September 15, 2012 (Vol. 32, No. 16)

Many pharmaceutical research studies seek to overcome the challenges associated with high-concentration drug formulations.

This work is also designed to improve the time and accuracy of acquiring and analyzing data that is critical throughout the development process.

Overlooking any one attribute can influence the productivity and profitably of a pharmaceutical company. Fortunately, new predictive methods and tools for product characterization have allowed scientific teams to better understand the state and behavior of novel drug formulations and to tweak formulation conditions toward manufacturing effective drugs with long shelf-life.

Delivering a promising drug at high concentration is a key goal, but challenges in formulation stability, viscosity, and solubility as well as data interpretation continue to be obstacles in formulation development. Recent research on predictive methods for formulation development provides a glimpse of possible solutions to these challenges.

One researcher addressing the issues is Mark Brader, Ph.D., senior scientist, protein pharmaceutical development at Biogen Idec. Presenting at IBC’s “Formulation Strategies for Protein Therapeutics” next month, Dr. Brader will discuss a rapid screening method to predict long-term stability that will help accelerate formulation development.

“The importance of the technique lies in its ability to answer the question” ‘How well does the molecular thermostability actually translate into predictive pharmaceutical stability,’” notes Dr. Brader.

Protein formulation is such a critical step in new drug development because any alteration to a protein’s structure may not only affect its therapeutic potential, but may also lead to unwanted side effects. [Health Protection Agency/Photo Researchers]

Molecular Thermostability

Molecular thermostability is the stability of a formulation at high temperatures whereas pharmaceutical stability is the stability of a formulation when stored at 4–5°C as well as the in-use stability at room temperature. Typically, pharmaceutical companies aim to create formulations with a two to three year shelf-life.

To predict long-term stability and determine the best conditions for this, Dr. Brader’s screening process involves monitoring the unfolding and folding kinetics of proteins, varying formulation conditions (e.g., temperature, concentration, pH, buffer), protein-protein interactions, and variations of a molecule.

“A major challenge in formulation development is the screening of highly concentrated formulations that are greater than 100 mg/mL,” says Dr. Brader. “At high concentrations, formulation scientists are often concerned with physical degradation, which includes aggregation and denaturation, and chemical degradation.”

One way to test for degradation is to incubate a drug for a long time and test viscosity. However, this approach is impractical and inefficient.

Daniel Some, Ph.D., principal scientist at Wyatt Technology, studies intermolecular interactions for the optimization of biotherapeutic formulations. He prefers the use of CG-MALS (composition-gradient multi-angle light scattering) as the method requires only a few hours to obtain results.

“CG-MALS relies on the theory of static light scattering and is capable of determining absolute molecular weight, which is one attribute that is considered when monitoring for protein degradation,” explains Dr. Some, who spoke last month at CHI’s “Bioprocessing Summit”.

“Conventional methods, such as dynamic light scattering (DLS) for studying aggregation and surface plasmon resonance (SPR) for investigating protein-protein interaction, are limited to low sample concentrations and do not provide the data necessary for interpreting the state and behavior of high concentrated formulations.”

With dynamic light scattering, characterization of formulations at high concentration is difficult as the method is best suited to interpret single-scattering events in a dilute solution. With surface plasmon resonance, the method cannot properly determine stoichiometry and is limited between 1:1 and 2:2 binding models.

Since an objective of current efforts in formulation development is to understand the stability and interaction of formulations at high concentration, which best resembles in vivo conditions, accurate and timely analysis is critical.

CG-MALS is a preferred method for protein characterization as it is rapid and convenient. The technique can be used to determine absolute stoichiometry, binding affinity, and nonspecific protein-protein interactions. In addition, CG-MALS can quantify self-association, hetero-association, and reversible self- and hetero-association, which constitute information that can provide clues about interactions.

“In CG-MALS, the attraction and repulsion between molecules is characterized by virial coefficients: self-virial coefficients and cross-virial coefficients,” continues Dr. Some. “These coefficients can provide insight into the attractive and repulsive forces between molecules and can be fine-tuned to determine the best conditions for formulation development, especially for optimizing a formulation for purification.”

Purification Step

Indeed, an important stage in formulation development is purification. At the IBC meeting next month, Haripada Maity, Ph.D., principal scientist, formulation development at ImClone Systems, an Eli Lilly subsidiary, will talk about carrying out biophysical analyses to understand and mitigate challenges in downstream purification.

Dr. Maity will present a case study on a monoclonal antibody to demonstrate the “exceptional utility of monitoring unfolding/refolding kinetics over other measurements.”

Dr. Maity explains that although optimization of stability and solubility of a protein at low pH is important for chromatography, “the challenge is to avoid damaging the protein during the purification process and after purification when formulations are brought back to neutral pH.”

“Otherwise, degradation of a formulation during purification can result in poor yield and, ultimately, lead to less profits.”

Dr. Maity will also address the challenges of using conventional methods for optimization and other aspects of biophysical analysis, including conformation stability, structural characterization, physical size, and concentration-dependent protein-protein interactions.

Protein stability is critical and is one aspect of formulation development in which the use of conventional methods may not be enough to understand the effects of various formulation conditions, he maintains.

However, one conventional approach that is invaluable is differential scanning calorimetry (DSC), which provides the formulation scientist with data on conformational stability, which is related to thermodynamic stability, of a protein formulation.

For example, if one wants to determine an appropriate buffer for downstream processing, DSC is a technique that allows one to determine which buffer would be optimal for maintaining protein stability.

“By monitoring the melting temperature, one can understand different buffer systems and pick the best buffer with low pH,” he continues. “Moreover, monitoring the melting temperature is a direct way to understand whether a particular protein is stable or not.”

In addition to monitoring thermodynamic stability with low pH buffers, monitoring the unfolding and refolding kinetics at low pH can provide answers to questions such as, “Has the structure changed?” and “How long does the protein take to undergo a change?” In other words, if one decreases the pH from 7 to 3.5 as a function of time, one would be interested in what structural changes occur and the duration of those changes.

However, the challenge that remains with conventional methods such as DSC is data interpretation. The data can be interpreted in multiple ways. First, a higher melting temperature does not mean higher stability, although, in most cases it is true. Second, DSC can provide two different kinds of data, exothermic and endothermic curves, and the interpretation can be complex.

“Nonetheless,” adds Dr. Maity, “the goal of finding a good buffer is to have a strong correlation between kinetic data and equilibrium stability data.”

Previous articleBayer HealthCare to Buy Teva Animal Health Business
Next articleFeatured Video: Mastering the DNA Transfection Workflow