Introduction
The quest for ensuring drug efficacy and safety is perennial in the complex realm of Drug Metabolism and Pharmacokinetics (DMPK). The pathway from drug discovery to development is fraught with uncertainties, making predictive accuracy not just an aspiration but a necessity. The evolution of computational models has ushered in a new era where predictions made in silico aid crucial decisions at every juncture of drug discovery and development. This blog post unravels how advanced predictive models are becoming the linchpin in bridging the understanding of drug behavior, thereby propelling the strides in DMPK.
Phase I reactions introduce or unmask a functional group on the drug molecule, making it more polar. These reactions are primarily catalyzed by the cytochrome P450 (CYP) enzymes. Phase II reactions involve the conjugation of the drug or its Phase I metabolite with an endogenous compound, such as glucuronic acid or sulfate, to increase its water solubility and facilitate its excretion (Rendic & Di Carlo, 1997).
The Paradigm of Predictive Modeling
In the epoch of data-driven science, predictive modeling has emerged as a cardinal tool in streamlining drug discovery and expediting drug development processes. The following segments elucidate how predictive modeling is being harnessed in these two critical phases:
1. Predictive Modeling in Drug Discovery:
Predictive ADME (Absorption, Distribution, Metabolism, Excretion): Utilizing computational models to predict drug-like properties is pivotal in filtering promising compounds early in the discovery phase. Predictive models offer insights into the ADME properties of new chemical entities, thus reducing the attrition rates associated with poor pharmacokinetics.
Target Identification and Validation: Machine learning algorithms sift through vast omics datasets to identify and validate potential drug targets. These algorithms can unveil hidden patterns and correlations, speeding up the target identification and validation processes.
2. Predictive Modeling in Drug Development:
Population Pharmacokinetics and Pharmacodynamics (PK/PD): Advanced modeling techniques are employed to understand how drugs behave in diverse populations. Population PK/PD models help understand the variability in drug response, thus aiding in dose optimization to ensure efficacy and safety across different population strata.
Clinical Trial Simulation: Simulation models are harnessed to optimize clinical trial designs by predicting the outcomes under different scenarios. This improves the efficiency of clinical trials and helps mitigate the risks associated with unforeseen adverse events.
DMPK studies have also played a pivotal role in optimizing the use of existing antimicrobials. For instance, DMPK studies have contributed to optimizing vancomycin dosing in patients with methicillin-resistant Staphylococcus aureus (MRSA) infections. These studies showed that a higher vancomycin dose is needed to achieve therapeutic concentrations in the serum and at the site of infection, leading to a revision of the dosing guidelines (Rybak et al., 2009).
Pioneering Case Studies
1. Mavrilimumab Development: A Model-Based Approach: The development of Mavrilimumab showcased a “learning-predicting-confirming” continuum implemented at every decision point, highlighting the essence of model-based approaches in rational discovery, preclinical development, clinical study design, and dose selection.
2. Predictive PBPK Models for Drug-Drug Interactions: The integration of data for Physiologically-Based Pharmacokinetic (PBPK) model development has opened new vistas in understanding metabolic drug-drug interactions, marking a shift in perspectives and emerging trends in DMPK studies.
3. Collaborative Project Between FCC and AstraZeneca: The synergy between FCC and AstraZeneca in a three-year project exemplified how collaborative endeavors could refine predictive models and methodologies, augmenting the precision in drug discovery and development.
4. Physiologically Based Absorption Modeling: Diverse case studies elucidate the impact of physiologically based absorption modeling in addressing biopharmaceutics or formulation questions, showcasing its pivotal role in drug development.
The advancements in predictive modeling significantly propel the frontier of drug discovery and development in DMPK. As elucidated through various case studies, these models are not merely theoretical constructs but have demonstrable impacts in real-world drug development scenarios. They are instrumental in unraveling drug candidates’ complex pharmacokinetic and pharmacodynamic behaviors, thus providing a solid foundation for making informed decisions through the drug development pipeline.
The collaborative ethos between academia, industry, and regulatory bodies is quintessential in further honing these models to accurately mirror the complex reality of drug behavior. As the DMPK community continues to embrace and advance predictive modeling, the horizon is expanding for more rapid, precise, and successful drug development endeavors.
The impetus is to foster a culture of continuous learning, collaboration, and innovation to refine predictive models further. The trajectory is promising, and the onus is on the DMPK community to continue pushing the boundaries of what’s possible with predictive modeling, ensuring the translation of these advancements into tangible benefits in drug discovery and development.
References:
- - Abouir K, Samer CF, Gloor Y, Desmeules JA, Daali Y. Reviewing Data Integrated for PBPK Model Development to Predict Metabolic Drug-Drug Interactions: Shifting Perspectives and Emerging Trends. Front Pharmacol. 2021 Oct 28;12:708299. doi: 10.3389/fphar.2021.708299. PMID: 34776945; PMCID: PMC8582169.
- - Kesisoglou F, Chung J, van Asperen J, Heimbach T. Physiologically Based Absorption Modeling to Impact Biopharmaceutics and Formulation Strategies in Drug Development-Industry Case Studies. J Pharm Sci. 2016 Sep;105(9):2723-2734. doi: 10.1016/j.xphs.2015.11.034. Epub 2016 Jan 23. PMID: 26886317.
- - Prof. Johan Gabrielsson, Swedish University of Agricultural Sciences, Prof. Bert Peletier, Leiden University, Christine Ahlström, AstraZeneca, https://www.fcc.chalmers.se/focus-areas/life-science/modeling-and-simulation-for-predictive-model-based-drug-discovery-and-development/#:~:text=,Pharmacodynamics%20%28PKPD%29%20scientific%20areas
- - Litsa, E. E., Das, P., & Kavraki, L. E. (2021). Machine learning models in the prediction of drug metabolism: challenges and future perspectives. *Journal of Pharmacological Sciences, 145(2), 81-92.* doi:10.1016/j.jphs.2021.01.007.