February 15, 2024 PAO-02-24-CL-02
Formulation development of any drug molecule is an inherently challenging process that aims to successfully pair all the added excipients with the active drug, while maintaining its biological activity and bioavailability. The complexity of biologic drugs presents a unique challenge driven primarily by the intrinsic dynamic structure of the drug substance. In fact, formulation development for biopharmaceuticals, as well as the broader drug product development, is an interdisciplinary pursuit requiring expertise in physics, mechanics, engineering, structural biology, chemistry, and pharmaceutical sciences.
Each molecule is unique, with different chemical and physical characteristics and therefore different formulation needs. Thus, each molecule must be evaluated for stability, aggregation, critical quality attributes, and potential interactions with excipients. It takes time to adequately investigate a molecule, identify potential hurdles and devise an effective formulation that overcomes these difficulties. When working with multiple products, the time needed to select an effective drug formulation can present a significant bottleneck in the context of today’s accelerated drug development timelines.
The ever-changing diversity of molecular formats being advanced as potential drug candidates contributes to the inherent complexity of biopharmaceutical formulation development. Just within the antibody arena, there are now bi-specific, tri-specific, and other multi-specifics, antibody fragments, nanobodies, enhanced expression vectors (EEVs), fragments, and so on. Beyond antibodies, biotherapeutic formats have expanded to include enzymes, cytokines, polypeptide hormones, recombinant fusion proteins, and others. This considerable spectrum of molecule classes makes it challenging to even know where to begin a formulation project because the information available for commercial products cannot be reliably transferred to new molecules in development.
One of the leading causes of candidate failure within the bounds of formulation development is instability of the drug substance –– particularly for molecules that are not traditional recombinant antibodies or proteins. The newer formats present a new range of issues with respect to colloidal, physical, and chemical stability, all of which determine whether a drug substance will retain its integrity once formulated. Stability directly impacts the safety and efficacy of formulated drug products.
Furthermore, these newer formats often require the use of different sets of excipients than those used for traditional antibodies. A decade ago, there was a single set of excipients for the formulation of biologic drugs that could be screened quickly and easily. Today, with so many different molecular formats, the number of excipients has expanded greatly. For biologic drugs, there are buffering agents, surfactants, viscosity-lowering agents, pH-adjusters, stabilizers, complexing agents, microbial preservatives, and many more. A matrix of excipients must be used to address all the needs of these biological drug substances; however, screening the many different options in each category, in a relatively quick and effective manner, to obtain a clear picture of the impact on the colloidal, physical, and chemical stabilities of formulated biopharmaceuticals, is not a simple task.
A proper developability assessment can help address many of these challenges, and the use of appropriate computational methods is an important aspect of such an assessment. Algorithms have been developed that can predict the behavior of a specific biologic drug based on specific physicochemical characteristics. While access to structural information obtained using cryo-electron microscopy and crystallography is ideal, most biologics can still be modeled with high precision using key property data. For instance, identifying regions of high negative charge or high hydrophobicity that can contribute to degradation can now be achieved for most biologics.
Most developability software packages include specified sets of excipients and use structural information about the biomolecule to identify potential opportunities for interaction between different excipients and the drug substance, as well as potential impacts on stabilization versus degradation, thermodynamic behavior, and other properties.
It is worth noting, however, that artificial intelligence (AI) and machine learning (ML) are beginning to be leveraged in formulation development. The application of these technologies to formulation development is still in its early stages, but solutions will ultimately be developed that will fully predict the optimal formulation for a given biologic drug substance. Indeed, within the next 10 years, AI is poised to be widely used in formulation development. It can be expected, for instance, that as AI programs advance, we will be able to identify novel surfactants that will safely and effectively stabilize antibodies to replace the commonly used polysorbates.
In many cases, excipients that improve one aspect of a formulation can have a detrimental impact on another. The multifactorial nature of formulation development and the need for diverse excipient types greatly contributes to the challenges associated with formulation development. It is essential to use a quality-by-design (QbD) approach that takes into consideration the potential impact of excipients on not only the drug substance but also the effectiveness of other excipients. Such an approach allows for concurrent monitoring of multiple excipients, identifying which interactions between the excipients and the drug substance — and among the excipients themselves — may influence the stability of the biomolecule.
Predicting such behavior is merely the first step. Predictions must be confirmed, particularly when molecules deviate from the standard monoclonal antibody format, thus increasing the challenge to generate accurate predictions. Once computational modeling has been employed, it is also necessary to conduct physical experiments to confirm (or disprove) the predicted outcomes empirically and facilitate further fine-tuning.
For molecule classes — such as enzymes — that present challenges with respect to establishing effective predictive models of developability, physical screening becomes paramount. Rather than evaluate a smaller number of excipients identified using computational methods, a minimum of several hundred excipients should be screened in high-throughput fashion to successfully identify those that may be effective at stabilizing the target biologic drug substance.
Before initiating formulation development for a drug candidate, it is critical to consider the optimal route of administration. Formulations and excipients for oral administration are very different from those intended for subcutaneous or intravenous injection, as well as other means of administration. Excipients for stabilizing biologics in liquids will be different from those used to protect the drug substance through the freeze-drying process to generate lyophilized drug products.
Choosing the optimal route of administration requires the best possible understanding of the drug substance, the disease being treated, and the needs of the patient population. Not all molecules are suited for all forms of administration. Some are too unstable in solution, while others degrade during lyophilization. Some drugs must be delivered orally and others directly into the eye or the spinal column. Similarly, not all forms of administration are suitable for all patient groups. Many elderly patients with arthritis and other sources of limited movement have difficulty self-administering prefilled syringes, while people located far from healthcare centers cannot easily visit a doctor’s office for regular injections.
In addition, there may be manufacturing and economic drivers for the selection of a particular delivery approach. Lyophilization, for example, generates more stable products with longer shelf lives, which can enable less frequent production of larger batches, leading to lower overall costs.
Biosimilars present some additional unique challenges with respect to formulation development. While it is possible for the biosimilar formulation to differ slightly from that of the innovator drug it is intended to replace, it can only be slightly different: changes in certain excipients, adjustment of the pH and the protein concentration, and so on.
These changes are allowed because the original formulation often was developed with earlier technology and there is room for improvement applying modern platforms. In addition, some formulations are specifically protected by patents and cannot be copied. Third, it is nearly impossible to completely reproduce the original cell line used to make the branded biologic. As a result, the composition of the biosimilar may differ to some degree — and those differences may require a slightly different formulation. Finally, there is often very limited information available beyond the formula. Data about how different excipients impact long-term stability, degradation pathways, shelf life, and so on are not available and must be generated to meet regulatory requirements.
As with formulation of any type of drug product, the focus is on using excipients that have previously been used in approved medications, preferably those that are similar. Using novel excipients creates an additional level of risk: in the United States, the FDA requires extensive data supporting their use and demonstrating their safety.
If developing a drug for the U.S. market for which previously approved excipients do not enable effective formulation, it is possible to consider excipients previously approved by regulatory agencies in other countries with which the FDA has good relations, such as those in Europe, Japan, and South Korea. A second option is to consider a derivative of an approved excipient. Novel excipients are generally a last resort, and in this case the preference is to use molecules for which safety data is already available.
The strength of Tanvex CDMO derives from multiple factors. We have a dedicated team that spans all the scientific expertise that must converge on efficient formulation development, including strong physics, chemistry, biology, and analytical groups, as well as an excellent and visionary team. This expertise is backed by cutting-edge technology that enables Tanvex CDMO to conduct formulation screening in a manner that is unique to the company and lowers the risk of failure for our clients. Biomolecules are examined by applying defined physical principles to predict and later confirm various behavioral factors, and that information is used to identify more stabilizing agents that can lead to biologic products with long-term stability.
The unique platform is proprietary and was built in-house at Tanvex CDMO. It leverages our assets of biophysical tools — including light-scattering, circular dichroism, fluorescence, and IR or NMR spectroscopy, among other techniques –– to generate data that are then used to predict the behavior of molecules within formulations. For instance, our technology can predict if a biomolecule will aggregate owing to intra- or intermolecular interactions and whether a specific excipient will inhibit that type of activity.
A biophysical approach was selected because it provides greater prediction accuracy and thus a higher likelihood of identifying truly developable candidates. This benefit is also realized using smaller amounts of material in less time than traditional methods, and thus is more cost-effective and can accelerate project timelines.
Not only does Tanvex CDMO have outstanding scientists, but those scientists possess an understanding of many complicated biological mechanisms not necessarily available at other CDMOs. The combination of expertise with our unique and rapid predictive/screening technologies is the foundation of formulation development strength in our company. Tanvex CDMO develops formulations in weeks rather than months, which is the industry norm. Those formulations are backed by pre-formulation studies that generate predictive colloidal, chemical, and stability data. The performance of the selected excipients is then confirmed using a wide variety of advanced analytical techniques performed on state-of-the-art instruments.
Customers of Tanvex CDMO also benefit from the company’s location in San Diego, California, which has become an important biotechnology hub in the United States. Access to top talent and networking opportunities with other technology leaders directly drives further innovation within the company. As a benefit, there is a vigorous flow of scientific knowledge and understanding, and clients are assured that the science behind our recommended formulations will always be of the highest standard.
Finally, the leadership at Tanvex CDMO makes a point of providing a positive work environment and valuing employees in terms of both financial compensation and growth opportunities. Happy employees significantly contribute to the success of the company. Indeed, hiring people who learn from experience and have a passion for science is a recipe for success. Tanvex CDMO has achieved that recipe. This team of intelligent, satisfied people — all passionate about what Tanvex CDMO is doing and working together toward a common goal — is truly helping advance safe and effective biotherapeutics to improve the lives of patients around the world.
Pawel is Tanvex CDMO's drug substance and drug product development director, bringing nearly two decades of experience in early R&D, as well as early and late stage of process development to Tanvex. Pawel started his career at The Scripps Research Institute, working under legendary Nobel Prize laureate Kurt Wüthrich and Raymond Stevens, exploring biophysical and biochemical approaches to some of the most challenging pharmaceutical targets, GPCRs and other membrane proteins. Further industrial experience and growth in leadership roles encompasses Dart Neuroscience, AnaptysBio and Bristol Myers Squibb. Pawel’s strength and competence in biologics development lies in his deep understanding of structural biology, analytical chemistry and biophysics of proteins and their non-linear interactions with small molecules. Pawel has published several dozens of publications in peer-reviewed journals and played a key role in several IND and BLA fillings, as well as commercial product launches. Pawel’s real scientific muscle is his colossal passion for science and helping other people in their growth and personal evolution.
.svg)