Send to printer »

Feature Articles : Sep 15, 2011 (Vol. 31, No. 16)

Timing Is Everything in Protein Formulation

Choosing When and How to Proceed Depends on Available Resources and Competition
  • Angelo DePalma, Ph.D.

Formulation can make or break a development-stage protein drug, and add years of IP protection and value for post-patent molecules. Yet significant disagreement exists on the best time to undertake the formulation investment.

Robert E. Zoubek, Ph.D., head of protein characterization and preformulation at Formycon, a business unit of Scil Technology, advises customers to complete final formulation before Phase I.

He recognizes the dilemma of investing in formulation development before knowing if the drug will even make it to Phase II. “However, the reason many monoclonal antibodies fail is specifically because they are improperly formulated in the first place.”

Moreover, the moving-target approach to formulation, whereby tweaks are introduced over time, can be risky from a regulatory perspective. “Regulators have been known to ask for more tests on suboptimal formulations,” Dr. Zoubek notes.

“I know of one company that was asked to redo preclinical studies just as they were preparing to enter Phase II. Regulators wanted proof that the product as it was formulated at the time would not harm pregnant animals.”

He also cites ethical reasons for formulating from the start. It’s not right to subject Phase I subjects to a substandard formulation, he says, “just to save money.”

Dr. Zoubek believes the time for considering reformulation as a lifecycle-extending strategy depends on the therapeutic area and level of competition. Suboptimal formulations put companies at a competitive disadvantage, so it is never too early to consider reformulation.

However, all things being equal, companies need to factor in time-to-market, including any human studies that will be required. One and a half years before patent expiration is not too early, he says.

Sponsors should add time, possibly more than a year, when lifecycle extension includes more exotic, specialty-type extenders including nanoparticles, liposomes, and bioresorbable ceramics. These formulations will require more testing and more sophisticated analytics than straight-up liquid formulations. The same is true for lyophilization.

Since formulation development typically occurs during early clinical stages (or, optimally, even earlier)—when sponsors are scrambling to manufacture product for testing—material limitations can pose serious challenges. This is particularly true for products that will be formulated at high concentration.

High Throughput or "Rational"?

From his perspective as principal scientist at Cook Pharmica, Zachary Yim, Ph.D., is seeing more combination products than ever before, such as Fc fragments coupled with enzymes, small molecules, or peptides linked to proteins. “Formulation of these products is complex and multifaceted because you have to deal with both components.”

As a CMO/CRO, Cook sees formulation-development strategies of every type, duration, budget, and timeline. Dr. Yim says that formulation is often defined by when toxicology studies or clinical trial manufacturing must be completed. Cash-strapped sponsors do the best they can to meet minimal stability requirements.

“On the other hand,” says Dr. Yim, “those with more resources can think longer-term, because ideally you don’t want to change the formulation unless you have to.”

Whatever the strategy, many formulation developers today employ some type of design of experiment and high-throughput method to arrive at formulation candidates and, eventually, to select among and optimize these choices. Yet high-throughput methods can still take considerable time and always generate hundreds or thousands of formulations.

Dr. Yim suggests taking a tiered or stepwise approach that might begin with testing critical factors like pH or ionic strength, identifying optimized values and trouble spots, then tackling other conditions. “Doing a pH screen at the beginning could save a considerable amount of work.”

Then there is the issue of breadth vs. depth. When evaluating a very large number of conditions at once through high-throughput methods, it’s difficult to devote as much time per condition as, say, through a more rational or stepwise approach to formulation development.

“Detailed observation is often lost with robotics,” Dr. Yim explains. “Something can occur transiently and you miss it. It becomes a tradeoff. In addition, since all parameters are linked, you cannot change just one parameter at a time.”

High-Concentration Proteins

The majority of protein therapeutics are formulated for injectable or infusible applications, explains Vadim Klyushnichenko, Ph.D., vp for preclinical services and process development at Paragon Bioservices.

“Since alternative delivery methods like transdermal, oral, and inhalable are limited by protein size, stability, and bioavailability, and many biotech drugs are administered at high dose, formulations must accommodate very high drug concentrations.”

Formulation developers need to navigate the thin line between concentration and aggregation or precipitation, both of which are common with monoclonal antibodies. High-dose therapeutic proteins can also cause side effects, for example infusion reactions.

Paragon originally focused on development and manufacturing for therapeutic proteins, monoclonal antibodies, viruses, and virus-like particles. Formulation development was part of these services.

When their clients began asking for full biologics development, including formulation development of drug substance and drug product, short- and long-term stability studies, and fill/finish, the company doubled the size of its facilities to more than 50,000 square feet. The new space, which includes new formulation labs, stability chamber area, and cGMP fill/finish, is expected to come online next month.

From Paragon’s perspective, approaches for achieving stable, high, and effective concentrations for protein drugs lie in sustained-release technologies such as micro- and nanoparticles, liposomal complexes, or biodegradable matrices for implantable depot delivery.

“Development of such systems for targeted delivery of therapeutic proteins is the goal of formulation groups across the biopharmaceutical industry,” Dr. Klyushnichenko adds.

Formulation development should be initiated as early as possible, he says, if only through preliminary preformulation studies or protein characterization, analytical development, degradation studies, short-term stability studies, and development of a preliminary drug substance formulation.

According to this model, developers will investigate several “preformulations” for each potential formulation type and delivery mechanism. Based on data derived from actual formulation, long-term stability studies, and preclinical data, the number of candidate formulations is reduced by the time the product enters Phase I studies.

But Dr. Klyushnichenko warns companies to avoid a common catch-22. “If you do not know the properties of the protein, you cannot define the formulation and calculate accurately the quantity of drug substance required for preclinical or early clinical trials.”

This will lead to a “simple and reliable” formulation for Phase I, and perhaps Phase II as well, he says. The formulation changes subsequently as data from human studies become available.

Formulation development continues through the entire development period and, in many cases, throughout the drug’s entire lifecycle. Ideally the same protein could be formulated to cover several drug forms, administration routes, or indications. Each additional formulation provides extra IP protection and market coverage.

“There is always room for improvement in terms of additional patient safety and stable revenue for the company,” Dr. Klyushnichenko adds.

Crystalline Work-Around

Althea’s protein crystal technology supports the switch from intravenous to highly concentrated subcutaneous therapeutic protein delivery, and with it self-administration, long-acting formulation, and lifecycle extension. Protein crystals are related to but should not be confused with CLECs (crosslinked enzyme crystals), which Althea inherited through its acquisition of Altus Biologics.

Protein crystal formulations are delivered as suspensions of 5–10 micron particles, not solutions. When formulated as suspended solids, proteins are far less prone to stability and aggregation issues. Since the drug is significantly less viscous than highly concentrated protein solutions, it may be delivered through a fine-gauge needle.

For example, it takes less than one minute to load a 1 mL syringe with crystalline infliximab (Remicade) at a nominal concentration of 200 mg/mL. By comparison, a 1 mL syringe of soluble infliximab at only 150 mg/mL takes 20 minutes to load. Similarly, injection of the crystalline product takes 20 seconds compared with 100 seconds for the solubilized protein.

John Hicks, director of corporate development at Althea, is quick to note that alternative subcutaneous injection formulations exist, “although there aren’t many, and few achieve high drug concentrations.” Most notable is Halozyme’s partnership with Roche to develop subcutaneous Rituxan and Herceptin using a recombinant hyaluronidase technology.

“It appears that they can deliver subcutaneously,” Hicks says. “However with Herceptin, Roche had to invest $185 million on a proprietary delivery device to target 5–10 minute administration of injections as high as 12 mL. Pain becomes a serious issue with injections significantly larger than 1.5 mL.”

The switch to subcutaneous administration involves an IP or lifecycle component—companies hurrying to switch before their IV patents expire. But healthcare economics is undoubtedly the true driving force. “Why bring a patient into a hospital or an infusion center when the treatment can be administered in an office or even self-administered?”

Crystalline suspensions don’t necessarily require longer development times than conventional platforms, but they do require a somewhat different skill set. Chemists may use any reagent they like to produce crystals, whereas bioprocessors are restrained to pharmaceutically acceptable buffers under conditions that promote rapid crystallization at high yield. Althea claims yields of greater than 90% in 24 hours.

Protein crystallization does not change the state of the native protein, yet Althea has met with resistance to the technology.

“There’s a knee-jerk reaction against injecting crystals, most likely based on the potential for immunogenicity. We haven’t seen any in the weekly hGH trials so far,” says Hicks. “Besides, insulin crystals have been used for years on a daily basis, for example Lilly’s Humalog, which is a mix of soluble and crystalline insulin.”

The Super-Excipient

Albumin is a critical ingredient for cell culture, formulated proteins, and other biotech products. It binds to and stabilizes important molecules and ions, prevents aggregation and nonspecific adsorption onto glass, and serves as an antioxidant.

Aralast (Baxter), Kogenate (Abbott), Intron A (Merck), and Avonex (Biogen Idec) are just a few of the products formulated with albumin as an excipient.

Because of cost, bovine albumin was at one time preferred over human albumin. However, the trend toward serum-free and animal component-free media, and injected/infused products, has led to significant demand for recombinant human albumin (rAlbumin).

“Albumin is an extremely useful excipient in protein drug formulations, particularly for drugs that are self-administered,” says Dermot Pearson, marketing director at Novozymes Biopharma.

“There is no such thing as a universal excipient, but albumin achieves several formulation objectives compared with SADs.” Pearson is referring here to the acronym for common bioformulation excipients: sugars, amino acids, and detergents.

SADs are known to stabilize products in lyophilized forms, but their ability to stabilize proteins formulated as liquids has yet to be proved on a broad basis, according to Pearson.

Novozymes specializes in recombinant human albumin manufactured in yeast. Recombumin®, the firm’s lead product, and albucult® are used in drug and vaccine manufacture.

Recombinant human albumin can assist in both initial formulation and lifecycle-management reformulations. Novozymes’ customers are becoming interested in fine-tuning their formulations at earlier stages in the clinical development process.

For companies producing monoclonal antibodies, recombinant albumin can help resolve formulation issues earlier, according to Pearson, particularly when initial efforts using SADs fail.

Similarly, he maintains that the excipient “can help build security and maintain market share and exclusivity” for products nearing patent expiration.