Turning the Tide for Protein Formulation and Delivery

Protein formulation specialists have long sensed that something big could be just around the corner. Over the past few decades, countless companies have attempted to bring to market new protein therapeutics that offer improvements—be they more patient friendly, more effective, or easier to manufacture—over traditional formulations. Earlier this year, the launch of Pfizer's "Exubera" pulmonary insulin met this anticipation head on. The fast-acting, inhaled-powder form of recombinant human insulin brought hope to the millions of diabetic patients waiting for an alternative to injections.

Industry, too, gave Pfizer a lot of credit for being the first to market with a pulmonary insulin product. But at the same time, it recognized that many questions remain unanswered about potential long-term side effects, bioavailability, even its efficacy.

"How well is that product doing in a therapeutic sense and therefore commercial sense? I think we're going to need a little more time to know," one industry observer told Pharmaceutical Technology.

These concerns are not new to protein therapeutics specialists. Soluble proteins lack chemical stability in the body: they have very short half-lives and may degrade upon storage by aggregation, oxidation, or deamidation mechanisms. And, drug delivery options are limited. Barring a few exceptions, most proteins are injected.

Now, however, it seems that the tide may be turning. Drug delivery, which often is thought of as a product life-cycle management technique, is now being considered much earlier in the development process—especially for proteins. "We see many more dosage forms for which proteins are the active ingredients. It's not a huge market at this point, but compared with before, it's really a trend," says Jean-Marc Bovet, executive senior vice-president of Cirrus Pharmaceuticals (Research Triangle Park, NC, www.cirruspharm.com).

Of course, protein molecules can be stabilized and delivered in numerous ways and no single method will be appropriate for all applications. Several new formulation and delivery techniques have made strides along the development pathway.

By way of mouth

Fundamental challenges. Oral delivery is probably the most natural route to consider for drug delivery. But, digestion hinders researchers attempting to deliver proteins through the gastrointestinal (GI) tract. "The stomach is a bad place for proteins. From the standpoint of a therapeutic place through which you want to move proteins, it isn't effective," explains Bovet.

Several companies have investigated additives that improve absorption in the GI tract. Despite positive test results, says Michael J. Pikal, PhD, Pfizer distinguished endowed chair in pharmaceutical technology at the University of Connecticut's School of Pharmacy (Storrs, CT, www.uconn.edu), the sobering fact is that nothing has come to market. "I see the oral delivery of proteins as the magic bullet, but it's almost mission impossible," he says.

Pikal points out that in addition to formulation hurdles, several manufacturing science challenges may block the development of a viable oral dosage form for proteins from being developed—at least with today's technologies.

"Proteins are amorphous, very complex, and are very hydroscopic," he says. "At most manufacturing facilities, 'low humidity' means 20%. In fact, many sites operate at 40% or so. That's just way too high for this kind of dosage form." During normal manufacturing processes, the materials absorb large quantities of water, which results in protein degradation during distribution and storage. There may be ways around the humidity problem, but "they're really out of the box and move you into different kinds of manufacturing environments entirely," Pikal notes.

Other skepticism centers on repeatability and wasted administered proteins. In the case of oral insulin, glucose levels can be lowered for a short time, but "no report in the literature has shown the repeated administration of insulin resulting in effective control of the glucose level and has induced the insulin effect at the right time" (1).

In addition, "With most new delivery technologies, bioavailability is a main issue. You observe only 10–20% at the most," notes Byeong S. Chang, PhD, founder of Integrity Biosolution (Camarillo, CA, www.integritybio.com), a specialist in the formulation and delivery of proteins and peptides.

Carriers. But, if these challenges can be addressed, oral delivery might offer better options for dosing regimens. "The biotech industry, which has had tremendous growth within the past 10 years, has developed several protein therapeutics. If you look at the potential for technologies that can cause these products to be taken orally, you open up the possibility of using those therapeutics in such a more powerful way because you can administer therapies more frequently if need be. You can expect greater patient compliance and you can target the liver and the GI tract if necessary," says Lewis H. Bender, senior vice-president of business development at Emisphere Technologies, Inc. (Tarrytown, NY, www.emisphere.com).

Under development at Emisphere are functional excipients that serve as delivery agents (sometimes referred to as carriers) for large molecules such as proteins. A therapeutic drug is mixed with a delivery agent (a synthetic chemical or compound), which temporarily changes the shape of the macromolecule through noncovalent interactions and partially exposes a drug's more lipid-like core. The GI tract membrane (being lipid-based itself) absorbs the delivery agent drug complex and allows it to pass into the bloodstream. Then, the drug and the delivery agent separate, and the carrier is excreted from the body. Studies of this "eligen" technology "show reproducible delivery of the drug and that the drug maintains its full activity. You can't find any measurable changes in the drug," says Michael M. Goldberg, MD, chairman and chief executive officer of Emisphere.

Because the delivery agent is used as an excipient, nothing unusual is required to manufacture the oral formulations in tablets or capsules. "It's basically blending processes and making sure you have content uniformity. Standard unit operations can do it," explains Bender. Emisphere currently has several products undergoing clinical trials, including oral salmon calcitonin (nearing Phase III, with Novartis), oral insulin (Phase II, self-developed), and oral heparin (Phase III, self-developed).

Crystallization. For years, small-molecule drugs have been delivered orally in a crystalline state, which protects the active ingredient from being destroyed by the stomach's temperature and pH. Crystallization has not been used for oral protein delivery, however, because the manufacturing process is difficult to scale up and the protein would not be absorbed through the intestinal wall.

But according to Robert Gallotto, vice-president of strategic planning and alliance management at Altus Pharmaceuticals (Cambridge, MA, www.altus.com), new crystal technologies can be used to manage protein purity without significantly reducing yields (which increases the cost of goods). And, crystallization can increase protein concentrations. "You can fit billions of particles of proteins in a 10-μm crystal. You don't have interaction going back and forth which thereby improves stability and reduces aggregation," says Gallotto. "It confers far more delivery options, whether it's on the oral side or the injectable side."

A crystallization technology developed by Altus, according to Gallotto, produces highly stable, potent proteins that can be delivered orally in a capsule. The crystals help the proteins survive the harsh pH environment in the stomach and are effective without systemic absorption. In addition, molecules can be cross-linked in a crystal lattice for better stability and control over where the drug is delivered in the body. Applications include the delivery of enzymes that exert a therapeutic effect in the stomach.

The crystal production technique does not involve x-ray diffraction, which could help keep down the cost of goods. Says Gallotto, "You're not going to lose yields to some of the columns and purification steps that are sometimes used. You can use crystallization as a purification technique."

Tackling sensitive issues: inhalation routes

Because the lung does not destroy proteins, as the stomach does, many groups are looking into inhalation delivery techniques. Various organic and chemical excipients have been investigated to stabilize protein formulations delivered through the nose and the lungs . Unfortunately, these additives are problematic.

"Many stabilizers work great in a laboratory, but they're not something you'd want to inhale," says Edward T. Maggio, PhD, CEO of Aegis Therapeutics (San Diego, CA, www.aegisthera.com). "If you look at the literature, more than 100 mucosal adsorption enhancement agents have been studied, and they all have the same two limitations: either they don't increase bioavailability significantly or they cause irritation of mucosal tissue."

For example, says Maggio, every attempt to produce nasal insulin has failed, mostly because of unacceptable nasal irritation caused by large amounts of excipients and other factors. "The first tries were almost disastrous," he says.

Pulmonary delivery of proteins has had similar problems. "You're putting excipients and formulations into the lung that don't have a history of being in the lung," says Larry Brown, ScD, chief technology officer of Epic Therapeutics, a subsidiary of Baxter (Norwood, MA, www.baxter.com). "While a particular protein may not cause irritation, other ingredients in a formulation have to be evaluated carefully to ensure that they don't cause irritation."

Irritation to the lung is one challenge facing formulators. Regulatory approval is another. In fact, very few excipients are currently approved for lung delivery. "You basically end up doing formulations with very few molecules to play with. That's really the challenge," Bovet explains.

Massachusetts Institute of Technology (MIT, Cambridge, MA, www.mit.edu) Associate Professor of Chemical Engineering Bernhardt L. Trout, PhD, agrees. "Companies are looking for new ideas. The current space of additives is limited and they don't work very well," he says.

For good reason, many toxicity studies must be performed on new excipients—probably at a reasonably high price. The lung is a fragile organ and formulators must be cautious about what goes through it. Nonetheless, this cost and caution have put the brakes on the development of some new excipients and, in turn, some pulmonary protein products.

Says Bovet, "It's like making a concerto with just a few notes. You don't have the full orchestra to play with."

New additives and microspheres. But, some recent developments in nontoxic excipients may address these safety issues while also stabilizing proteins. For example, nonionic, nontoxic excipients under development by Aegis Therapeutics can be added to a protein solution during purification and concentration to reduce or eliminate aggregation. The excipients can be used not only in the final formulation, but also during the manufacturing process to prevent aggregation.

"If you're going to use the excipients in the final formulation, it's great to have it in production because you don't have to worry about taking anything out at a later stage," says Maggio. The group can stabilize formulations containing 0.1% excipients. Other formulations have needed 40% excipients to maintain stability.

According to Maggio, the excipients do not irritate the lungs and metabolize CO₂ in water. "The only atoms involved are carbon, oxygen, and hydrogen. They're very simple structures," he explains. The company has used its technology to stabilize proteins such as a human growth hormone currently being applied in transmucosal systems. "With growth hormones, we can do nasal. For pediatrics, that would be a dramatic improvement," notes Maggio. The excipients also have been used in a room-temperature-stable injectable insulin currently under development.

Work on lyophilized, water-based microspheres by Epic Therapeutics and Baxter also could help address these issues. "Promaxx" microspheres are made out of the protein itself and thus require little (if any) excipients (see Figure 1). "We take a water-soluble polymer such as polyethylene glycol, and we have developed a method whereby in an aqueous solution, we're able to precipitate the protein into 1 to 3-μm microspheres," says Brown. "In the case of pulmonary insulin, even though we don't have any of these classic protein stabilizers in them, we've shown that our microsphere process can significantly reduce the amount of chemical degradation of the protein." According to Brown, the drug loadings in the Promaxx microspheres are usually more than 90%.

Figure 1: Promaxx inhalable molecules. (Baxter Healthcare Corp./Epic Therapeutics)

Although the company precipitates the microspheres using water-soluble polymers, the polymers are washed away, leaving undetectable amounts of the residue. "Because we add no excipients or substances that the lung has never seen before, we believe this may reduce the risk of side effects," explains Brown.

Getting under the skin: parenterals

Monoclonals. Technologies such as Baxter's Promaxx microsphere technology as well as Altus's crystallization technology also can be used for highly concentrated injectable products, which is important from a biologics perspective.

With the boom of the biotech industry, the monoclonal antibodies market has grown at a rapid pace. Sales approached $15 billion in 2005 and the market isn't stopping there (2). By 2010, market analysts predict that the worldwide market for monoclonals will be $26 billion (3).

"The monoclonal antibodies and antibody-related structures have provided much more scope for biologics beyond the native proteins and the soluble receptor domains," says Susan Hershenson, PhD, vice-president of pharmaceuticals at Amgen Inc. (Thousand Oaks, CA, www.amgen.com). "Industry is still exploring all the aspects of understanding those structures—how to engineer them for the best performance and it is really developing a very broad understanding of formulation and the delivery potential of those structures."

Monoclonal antibodies are engineered in the laboratory to recognize and target specific antigens or cells. Potential therapeutic applications for the proteins could include the delivery of drugs or toxins to treat cancer, rheumatoid arthritis, or autoimmune and infectious diseases. Though they have several advantages, the doses are relatively high.

"If you're going to deliver a large dose, and if you're going to do so parenterally, you must have a concentrated solution. A large dose and low concentration mean large volume, which is very painful," explains Pikal. "Patient compliance is not going to be good. It's not going to sell."

"I think it's something everyone is grappling with," says Hershenson. "How high can we push the concentrations?"

As industry observers point out, the only larger protein that was ever approved as an injectable was the "Nutropin Depot" human growth hormone from Genentech. The product, launched in 2000, did not achieve great commercial success. Many attribute this to the large, 21-gauge needle required to deliver the formulation, which was unpopular with patients.

Higher concentrations also increase the chance of degradation and can lead to processing problems. For example, the highly viscous formulations are difficult to sterile filter without clogging standard membranes.

According to Brown, it is possible to use Promaxx microspheres in an aqueous suspension for injection through a small-bore needle. "We are able to inject 200–300 mg per mL and inject them through a needle that is much smaller than the needle used to inject classic microspheres," he says. The company has collected stability data on monoclonal formulations that show good stability without large amounts of stabilizing excipients.

The Altus crystallization technology also can be applied to parenterals, especially more stable, concentrated formulations such as for monoclonal antibodies. "We stabilize formulations so they can be delivered subcutaneously with a 29 or 30-gauge needle with one injection versus delivery with transfusion or through multiple injections or having issues with viscosity. We can increase the concentration to 400 mg per mL without a significant increase in viscosity," says Gallotto. And, Altus can improve stability at room temperature. The company has a growth hormone currently in clinical trial that can be left without refrigeration for 2–4 weeks.

Neutral crowders. Another option could be using additives larger than the primary solvent to hinder protein aggregation by significantly increasing the free energy barrier for protein–protein interactions.

According to the "Gap Effect Theory" developed by MIT researchers, as proteins associate, molecules between them must be removed from their interface. Additives in the gap between the proteins must be removed, and if they are larger than water, their removal creates a free-energy penalty caused by osmotic stress. This free energy penalty leads to slower rates of association, and hence slower aggregation (4, see Figure 2).

Figure 2: "Gap Effect Theory." (a) Protein's (P) channels are too small for large additives (black dots) to enter. The channels collapse and water (grey dots) is released. (b) In protein–protein interactions, the gap effect slows isomerization and aggregation. (Source: reference 4)

"We're selectively increasing the barrier of native proteins to aggregate," says Trout.

The group is working to synthesize and test potential neutral crowders, which could be used for many proteins in high-concentration formulations for injection. Trout's group also is developing quantitative models of degradation (4–6). "If we can quantify it, we can rationally manipulate it to stabilize it," he says.

Analytical needs

Demonstrating that two biopharmaceutical products are identical is difficult with current analytical technology. "We want to develop protein products in an absolutely pure form, and for that, we need good analytical methods to support them," notes Chang of Integrity Biosolution. "If you don't have a good analytical method, you can't even detect degradation."

Analytical tools are rapidly evolving to help address this need, including protein mass spectrometry, analytical ultracentrifugation, and capillary electrophoresis, which offer better ways of looking at the chemical and physical structures of these biomolecules. Industry also may be on the verge of seeing greater applications of techniques not used in today's formulations.

"These techniques are very critical," Amgen's Hershenson points out. "The science that underlies the regulatory issues is the challenge of being able to characterize these very large, complex, and fragile molecules for which seemingly slight changes can lead to big differences in performance."

Though industry does not yet have the tools it needs to fully characterize these complex molecules, it's clear we're headed in the right direction. Says Hershenson, "There are definitely improvements with many analytical technologies and I think they are, in many ways, as critical as other technical developments."

References

1. J.G. Rocca and K. Park, "Oral Drug Delivery: Prospects and Challenges," Drug Deliv. Technol. 4 (4), (2004).

2. Business Insights, The Future of Monoclonal Antibodies Therapeutics: Innovation In Antibody Engineering, Key Growth Strategies and Forecasts to 2011 (Business Insights, London, UK).

3. "Worldwide Antibody Market to Reach $26 Billion," Drug Researcher Feb. 6, 2005, http://www.drugresearcher.com/news/ng.asp?n=60408-worldwide-antibody-market, accessed Sept. 13, 2006.

4. B.M. Trout and B.L. Trout, "Rational Design of Solution Additives for the Prevention of Protein Aggregation," Biophys. J. 87 , 1631–1639 (2004).

5. B.M. Baynes and B.L.Trout, "Proteins in Mixed Solvents: A Molecular-Level Perspective," J. Phys. Chem. B, 107 , 14058 (2003).

6. B.M. Baynes, D.I.C. Wang, and B.L. Trout, "The Role of Arginine in the Stabilization of Proteins Against Aggregation," Biochem. 44 , 4919 (2005).