Differences Between a GMP and Non-GMP Lab – Part 1

Greetings! I was hired on with Velesco earlier this summer after several years of work in cGMP pharma labs, both contract and manufacturing-related. Coming from a cGMP (current Good Manufacturing Practices) background, I’ve noticed a few differences in the way the lab operates when not confined to GMP regulations.  In the next couple of posts, I will discuss these differences.

Coming from cGMP pharma, which I will focus on in this post, I have been accustomed to all of the rules and regulations that are required by the FDA and they are second nature to me now.  Rules and regulations such as full documentation of all analytical work and results, audit trails and checks upon checks. See the FDA’s website for the full list of cGMP requirements. Analysts are required to train on each testing discipline, if not each individual test, before being able to perform the test alone. Countless Standard Operating Procedures (SOPs) are in place to provide step-by-step instructions for everything that needs to be done. SOPs must be read with reading documented before performing the test.  Checks and balances are in place for everything. Each piece of instrumentation or equipment is calibrated (checked to ensure it is working properly and accurately) before use.  Balances are checked with external weights that have been verified for accuracy. Thermometers and water baths are compared to known temperature standards (National Institute of Standards and Technology – NIST-traceable) to make sure they give the correct reading. Standard preparations are made and run alongside the samples to verify system suitability. No one thing is independently measured.

In cGMP labs, everything that is done is required to be fully documented and then reviewed by another qualified person to ensure the test was performed according to the method (usually SOPs, USP/NF, or EP) with all of the correct documentation.  Methods are also validated to make sure they work correctly for the samples and standards and give accurate results.  Dates and times are very important, and all data is signed off when completed.  All equipment used for any testing is validated, calibrated and documented. Each piece of equipment is assigned an asset number that is used to track calibrations, usage, etc. That number and the calibration information has to be recorded with the testing. In addition to that, each piece of equipment also has a logbook where the testing is documented, so reviewers can verify which piece of equipment was used for which tests. It becomes a web of documentation and I’m just scratching the surface.  For example, when an analyst gets water for a test, the water system used has been validated: checked to ensure it works right and puts out water (distilled, de-ionized) at the correct resistivity (usually 18.2MΩ), microbe-free. The water system is documented along with the analytical testing. The reviewer has to ensure that the water system information (asset number, calibration information and resistivity of the water) is documented in the testing write-up/logbook and that it was entered correctly.  That’s an easy example.

With recent innovation, most companies are headed to a paperless system, utilizing electronic notebooks, LIMS (Laboratory Information System) and software systems for data review. The GMP guidelines (21CFR part 11) now mandate that any change that is electronically made in software is captured in an audit trail – a part of the software system that tracks all changes made and who made the change – and those changes must be reviewed.  Most software systems require electronic signature (username and password) in order to save any change, especially when the change affects the test result or the test method.  In order to make even a small change to a sample name, etc. a username and password must be entered with a typed description of the change. All of this required documentation is quite time-consuming.

The most challenging part of documentation for cGMP’s is when a test does not go according to plan, or a piece of instrumentation/equipment fails.  That brings on the dreaded investigation – a large file of paperwork describing what happened, any impact it may have, and corrective and preventative actions to try to keep things from happening again. The purpose of an investigation is to outline what happened, try to identify the source of the issue and resolve the issue.  Mistakes may not be easily identifiable, and tests may not always be simply repeated. Retests are planned out, and often a one-for-one replacement for a result is insufficient.  Even a known error can take a day or so of paperwork and then retesting to replace the questionable result.

A large portion of my experience has been in product stability. I worked with quality control to identify and confirm expiration dates for various products and track their stability over time.  Part of the GMP requirement for commercial products is tracking label claim and impurity levels through the product expiration and checking for trending.  That means verifying that the product potency is not dropping out of the specified claim before the expiration and that it is not degrading beyond the allowable limit.  All of this data is summarized and reported to the FDA in an annual product report. Any product that is not meeting expiration requirements or that produces large degradants must be examined. Degradants that are too large must be identified, which is where the research aspect comes in.  A lot of the research involved comes from the method development work for new products.  Method Development is one thing that Velesco specializes in, which I will continue to discuss in Part 2 of this topic.


  1. …and part 2?

    • Sarah DeBoer

      Part 2 has been written and is being reviewed before posting! Should be posted soon!

  2. Hi Sarah, would be great to read part 2.

  3. Hi, this is a very good summary of the requirements of a GMP lab, but how does a non-GMP lab differ? I also have spent many years in cGMP and I’m now tasked with developing a quality system for a non-GMP/research lab. I am trying to convince upper management that we should strive to be ‘GMP-like’, in other words we should strive to adhere to the principles of GMP if we can. Personally I feel all that you have mentioned above is simply good practice regardless of the label. Calibrated equipment, SOPs, robust training, peer reviews – these are critical elements of a good QMS. But what would you say is not necessary for a non-GMP lab?

Leave a Reply