Transforming Preclinical Research with Mechanistic PK/PD Modeling Approaches
What is mechanistic modeling and why is it important for drug development?
Mechanistic modeling, or mechanistic PK/PD modeling, or systems pharmacology takes traditional PK/PD modeling to the next level. In traditional PK/PK modeling, you might have very well-defined curves that are fit to individual data, which can then be allometrically scaled from one species to another. With the next level in systems modeling, we look at the real concentration data and PD, but we also incorporate the mechanism of action of the drug and the mechanism of action of the target biology.
For example, if you are working on a gene therapy delivered by a lipid nanoparticle, the model would consider the dosing of the LNP, payload within the LNP, targeting, distribution, and interaction with tissues, among other things. It helps us understand the dynamics of the entire system, from dosing to downstream effects, offering a deeper understanding of drug behavior.
| [Article] What is Mechanistic Modeling and How Can it be Applied in Drug Discovery? Learn More >> |
“A way to think about [...] mechanistic PK/PD modeling, or systems pharmacology, is to think about PK/PD maybe at the next level.”
What comes after developing the preliminary model?
Once you have developed the initial model, you want to keep it relatively simple to determine if the drug's development is viable. This simplification helps prioritize experiments and go/no-go decisions. If the model parameters appear favorable, indicating a high likelihood of success, you move on to refining the model to design the best-in-class drug. This involves considering downstream biology, such as cell interactions, safety, and various parameters that affect the drug's effectiveness. The goal is to identify the most important experiments and parameters to control, which ultimately streamlines the drug's development.]
Over the model development process, during sensitivity analyses, you may find that changing some parameters impacts the output drastically while other parameters do not make a significant impact. It is important to assess this variability in the PK/PD because this is observed in humans as well. Not every disease patient has the same number of cells, the same number of sites per cell, or protein synthesis rates. The strength of mechanistic modeling is in handling the uncertainty and non-linearity. This information will help create model-informed next preclinical experiments.
“It's a continuum, but the purpose is trying to figure out the most important experiments and the most important parameters that you can control or select for, such that hopefully you can get to the clinic faster, or hopefully you will be a best in class therapeutic.”
How many iterations of the model are typically needed during the drug development process?
The number of iterations depends on the specific drug and its development plan. You want to be strategic and not update the model too frequently. You can continue to update the model using parameter estimation to further constrain some of the unknown parameters. With each iteration, you will obtain more certainty on your human dose predictions, which will impact your clinical candidate selection early on, and hopefully finalize the best in class. The goal is to continuously refine and improve your model based on new data and insights. The number of updates will vary, but they should align with the critical decision points in the development process.
“[...] every time you do one of these studies, we're getting more and more certainty on your human dose predictions, which early on will impact your clinical candidate selection.”
Try out a 2-week trial with predefined models or request a demo today
Can the competition factor into these analyses, and how?
Competition plays a significant role in drug development. Companies can use mechanistic modeling to gain a competitive edge by comparing their drug with existing treatments or competitors' drugs. This allows you to identify advantages and disadvantages and make informed decisions about the drug's viability. For example, you are working on a gene or cell therapy based on information from an existing antibody. The information on the existing antibody in the public domain can be used to inform the model. Additionally, you can anticipate potential therapies that might emerge in the future and assess how your drug stacks up against them. Mechanistic modeling helps you make strategic decisions about the development and positioning of your drug in a competitive landscape.
| [Poster] Quantitative modeling predicts competitive advantages of a next generation anti‐NKG2A therapy over monalizumab for the treatment of cancer Download Poster >> |
Can the competition factor into these analyses, and how?
Competition plays a significant role in drug development. Companies can use mechanistic modeling to gain a competitive edge by comparing their drug with existing treatments or competitors' drugs. This allows you to identify advantages and disadvantages and make informed decisions about the drug's viability. For example, let us you are working on a gene or cell therapy that is based on information from an existing antibody. The information on the existing antibody in the public domain can be used to inform the model. Additionally, you can anticipate potential therapies that might emerge in the future and assess how your drug stacks up against them. Mechanistic modeling helps you make strategic decisions about the development and positioning of your drug in a competitive landscape.
| [Poster] Quantitative systems pharmacology (QSP) model for preclinical to clinical translation of antibody-drug conjugates (ADCs) Download Poster >> |
How does mechanistic modeling help companies prepare for clinical trials and the IND (Investigational New Drug) application process?
Mechanistic modeling is particularly valuable in preparing for the IND stage. By creating detailed models, you can simulate patient variability and assess the impact of variability in human parameters on safety and efficacy predictions. This allows you to determine a safe yet efficacious starting dose for clinical trials. In the case of gene therapies and advanced treatments, mechanistic modeling is essential because traditional methods may not provide accurate human dose predictions. The FDA's Project Optimus recognizes the importance of quantitative, model-informed drug discovery to optimize dosing across oncology, and mechanistic modeling plays a vital role in achieving this goal.
[Article] How Mechanistic Modeling Informs First-in-Human Dose Selection |
Recently, the FDA released Project Optimus, which is a new approach to dose optimization across oncology. How does Applied BioMath help with this goal?
The FDA is counting on the industry to bring the best science to make the best predictions and as accurate as possible, for safety and efficacy, driven by quantitative model-informed drug discovery. Using traditional MABEL/NOEL approaches may still offer doses that are starting too low, and this is where mechanistic modeling provides an advantage. We can start thinking about how the possible uncertainty and variability and patient parameters coupled with the new therapy parameters, impact predictions on safety and efficacy. We may also be able to justify a higher safe starting dose when compared to traditional methods.
Mechanistic modeling for preclinical development can transform the drug development process. It enables companies to make informed decisions, build competitive products, better plan preclinical experiments, and optimize dosing strategies for clinical trials. All in all, mechanistic modeling offers a promising approach to improving the success rate of new therapeutics.
A more detailed discussion of mechanistic modeling for preclinical development can be heard on this podcast.