NSF International auditors share their insights on common GMP problem areas.
The dietary supplements industry has grown tremendously over the last several years, thanks in part to a regulatory structure that allows new products quick time-to-market. In turn, consumers have benefitted from the wide range and availability of dietary supplements that help support a healthy lifestyle. With 68% of U.S. adults labeling themselves as supplement users,1 nutritional product manufacturers have significant opportunity to grow their operations. However, with this opportunity and ease of the open market comes a great deal of responsibility to ensure that dietary supplements are safe and meet the quality standards that consumers, and regulators, expect.
FDA clearly defines these safety and quality standards in the Code of Federal Regulations Title 21 (21 CFR111), otherwise known as Good Manufacturing Practices (GMPs). Unfortunately, as evidenced by the alarming number of FDA-issued warning letters in recent years, many firms still struggle to comply with GMPs.
Only through rigorous compliance with GMPs can a company be confident in the products it produces. Sometimes, this means going beyond what is spelled out in the regulations.
Auditors from NSF International-who since 2000 have registered more than 350 facilities to NSF’s GMP program-share their insights, as well as tips, on what companies continue to struggle with the most: testing. Inadequate testing is the most common GMP infraction. Testing is broken down into three parts-identity testing, supplier qualification testing, and finished-product testing.
According to FDA reports and warning letters, many firms are still not spending enough time developing appropriate test methods to validate the identity of incoming ingredients.
To satisfy this GMP requirement, many companies use in-house verification technology, such as Fourier transform infrared (FTIR) and near-infrared (NIR) spectrometers. NSF auditors find these technologies commonly used, as they are relatively inexpensive and quick, hence their appeal; however, they may not be appropriate for identifying all incoming materials, such as blends or botanicals. The reason is that when ingredients are mixed with other compounds or are derived from plant material, and if the proper validated methods are not being used, these verification techniques could potentially misidentify the botanical species or fail to detect an adulterant.
Also, FTIR and NIR spectrometers create a “fingerprint” or data pattern of the ingredient. The more complex the ingredient, the more complex the fingerprint. The most important aspect of using this method-and where many companies continue to struggle-is establishing a reference standard of what a sample of good, high-quality material should look like. Companies need these reference standards with which to compare incoming ingredients. An FTIR and NIR spectrometer will produce a scan for virtually any ingredient, but without an established standard to compare an ingredient against, it’s hard to tell whether the incoming ingredient meets the manufacturer’s specification.
To properly develop these reference standards, companies should have several samples (six or more, preferably) that they’ve qualified using other techniques, to develop a library of good ingredient data sets. NIR instruments in particular are sensitive enough to pick up very subtle differences in various ingredient lots, which may be particularly critical when dealing with botanicals. Factors such as moisture level, particle size within the sample, and even what time of year the ingredient was harvested can have an effect on the resulting fingerprint or data pattern. For this reason, having multiple reference samples to compare against will accommodate these very miniscule variations and still provide a means of identifying the correct match to the ingredient specification.
Properly identifying incoming ingredients also enables a company to protect itself against bad players in the supply chain. For instance, NSF auditors have observed cases in which chondroitin sulfate has been diluted with a cheaper substitute called carageenan. If misidentified ingredients make their way into a finished product, they compromise the integrity and accuracy of the formulation and label claims. Again, testing is crucial.
GMPs require that dietary supplement companies qualify their suppliers. While the GMP regulations don’t exactly spell out what “qualify” means exactly, they do go as far to say that companies must verify the Certificate of Analysis (CofA) of incoming ingredients. But ingredients and their CofAs pass through many hands before making their way to a company’s facility for processing, providing plenty of opportunity for contamination or fraud along the way. Depending on the nuances of a company’s supply chain, it may not be enough to just compare a CofA against specifications and call it a day. CofAs may not always be a reliable testament to the quality of the ingredient. Often, CofAs for a particular ingredient are a reproduction created by the ingredient distributor rather than the true ingredient CofA created by the manufacturer of the ingredient. NSF auditors have seen synthetic vitamin E being sold and represented as the more expensive natural vitamin E, as well as instances in which a company sourced a high-quality version of glucosamine but after testing found that it was a much cheaper form that didn’t match the CofA. Proper testing to verify the source eliminates this type of fraud. This is why many leading dietary supplement companies have heeded the good advice of the FDA and verify CofAs through proper testing.
Additionally, NSF auditors continue to see facilities struggle with properly identifying their suppliers. Supply chains are usually long and complex, making it hard to determine who originally manufactured the ingredient. Known cases of adulterated raw materials making their way into the United States have even sparked new regulations. The Food Safety Modernization Act in particular keys in on suppliers as a control point for preventing contaminated products from entering U.S. facilities.
NSF auditors recommend keeping the paper trail as clean and clear as possible. To ensure continued compliance to specifications and to GMPs, it is highly recommended that companies perform periodic re-qualification of suppliers and testing to verify CofAs.
Consumers buying a product place a great deal of trust in that brand and expect product labels to be accurate. Testing finished products to verify the accuracy of their labels, as well as the absence of harmful levels of contaminants such as lead, is paramount to GMP compliance; however, testing complex finished products requires high-tech instrumentation, test methods, and expertise that many companies do not have in house.
Some finished products have upwards of 20 to 50 different ingredients and label claims. This becomes even more complex as testing and verification involves trace amounts-microgram and milligrams in dosage quantities. The challenge can sometime lie in selecting the appropriate test methods and instruments. High-performance liquid chromatography (HPLC) is often most suitable for the analysis of various phytochemicals. Inductively coupled plasma (ICP) instruments are ideal for confirming mineral levels such as calcium, while ICP mass spectrometry (ICPMS) is needed for measuring trace heavy metal contaminants. For products containing fatty acids, such as fish oil, gas chromatography (GS) instruments are required.
Companies that manufacture complex proprietary blends have an even greater challenge, as there are no compendial standards to compare their finished products against (as opposed to a product containing a singular ingredient). This is why many companies seek the expertise of outside, third-party accredited laboratories to help review their formulas and develop test methods, or even perform the testing for them.
According to Kerri LeVanseler, technical manager of NSF International’s Chemistry Laboratory, sometimes it’s not just an analytical testing challenge. Separating the different compounds in the finished products and then controlling the cost of the various testing required can also be difficult. To help control the cost of testing, LeVanseler recommends performing a majority of the contaminant testing on incoming ingredients and then performing label claim verification testing on finished products. For complex finished products, this can still add up to a lot of tests. By always ensuring the quality of the ingredients and monitoring the ingredient inputs during the manufacturing process, the focus in the finished-product testing can be on those parameters that are most sensitive to change or degradation. Building quality into the system can save on overall test cost and goes a long way toward achieving the ultimate GMP goal of producing and releasing only high-quality products.
A reasonable mix of testing in all three of these aforementioned areas-ingredient testing, supplier qualification testing, and finished-product testing-is the key to GMP success. NSF auditors recommend the following:
Other areas in which companies struggle to comply with GMP regulations include developing master manufacturing records (MMRs) and batch records (BRs) that include all 14 requirements outlined the Code of Federal Regulations. MMRs should identify specifications for the control points, steps, or stages in the manufacturing process for which control is necessary to ensure the quality of the dietary supplement. This includes all processing steps, from production through packaging and labeling. Companies may see these as just a lot of paperwork, but without traceable, accurate records, product consistency can be a real challenge.
Finally, the ensured recording and review of customer complaints is also an area of GMPs. Serious adverse events are vastly underreported, according to the FDA. And recording customer complaints and adverse events is just the beginning; companies must then investigate the root cause of the deficiency and develop appropriate corrective actions to be implemented. Without this process, continuous improvement cannot be achieved.