Late-stage failures in drug development often result from advancing molecules that are difficult to formulate or manufacture. Coriolis Pharma addresses this challenge by combining laboratory-based and in silico developability assessments to provide early insights into molecule characteristics and potential liabilities, optimize formulations, and mitigate risks. By streamlining decision-making processes and predicting stability and manufacturability issues early on, Coriolis helps developers reduce costs, accelerate timelines, and improve the likelihood of success from discovery to commercialization.
Evolving the Drug Development Playbook with Early-Stage Developability Assessments
The biopharmaceutical industry faces mounting pressure to accelerate the timeline for bringing new therapies to market. However, advancing molecules through development without proper scrutiny often leads to costly failures in later stages. A significant portion of these failures can be traced back to advancing candidates that ultimately prove difficult to formulate or manufacture at scale.
Given the immense financial and time investment required to develop new drugs, the importance of evaluating a molecule's developability early in the process cannot be overstated. Developability assessments identify the inherent properties of a candidate that predict its potential to be developed into a safe, effective, and commercially viable drug. By conducting these assessments during early development, companies can proactively manage risks associated with stability, manufacturability, and scalability.
For biologic therapies, particularly monoclonal antibodies (mAbs) and related molecules, developability assessments focus on identifying critical biophysical properties. Stability, formulatability, and manufacturability are key considerations. Stability directly impacts the therapeutic’s long-term effectiveness and safety, while formulatability provides information on the degree of stability optimization achievable through formulation development. Manufacturability addresses compatibility with existing production technologies and processes.
Beyond merely identifying suitable candidates, early developability assessments streamline the entire drug development process. Computational modeling plays a vital role in this strategy by providing early insights without the need for extensive material testing. These models predict potential challenges, such as aggregation, chemical degradation, or viscosity issues, allowing developers to address risks before they become critical obstacles. This approach not only reduces the likelihood of late-stage failures but also significantly shortens development timelines and reduces costs.
Incorporating computational modeling into developability assessments allows for the evaluation of a broader range of molecules at the earliest stages, enabling the selection of candidates with the highest potential for success. This practice aligns with the growing industry trend toward quality by design, where every decision is driven by data to reduce risks and improve outcomes.
Ultimately, comprehensive developability assessments empower drug developers to make smarter, more informed decisions, increasing the likelihood of bringing effective therapies to market faster and with fewer setbacks.
Material Constraints in Early-Stage Assessments
While developability assessments are essential for reducing risk and streamlining drug development, conducting these evaluations in the early stages presents unique challenges. A major limitation is the quantity of drug substance or formulated product available for testing, which is often minimal at this stage. This scarcity of material can restrict the number of tests that can be performed and limit the depth of analysis.
Moreover, the material produced during early development often has lower purity and concentration compared with the final commercial product. These differences can impact the reliability of experimental results, making it difficult to accurately predict a molecule’s behavior in later stages of development. As a result, traditional laboratory-based assessments may be constrained by the availability and quality of materials, potentially leaving critical questions unanswered until more material is produced.
To address these challenges, specialized laboratory methods requiring limited input material can be used. A number of approaches have been developed to predict developability, formulatability, and long-term stability by evaluating key stability-indicating biophysical properties in miniaturized assay formats. In addition to traditional laboratory experimentation, in silico modeling provides a complementary approach that eliminates the dependency on physical material for initial assessments. Computational tools can predict key properties, such as chemical stability, aggregation propensity, and formulatability based solely on the molecular sequence, offering valuable insights even when physical samples are limited. By integrating digital and laboratory methods, developers can achieve a more comprehensive understanding of a candidate’s potential without the need for extensive physical testing at early stages.
Unlocking New Insights with Computational Modeling
Computational models serve as a powerful complement to laboratory experimentation, enabling more efficient and thorough evaluation of early-stage drug candidates. These models require minimal input —just the primary sequence— and can predict critical biophysical characteristics that influence stability, aggregation, formulatability, and manufacturability.
At the earliest stages of development, in silico modeling is invaluable for screening large libraries of candidates to identify those with the best potential for success before any physical testing is conducted. Later in the development process, computational models help predict how lead candidates will behave under various formulation, manufacturing, and storage conditions. This insight reduces the need for extensive physical testing, enabling faster progression through key development milestones.
The speed and efficiency of computational calculations offer distinct advantages, particularly in high-throughput environments. Machine learning algorithms can rapidly analyze vast amounts of data to identify patterns and relationships that might not be immediately evident through traditional analysis. These models can predict issues, such as aggregation, chemical degradation, or high viscosity, providing actionable insights that inform experimental design and decision-making.
Moreover, in silico modeling generates a wealth of early-stage data, which supports more informed decision-making and leads to higher-quality product development. By integrating computational tools with traditional laboratory methods, developers can achieve a more comprehensive assessment of drug candidates, ultimately improving the likelihood of success while reducing costs and timelines.
Matching Models to Applications: A Multiscale Approach
Achieving the greatest value from in silico modeling requires the application of the right computational techniques for specific challenges. Different types of models excel in distinct areas of developability and formulation assessments, making a multiscale approach essential.
For identifying potential stability issues of biologic molecules, machine learning models are particularly effective. These models use large datasets to identify patterns and predict biophysical characteristics with high accuracy. On the other hand, mechanistic molecular dynamics models are better suited for examining interactions between the target molecule and various excipients during formulation development, providing insights into potential formulation corridors (i.e., suitable pH ranges and types of excipients). Kinetic models come into play when predicting long-term behavior during storage. These models simulate how a molecule degrades over time, offering valuable information about shelf-life and stability.
The multiscale modeling approach integrates models operating at different levels — from quantum-scale interactions, which capture chemical reactions occurring in picoseconds, to mechanistic models that simulate a molecule’s behavior over months or even years. By combining these models, developers can obtain a more holistic understanding of how a drug candidate will perform across various conditions and timescales, ensuring more robust formulation and manufacturability decisions.
In Silico Development Assessment at Coriolis Pharma
Coriolis Pharma’s developability platforms integrate in silico modeling to identify lead candidates with the highest potential for success and to outline formulation corridors. This approach provides early insights into critical risks, such as colloidal and conformational instability, aggregation, high viscosity, and chemical degradation pathways, including oxidation and deamidation — all without requiring physical material.
A key strength of Coriolis Pharma’s multiscale methodology is its ability to simulate molecular behaviors across varying timescales. Molecular dynamics models offer detailed information on interactions occurring in microseconds, while kinetic models predict long-term stability over months or years. This comprehensive coverage helps assess how candidates will perform under diverse conditions and in various formulations.
By combining computational insights with laboratory-based testing, Coriolis Pharma delivers thorough and reliable assessments. This integrated approach predicts chemical and physical stability, providing clients with actionable data for smarter decision-making, tailored formulation strategies, and streamlined development timelines.
Ongoing Data Development
The successful application of computational modeling hinges on access to sufficient, high-quality data. The relevance and volume of data directly impact the accuracy and interpretability of predictions, making robust data collection essential.
Coriolis Pharma leverages a wide array of data types and sources to strengthen its in silico models. These include data from scientific literature, public databases containing information on approved molecules and emerging modalities, and proprietary internal data sets. This diverse pool of data allows for more comprehensive modeling and improved prediction accuracy.
In the context of excipient selection, Coriolis primarily focuses on ingredients that have established regulatory acceptance and a track record of use in commercial drug formulations. While novel excipients offer potential advantages, they also introduce regulatory uncertainties that many drug developers prefer to avoid. However, Coriolis remains open to incorporating data on new excipients of interest, ensuring that its models remain adaptable to evolving industry needs.
A Combined Approach to Developability at Coriolis Pharma
Developability assessments at Coriolis Pharma integrate both in silico and laboratory-based approaches to provide comprehensive support to clients and accelerate drug development timelines. The unique combination of computational and physical experimentation allows for a more efficient and thorough evaluation process.
Although physical testing remains necessary to validate predictions and confirm real-world behavior, the combined platform guides the identification of promising lead candidates, thus significantly reducing the number of molecules needing experimental evaluation.
Coriolis Pharma’s laboratory-based developability platform was launched in mid-2024, with the in silico platform slated for launch in early 2025. Designed to assess the potential of drug candidates with minimal material requirements, these platforms are particularly valuable for biopharma companies that may have limited quantities of a candidate available at early stages of development.
The newly added computational modeling capability enables even earlier-stage developability assessments, helping clients make crucial decisions sooner and avoid costly late-stage failures. In addition, in silico modeling serves as a crucial supplement when physical samples are insufficient for comprehensive lab-based evaluations.
Coriolis Pharma’s developability approach is complemented by a high-throughput preformulation screening platform, which systematically explores a wide range of excipients and conditions tailored to antibody development. The data generated through this platform supports informed decision-making aligned with the quality target product profile (QTPP) of the therapeutic molecule being investigated, streamlining regulatory submissions and enhancing product predictability in real-world settings.
Reducing Risk and Streamlining Development
With its combination of in silico and laboratory-based assessment capabilities, Coriolis Pharma is well positioned to support clients throughout the entire drug development process — from early development through commercialization and life cycle management. The company’s service offerings extend well beyond formulation development to include all kinds of stability and forced degradation studies, clinical in-use and formulation robustness studies, drug product manufacturing support, analytical method development, and GMP-compliant validation and testing.
These comprehensive services enable drug developers to make informed decisions at critical points in the development timeline. By identifying promising candidates early, optimizing formulation strategies, and addressing potential manufacturability challenges, Coriolis helps reduce the risk of unexpected issues arising during later-stage development.
The proactive management of risk offered by Coriolis Pharma ensures smoother progression through key development milestones, minimizing delays and unexpected costs. This streamlined approach accelerates the transition from discovery to clinical trials and, ultimately, to market launch, helping clients achieve faster, more predictable, and more successful outcomes.
