To Test Or Not to Test: Justifications

In the past, performing medical device testing was like checking off boxes: Neither the device type, type and duration of contact, nor materials mattered. While G95 was used as a kind of checklist, the new guidance teaches testing professionals how to change materials or how to view testing in a more scientific way. It puts more emphasis on scientific analysis, thus compelling professionals to have more knowledge and background about the tests they perform. In addition, it contains good ideas on how to justify biocompatibility testing—specifically animal testing—and presents ways for limiting the use of such tests for certain devices.

Weighing Device Benefits and Potential Risks. Failing a cytotoxicity test does not mean that a medical device should not be used in the body. The failure must be weighed against the benefits of the device. For example, while chemotherapy drugs are highly toxic, their benefits outweigh the risks.

Applying Scientific Judgments to Animal Test Data. Many good tests—in vitro or chemical characterization tests, for example—can be used to help justify certain animal tests that have been around for a long time but may not be as good as patient tests. Thus, testing professionals are encouraged to apply scientific judgment. On the other hand, medical device manufacturers are encouraged to apply scientific judgments to all the tests they perform, despite the difficulties involved in doing so. Based on the device in question, it behooves medical device experts to analyze the tests they have performed in order to determine where scientific judgment can take the place of some tests.

Literature vs. Correct Data. Scientific judgments can also be used to justify which tests are performed. However, the correct end points of toxicity tests must be known, and data derived from the literature must correspond to these end points.

Biocompatibility vs. Animal Testing. Biocompatibility testing or adequate chemical characterizations in conjunction with supplementary biocompatibility testing may be acceptable to minimize animal testing.

Predicate Samples. If a medical device’s intended use, physical form, formulation, processing method, component interaction, and storage conditions are the same as those of a predicate sample, no testing or limited testing is required. Clients often say, “I have a device that’s made from the same material as my competitor’s device, and they have a 510(k), so I’m not going to do any biocompatibility testing.” However, there are many factors that medical device manufactures must consider when they are planning to forego biocompatibility testing based on predicate samples or the existence of the same device used for other purposes.

Functionality Testing vs. Animal Testing. To help reduce animal testing, functionality tests can be designed to determine some of the end points normally derived from biocompatibility testing, particularly hemocompatibility end points. When medical devices are subjected to functionality tests in large animals to analyze their performance, many hemocompatibility end points can be incorporated into these tests to reduce animal tests for hemocompatibility. A similar approach can be applied to implantation or chronic toxicity end points. However, before commencing with functionality testing, it is crucial to incorporate the end points into the functionality protocol to help justify which tests will be performed.

The New Guidance and You

The 2013 guidance document incorporates changes that are new to the medical device industry. The first such change is that devices subjected to multiple types of exposures should be tested for all of them.

For example, if a short-term drug-delivery device or IV bag features a long-term patient-access port through which multiple devices might pass, both the port and the device itself should be evaluated. In the past, some device manufacturers performed tests on long-term-contact devices that incorporate both short- and long-term components, relying on the worst-case scenario. But such tests are based on surface area, and the greater a device’s surface area, the more volume it has. Thus, by testing both short- and long-term components, the material’s testable surface area increases. If the short-term parts of the device are nontoxic while the long-term parts are toxic, the increased surface area will skew the test results. Thus, it’s important to separate the two types of components so that the surface area of the safe materials does not obscure that of the toxic materials.

Another change in the new guidance documents stipulates that if a device includes components that contact the body for different lengths of time, such as an implant and its delivery system, extract tests should be conducted on the components separately. For example, a stent delivery system cannot be tested together with the stent.

The document states that “implants, as well as sterile devices in contact directly or indirectly with the cardiovascular system, the lymphatic system, or cerebrospinal fluid, regardless of duration of contact,” should meet pyrogen limit specifications. In the past, guidance, especially from the U.S. Pharmacopeial (USP) convention, stressed that such implantable devices should undergo limulus amebocyte lysate (LAL) testing. A pyrogenicity test—or a test for fever—the LAL test specifically looks for a pyrogen called endotoxin, a compound found in Gram-negative bacteria cell walls. Whether all implants—not just those that make contact directly or indirectly with the cardiovascular system, the lymphatic system, or cerebrospinal fluid—have to undergo LAL testing is still unclear.

In the subsection on pyrogenicity, the draft guidance recommends that regardless of contact duration, material-mediated pyrogenicity should be assessed at 50°C or higher (except in the case of heat-labile or heat-sensitive materials) using the rabbit pyrogen test. While this statement makes sense from a scientific standpoint, why does FDA single out the rabbit pyrogen test if it wished to stress that devices should be extracted at 50°C where possible? Since it had already addressed testing involving prolonged contact at 50°C or higher elsewhere in the document, it would appear to be redundant here unless FDA is concerned specifically about using higher temperatures in the rabbit pyrogen test.

In the past, FDA preferred that devices be tested for genotoxicity using the mouse micronucleus, Ames, mouse lymphoma, or chromosomal aberration assay. Now, the new guidance document has settled on the mouse lymphoma test because it detects the broadest set of genotoxic mechanisms associated with carcinogenic activity. However, contradicting data suggest that the chromosome aberration test can detect a broader set of genotoxic mechanisms. In fact, Japanese experts presented data three years ago indicating that the chromosome aberration test might be the most sensitive genotoxicity test. Thus, more guidance is required in this area.

“Initial Evaluation Tests for Consideration,” Table 1 in the new guidance document, may have the greatest impact. In G95, the table contained little dots to designate required tests and diamonds for others that may be applicable, enabling an FDA reviewer to require the additional tests, depending on the device.

Emphasizing the importance of the tests designated with diamonds, the new document states, “These additional evaluation tests should be addressed in the submission, either by inclusion of the testing or a rationale for its omission.” In the spirit of not treating the guidance document as a checklist, this table highlights the need to evaluate each one of these tests carefully, addressing them in the submission either by way of testing or by justifying the decision to forego testing. The new draft document puts greater emphasis than G95 did on the importance of knowing the scientific rationale behind the tests that are performed and devising a meaningful test outline for the device.

Thor Rollins is in vivo biocompatibility section leader at Nelson Laboratories Inc. A certified microbiologist with the National Registry of Certified Microbiologists, he is a participating member on all AAMI 10993 ISO committees and plays an active role along with FDA and regulatory committees in developing, discussing, and voting on changes to standards. Rollins received a BS in biology with an emphasis in cell biology and genetics from Idaho State University. E-mail him at [email protected].

? Return to Part 2