7196 research outputs found
Sort by
Machine Learning for Toxicity Prediction Using Chemical Structures: Pillars for Success in the Real World.
No full textMachine learning (ML) is increasingly valuable for predicting molecular properties and toxicity in drug discovery. However, toxicity-related end points have always been challenging to evaluate experimentally with respect to in vivo translation due to the required resources for human and animal studies; this has impacted data availability in the field. ML can augment or even potentially replace traditional experimental processes depending on the project phase and specific goals of the prediction. For instance, models can be used to select promising compounds for on-target effects or to deselect those with undesirable characteristics (e.g., off-target or ineffective due to unfavorable pharmacokinetics). However, reliance on ML is not without risks, due to biases stemming from nonrepresentative training data, incompatible choice of algorithm to represent the underlying data, or poor model building and validation approaches. This might lead to inaccurate predictions, misinterpretation of the confidence in ML predictions, and ultimately suboptimal decision-making. Hence, understanding the predictive validity of ML models is of utmost importance to enable faster drug development timelines while improving the quality of decisions. This perspective emphasizes the need to enhance the understanding and application of machine learning models in drug discovery, focusing on well-defined data sets for toxicity prediction based on small molecule structures. We focus on five crucial pillars for success with ML-driven molecular property and toxicity prediction: (1) data set selection, (2) structural representations, (3) model algorithm, (4) model validation, and (5) translation of predictions to decision-making. Understanding these key pillars will foster collaboration and coordination between ML researchers and toxicologists, which will help to advance drug discovery and development
Improving accuracy in food effect Predictions: Application of In-Vitro absorption experiments as a useful tool for the evaluation of ten drug products.
No full textPredicting the magnitude and direction of food effects on oral drug delivery can be challenging, especially for compounds with absorption limited by changes in the permeation rate. Currently available in-vitro tools assess the impact of increased bile flow and food on drug solubilization, potentially leading to increased absorption under fed conditions. However, the presence of bile can sequester the drug within bile/food colloids, reducing free drug availability and resulting in unanticipated absorption. The aim of this study is to explore the application and outcome of a combined dissolution/permeation (MacroFLUX™) assay of ten drug products for a more accurate prediction of clinical food effects in the context of given dose and formulation. The ratio of the fed-to-fasted dissolution and Flux were used to correlate each experimental model to clinical food effect in humans. Assessing the flux across a biomimetic artificial membrane provided superior predictability over dissolution alone. Food effects were predicted accurately for 60% of compounds within 1.25-fold based on flux analysis, while dissolution analysis only predicted 30% of compounds evaluated. The most interesting outcome is that dissolution did not pick up on any of the negative food effects. Notably, the study revealed that the common assumption of compounds exhibiting a positive food effect due to increased dissolution/solubility from fasted to fed state does not always hold true. This in-vitro absorption experiment proved to be a valuable in-vitro biopharmaceutic tool that can predict clinical food effects, support (pre-)formulation development, and guide the design of dedicated clinical pharmacology studies
Proceedings of the 15th European Immunogenicity Platform Open Symposium on Immunogenicity of Biopharmaceuticals
No full textThis year, the European Immunogenicity Platform celebrated the 15th edition of its Open Symposium on Immunogenicity of Biopharmaceuticals and its associated one-day workshop. The meeting attracted experts and newcomers across industry, regulatory agencies and academia, who actively participated in 3 days of discussion on risk assessment, monitoring and mitigation of unwanted immunogenicity of biologics. Besides oral presentations, poster sessions were held to maximize scientific exchange and networking opportunities. Here, we report the discussions that took place April 22-24 in Lisbon
Evaluating teratoma formation risk of pluripotent stem cell-derived cell therapy products: a consensus recommendation from the Health and Environmental Sciences Institute's International Cell Therapy Committee.
No full textHuman pluripotent stem cells (hPSCs) can differentiate into any cell of choice and hold significant promise in regenerative medicine and for treating diseases that currently lack adequate therapies. However, hPSCs are intrinsically tumorigenic and can form teratomas. Therefore, the presence of residual undifferentiated hPSCs must be rigorously assessed using sensitive methodologies to mitigate the potential tumorigenicity risks of hPSC-derived cell therapy products (CTPs). In this comprehensive review, we describe methods for detecting residual undifferentiated hPSCs and discuss the relative value of current in vitro assays versus conventional in vivo assays. We highlight that in vitro assays such as digital PCR detection of hPSC-specific RNA and the highly efficient culture assay, have superior detection sensitivity. Additionally, we outline important considerations for validating in vitro assays when applying them to assess each product. This article lays the groundwork for guiding internationally harmonized procedures for evaluating the potential teratoma formation risk of hPSC-derived CTPs and increasing confidence in the safety of these products
Discovery and Optimization of Selective Inhibitors of Large Tumor Suppressor Kinases LATS1 and 2 for In Vivo Investigation of the Hippo-YAP Pathway in Tissue Regeneration.
No full textLarge Tumor Suppressor kinases LATS1 and 2 (LATS1/2) are serine/threonine kinases and core regulators of the Hippo-YAP pathway. Inhibition of LATS1/2 promotes nuclear translocation of nonphosphorylated YAP, thereby initiating a downstream cascade promoting cell proliferation. We set out to investigate the potential of LATS inhibition as a therapeutic approach to enhance tissue regeneration and hereby report a structure-guided optimization of screening hit 1 for potency, binding efficiency, and physicochemical properties, leading to a highly selective, cellularly active, and orally available tool compound 7 (NIBR-LTSi) that demonstrated ex vivo target engagement and in vivo YAP target gene activation in rodents
The changing landscape of medicinal chemistry optimization.
No full textThe goal of a small-molecule drug discovery campaign is the development of chemical entities that fulfil the criteria of the target product profile for progression into clinical trials. This objective is realized through multiparameter medicinal chemistry optimization, typically by identifying the compounds at the hit stage with molecular properties that provide a high chance of subsequent success, and then iteratively optimizing the properties, often in parallel, to identify leads and, ultimately, drug candidates. To assess the impact of medicinal chemistry optimizations on molecular properties, a set of new drug candidates reported in the literature between 2015 and 2022, and their corresponding hit and lead compounds, were analysed, and compared with a set of drug candidates identified between 2000 and 2010, and their corresponding hits and leads. This analysis was complemented by similar analyses of the internal medicinal chemistry programmes pursued at AstraZeneca and Novartis. Here, we highlight and discuss the implications of the observed trends, which include shifts in key physicochemical properties and strategic changes in medicinal chemistry programmes
InterFace/Off: characterization of competitive adsorption of novel surfactants and proteins at the solid-liquid and oil-liquid interfaces.
No full textInterfacial stress encountered by biopharmaceuticals is often opposed by employing surfactants in their formulations. Surfactants protect proteins from this stress by either shielding the interface or displacing adsorbed proteins. Most previous studies were dedicated to the air-liquid interface, and to characterize monoclonal antibodies or non-pharmaceutically relevant proteins in combination with established surfactants (polysorbates and poloxamer 188). Herein, we employ quartz crystal microbalance with dissipation (QCM-D) and tensiometry to investigate the adsorption behavior of established surfactants as well as the novel surfactants VEDS and VEDG-3.3 at the solid-liquid and silicon oil-liquid interfaces in presence and absence of three model biotherapeutics of different modalities. Our study shows that the individual adsorption behavior is molecule-dependent, as expected. When mixed either simultaneously (co-adsorption) or sequentially (shielding and displacement), both proteins and surfactants were detected to co-adsorb at the interface. Compared to the established surfactants, VEDS and VEDG-3.3 showed a slower adsorption followed by molecular rearrangements that resulted in a denser packing, supporting the mechanistic explanation of their favorable protein stabilization effect previously reported. Collectively, our results support the generation of a unified thermodynamic description of the adsorption of protein-surfactants mixtures in pharmaceutically-relevant conditions
Effect of Neprilysin Inhibition on Plasma Proteins in Heart Failure With Mildly Reduced or Preserved Ejection Fraction.
No full textno abstrac
Normalization of Cerebral Hemodynamics After Gene Therapy in Adults With Sickle Cell Disease.
No full textNo abstrac
Multiparametric grading of glaucoma severity by histopathology can enable post-mortem substratification of disease state.
No full textNeurodegeneration in glaucoma patients is clinically identified through longitudinal assessment of structure-function changes, including intraocular pressure, cup-to-disc ratios from fundus images, and optical coherence tomography imaging of the retinal nerve fiber layer. Use of human post-mortem ocular tissue for basic research is rising in the glaucoma field, yet there are challenges in assessing disease stage and severity, since tissue donations with informed consent are often unaccompanied by detailed pre-mortem clinical information. Further, the interpretation of disease severity based solely on anatomical and morphological assessments by histology can be affected by differences in death-to-preservation time and tissue processing. These are difficult confounders that cannot be easily controlled. As pathogenesis and molecular mechanisms can vary depending on the stage and severity of glaucoma, there is a need for the field to maximize use of donated tissue to better understand the molecular mechanisms of glaucoma and develop new therapeutic hypotheses. Further, there is a lack of consensus around the molecular RNA and protein markers that can be used to classify glaucoma severity. Here, we describe a multiparametric grading system that combines structural measurements of the retinal nerve fiber layer with linear regression and principal component analyses of molecular markers of retinal ganglion cells and glia (RBPMS, NEFL, IBA1 and GFAP) to stratify post-mortem glaucoma eyes by the severity of disease. Our findings show that a quantitative grading approach can stratify post-mortem glaucoma samples with minimal clinical histories into at least three severity groups and suggest that this type of approach may be useful for researchers aiming to maximize insights derived from eye bank donor tissue