices, it is challenging to determine actual leachables.

A simulated-use study can be designed to evaluate quantities of chemical constituents released from the medical device over time in a simulating solvent, at a clinically relevant temperature (e.g., 37°C). The time periods for analysis will be based on the duration of clinical exposure, and more than one timepoint. For prolonged and long-term devices, often the 24-hour timepoint will provide the worst-case daily exposure estimate, though there are exceptions.

Using a long-term implant as an example, if we conduct an exhaustive extraction and assume that all the chemicals will be released in 24 hours, and a patient will be exposed to this amount daily for the duration of device exposure, it will likely lead to a low MOS for many chemicals. A proposed new approach in the draft version of ISO 10993-17 will allow for estimation of assumed worst-case release of chemicals over time that could result in a conclusion that risk is acceptable for subchronic and chronic exposure durations but is equivocal for acute/subacute exposure to the implant.

A simulated-use study may be used to determine release kinetics over multiple timepoints during this short-term exposure duration. For example, to establish a short-term kinetic release curve, the device may be incubated in the selected simulating solvent and chemical analysis conducted at 24 hours, 3 or 7 days, and 30 days. Estimated exposure based on release kinetics data may provide a more accurate assessment of toxicological risk for the short-term duration; and support the toxicological risk for a long-term duration when levels of extractable chemicals decrease over time.

The biggest challenge with simulated-use studies is selecting the appropriate solvent or solvents to simulate the clinical use environment. This is an area where research and guidance are lacking and will be an important area of future development among the medical device community.

A Final Word

ISO 10993-18:2020 first mentioned release kinetics as a consideration for chemical characterization. The upcoming revision to 10993-17 proposes the use of release kinetics data to support toxicological risk assessments. As this information becomes more important, it should stimulate the advancement of science in this area, hopefully leading to better published scientific information, and guidance on designing release kinetics studies for medical devices.

Manufacturers should expect that release kinetics data will be needed to help mitigate the risks identified in exhaustive extraction studies. The additional chemical information will likely need to be included in their materials characterization and toxicological risk assessment timeline, particularly for long-term devices. The proposed approaches in the draft version of ISO 10993-17 give toxicologists additional tools in their toolbox to evaluate the toxicological risks of chemical constituents, which will better address biological safety.