Send to printer »

Feature Articles : Sep 15, 2009 (Vol. 29, No. 16)

Strategies for Successful Formulation

Tips on Streamlining Preformulation Projects, Particle Characterization, and Protein Aggregation
  • Angelo DePalma, Ph.D.

Next to the active ingredient, formulation is arguably the most critical component of a biopharmaceutical product. Formulation scientists from leading biopharmaceutical companies, vendor firms, and universities will be discussing challenges in the field and sharing tips on overcoming those obstacles at the IBC conference “Formulation Strategies for Protein Therapeutics” to be held in Raleigh, NC, next month.

Dosage form elements, especially formulations, can evolve as products undergo clinical testing. Angela Kantor, principal research scientist at Wyeth, uses a Cheshire Cat analogy. When Alice asked the Cat which way she should go, the feline hedged: “Well that depends a good deal on where you want to get to.”

The same applies to formulation strategies, Kantor says. “It’s all about decisions. Formulation strategies have to be based on goals, on destinations. Not every biological belongs in, say, a needle-free device for home administration. Along the way, you have to make sure you’re doing the right thing for the patient and the product. You need a strategy for arriving at a good product, and how you get there matters.”

A life cycle focused formulation strategy will better position the product for a competitive future and define, at every point on the development continuum, where the product is going.

One critical period along that continuum is clinical testing. In 2006, Wyeth adopted the Learn & Confirm clinical development paradigm, a flexible approach that differs from conventional Phase I–III trials. The Learn phase is geared toward understanding a drug early in clinical development, before committing resources to a confirmatory Phase III trial.

To achieve this, testing is initiated in representative patient populations rather than in normal human subjects. This adaptive trial design permits an earlier assessment of efficacy. “While doing safety and tolerability studies, you’re also getting an earlier read on efficacy, so when you enter late-stage trials you know more than you would have after conventional Phase I/II studies.”

Learn & Confirm has the potential to compress clinical development and guide the life-cycle formulation strategy, since it provides earlier insights into maximizing Wyeth’s five points in the product life cycle: safety, efficacy, tolerability, convenience, and affordability. “These are our destinations,” Kantor notes, “and as formulators, our goal is to maximize these for the patient while delivering a product that maintains value throughout its life cycle.”

Because drug development is risky business, the formulation strategy must be based on clinical and marketing knowledge gained during development, and the earlier this occurs, the better. “That’s where the Learn & Confirm paradigm could give us an edge,” Kantor notes. “Making too much of an investment before knowing if a drug will be successful diverts resources that could have been used to develop other promising candidates. The risk is investing too much, too soon.”

For example, manufacturers often prepare biologicals in lyophilized form because that is the easiest way to get them into Phase I although it may not be the ideal market formulation. A liquid may be more attractive because of convenience or cost, and convenience is preeminent; developers might even consider an auto-injector or needle-free device.

“As you learn more about a drug, reformulation is a life-cycle change that may be desirable to achieve better storage, stability, or performance—things you always want to improve.”

Particularly for biologicals, life-cycle considerations should focus on making parenteral administration more palatable. “Nobody likes injections. But if we can make them more convenient, less frightening, administer them less frequently, or make them less expensive, then we’ve made a better drug.”

Earlier Is Better

Applying analytical methods early on in development can help scientists anticipate and mitigate formulation problems that arise later on. This is the key message of Robert Simler, Ph.D., staff scientist at Genzyme. Dr. Simler applies conventional methods like differential scanning calorimetry, light-scattering, size-exclusion chromatography, and analytical ultracentrifugation, as well as spectrometric techniques like circular dichroism, fluorescence, and FTIR to preformulation projects with the idea of establishing methods as early as possible.
The techniques themselves are not extraordinary, but the point at which they are applied and worked out is. Once the methods are in place, Genzyme scientists can draw on them for quality control and stability studies, during clinical development and beyond. “We use those techniques, early on, to identify potential problems, or to get better characterization at the front end of development rather than waiting and reacting to what happens down the road.”

Obtaining formulation information early requires communication between groups to determine the most appropriate techniques. The two principal hurdles at this stage are material availability and time lines. Preclinical-stage biologicals are in short supply because production batches are small while demands for animal testing are substantial. “You can’t do everything,” Dr. Simler observes. “Animal testing chews up a lot of material, as do analytics.” The time constraint, particularly getting through preclinical studies as rapidly as possible, is common to all pharmaceutical development, but especially acute for biopharma.

“But when the analytical work is successful, and you’ve identified a protein’s main stability issues, and you have the techniques lined up, you have powerful tools to move forward with, as the molecule is being developed,” Dr. Simler adds.

One technical hurdle to overcome, he says, is particulate testing. “There’s a mandate from FDA to look closely at this.” Since every biomolecule is unique, even within a class, a one-size-fits-all approach works as poorly for particulates as it does for formulations. Genzyme has been experimenting with instrumentation to characterize particulates, and works closely with vendors to improve and fine-tune those products.

Characterizing Particles

One of those vendors is Brightwell Technologies, which will also be present at the IBC meeting. Deepak Sharma, Ph.D., senior research scientist, will host a workshop that describes how Micro-Flow Imaging™ (MFI) applies to biopharmaceutical formulations.

Particle characterization is a standard method for measuring the stability of biotech products, particularly for protein aggregates and other species indicative of degradation, oxidation, or other undesirable chemical events.

Conventional particle-analyzing instrumentation works best with opaque, spherical particles, but protein-based particles are highly irregular in shape, transparent to light, and usually unstable. For example, manual light microscopy and light obscuration, the default techniques for analyzing particulates in parenteral drug formulations, both have drawbacks that are becoming obvious as more biotech products hit the market.

Manual methods are tedious, error-prone, and slow. Moreover, they consistently undercount protein-based particles that are almost always transparent and gel-like. Developed in the 1970s, light obscuration (LO) was designed to detect extrinsic particulates (e.g., glass, fibers, metal) larger than about 10 microns, in drug formulations. “Since LO was not designed to study intrinsic particles originating in bioformulations, its performance for these particle types is limited,” explains Dr. Sharma. Scattering and obscuration do poorly in sizing nonspherical or optically transparent species. When particles are detected, the information gleaned from them is limited.

MFI, which was developed by Brightwell, and is currently under investigation at Wyeth, analyzes images of particles captured in succession as a sample stream passes through a flow cell. Size, shape, and intensity of each individual particle are measured, as well as the count and concentration of the entire particle population. MFI can also operate in a time-resolved mode to monitor dynamic processes.

Maintaining a correlation between each particle’s morphological parameters and its image, aids in the classification of unique particle species, which may be isolated and characterized through software. MFI’s strength is monitoring protein aggregation or precipitation during formulation development, stability testing, production processes, and final-lot validation. It also detects and characterizes extrinsic particles such as air bubbles, silicone oil droplets, and foreign particles like glass, rubber, or metal.

“The technology is already widely employed as an orthogonal technique in the formulation development phase of drug discovery by big pharmaceutical companies,” notes Dr. Sharma. “FDA has also started showing interest in MFI as an orthogonal technique.”

Protein aggregation is a fundamental problem with implications in diseases, as well as biopharmaceutical formulations. One hallmark of Alzheimer’s disease, for example, is aggregation of amyloid plaques. Jennifer Laurence, Ph.D., professor of pharmaceutical chemistry at the University of Kansas, is planning to discuss her group’s investigations into factors and conditions that affect stability and aggregation, including mechanisms implicit in the formation of amorphous protein aggregates. “It’s an area where not a lot is known at the molecular level,” Dr. Laurence says.

The Disease-Aggregate Connection

Dr. Laurence combines higher-resolution analytic techniques like solution NMR (nuclear magnetic resonance), combined with molecular engineering of proteins, to assess which regions of the protein are responsible for amorphous aggregation, and under what conditions. Low-resolution methods such as circular dichroism, Fourier-transfer infrared spectroscopy, and light scattering provide information on the system as a whole rather than specific regions.

Solution NMR detects changes in chemical environment for resonant nuclei (typically uniformly labeled 13C- and 15N-enriched proteins) in response to changes such as heat or a rise in pH. NMR allows investigators to perturb the protein, observe which regions are most affected, and then do the same with a protein in which amino acids have been swapped out through protein engineering. The changes are then correlated with aggregation behavior.

Parallel experimentation has become a popular method for numerous process- and development-related activities. Byeong Chang, Ph.D., CSO at Symyx Technologies, will talk about microscale, parallel techniques for formulation optimization. The Symyx core formulation module tests hundreds of liquid formulations automatically, including active ingredients, buffers, and excipients. The module has components for excipient compatibility, forced degradation, stability, and solid and liquid formulations.