Issue Date: September 27, 2010
Detecting Low-level Impurities
In the fairy tale, a supposed princess detects a single pea hidden under a pile of 20 featherbeds and 20 mattresses. Her remarkable pea-per-mattress sensitivity proves her royal identity. In the real world, analytical chemists are akin to suspicious royals in the tale, looking for highly sensitive measures to prove a truth.
Since 2007, guidelines from the European Medicines Agency and a draft version from the U.S. Food & Drug Administration have called on the drug industry to reduce DNA-reactive genotoxic impurities (GTIs) in pharmaceuticals. The EMA guidelines suggest only that GTIs should be detected and quantified by “state-of-the-art analytical techniques.” But the result has been a paradigm shift in pharmaceutical impurity profiling and control, says Fenghe Qiu, leader of the trace analysis group at drugmaker Boehringer Ingelheim.
Although the guidelines broadly affect process chemistry and drug development work (see page 16), the analytical science groups at pharmaceutical companies have been hit hard. Analytical chemistry methods must be brought to bear across all stages of development, from first identifying any potential GTIs through final product testing and, later, manufacturing. The challenge is finding the right technique for the task at each stage.
Regulators have set the limit, or threshold of toxicological concern (TTC), for each GTI at 1.5 μg per day. For an average daily dose, the impurity will be just a few parts per million in the drug product, or about 1,000 times lower than the fraction-of-a-percent levels allowed for general impurities.
This restriction can increase the sensitivity requirement for analytical methods by two to three orders of magnitude, Qiu says, and can push the limits of routine analytical techniques. At the same time, the number of impurities that must be assessed and controlled has grown dramatically.
“During development, the risk of potential or actual GTIs is assessed chemically, analytically, and toxicologically as soon as a route or representative route intended for the synthesis of clinical drug substance is available,” Qiu explains. This assessment is repeated whenever a change is made in the synthetic process, and again during late-stage development when clinical doses are considered for regulatory filings and practical analytical techniques are evaluated for the manufacturing setting.
FDA and EMA allow impurity levels to be higher in medicines in clinical trials because the drugs will be administered briefly. The level is adjusted depending on the duration of the trial. This staged TTC lessens some of the constraints on what otherwise would be a laborious and technically arduous analytical process, according to Larry Wigman, former associate director of analytical sciences at Sanofi-Aventis and now principal consultant at Pennsylvania-based Regulitics.
The staged TTC allows analytical method development to evolve along with a drug candidate as it turns into a marketed product. Drug developers want to avoid having to test the final product, where the impurity limits are lowest and the most sophisticated and expensive analytical techniques would be needed. Potentially genotoxic materials can arise at any point in drug synthesis, including among starting materials, reagents, intermediates, by-products, and impurities.
Many GTIs are intrinsically reactive, and process chemists often don’t expect them to persist through a multistep synthesis. By understanding the chemistry and the fate of a potential impurity during the process, drug developers hope to control GTIs in the final product. Paper arguments alone haven’t been enough to convince regulators, however, so the theory is backed up by an analytical assessment.
The reactivity of GTIs is one of the biggest challenges for analytical scientists, explains David K. Robbins, senior research scientist at Eli Lilly & Co. Although this instability is good for the ultimate demise of the impurity, it presents significant analytical headaches. Problems can arise in recovering impurities and in the reproducibility of the analysis. The sample matrix itself, often poorly soluble, can affect selectivity or sensitivity. And sample preparation can require extraction, deactivation, and derivatization.
“Significant resources have to be invested in identifying potential impurities that are likely to be carried over into the drug substance,” Qiu explains. “Chemical purifications and spiking studies have to be conducted to understand the fate of potential GTIs or to reduce the levels of such impurities. If genotoxicity is confirmed, purification and control have to be continued during late development stages and beyond.”
Impurity fate mapping, or spike-and-purge testing, to track impurity levels is usually done using quantitative analytical methods, explains Alireza S. Kord, director of analytical sciences at GlaxoSmithKline’s King of Prussia, Pa., R&D site. “If you can show exactly what the level was, it demonstrates the power of your purging much better,” he says. The data collected also serve as a reference if process changes are made.
In some circumstances, analytical scientists will have seen an impurity before, and the work will go quickly, Kord says. When that’s not the case, it’s good to have a planned approach. Kord and colleagues laid out their systematic approach for method development in a recent publication (Org. Process. Res. Dev. 2010, 14, 977).
Technology advances have significantly improved pharmaceutical companies’ ability to analyze trace-level impurities. The most sophisticated methods come into play in the early stages of clinical development.
“We can use liquid chromatography combined with mass spectrometry or with nuclear magnetic resonance spectrometry,” Kord says about detection in the R&D setting. LC/MS, LC/NMR, and other such combined technologies, often costing $1 million or more per instrument, are operated by trained scientists. “These methodologies are meaningless, however, for people in manufacturing,” Kord adds.
During production, quality-control analysis typically involves simple dissolve-and-shoot sample preparation in conjunction with high-performance LC and ultraviolet detection or with gas chromatography and flame ionization detection. Knowing that a GTI is effectively purged by the end of the process allows a specification limit to be set at the point where the impurity initially appears. Because the control level is relatively high at this upstream point, it alleviates the need for sophisticated testing. As a result, simpler methods can be used, often to get yes-or-no answers about whether specification limits are met.
Once a robust control strategy is developed and validated, it must be transferred to a company’s manufacturing operations or to an external contractor. “We always keep in mind the capabilities of our partners,” Lilly’s Robbins says. “While we are developing an analytical method, we will use the most sophisticated techniques we can to provide information quickly for our process chemists.
“But knowing, for example, that our third party doesn’t have MS/MS capability, we’ll evaluate that method on an instrument similar to what they have to demonstrate that it is feasible for them to use,” he says. The level of expertise and analytical capabilities will differ among contract manufacturing organizations.
Alternative analytical techniques for more sensitive and specific detection, along with more sophisticated sample handling, may be needed as part of the overall analytical strategy for controlling GTIs. “Although it may not be desirable, if there is no practicable alternative, techniques such as LC/MS can be and have been successfully transferred to manufacturing operations,” Qiu adds.
Because of the variety of pharmaceutical impurities, even the sophisticated techniques can have detection selectivity issues. To achieve a balance between efficiency and practicality, Boehringer Ingelheim has employed multiple approaches for compound-specific analysis, according to Qiu. Alternative techniques have included HPLC with charged aerosol, chemiluminescent, or fluorescence detection. These techniques can be used in conjunction with chemical derivatization and solid-/liquid-phase extraction to further enhance sensitivity or selectivity, he adds.
FDA officials have acknowledged in articles that the proposed limits on impurities are not without their corresponding analytical challenges. In many cases, multiple impurities may have to be controlled. When these are structurally similar and have the same toxicological mechanisms and potency, GTI guidelines indicate they are to be handled as a group and subject to a single TTC limit as though they were a single compound.
In practice, this could mean having to control the individual GTIs that make up the group at below part-per-million levels. This requirement can push the limits of analytical capabilities and create significant extra work for analytical chemists, Kord says. “To some degree, but not all the time, handling GTIs pushes current technology. So far it is manageable, but it is not always easy.”
- Chemical & Engineering News
- ISSN 0009-2347
- Copyright © American Chemical Society