Published in last 50 years
Articles published on Potent Kinase Inhibitor
- New
- Research Article
- 10.1161/circ.152.suppl_3.4369238
- Nov 4, 2025
- Circulation
- Sabindra Pradhananga + 8 more
Introduction: SGLT2 inhibitors improved clinical outcomes in patients with heart failure (HF), highlighting metabolism-related pathways as a therapeutic target. Genetic inhibition of SGK1, a stress activated kinase, has been previously shown to be cardioprotective, identifying SGK1 as a promising molecular target in HF. Research Questions: Evaluate the efficacy of a potent, selective SGK1 inhibitor (THRV-1268), and empagliflozin (EMPA), and their combination in a thoracic aortic constriction (TAC) model of HF. Methods: Male Sprague Dawley rats (5 weeks old) that underwent either TAC or sham surgery were dosed once daily via oral gavage from Day 3 until the end of the study (Day 56) with vehicle, THRV-1268 (30 mg/kg/day), EMPA (10 mg/kg/day), or THRV-1268 and EMPA (COM). Cardiac function was serially assessed by echocardiography to monitor disease progression. Tissues were stained by picrosirius red, and collagen was quantified using ImageJ software. Results: Compared to sham, TAC rats showed impaired systolic function marked by reduced LVEF (Figure 1), and CO (103.17 ± 14.89 mL/min vs 69.37± 7.57 mL/min, p≤ 0.001), and increased LVESV (Figure 3). Ventricular hypertrophy was also observed, with elevated PWT-ED (Figure 4) and heart weight/tibia length (33.79 ± 0.0 mg/mm vs 42.91 ± 0.0 mg/mm, p=0.3). EMPA improved LVEF at Day 21 (Figure 1), but this effect was not sustained on Day 56 (Figure 1 and 2) with similar trends observed for other functional parameters. Interestingly, THRV-1268 and COM, produced sustained protection from Day 21 onward. At Day 56, higher LVEF (Figure 1), and CO (THRV-1268: 90.98 ± 4.49 mL/min, COM: 89.48 ± 10.93 mL/min, p ≤ 0.01), with lower LVESV (Figure 3) and PWT-ED (Figure 4) were observed. HW/TL also trended to lower (THRV-1268: 36.65 ± 0.0 mg/mm, COM: 37.38 ± 0.0 mg/mm). Histological analysis revealed increased fibrosis in vehicle-treated rats compared to sham, with reduced levels in treatment groups, though not statistically significant. Conclusions: THRV-1268 demonstrated superior cardioprotective effects compared to EMPA and in combination with EMPA in the TAC-model of HF. These data support continued development of THRV-1268 as a single or additive therapeutic approach for the optimal management of HF.
- New
- Research Article
- 10.1016/j.rechem.2025.102727
- Nov 1, 2025
- Results in Chemistry
- Jay B Maheta + 7 more
“Triazole-oxadiazole-based schiff bases as potential kinase inhibitors: synthesis, DFT studies, molecular docking and anticancer evaluation”
- New
- Research Article
- 10.1016/j.molstruc.2025.143117
- Nov 1, 2025
- Journal of Molecular Structure
- Samir Hmaimou + 9 more
One-pot synthesis via [3+2] cycloaddition reaction, structural analysis, and computational insights of obtained oxadiazole-benzodiazepine derivatives as potential kinase inhibitors
- New
- Research Article
- 10.1016/j.molstruc.2025.144652
- Nov 1, 2025
- Journal of Molecular Structure
- Salima Mosbah + 4 more
Synthesis, Crystal Structure, DFT Analysis, and Molecular Docking Studies of a Novel Thiophene- Containing Acrylonitrile Schiff Base as a Potential Tyrosine Kinase Inhibitor
- New
- Research Article
- 10.2174/0115734064310145240822060730
- Nov 1, 2025
- Medicinal chemistry (Shariqah (United Arab Emirates))
- Haruna Isiyaku Umar + 10 more
Oxidative stress is strongly linked to neurodegeneration through the activation of c-Abl kinase, which arrests α-synuclein proteolysis by interacting with parkin interacting substrate (PARIS) and aminoacyl tRNA synthetase complex-interacting multifunctional protein 2 (AIMP2). This activation, triggered by ataxia-telangiectasia mutated (ATM) kinase, leads to dopaminergic neuron loss and α-synuclein aggregation, a critical pathophysiological aspect of Parkinson's disease (PD). To halt PD progression, pharmacological inhibition of c-Abl kinase is essential. Despite three generations of tyrosine kinase inhibitors (TKIs) being explored for PD treatment, they present significant concerns including poor blood-brain barrier penetration, off-target effects, and severe side effects. Notably, there are currently no FDA-approved c-Abl kinase inhibitors in clinical usage for PD treatment, highlighting the urgent need for potent, safe, and cost-effective alternatives. This study aims to identify potential c-Abl kinase inhibitors from plant-derived compounds with reported anti-Parkinson's potential and their derivatives using molecular docking, molecular dynamics simulations (MDS), and in silico pharmacokinetics and toxicity profiling. Seventy-eight compounds sourced from literature were docked against c-Abl kinase using Maestro 12.5. The top three hit compounds, along with nilotinib (control drug), were subjected to drug-likeness, ADMET profiling using the AI Drug Lab server and 100 ns MDS using Desmond. Amburoside A, diarylheptanoid MS13, and dimethylaminomethyl-substituted-curcumin showed binding affinities close to nilotinib, with values of -12.615, -12.556, and -11.895 kcal/mol respectively, compared to nilotinib's -16.826 kcal/mol. The three plant-derived compounds exhibited excellent structural stability and favorable ADMET profiles, including optimal blood-brain barrier permeation Conclusion: The three hit compounds identified in this study show potential as c-Abl kinase inhibitors. Given the absence of FDA-approved c-Abl kinase inhibitors for PD, these findings are significant as they could contribute new therapeutic options for the treatment and management of PD. However, further in vitro and in vivo experiments are necessary to validate these findings.
- New
- Research Article
- 10.1007/s12010-025-05375-1
- Oct 28, 2025
- Applied biochemistry and biotechnology
- Amal Kadimi + 12 more
Inhibiting c-Abl kinase pharmacologically is necessary because of its role in oxidative stress and neurodegeneration. When activated, it causes the accumulation of α-synuclein and dopaminergic neuron damage, leading to Parkinson's disease (PD). Reports of the effectiveness of c-Abl inhibitors repurposed for PD were accompanied by both hope and numerous concerns. Therefore, there is an urgent need for alternative c-Abl inhibitors. We employed a machine-learning-based QSAR model to identify potential actives against c-Abl kinase, screening selected FDA-approved and phase 1 drugs; optimizing the compounds' structures through bioisostere replacement; conducting molecular docking algorithms (HTVS, SP, XP, and Prime's Molecular Mechanics with Generalized Born and Surface Area (Prime-MMPBSA)); in silico pharmacokinetic profiling; and structural stability and dynamics studies for 200ns. From 3605 drugs and 1456 bioisosteres, two bioisosteres of indobufen (indobufen 25 and 22) showed promising potential against c-Abl kinase. As the two bioisosteres returned the closest docking scores (14.880 and - 14.265kcalmol-1, respectively) to the control drug (nilotinib, - 15.312kcalmol-1), the Prime-MMGBSA calculations returned - 81.92 and - 84.07kcalmol-1, respectively; MMPBSA calculations after a 200-ns MD simulation run returned - 48.20 ± 3.69kcalmol-1 and - 49.94 ± 3.05kcalmol-1, respectively. This indicates their stability compared to other test compounds, as supported by the RMSD, RMSF, PCA, and DCCM results. Finally, both bioisosteres interacted with MET 318, ASP 381, TYR 253, ALA 269, and PHE 317 in the c-Abl active site. We present these bioisosteres as potential candidates for the treatment or management of PD targeting c-Abl kinase. However, in vitro and in vivo experiments to validate the findings are urgently required.
- New
- Research Article
- 10.1007/s10822-025-00690-5
- Oct 25, 2025
- Journal of computer-aided molecular design
- Ankit Kumar Singh + 10 more
BRAF mutations were first discovered by Davies et al. in 2002. BRAFV600E mutation is the most prevalent, accounting for approximately 90% of all BRAF mutations. BRAFV600E mutations have been identified at varying frequencies across multiple human cancers, including malignant melanoma (70-90%), thyroid cancer (45-50%), colorectal cancer (5-20%), and others. In this study, we designed a series of pyrimidine-sulfonamide hybrids, inspired by first- and second-generation FDA-approved BRAF inhibitors such as sorafenib, dabrafenib, and vemurafenib. The designed compounds were intended to target the αC-OUT/DFG-IN conformation of the BRAFV600E mutant protein. Eighteen compounds (B1-B18) were synthesized and characterized using spectral techniques. Molecular docking and MD simulations were carried out to assess their binding affinity and stability with the BRAFV600E protein. Kinase inhibition was assessed using a BRAFV600E specific assay, and anticancer activity was tested against HCT-116, A375, HT-29, and TPC-1 cell lines. Among the tested derivatives, B14 exhibited the highest cytotoxicity against HCT-116, B8 was most effective against A375, B18 showed potent inhibition in HT-29, and B3 demonstrated the strongest activity in TPC-1 cells. All four compounds exhibited activity comparable to sorafenib. Notably, B4 emerged as the most potent BRAFV600E kinase inhibitor in assays.
- New
- Research Article
- 10.1002/ajoc.70235
- Oct 25, 2025
- Asian Journal of Organic Chemistry
- Yun A Yum + 5 more
Abstract The small nucleoside molecule LJ‐4827 exhibited potent kinase inhibition and in vitro anticancer activity. Driven by bio‐isosteric principles, we synthesized LJ‐5157, which also demonstrated high kinase selectivity and anticancer efficacy. To further expand the structural diversity of bioactive nucleoside analogs, we developed a metal‐free, late‐stage strategy for converting terminal alkynes to nitriles using tert ‐butyl nitrite ( t BuONO). By substituting 2‐picoline N ‐oxide with N ‐methylmorpholine N ‐oxide (NMO) as the oxidant, we accomplished efficient nitrile formation in nucleoside scaffolds. This approach broadens the scope of alkyne cleavage chemistry and offers a practical route for analog diversification.
- Research Article
- 10.1016/j.ejmech.2025.117885
- Oct 1, 2025
- European journal of medicinal chemistry
- Lidong Gong + 8 more
Molecular hybridization-driven FAK inhibitors: N-2,4-diarylaminopyrimidine-3-sulfamoyl-benzamide derivatives with improved antitumor potency.
- Research Article
- 10.2174/0118756921378958250910193114
- Sep 23, 2025
- Current Pharmacogenomics and Personalized Medicine
- Sonali S Shinde + 2 more
Introduction: Cancer remains one of the most prevalent and deadly illnesses, even with notable advancements in novel treatment alternatives. One important strategy for targeted cancer therapy is the EGFR (epidermal growth factor receptor) signaling pathway. Cancer treatment has greatly improved when the EGFR-driven pathway is inhibited by targeting the tyrosine kinase domain of EGFR. Several small compounds, particularly quinazoline-containing derivatives, have been developed using simulation studies to determine EGFR tyrosine kinase inhibitors (TKIs). The design process also assessed the appearance of epigenetic alterations and resistance issues, which limited the effectiveness of medications over time and clarified the necessity for additional research in this area. In recent decades, extensive research has investigated the genetic alterations occurring in the EGFR tyrosine kinase domain. These alterations have paved the way for the development and production of highly potent and efficacious inhibitors. Methods: This review highlights the structure-activity relationship (SAR) and biological activity of different quinazoline derivatives that have been reported to have EGFR-TK inhibitory antiproliferative activity. We searched Embase, Cochrane, and PubMed for literature on quinazoline derivatives showing EGFR inhibition as anticancer agents. Results: We confirmed that quinazoline derivatives with different substitutions are useful pharmacophores as EGFR TK inhibitors for achieving strong anticancer activity after a thorough review of the literature. Discussion: This review offers insights into the latest advancements in developing novel and potent quinazoline derivatives as potential EGFR tyrosine kinase inhibitors, which could be further optimized to develop new EGFR inhibitors in the future. Conclusion: Quinazoline-based scaffolds remain promising leads for developing nextgeneration EGFR tyrosine kinase inhibitors with enhanced anticancer efficacy.
- Research Article
- 10.1021/acs.jmedchem.5c00107
- Sep 11, 2025
- Journal of medicinal chemistry
- Liang Long + 10 more
This work describes the discovery of a new series of Aurora kinase inhibitors based on quinazoline skeleton derived from ENMD-2076, as well as the first X-ray cocrystal structure complexes of vinyl-quinazoline 9h with Aurora A. Replacing pyrimidine with quinazoline improved anticancer activity and facilitated cocrystal formation. Compounds 9a and 9h showed excellent Aurora A kinase inhibition, with IC50 values of 6.0 and 2.8 nM, respectively. 9h demonstrated superior activity against TNBC MDA-MB-231 cells with an IC50 value of 48 nM and achieved 59% tumor growth inhibition in xenograft models, vs ENMD-2076's 33% with no observable toxicity. Mechanistic studies using immunoblotting, immunofluorescence staining, and flow cytometry showed that 9h outperforms ENMD-2076 in inhibiting Aurora A kinase activation, preventing spindle formation, arresting the cell cycle, and inducing cell apoptosis. Thus, 9h has the potential for further optimization and is a promising anticancer drug candidate.
- Research Article
- 10.1080/10426507.2025.2558683
- Sep 9, 2025
- Phosphorus, Sulfur, and Silicon and the Related Elements
- Israa Taha + 3 more
Research into breast cancer treatments is progressing and contemporary, particularly with a focus on tyrosine kinase receptors, which are crucial for regulating various cellular functions. The chemistry of the 2-pyridone scaffold allows for the creation of diverse molecular structures. Converting 2-pyridone into a hydrazide form enhances its biological activity. Using polyphosphoric acid enables effective electrophilic attacks, resulting in various polycyclic compounds as naphthyridine and pyridopyrimidine. These compounds are being evaluated as potential tyrosine kinase inhibitors, refined through the “rule of five” to improve their drug-like properties. In assessing their cytotoxicity, the MTT assay was performed on MCF-7 cell lines using different concentrations. Compounds 2, 4, 8, 10, 12, and 16 have demonstrated notable IC50 values, surpassing the efficacy of doxorubicin, which has an IC50 value of 11.86 µM. Among these compounds, 2, 4, and 8 are particularly distinguished by their favorable molecular docking simulation profiles. These compounds represent promising candidates for further investigation in the treatment of breast cancer.
- Research Article
- 10.1016/j.bpc.2025.107470
- Sep 1, 2025
- Biophysical chemistry
- Beom Soo Kim + 1 more
Identification of potential Abl kinase inhibitors using virtual screening and free energy calculations for the treatment of chronic myeloid leukemia.
- Research Article
- 10.4143/crt.2025.279
- Aug 20, 2025
- Cancer research and treatment
- Chaelin Lee + 8 more
Non-small cell lung cancer (NSCLC) frequently harbors targetable EGFR mutations. However, rare variants such as EGFR L858M or L861R remain poorly characterized. This study aimed to elucidate the oncogenic potential and EGFR tyrosine kinase inhibitors (TKIs) sensitivity of the EGFR L858M/L861R mutation to inform personalized treatment strategies. Tumor samples from a NSCLC patient were analyzed using targeted panel sequencing and confirmed with the FoundationOne Liquid CDx assay. EGFR-mutant constructs, including L858M, L858R, L861R, L861Q, L858M/L861R, and L858R/L861Q, were generated and transduced into various cell lines. Cell viability, immunoblot, and soft agar colony formation assays were conducted to assess the oncogenicity and drug sensitivity, while computational protein modeling and docking simulations evaluated the drug-binding affinities of EGFR TKIs. Ba/F3 cells expressing the EGFR L858M/L861R mutation exhibited robust IL-3-independent proliferation accompanied by markedly increased EGFR phosphorylation, while NIH-3T3 cells showed anchorage-independent colony formation. Compared to other mutations, cells expressing EGFR L858M/L861R mutation were less sensitive to first-generation EGFR TKIs (gefitinib, erlotinib) and third-generation EGFR TKIs (osimertinib, lazertinib), whereas second-generation EGFR TKIs (afatinib, poziotinib) demonstrated potent inhibitory effects. Computational modeling revealed a narrower drug-binding efficiency of first-generation inhibitors. The EGFR L858M/L861R mutation drives strong oncogenic signaling and exhibits preferential sensitivity to second-generation EGFR TKIs. These findings underscore the importance of accurate molecular diagnosis for guiding effective, personalized therapeutic strategies in NSCLC.
- Research Article
- 10.3390/biom15081180
- Aug 18, 2025
- Biomolecules
- Jixia Yang + 5 more
This study reports the rational design and systematic evaluation of a novel series of 2-substituted aniline pyrimidine derivatives as dual Mer/c-Met inhibitors. Among the synthesized compounds, 17c demonstrated potent dual kinase inhibition, with IC50 values of 6.4 ± 1.8 nM (Mer) and 26.1 ± 7.7 nM (c-Met). The compound exhibited significant antiproliferative activity across multiple cancer cell lines (HepG2, MDA-MB-231, and HCT116), while showing minimal hERG channel inhibition (IC50 > 40 μM), indicating favorable cardiac safety. Pharmacokinetic profiling revealed high metabolic stability in human liver microsomes (t1/2 = 53.1 min) and moderate oral bioavailability (F: 45.3%), with strong plasma protein-binding affinity (>95%). Mechanistic studies further demonstrated that 17c dose-dependently suppressed HCT116 cell migration and induced apoptosis. These integrated pharmacological properties position 17c as a promising therapeutic candidate for dual Mer/c-Met drive malignancies.
- Research Article
- 10.1080/14756366.2025.2538673
- Aug 5, 2025
- Journal of Enzyme Inhibition and Medicinal Chemistry
- Chi-Hsiu Chung + 5 more
Pancreatic cancer is among the most lethal malignancies, with a five-year survival rate of only 6%. For patients with metastatic disease, current treatments extend median survival by merely four months. This study addresses the urgent need for targeted therapies, as no specific drugs are currently available. Clinical analyses revealed significantly elevated RSK2 expression in pancreatic cancer tissues, associated with shorter survival. We aimed to identify a novel RSK2 inhibitor for metastatic pancreatic cancer. Through structure-based virtual screening, we identified NSYSU-115 as a promising candidate with an IC50 of 45.5 nM. At low concentrations, NSYSU-115 significantly suppressed colony formation, while higher concentrations reduced cell viability and proliferation. It also inhibited phosphorylation of IκBα, a known RSK2 substrate, in a dose- and time-dependent manner. Furthermore, NSYSU-115 impaired cell migration and altered epithelial-mesenchymal transition (EMT) markers. These findings highlight NSYSU-115 as a potent kinase inhibitor with promising therapeutic potential for pancreatic cancer treatment.
- Research Article
- 10.1002/prp2.70153
- Aug 1, 2025
- Pharmacology research & perspectives
- Xiaochen Zhang + 6 more
Tropomyosin receptor kinase (Trk) inhibitors are an essential class of anticancer drugs treating NTRK gene fusions-positive cancer. However, the potential for the emergence of on-target resistance suggests newer Trk inhibitors with low drug resistance risk are needed. LPM4870108 is a novel Trk inhibitor with robust anticancer efficacy in preclinical studies. To support future clinical development, this study systematically evaluated the acute and subacute (4-week) toxicity, toxicokinetic, genotoxic, and safety pharmacology of LPM4870108. The acute toxicity study revealed the maximum tolerated dose of LPM4870108 was 300 mg/kg, whereas subacute studies determined its STD10 in rats was 10 mg/kg/day. The toxicological effects of LPM4870108 were consistent with its pharmacodynamic efficacies as a Trk inhibitor, including corneal inflammation, splenic lymphocytopenia, hepatocyte vacuolar degeneration, scab formation, and increased food consumption and body weight. These changes were partially or fully recovered after 4 weeks of recovery. In rats treated with 10 or 20 mg/kg/day, 2/30, or 6/30 rats died or were moribund, and the primary organs affected by treatment-related toxicity included the eyes, liver, and skin. Rat toxicokinetic findings were consistent with a dose-dependent effect of LPM4870108. There was no evidence of LPM4870108-related genotoxicity, nor did it affect respiratory function or neurobehavioral activity in rats or blood pressure or electrocardiogram results in rhesus monkeys. The IC50 of LPM4870108 for hERG current inhibition was 18.2 μM. Together, these results demonstrate that LPM4870108 exhibits a satisfactory safety profile which is appropriate for further clinical development.
- Research Article
- 10.3389/fimmu.2025.1643676
- Jul 31, 2025
- Frontiers in immunology
- Chunmei Ren + 10 more
Apical periodontitis, caused by bacterial infection through the root canals, is characterized by chronic inflammation and bone resorption around the root apex. Metformin, a first-line therapeutic drug for type 2 diabetes mellitus, has attracted attention for its potential anti-inflammatory properties and role in regulating bone homeostasis. The hypothesis in this study was that metformin inhibits bone destruction in apical periodontitis by suppressing macrophage-mediated inflammatory responses. The aim of this study was to evaluate the effect of systemic metformin administration on experimentally induced apical periodontitis development in an animal model and clarify the underlying anti-inflammatory mechanism of metformin in lipopolysaccharide-stimulated mouse macrophages. Evaluations on the effects of metformin on the progression of periapical lesions were conducted in experimentally induced mouse apical periodontitis in vivo, and its anti-inflammatory effects in lipopolysaccharide-stimulated RAW264.7 macrophages in vitro were analyzed. Metformin significantly reduced periapical bone destruction on postoperative days 21 and 28, and decreased the number of osteoclasts on the periapical alveolar bone on postoperative day 28. It also suppressed pro-inflammatory cytokine expression and nuclear factor kappa B signaling in lipopolysaccharide-stimulated RAW264.7. RNA-sequencing data revealed the downregulation of the mammalian target of rapamycin signaling after metformin treatment, which was confirmed by the downregulation of the mammalian target of rapamycin phosphorylation by metformin. Furthermore, metformin activated adenosine monophosphate-activated protein kinase, a potent negative regulator of mammalian target of rapamycin complex 1. The suppression of inflammatory cytokine expression by metformin was abolished by compound C, a potent adenosine monophosphate-activated protein kinase inhibitor. This study revealed that metformin suppressed inflammatory bone destruction in periapical lesions. The mechanism partially involves inhibiting the mammalian target of rapamycin/nuclear factor-kappa B signaling in macrophages through adenosine monophosphate-activated protein kinase signaling activation. Findings from this study show that metformin has therapeutic potential in inflammatory bone destruction, such as apical periodontitis.
- Research Article
- 10.54105/ijpmh.d1079.05050725
- Jul 30, 2025
- International Journal of Preventive Medicine and Health
- Vinit Rajiv Yedatkar + 2 more
Accelerated development in artificial intelligence (AI). The phrase has encouraged advancements in drug discovery and development. In this study, we probe the constraints of AI models—AlphaFold, AtomNet, and Insilico GANs—on predictive precision and cross-therapeutic generalizability. We propose HybridAI, a hybrid AI framework that combines geometric deep learning (GDL), reinforcement learning (RL), and federated learning (FL) for improved predictive modelling of drug-target interactions. They were evaluated against metrics such as ROCAUC, RMSD, and hit-rate accuracy across four therapeutic categories: oncology, antimicrobial resistance, neurodegenerative disease, and autoimmune disease. HybridAI was implemented and validated on a dataset of 150 structurally diverse compounds from ChEMBL and DrugBank. The model outperformed current AI frameworks, achieving 92 parcent accuracy in predicting drug-kinase interactions, with a 34 parcent reduction in toxicity prediction error compared to conventional ADME models. A case study involving non-small cell lung cancer (NSCLC) illustrated the in vitro applicability of Hybrid AI. The system correctly identified afatinib as a potent kinase inhibitor, with a predicted binding affinity of 89 parcent. The prediction was confirmed by molecular docking and in vitro assays within 14 days. Our findings highlight the limitations of single-purpose AI models and underscore the need for hybrid systems, such as Hybrid AI, to enhance precision, flexibility, and scalability. The research supports the use of advanced learning methodologies to facilitate personalised medicine and expedite the drug development process. By integrating various AI methods, HybridAI raises the bar for intelligent drug discovery architectures. The rapid growth of artificial intelligence (AI) in drug discovery necessitates a critical evaluation of its predictive validity and therapeutic applicability. The current study aims to compare the predictive performance of different AI-based models for predicting the success of drug therapy and to introduce a novel combinational AI method, HybridAI, to enhance predictive strength and cross-therapeutic applicability. Seven AI models, such as AlphaFold, AtomNet, and Insilico GANs, were thoroughly assessed for drug efficacy, toxicity, and binding affinity prediction in four disease areas: oncology, antimicrobial resistance, neurodegenerative disorders, and autoimmune diseases. Normalized metrics such as receiver operating characteristic (ROC-AUC), root mean square deviation (RMSD), and hit-rate accuracy were used to evaluate the models. HybridAI, a new combinational model incorporating geometric deep learning GDL, reinforcement learning RL, and federated learning FL, was tested on a 150-structurally different compound dataset that was extracted from ChEMBL and DrugBank. Comparative analysis revealed that the existing AI models are 78– 85 parcent accurate in target-specific drug design but show extreme variability (12–28 parcent) in cross-therapeutic generalizability. Hybrid AI outperformed individual models by achieving 92 parcent drug-kinase interactions (compared to 79 parcent with AlphaFold) and a 34 parcent reduction in errors in toxicity prediction compared to conventional ADMET predictors. HybridAI was cross-validated through a case study by repurposing kinase inhibitors for non-small cell lung cancer (NSCLC), with a correct prediction of afatinib based on 89 parcent binding affinity, and subsequently confirmed in vitro within 14 days. The findings highlight the limitations of single AI models for drug discovery and underscore the importance of hybrid AI architectures in delivering greater predictive reliability. By utilising multi-modal learning frameworks, Hybrid AI provides an open and adaptable infrastructure that facilitates the acceleration of precision medicine, reduces inefficiencies in drug development, and personalises therapeutic strategies.
- Research Article
- 10.1134/s1068162025601648
- Jul 28, 2025
- Russian Journal of Bioorganic Chemistry
- E V Koroleva + 7 more
Chimeric Amides of Substituted Allyl- and Phenylcarboxylic Acids with Pharmacophoric Moieties of Aromatic and Heteroaromatic Rings as Potential Multitarget Protein Kinase Inhibitors: Design, Synthesis, Evaluation of Antitumor Activity, and In Silico Analysis