Integrated Lead Optimization: Translational Models as We Advance Toward the Clinic

Author index

Perfect drugs are potent, specific and nontoxic. Many compounds fail because of unexpected toxicity and lack of efficacy in later stages of clinical development. Therefore, more complete knowledge and understanding of the properties of a drug is needed at an earlier stage of drug development. DNA microarrays can yield gene expression profiles from cells or tissues treated with a compound. Such "expression fingerprints" are used in drug discovery for drug target identification and validation and for elucidating the mode of action of novel compounds during lead identification and optimization. Moreover, during drug development, DNA microarrays help in the discovery of new diagnostic and prognostic biomarkers, as well as in the prediction of resistance and toxic side effects. This review aims to assess to what extent the promise of gene expression profiling has already materialized for the different stages of drug discovery and development. (c) 2002 Prous Science. All rights reserved.

Bioanalysis in a drug metabolism environment can be defined as the analytical techniques utilized to assay various samples that are the result ofin vitroorin vivostudies of one or more new chemical entities (NCEs). Bioanalysis has continued to evolve in response to the needs of the drug discovery and drug development teams, as well as the continuing improvement of the analytical technology that is used for the various assays. This chapter provides an overview of the various assays that are commonly used in various stages of drug discovery and drug development. The chapter is divided into three main sections: bioanalytical strategies,in vitroassays, andin vivoassays.For most of the assays discussed in this chapter, the analytical tool is high performance liquid chromatography (HPLC) combined with tandem mass spectrometry (MS/MS). It is common to have different levels of analytical acceptance criteria that become stricter as one moves from the early lead optimization arena into the development arena. By using this strategy, the higher throughput assays that are used in the screening stage would have minimal standards needed as part of the assay. As one goes to the higher levels, relatively smaller numbers of compounds are being assayed and the number of samples per compound is higher.There are multiplein vitroassays that are used for drug discovery screening of NCEs. The basic strategy for utilizing these variousin vitroassays is to use them to screen NCEs in order to select those that have druglike properties. These tests are commonly used as part of the absorption, distribution, metabolism, and excretion (ADME) optimization phase of new drug discovery. The most commonly used assays are designed to measure one of the following parameters: permeability, metabolic stability, enzyme inhibition, or metabolism. Each of these assay types are described in this chapter as a subtopics.In a drug discovery setting,in vivoassays are similar toin vitroassays in terms of the need for providing the results in a timely manner. The main difference betweenin vitroassays andin vivoassays is thatin vivoassays are often more challenging because the sample matrix is more complex. Mostin vivoassays rely on some form of HPLC‐MS/MS for the analytical step. The mainin vivotopics covered in this chapter are PK screening, PK studies, and metabolite identification.

The drug discovery process has been time-consuming and expensive in the past. Also, previous drugs were not formulated as well as the drugs being expected and developed using AI and ML technologies. This article shall elaborate on the stages of drug discovery and development where AI and ML modelling have revolutionized the traditional methods of drug development. AI, in the broader context, has been taken up and implemented at each stage of drug discovery, from target identification and validation to hit finding and the progression from hit to lead optimization, with a vital role in streamlining the formerly time-consuming process of drug screening. Several methods based on machine learning are being used to predict drug targets, predict the structures of drug targets, predict binding sites, perform ligand-based similarity searches, design ligands with certain desired properties de novo, develop scoring functions for molecular docking, build Quantitative structure-activity relationship (QSAR) models for the prediction of biological activity, and predict the pharmacokinetic and pharmacodynamic properties of ligands.

ABSTRACTPharmacokinetic studies play an important role in all stages of drug discovery and development. Recent advancements in the tools for discovery and optimization of therapeutic proteins have created an abundance of candidates that may fulfill target product profile criteria. Implementing a set of in silico, small scale in vitro and in vivo tools can help to identify a clinical lead molecule with promising properties at the early stages of drug discovery, thus reducing the labor and cost in advancing multiple candidates toward clinical development. In this review, we describe tools that should be considered during drug discovery, and discuss approaches that could be included in the pharmacokinetic screening part of the lead candidate generation process to de-risk unexpected pharmacokinetic behaviors of Fc-based therapeutic proteins, with an emphasis on monoclonal antibodies.

Reaction phenotyping is a process to identify enzymes responsible for the metabolism of a drug. It allows for the understanding of the role of metabolizing enzymes in drug elimination through quantifying the contribution of major enzymes to the total clearance of a drug. This is an important step in the prediction of pharmacokinetic and drug–drug interactions (DDI) of a new chemical entity (NCE). This chapter discusses the commonly usedin vitrocytochrome P450 (P450) reaction phenotyping assays with highlights on key experimental conditions, main deliverables, and differentiation of each approach. The importance in the translation ofin vitroreaction phenotyping data toin vivois emphasized and highlighted through several published examples. Strategies in the application of CYP reaction phenotyping studies to support clinical pharmacokinetic and DDI assessment at different stages of drug discovery and development are also discussed and recommended.

ABSTRACTThe most strategic weapon in drug discovery in the recent past has been artificial intelligence (AI)—bringing new approaches to one of the toughest areas of the pharmaceutical industry. Various AI approaches such as DL and ML methods utilized in various stages of drug discovery and development including but not limited to virtual screening and target identification are also discussed here. Employing this approach, this review looks at AI programs and platforms that exist in drug discovery today in a bid to outline what a future with AI in this field has in stock. In addition to this, this review does not only give a momentary state of the state of affairs of the AI in the space, but also briefly discusses what is in store next, along with the drawback and the opportunity more so from this perspective.

A sequence of 6-(6-(methylthio)-3-phenyl-[1,2,4]triazolo[3,4-b][1,3,4]thiadiazole [5a], 6-(6-(ethylthio)-3-phenyl-[1,2,4]triazolo[3,4-b][1,3,4]thiadiazole [5b], 3-phenyl-6-(propylthio)-[1,2,4]triazolo[3,4-b][1,3,4]thiadiazole [5c] has been designed. The spectral analyses validated the synthesis of novel 6-(Alkylthio)-3-phenyl-[1,2,4]triazolo[3,4-b][1,3,4]thiadiazole, which was examined using FT-IR, 1H-NMR, 13C-NMR, and mass spectrometry. The ADME profiles demonstrated favorable oral bioavailability and pharmacokinetic characteristics. These parameters were investigated because many synthesized compounds often fail at later stages of drug discovery and development due to poor pharmacokinetic behavior. The SwissADME analysis provided theoretical insights into pharmacokinetics, revealing that most of the designed compounds exhibit high gastrointestinal absorption and do not cross the blood–brain barrier (BBB). This finding is particularly relevant for oral drugs targeting the gastrointestinal system, indicating their potential for good oral bioavailability. Overall, the newly synthesized compounds show promising pharmacokinetic properties, being well absorbed, distributed, metabolized, and excreted according to the SwissADME predictions. Therefore, these compounds may serve as viable candidates for further drug development. An in silico ADME evaluation was also conducted to assess their drug-likeness and support their potential as novel therapeutic agents. Anticancer evaluations demonstrated that the compounds had action against HepG2 and WRL68 cell lines. The synthesized chemical exhibited a pronounced cytotoxic impact on the HepG2 cell line (μg/mL 99.73), whereas little cytotoxicity was detected in normal human cells (WRL68 cell line).

A goal of preclinical discovery is the identification of drug candidates suitable for clinical testing. Successful integration of in vitro and in vivo experimental data sets can afford projections of human dose regimens anticipated to be safe and therapeutically beneficial. While in vitro experiments guide new chemical syntheses and are essential to understanding drug action and disposition, in vivo characterizations provide unique insight into complex biological systems that control concentrations at the site of action and pharmacologic response. Pharmacokinetic and pharmacodynamic (PK/PD) concepts underlying drug disposition and response provide a quantitative framework with which to identify potential clinical candidates. To improve throughput in earlier stages of drug discovery, in vivo pharmacokinetic study designs such as cassette dosing and sparse sampling schemes have been utilized. In later stages of discovery, pharmacokinetic studies using chemical inhibitors or surgical and genetic animal models are used to characterize the underlying determinants of drug disposition. In a complimentary fashion, modeling of in vivo pharmacodynamic effects may quantitatively link biomarkers to pharmacological response, validate in vitro to in vivo correlations and underwrite predictions of efficacious exposure targets. When applied to in vivo discovery data, PK/PD models have aided in understanding mechanisms of pharmacological response such as receptor theory in the central nervous system and cell turnover concepts in infectious disease and oncology. This review considers the role of in vivo testing toward understanding the pharmacokinetic and pharmacodynamic attributes of lead candidates in drug discovery.

An alternative biomimetic tool – Dual hydrophilic/reversed-phase interaction mode

Surface plasmon resonance (SPR) is a powerful technology that allows an information-rich assessment of the interactions of tethered protein molecular targets with their interacting ligands in real time. As such, SPR has proven useful at all stages of drug discovery, from initial hit finding (particularly the screening of fragment libraries) to mechanistic kinetic and thermodynamic profiling during hit validation and lead optimisation phases. Each of these stages of drug discovery involves somewhat different considerations in the optimal design and interpretation of SPR experiments. These factors will be discussed along with other relevant applications of SPR technology for drug discovery, such as in DMPK and toxicology assessments. Finally, the future developments in SPR for drug discovery will be considered.

The prediction of drug-target interactions (DTI) is a crucial preliminary stage in drug discovery and development, given the substantial risk of failure and the prolonged validation period associated with in vitro and in vivo experiments. In the contemporary landscape, various machine learning-based methods have emerged as indispensable tools for DTI prediction. This paper begins by placing emphasis on the data representation employed by these methods, delineating five representations for drugs and four for proteins. The methods are then categorized into traditional machine learning-based approaches and deep learning-based ones, with a discussion of representative approaches in each category and the introduction of a novel taxonomy for deep neural network models in DTI prediction. Additionally, we present a synthesis of commonly used datasets and evaluation metrics to facilitate practical implementation. In conclusion, we address current challenges and outline potential future directions in this research field.

Importance of the field: Water molecules often appear around ligands in protein crystal structures. Reliable prediction of the effects of water on ligand binding remains a challenge. Solvation effects are crucial for lead optimization where a 100-fold difference in binding affinity is significant but correspond to only ∼3 kcal/mol in binding free energy. Well-known examples, such as nonpeptidic urea inhibitors of HIV protease, prove that careful examination of water molecules and their energetics can contribute significantly to a successful drug design campaign.Areas covered in this review: In this review, we examine methods to account for the effect of water in ligand binding at two stages of drug discovery: lead identification via docking calculations and lead optimization. We provide a survey of the models and techniques available to account for water in drug design.What the reader will gain: The reader will become aware of common practices and pitfalls in dealing with water molecules in structure-based drug design.Take home message: Although solvation effects are not fully understood, some pragmatic recommendations at the end of the article provide guidance for modelers in this area as well as new practitioners.

For the pharmaceutical industry, one of the challenges in evaluating the risk of future compound attrition at the discovery stage is the successful prediction of the major routes of clearance in humans. For compounds cleared by metabolism, such information will help to avoid the development of compounds that will exhibit large interpatient differences in pharmacokinetics via 1). routes of metabolism catalyzed by functionally polymorphic enzymes and/or 2). clinically significant metabolic drug-drug interactions, in the later stages of development. The degree of intersubject variability that is acceptable for a drug candidate is uncertain in the discovery stage where knowledge of other important factors is limited or unavailable (i.e. therapeutic index, pharmacodynamic variability, etc). Reaction phenotyping is the semi-quantitative in vitro estimation of the relative contributions of specific drug-metabolizing enzymes to the metabolism of a test compound. However, reaction phenotyping in the discovery stage of drug development is complicated by the absence of radiolabelled parent compound or metabolite bioanalytical standards relative to later stages of development. In this commentary, some of the approaches, based on published data, which can be taken to overcome these challenges are discussed. In addition, knowledge of the molecular structure (i.e. specific chemical substituents), physicochemical properties, and routes of clearance in animals can all help in making a successful prediction for the routes of clearance in humans. In combination, the objective of these studies should be to reduce to a minimum the risk of finding significant inter-patient differences in pharmacokinetics at a later stage in development due to significant metabolism by polymorphic enzymes or drug-drug interactions. Consequently, this data should be used to avoid costly late stage attrition.

Drug discovery is a multifaceted and resource-intensive process traditionally marked by long development timelines, high financial costs, and significant attrition rates. The integration of Artificial Intelligence (AI) has introduced transformative opportunities to overcome these barriers by enhancing data-driven decision-making, accelerating molecular design, and improving clinical translation. This chapter explores the foundations, applications, and case studies of AI across all stages of drug discovery - from target identification and validation to lead optimization and clinical development. Real-world examples, including BenevolentAI, Atomwise, Insilico Medicine, and the Pfizer–IBM Watson collaboration, illustrate AI’s role in accelerating drug design, repurposing, and optimization. Despite its promise, challenges such as data quality, model interpretability, regulatory uncertainty, and intellectual property disputes continue to impede widespread adoption. The chapter concludes by emphasizing the future directions of AI in synergy with quantum computing, multi-omics integration, and personalized medicine. Overall, AI represents a paradigm shift in pharmaceutical innovation, poised to make drug discovery faster, more precise, and more cost-effective while promoting a human–AI collaborative ecosystem for sustainable biomedical advancement.