Scrutinizing the Subvisible

April 2, 2011
Pharmaceutical Technology

Volume 35, Issue 4

Regulators question whether particles that they can't see hurt patients.

Protein aggregates in biological drugs have the potential to trigger an immune response in the patient, which, at the very least, can decrease the drug's effectiveness, and at worst could cause potentially harmful side effects. Part of the industry's efforts to limit the presence of aggregates in their therapies entails testing for subvisible particles, which potentially are the most immunogenic aggregates. Subvisible protein particles are relatively large assemblies that contain anywhere from thousands to millions of protein molecules. US Pharmacopeia <788> "Particulate Matter in Injections" limits the number of particles equal to or larger than 10 µm that are allowed per container of drug (1).

The disregarded particles

USP does not address particles smaller than 10 µm in parenteral drugs, however. This omission drew little comment until it was mentioned in a 2009 article by John F. Carpenter, associate professor of pharmaceutical sciences at the University of Colorado Health Sciences Center. "If only particles > 10 µm were quantified in a given product, there could be gaps in understanding of important degradation products and in product-quality assessment," wrote Carpenter (2).

Carpenter and his coauthors, including eight officials at FDA's Center for Drug Evaluation and Research, raised the possibility that particles smaller than 10 µm could affect the safety and efficacy of therapeutic protein products over their shelf lives. Manufacturing operations sometimes create hundreds of thousands of particles 1.5–3 µm in size, the authors noted, and protein particles can accumulate over time during storage of the final product. Yet recommendations for detecting such particles are lacking.

Carpenter called for industry and academia to define current particle-counting instruments' capabilities to observe particles as small as 0.1 µm, recognizing the potential need for new instruments. The effect of protein aggregates on immunogenicity also should be examined, including "studies of the role of protein class, amount of aggregate, size of aggregates, and protein conformation in aggregates," wrote Carpenter (2).

The regulators' response

Carpenter's article attracted the attention of drug companies and regulatory bodies around the world. To gather information about the topic, USP held a discussion about particle characteristics and their effect on liquid and aerosol products during their workshop on particle detection and measurement on Dec. 8–10, 2010.

Participants spoke about sampling techniques and methods for data expression and interpretation, according to Scott Aldrich, principal consultant for Ultramikro and member of the 2010–2015 USP Dosage Forms expert committee. Much of the workshop discussion focused on the difficulties in adequately measuring the sizes and concentrations of particles in the sub-10-µm population for biotherapeutic formulations.

USP is assembling an expert panel to determine whether to establish a new particle-limits chapter for biotherapeutic pharmaceutical injections. "We anticipate any new chapter to provide methods tailored to the sensitivities of these formulations, with options for methodologies and a discussion of typical effects upon particle size, yet with no plan for limits specific to the biomolecular formulation," says Aldrich.

FDA has not published regulations about particles smaller than 10 µm, partly because so few data are available. But the agency has asked firms to start assessing the background and control of subvisible particles in that size range, which previously had been ignored. FDA is interested in determining whether particles smaller than 10 µm correlate with adverse reactions reported through USP's or FDA's medical-awareness systems, says Cherris. The data also could help the agency decide whether to issue a guidance.

FDA primarily is approaching biopharmaceutical firms because they must monitor their products for intrinsic protein-based particles and control the extrinsic particles that might infiltrate their process, according to Roy Cherris, managing partner of Bridge Associates International. Large biopharmaceutical companies are highly interested in trying to determine whether particles smaller than 10 �m are a major cause for concern, says Cherris.

The particles' origins

Various types of particle smaller than 10 µm can be present in a parenteral drug, but regulators' primary concern is with proteinaceous particles because they are more likely to cause immunogenicity than extrinsic particles. Denaturation can lead to aggregation that causes the native monomeric therapeutic protein to form dimers, trimers, or polymers. During development, proper consideration of the formulation, the way the material is suspended in the formulation, and the formulation's dynamics over time can help prevent intrinsic particles from forming or agglomerating.

Stresses during product manufacture (e.g., freezing, thawing, agitation, and foaming) also could lead to the formation of aggregates. These concerns, too, can be addressed during formulation and process development, and the industry already knows a good deal about these concerns. "The risk for aggregate and potential proteinaceous particle formation is really product dependent. Certain molecules are more sensitive than others," says Satish K. Singh, research fellow at Pfizer. "It is imperative that the formulation-development scientist understand the weaknesses of the molecule and the impact a drug-product manufacturing process can have on it."

The most common type of extrinsic particle that is present in this size range is silicone oil, which manufacturers use to coat cartridges, syringes, and stoppers. Particles of glass and stainless steel, common materials in pharmaceutical manufacturing, conceivably could migrate into parenteral products, too. Piston fillers and filters reportedly have shed particles into parenteral products, and it is possible for hair or fibers shed from wipes to get into the product, too.

Although the potential for extrinsic contamination always exists, industry professionals believe that current controls for preventing it are adequate. "In aseptic processing, liquids pass through one or more 0.22-µm filter media. In our current environment, control of visible to subvisible particle content is believed to control the sub-10 µm content," says Aldrich.

The industry's perspective

Pharmaceutical manufacturers generally view FDA's concerns about these subvisible particles as legitimate. Scientists suspect that proteinaceous particles smaller than 10 µm may have caused certain reported immunogenicity issues. In addition, glass and stainless-steel particles in that size range have spurred several recent product recalls, says Cherris.

Yet drugmakers are not necessarily convinced that the particles are always hazardous. Companies agree with regulators about the need to monitor protein-based particles smaller than 10 µm because of their potential immunogenicity. But human clinical data on the connection between protein aggregates and immunogenicity are equivocal because a multitude of factors determines a patient's immune reaction, and it is difficult to identify the effect of a specific quality parameter, says Singh. "Animal-model data indicate that immunogenicity can be triggered by protein aggregates or particles, but it is also clear that not all aggregates or particles cause immunogenicity," he says.

Most of the standard, small-volume parenterals probably are not a cause for concern, says Cherris. Many of them either are not proteins or are proteins that do not cause problems. For example, vaccines generally have not been indicated in any of the reported immunogenicity problems, he adds.

Whether these subvisible particles can affect drugs' efficacy also is a matter of debate. "These particles cannot decrease the effectiveness of protein drugs directly because they often represent only nanograms or micrograms of protein" and would not measurably decrease the amount of available drug, says Singh. If the particles cause an immune reaction, however, the resulting antibodies can either have no effect, have an indirect effect through pharmacokinetics or pharmacodynamics, or directly negate the drugs' efficacy. "The worst-case scenario is if these antibodies are neutralizing against a nonredundant endogenous protein also, in which case, the antibodies become a safety risk," says Singh.

The nature of the subvisible particle may determine whether it causes adverse reactions. Not every biological product necessarily is associated with immunogenicity, says Cherris. "I would say that there are only subtle indicators of that [problem] in specific product cases, many of which in the biotech world are still in development," he adds.

Others agree that the risk of immunogenicity appears so far to be associated with certain products. "Specific characteristics in the aggregates, probably related to the structure of the protein in the aggregate and the nature of the exposed epitopes, cause some aggregates to be immunogenic whereas others are not, says Singh. Likewise, the risk that particles smaller than 10 µm will decrease a drug's efficacy also is likely to be limited to specific products or formulations.

Some drugmakers point out that no injectable solution is required to be completely free of subvisible particles, which have been present in marketed products for some time. Nevertheless, the link between immunogenicity and particles smaller than 10 µm has not been studied thoroughly, and our understanding of what these particles do is limited.

"I think it would be in everyone's best interest for people to develop an understanding of subvisible particles, their generation, and analysis, so as to respond if we determine that they present any type of risk," says Morrey Atkinson, chief scientific officer and vice-president of research and development at Cook Pharmica. Companies can begin by generating the data during product development, while the drug is still in the clinic. If these subvisible particles are stable over time, they are not likely to pose a risk.

The appropriate methods

Compendial methods. Measuring and analyzing particles smaller than 10 µm is challenging for quality-control personnel because the standard tests on parenteral products don't examine particles in this size range. But the required technology is relatively straightforward and readily available, according to industry sources.

The preferred method for the first round of testing is light obscuration, which is described in USP <788> (1). Light obscuration detects particle sizes between 2 µm and hundreds of micrometers. Companies commonly calibrate the instruments to 2 µm, yet most firms outside the biopharmaceutical industry generally don't examine the region of particles smaller than 10 µm, says Aldrich.

The technique has known limitations, however. Air and immiscible oils in a product could lead to an artificially high particle count using light obscuration. And the method sometimes fails to properly size nonspherical particles and those with a refractive index close to that of the liquid formulation, thus resulting in inaccurate and low counts.

Scientists can use the membrane-microscopic technique as an alternate or secondary method. This method is a more direct revelation of particle content than light obscuration is because it isolates solids or semisolids from the sample liquid onto microporous media and counts particles in size thresholds. Membrane microscopy has a wide detection range (i.e., from 5 µm to many millimeters).

The technique also has shortcomings, though. "Because it is [an] optical-microscopy evaluation at 100 × [magnification] with reflected illumination, particles below 10 µm are difficult to resolve on the porous membrane surface," thus making accurate counting difficult, says Aldrich. Inherent proteinaceous particles also may be difficult for the microscopist to count or size. Air and immiscible oils are not problematic, however, because particles are observed in a dry state and may appear as stains.

In addition, the membrane microscopic method is tedious, and no company will resort to it "unless they have a real need to, or if they had to use that methodology to verify the counts they received during light-obscuration counting," says Cherris. Some equipment automatically counts and identifies particles on a filter surface, but it is much more expensive than light-obscuration particle counters are.

The most economical way to test for particles smaller than 10 µm is to use automated particle counting first, then to use microscopic and spectroscopic techniques to identify the particles on a membrane surface, says Cherris. "Most companies that have recently purchased new light-obscuration particle counters have the sensor technology now to start counting down in that range."

Noncompendial methods. Many biopharmaceutical companies use alternate, noncompendial methods to understand the subvisible particle population. These techniques are reliable, but have a measurement gap from 0.1 to 1 µm, says Aldrich.

One such approach is the Coulter principle, which uses a dilution of a product in conducting liquids to measure particles between 0.4 and 50 µm, depending on instrument setup. Formulation buffering and preparation affect the measurements, but the technique measures particles and aggregates in a solution to reveal a relevant particle–product state.

The flow-imaging approach is an extension of optical microscopy that uses cameras to detect and record in situ particles in product fluid. The method detects particles from 0.7 µm to 1 mm, depending on their shape. This technique's ability to capture particle images for subsequent analysis allows the evaluation of the particle population under various shape and size contexts, says Aldrich.

Investigating particles in this size range is important and requires more than one method. Monitoring particles smaller than 10 µm "can provide valuable information on changes in product stability, protein aggregation, and product attributes that affect patient safety," says Dan Berdovich, manager of quality assurance and regulatory affairs at Micro Measurement Laboratories.

The next steps

Certain drugmakers are collecting information about particles smaller than 10 µm at FDA's request. Although its goal is sound, FDA must agree with industry on the most appropriate standardized testing methodology, according to Cherris. "If we don't collect data in the 2–10-µm range in a comparable way, it will be a long road, and probably the wrong road to follow in understanding these particle populations," he says.

The industry, through large nonprofit organizations focused on bettering pharmaceutical quality, should sponsor this data-collection program, not the government, according to Cherris. The Parenteral Drug Association or the US Pharmacopeia should create a scientific committee to collect and study the data, and possibly draft a new standard, he says. "Without that type of industry rally in an organized manner, the requests from FDA to collect this information will not be widely effective."

References

1. USP 23–NF 18 (US Pharmacopeial Convention, Rockville, MD, 1995), pp. 1813–1819

2. J.F. Carpenter et al., J. Pharm. Sci. 98 (4), 1201–1205 (2009).