GLP and Device Studies

How do Good Laboratory Practice (GLP) Regulations Apply to Medical Devices?

This article first appeared in Medical Product Outsourcing, September 2014. Please visit the MPO site to view the original article.

A question frequently asked by medical device designers is how and when GLP regulations apply to medical device studies. GLPs, short for Good Laboratory Practices, are the regulatory standards that define the minimum requirements for planning, conducting, and reporting nonclinical safety studies.

There is significant confusion about what constitutes “best practices” for applying GLP to medical device studies. Likely, this is because the regulations, which were written primarily to apply to nonclinical studies of chemicals and pharmaceuticals, are now being applied to a greater variety of studies – including PMA submissions and medical device biocompatibility studies in support of 510k submissions. Some study sponsors are even requesting that sterilization validation studies be conducted according to GLP. Unfortunately, the GLP regulations have not been modified to reflect this expanded scope, leaving us to use our best judgment as to when GLP should be applied, and how the language in the GLP regulations corresponds to medical device studies.

The two most common questions we are asked with regard to medical device GLP studies are:

  • What medical device studies should be conducted GLP?
  • What characterization data of a medical device is needed when conducting a GLP study?

Before we attempt to answer these two questions, let’s first examine what the GLP regulations are really about.

A Brief History of GLP in the United States

GLP regulations were created in response to a lack of consistency and quality in the conduct of studies performed to assess the safety of chemicals that humans might be exposed to, as well as the safety of drugs prior to initiation of first-in-human studies. During the 1970s, severe deficiencies in the conduct of nonclinical toxicological studies were uncovered in many labs – the most notable and egregious example being the Industrial Bio-Test Laboratories scandal, where 71% of studies audited by the FDA were invalidated for falsified or incorrectly gathered data. Evidence of toxicity was covered up by changing results.

To ensure that this type of laboratory misconduct would not occur again, Good Laboratory Practice regulations were adopted by the US FDA and made law on June 20, 1979.

Understanding FDA Regulations and the GLP Guidelines

Many regulatory bodies around the world have enacted GLP regulations; here, we will refer to the United States’ Food and Drug Administration version, spelled out in the Code of Federal Regulations (21 CFR part 58).

The GLP regulations govern specifically how nonclinical studies must be performed – which is important because many types of studies do not need to be conducted according to the strict GLP guidelines.

“A nonclinical laboratory study is an in vivo or in vitro experiment in which a test article is studied prospectively in a test system under laboratory conditions to determine its safety (21 CFR 58.3(d)).”

Essentially, GLP was created to govern nonclinical studies conducted to assess the safety of a chemical or drug, prior to any human exposure. Upon adoption by the FDA, the GLP regulations were also applied to PMA submissions for devices, and have since been extended to apply to studies supporting 510k submissions as well.

The GLP regulations are extensive, and lay out requirements for the controls that any facility performing GLP studies must enact to be compliant, and to ensure solid scientific data from nonclinical studies. The areas covered in GLP include:

  • Personnel and training
  • Definition and duties of a Study Director
  • Facilities and care of animals
  • Testing operations and data collection
  • Quality systems
  • Characterization of test articles and sample accountability
  • Disqualification of testing facilities

GLP was designed to guarantee that results from nonclinical studies could be verified, would be repeatable, and would produce sound, scientifically valid data. This is ensured by requiring that testing facilities conform to standards, have an independent Quality Assurance Unit to assure the integrity of studies, and appoint a knowledgeable and trained Study Director to oversee the study and interpret the test results.

Which Medical Device Studies should be Conducted According to GLP?

Now to answer our initial questions. First, which medical device studies must be conducted according to GLP, and when is GLP unnecessary?

It is clear as to what types of studies of pharmaceuticals and biopharmaceuticals must be conducted according to GLP. For medical device studies the situation is less clear. To better understand the issue, let’s consider the concept of intrinsic safety.

Intrinsic safety can be thought of as the safety of the device itself, in its final form. A device that is intrinsically very safe to use in humans (for instance, a plastic adhesive bandage) can be unsafe for use if it is not sterile when applied to a wound. An intrinsically unsafe device might, for example, be made of a material with a high toxicity. No amount of sterilization will render that device safe to use in humans. With this understanding, we will next look at three classes of studies, and discuss whether GLP should apply to those studies.

Biocompatibility Studies

Biocompatibility studies are done to determine safety – we are looking at the medical device’s effect on a biological system to better evaluate potential safety in humans. Therefore, FDA requires that biocompatibility for PMA submissions be done according to GLP and has strongly recommended that biocompatibility studies in support of 510K submissions also be conducted according to GLP.

Validation Studies

Do validation studies need to be conducted according to GLP? For instance, does a device sterilization validation require GLP treatment? The study is trying to validate the process that will render the device free of microorganisms (and thus safe for use in humans.) However, the study does not determine the intrinsic safety of the device, and by that measure is not a “nonclinical safety study” requiring GLP.

Yet there remains some debate as to whether validation studies need to be conducted according to GLP. Suffice it to say that in our more than 30 years of experience, we have never seen GLP as necessary for a validation study.

Rather than applying GLP standards, it is more important that the study be conducted at a facility with an adequate quality unit (including the performance of an independent QC review of the study results). The extra steps required by GLP are unnecessary, and can add to both cost and duration of a study.

Chemical Characterization Studies

The GLP regulations do mention other studies not classified as nonclinical laboratory studies and therefore not subject to GLP regulations. These include exploratory studies, and studies done to determine the physical or chemical properties of a device.

“…basic exploratory studies carried out to determine whether a device has any potential utility, or to determine physical or chemical characteristics of a device, are not subject to the GLP regulations (21 CFR 58.3(d)).”

Therefore, exploratory material and chemical characterization studies do not need to be conducted GLP. However, chemical characterization studies designed to gather data for a toxicological risk assessment as part of an overall biocompatibility program (for example, studies described in ISO 10993-18, Chemical Characterization of Materials) should be treated with more care. They likely do not need to be conducted according to GLP, but since these studies may be used in support of a PMA or 510k submission, strong quality systems and procedures, including independent QC review of data, are recommended. Additionally, it is prudent to include test article characterization data and information, as discussed in the next section, when performing these types of biocompatibility-supporting chemical characterization studies.

Recommendations for GLP/non-GLP conduct of device regulatory studies

Study TypeGLP, non GLP, or GMP
Biocompatibility (both in-vivo and in-vitro)GLP
Validation Studies (sterilization validations, reusable device cleaning validations)GLP (most likely) not needed
Exploratory StudiesGLP not needed (non GLP)
Chemical Characterization Studies (such as ISO 10993-18 Chemical Characterization)GLP (most likely) not needed
Routine Lot Testing (bioburden, endotoxin, sterility)Conduct according to GMP*

*GMP is short for Good Manufacturing Practices

GLP and Test Article Characterization

Once it is determined that a test does need to be performed according to GLP, it becomes necessary to gather and provide test article characterization data. The Study Director of the GLP study will need this information, as will any regulatory agency.

For medical devices, this can be somewhat complicated. Not surprisingly, the second most common question we are asked with regard to GLP medical device studies is what test article characterization data is needed.

According to the GLP regulations, the following data characterizing a test article is required:

“The identity, strength, purity, and composition or other characteristics which will appropriately define the test or control article shall be determined for each batch and shall be documented. Methods of synthesis, fabrication, or derivation of the test and control articles shall be documented by the sponsor or the testing facility (21 CFR 58.105(a)).”


“The stability of each test or control article shall be determined by the testing facility or by the sponsor either: (1) Before study initiation, or (2) concomitantly according to written standard operating procedures, which provide for periodic analysis of each batch. (21 CFR 58.105(b)).”

To understand how to decide what data is needed for a medical device, it’s important to consider the problems the GLPs were meant to address. A primary consideration was to ensure that the test article is fully representative of the final product. A pharmaceutical is defined by its identity, strength, purity, and composition. Stability data is required to ensure that the pharmaceutical has not degraded in the time frame in which testing will occur.

Medical devices are more complicated. They are not defined by a chemical identity, and most often “strength” and “purity” are not applicable when describing a medical device. So a reasonable analogue must be used.

We believe the most relevant data to gather for GLP characterization of a medical device are:

  • Description of the device
  • Types of materials the device is made of (method of manufacture and name of the manufacturer of any polymers, colorants, metals, etc.)
  • Methods of manufacture and synthesis of the final device (i.e. injection molding) and location of manufacturing facilities.
  • Lot number (if applicable).

If adequate records have been kept during the device development process, then manufacturers of medical devices should be able to find this information in the Device Master Record (DMR) or Device History Record (DHR), as required by 21 CFR part 820 subparts 181 and 184, respectively.

Stability data is also required, which, in the case of pharmaceuticals is sensible, as methods to determine pharmaceutical stability are well defined. But “stability“ of a device is less clear. However, it can be assumed that regulatory bodies are most interested in ensuring that the product being tested is equivalent to the final product, and hasn’t broken down or been rendered nonfunctional in some way.

Therefore, the “stability” data needed for a device GLP submission is most often a letter from the manufacturer with the date the test article was manufactured, lot or production number and information, and a statement that the product will be equivalent to its final form in the time that the GLP testing is being conducted. In some cases “stability” or “shelf-life” data is gathered later and inserted into the final regulatory submission.

A Final Word about the FDA

It’s always important to remember that the FDA is an agency staffed by intelligent people able to interpret and understand the reasoning behind what was done in studies. If one aspect of GLP is not followed 100%, yet there is sound reasoning for that deviation, it does not necessarily invalidate a regulatory submission. In fact, 21 CFR 58 subpart J requires any GLP deviations or “circumstances that may have affected the quality or integrity of the data” be noted within the final report of the study. The Study Director must explain why the deviation occurred and assess its impact on data integrity. If the reasoning is sound, a minor deviation should not negatively impact a submission.

Good luck and we hope this article provided some help in understanding and interpreting how to apply GLP to medical device studies.

For further reading, we suggest the FDA’s Draft Guidance for Industry and Food and Drug Administration Staff – The Applicability of Good Laboratory Practice in Premarket Device Submissions: Questions & Answers.


  1. “Industrial Bio-Test Laboratories” Wikipedia: The Free Encyclopedia. Wikimedia Foundation, Inc., 14 May 2014. Web. 4 June 2014.
  2. Schneider, Keith (1983-05-11). “IBT Labs’ trial reveals faked data”. In These Times. pp. 3, 6. Retrieved from
  3. Baldeshwiler, Anne. “History of FDA Good Laboratory Practices.” The Quality Assurance Journal Volume 7, Issue 3, Article first published online: 29 JUL 2003
  4. 21 CFR Part 58 – Good Laboratory Practice for Nonclinical Laboratory Studies CFRsearch.cfm?CFRPart=58
  5. 21 CFR Subchapter H – Medical Device Part 820 – Quality System Regulation, Subpart 181 Device Master Record.
  6. 21 CFR Subchapter H – Medical Device Part 820 – Quality System Regulation, Subpart 184 Device History Record CFRSearch.cfm?fr=820.184
  7. Draft Guidance for Industry and Food and Drug Administration Staff – The Applicability of Good Laboratory Practice in Premarket Device Submissions: Questions & Answers GuidanceDocuments/ucm366338.htm

Pacific BioLabs