The long-awaited change from USP to USP and has finally come. Though even upon implementation, there is still some ambiguity as to who should be responsible for testing, and which products should be tested: Should the final drug manufacturer have to test the finished product? Should excipient manufacturers provide this data to the drug manufacturers? The USP has published a few different approaches.

Control strategy. The aim of the control strategy is to track impurities that may contaminate pharmaceutical products where there is potential for contamination by multiple different sources. To ensure that all pharmaceutical product components and required production steps demonstrate regulatory compliance, performing a risk assessment is now a priority for every pharmaceutical manufacturer. The risk assessment approach to testing and documentation can become a major challenge, particularly when taking into consideration all the potential impurity sources. These include excipients, water, the active pharmaceutical ingredients (APIs), container systems and manufacturing processes. When the assessment process shows a potential risk, then additional data is required and testing for elemental impurities becomes the next challenge. 

Final drug product testing. The second option for implementation is to test the final drug product. This approach ensures that there are no elemental impurities in the final product that will be administered to patients. The analysis may be validated as a quantitative test or a limit test.

Drug product analysis option. The results obtained from the analysis of a typical dose can then be compared to the daily dose and the maximum permitted daily exposure (PDE). This is the most comprehensive way to guarantee that there are no elemental impurities in the final drug product, and is likely to be the least expensive option. However, if the final product is shown to have elemental impurities, it may be difficult to identify the source of the impurity. 

Summation option. The summation option proposes to add the amount of each elemental impurity present in each of the components of the drug product, and then compare the total of each impurity to the daily PDE limits. While this technique is costlier, it provides a higher level of control. Manufacturers can monitor individual components before they are introduced into the manufacturing process, and production can be stopped if an ingredient does not pass. To use the summation, approach, the manufacturer must ensure that additional elemental impurities are not added through the production process or the container-closure system of the product. If the impurities from each of the individual components added together are less than the specification, then the final product may not need to be tested. The validation for this analysis must be quantitative. However, there are other factors to consider when using the summation approach, such as: will the containers add any elemental impurities, or will the manufacturing add any possible contamination?

Risk assessment
The USP has created a table to assist with risk assessment of 24 elements to be considered. The elements are the 15 listed in USP , plus 9 others, and have been placed in classes based on their toxicity and likelihood of occurrence in the drug product: Class 1, 2A, 2B and C. The USP has also included a table (see Table 1) that should be considered in the risk assessment, based on the route of administration—oral, parenteral or inhalation. In this assessment, if an element is intentionally added during production, then the manufacturer must test for it. If the element is not added, then it may not need to be considered in the risk assessment. Arsenic, cadmium, lead and mercury will always need to be tested due to possible environmental contributions.

Assessing which elemental impurities must be tested for is an important step in keeping costs in check. When addressing the testing for elemental impurities, each element must be considered separately for its own unique risks and analytical challenges. There are problem elements, such as osmium. Osmium is a rare element and manufacturers must evaluate whether it has been intentionally added, or used as a catalyst in the process. If an elements is of interest, but has not been intentionally added to a product, depending on the type of drug and how it is administered, it may need to be tested for. One such element is copper, which only needs to be tested for in products that will go into, or are a parenteral drug, if it is not used in the process.

Each element poses its own set of potential issues for example, when testing for mercury, gold must be added, and so if it is also necessary to test for gold, it will not be possible to use the same sample preparation method. It is beneficial to use a hydrochloric acid matrix for some elements, although adding hydrochloric acid may cause interference with elements such as arsenic.

Sample preparation is key in validating these methods, consequently, multiple sample preparations may be needed to obtain a validated method for all 24 elements. It is important to consider if testing for all 24 elements is required from the start in order not to complicate the analysis unnecessarily.

If manufacturers can demonstrate compliance through process monitoring and supply-chain control, then further testing may not be needed.

Test methods
The USP describes two analytical procedures for the evaluation of elemental impurities; inductively coupled plasma optical emission spectrometry (ICP-OES) and inductively coupled plasma mass spectrometry (ICP-MS). The USP states that by means of a validation study the analyst will confirm that the analytical procedure described is suitable for the specified material. Unfortunately, that means that all sample types will need to have their own validation study performed. Many methods can be used to test multiple sample types, however, each one will need to be validated individually to determine that the method is appropriate.

It will be up to each manufacturer to determine which approach their company should be adopting. Many excipient and API manufacturers are beginning to test materials and will have the appropriate information to provide to the drug manufacturers. However, if this data is unavailable, or the process may contribute too many elemental impurities, then testing the final drug product may be the better option.

Time will tell how companies approach the new challenges brought about with USP. As analysts, the systematic and methodical training, and experience in testing give starting points for each individual drug product’s needs. The scenarios below demonstrate how requirements and challenges can differ, and how USP has guided the strategies adopted to overcome these.

Scenario 1. Finished drug product with no information gathering performed—water soluble. All 24 elements would have to be tested for with no additional information available initially. This is a water soluble drug, and so from the very beginning it is known that there will have to be at least two test methods, due to the fact that the presence of gold must be identified. As previously discussed, gold is also used to stabilize mercury, and therefore one method cannot be used to analyze all 24 elements. One set of standards is prepared in a dilute nitric acid matrix and spiking and sample preparation, while a second set is prepared in a dilute hydrochloric acid matrix. Not only does this process take a much longer time, but routine testing in the future will always require two methods as well.

Scenario 2. Finished drug product with no information gathering performed—not water soluble. Similar to the scenario above, all 24 elements would need to be tested for. However, not only do gold and mercury need two different sample preparations, there will be additional issues when microwaving and testing for osmium as nitric acid is needed in order to oxidize and digest the sample. Nitric acid will also create osmium trioxide, causing false high positive results for osmium. One way to counteract this effect is to add hydrochloric acid. Hydrochloric acid cannot be used when testing for arsenic, this has the potential to bind with the argon (used in the plasma on the analytical instrument) thus creating argon chloride which has the same mass as arsenic. Finding the balance between hydrochloric acid and nitric acid can be tricky, especially if other elements are involved.

Scenario 3. Finished oral product—with previous information gathering performed. Since this is an oral drug product with no additional catalyst, only the Class 1 and 2A elements would need to be analyzed—arsenic, cadmium, lead, mercury, vanadium, nickel, cobalt. All of these elements work well in a dilute nitric acid matrix, and the presence of gold in the solution does not interfere in this case. Therefore, only one sample preparation would be needed, even if the sample is not water soluble. 

References

1. United States Pharmacopeia and National Formulary (USP 40-NF 35). Rockville, MD: United States Pharmacopeial Convention; 2017. http://www.usp.org/sites/default/files/usp/document/our-work/chemical-medicines/key-issues/232-40-35-1s.pdf. Accessed January 31,2018.