Efficacy and safety of branebrutinib (BMS-986195), an irreversible Bruton's tyrosine kinase inhibitor, for the treatment of rheumatoid arthritis: a phase 2a, randomised, double-blind, placebo-controlled study

 Østergaard M, Haavardsholm E, Nowak M et al.

The Lancet Rheumatology, 2026

https://doi.org/10.1016/S2665-9913(25)00374-1

1. Background

   Branebrutinib, an oral, highly selective, and irreversible Bruton's tyrosine kinase inhibitor, is a potential candidate for rheumatoid arthritis treatment as Bruton's tyrosine kinase has a role in B-cell activation, autoantibody production, and proinflammatory cytokine release, all of which are implicated in rheumatoid arthritis disease activity and progression. This study assessed the efficacy and safety of branebrutinib in patients with rheumatoid arthritis and an inadequate response to methotrexate.

   Methods

   This phase 2a, randomised, double-blind, placebo-controlled study was designed to assess the efficacy and safety of branebrutinib in patients with rheumatoid arthritis, systemic lupus erythematosus, or primary Sjögren's disease. Here, we report the results of the rheumatoid arthritis substudy, done in the USA, Poland, and Spain across 24 sites. The study included a 12-week double-blind treatment period followed by an additional 12-week open-label period with abatacept treatment. Only data for the double-blind treatment period are reported here. Eligible patients were aged 18–75 years, met the 2010 American College of Rheumatology (ACR)–European Alliance of Associations for Rheumatology criteria for rheumatoid arthritis, had disease duration less than 4 years, and had inadequate response to methotrexate. Patients were randomly assigned (3:1) to receive branebrutinib 9 mg once daily or placebo for 12 weeks. Randomisation was carried out centrally according to a computer-generated block randomisation scheme using interactive response technology. All parties were masked to treatment allocation. The primary endpoint was the proportion of patients who had 50% improvement in the ACR response criteria (ACR50) at week 12, assessed in all participants randomly assigned to treatment (full analysis set). Safety was assessed in patients who received at least one dose of branebrutinib or placebo. This trial was registered with ClinicalTrials.gov, NCT04186871. Patients with lived experience of rheumatoid arthritis were not involved in the study design.

   Findings

   Between Jan 7, 2020, to Dec 5, 2022, 85 patients were randomly assigned to receive branebrutinib (n=64) or placebo (n=21). 63 (74%) of 85 patients were female, 22 (26%) were male, 80 (94%) were White, and the mean age was 49·1 years (SD 12·0). The primary endpoint of ACR50 response at week 12 was not met; 12 (19%) of 64 patients had an ACR50 response in the branebrutinib group compared with seven (33%) of 21 patients in the placebo group (p=0·16). Adverse events were similar between the two groups (30 [47%] of 64 in the branebrutinib group and 8 [38%] of 21 in the placebo group), with no reported serious adverse events or deaths.

   Interpretation

   There was no significant difference between branebrutinib and placebo for any clinical efficacy measures. The 12-week safety profiles were similar between treatment groups, and branebrutinib was well tolerated with a favourable safety profile.

   Funding

   Bristol Myers Squibb.

A Novel Covalent Inhibitor Fragment for the SARS-CoV-2 Main Protease Identified by Target-Specific Deep Learning

Weijun Zhou, Angel D′Oliviera, Xuhang Dai, Jeffrey S. Mugridge, and Yingkai Zhang

ACS Chemical Biology 2026

DOI: 10.1021/acschembio.6c00120

The SARS-CoV-2 main protease (Mpro, also known as 3CLpro) is an attractive antiviral drug target due to its essential role in viral replication and absence of human homologues. Development of new coronavirus-specific Mpro inhibitors will be important as SARS-CoV-2 continues to evolve. Leveraging the rapidly expanding pool of diverse, experimental Mpro-inhibitor data, we developed a target-specific deep learning workflow to accelerate the discovery of new Mpro inhibitor compounds and fragment-like starting points. This workflow combined a fine-tuned inhibitor prediction model with solubility (logS) and lipophilicity (logP) models, molecular similarity analysis, and literature mining to prioritize novel, drug-like candidates. Applied to a purchasable library of over 500,000 compounds, the approach rapidly identified 24 candidates for experimental testing. Biochemical assays revealed a novel, small covalent inhibitor fragment (A02) with an apparent IC50 of 1.5 μM, prior to any synthetic optimization or derivatization. A 1.76 Å crystal structure of Mpro bound to A02 confirmed covalent modification of the catalytic Mpro cysteine (C145), unique engagement of the underutilized Mpro S3′ pocket, and the potential for derivatives of this scaffold to interact with additional Mpro pockets in future optimization efforts. Together, these results demonstrate the potential for target-specific deep learning approaches to guide the rapid screening and discovery of new inhibitor leads or drug scaffolds.

Electrophilic compound screening identifies GPX4-dependent ferroptosis as a senescence vulnerability

Mariantonietta D’Ambrosio, Matthew E. H. White, Efthymios S. Gavriil, Laura Bousset, Jodie Birch, Aleksandra Gruevska, Emiliano Pasquini, Manuel Colucci, Winnie Fong, Simone Mosole, Aurora Valdata, Dimitris Veroutis, Katie Tyson, Vikas Ranvir, Sandra Prokosch, Joaquim Pombo, Aoki Ardisson, Sanjay Khadayate, George Young, Alex Montoya, Georgia Roumelioti, Jack Houghton, Jianan Lu, Pavel V. Shliaha, Elena De Vita, Santiago Vernia, Vassilis G. Gorgoulis, Suchira Gallage, Mathias Heikenwälder, Zoe Hall, Andrea Alimonti, Iain A. McNeish, Edward W. Tate, Jesús Gil

Nature Chemical Biology, 2026

Journal: Nature Cell BiologySenescent cells drive ageing and age-related pathologies, including cancer. Consequently, senolytics, drugs that selectively kill senescent cells, have broad therapeutic appeal. Here we report a senolytic screen of a library of 10,480 electrophilic compounds. Among 38 identified hits, we found a subset of chloroacetamides with broad senolytic activity. Activity-based protein profiling, coupled with functional assays, identified the glutathione peroxidase GPX4 as a target. We show that senescent cells are primed for ferroptosis, displaying high levels of oxidative stress and intracellular Fe2+, but also upregulate GPX4, which prevents the accumulation of oxidized lipids. Treatment with senolytic chloroacetamides or GPX4 inhibitors selectively kills senescent cells by ferroptosis. The combination of anticancer therapies with GPX4 inhibitors eliminated senescent tumour cells in models of melanoma, prostate and ovarian cancer. Our results show that senescent cells rely on GPX4 to prevent ferroptosis and that GPX4 inhibitors kill senescent cells.

Data-driven design of chiral covalent fragments using highthroughput chemoproteomics and machine learning

McCarthy, William J.; Nightingale, Luke; Biggs, George S.; Cawood, Emma E.; Dudley-Fraser, Jane; Werner, Thilo; Riziotis, Ioannis G.; Pillay, Timesh D.; Lambert, Hugues; Pogány, Peter; van der Zouwen, Antonie J.; Pettinger, Jonathan; Boulton, Simon J.; House, David; Skehel, J. Mark; Bush, Jacob T.; Rittinger, Katrin

ChemRxiv.  2026.
DOI: https://doi.org/10.26434/chemrxiv.15002030/v1

A significant barrier in translating biological insights into therapeutic targets is the limited availability of high-quality chemical probes for target validation. Chemoproteomic profiling of covalent small molecules has dramatically accelerated the discovery of ligandable binding sites across the human proteome. However, the limited specificity and selectivity of initial hits often hinders their effectiveness in evaluating the functional consequences of ligand binding. To address this challenge, we developed a data-driven strategy that integrates chemoproteomic profiling of enantiomerically pure pairs of cysteine-targeting electrophilic fragments (enantiopairs) with machine learning (ML) to design fragment libraries optimised for proteome-wide selectivity. ML-guided library evolution produced a second generation enantiopair library markedly enriched in selective and stereospecific interactions relative to the first generation library. This approach identified high-quality enantioselective binding events with 205 cysteines, the majority not previously liganded. These findings establish a general framework for designing covalent fragment libraries to deliver higherquality initial hits.

Deciphering covalent kinase inhibitor binding landscape through structural kinome profiling

Zheng Zhao, Philip E. Bourne

European Journal of Medicinal Chemistry, 312, 2026, 118872

https://doi.org/10.1016/j.ejmech.2026.118872

Significant progress in kinase-targeted drug discovery has been made over the past two decades, with 100 FDA-approved kinase-targeted drugs and a substantial number of bioactive kinase inhibitors under preclinical study. However, given that more than 180 kinases have been implicated in disease, there remains a considerable need for continued kinase-targeted drug discovery. Covalent kinase inhibitors (CKIs) are a class of kinase inhibitors that form covalent interactions with kinase targets, valued for the potential for enhanced selectivity through anchoring nucleophiles. Here, we collate all the kinase structures from the PDB into dedicated structural kinome resources, containing: (i) the kinase domain structure database (6969 PDB structures); (ii) the kinase ligand-binding structure database (6122 PDB structures); and (iii) the kinase-CKI complex structure database (325 PDB structures). With these data, we systematically investigate the binding modes of CKIs, the fingerprint characteristics of kinase-CKI interactions, 21 types of electrophilic warheads, and 64 nucleophilic amino acids distributed in 15 corresponding spatial positions in kinase domains. We also mentioned covalent degraders and multi-warhead CKIs. Together, our results offer a comprehensive structural kinase resource and in-depth insights into CKI binding properties, supporting future kinase-targeted drug design. The databases are freely accessible at https://zhengzhster.github.io/KinaseStructureDatabase/.

A Fragment Screen Identifies Acrylamide Covalent Inhibitors of the TEAD•YAP Protein-Protein Interaction

Khuchtumur Bum-Erdene, Mona K. Ghozayel, Mark J. Zhang, Giovanni Gonzalez-Gutierrez, Samy O. Meroueh

bioRxiv 2026.03.18.712694; 

doi: https://doi.org/10.64898/2026.03.18.712694

TEA domain (TEAD) proteins bind co-activator Yes-associated protein (YAP) to regulate the expression of target genes of the Hippo pathway. The TEAD•YAP protein-protein interaction is not druggable, but TEADs possess a unique and deep palmitate pocket with a highly conserved cysteine located outside the TEAD•YAP protein-protein interaction interface. Here, we screen a fragment library of acrylamide electrophiles and identify a fragment that forms an adduct with the conserved palmitate pocket cysteine and inhibits TEAD4 binding to YAP. Synthesis of a focused set of derivatives and time- and concentration-dependent studies with four TEADs provide reaction rates and binding constants. Co-crystal structures of fragments bound to TEAD2 and TEAD3 reveal reaction at the conserved palmitate pocket cysteine but also at another less conserved cysteine located in the palmitate pocket of TEAD2 closer to the TEAD•YAP interface. These fragments provide a starting point for the development of allosteric acrylamide small-molecule covalent TEAD•YAP inhibitors.

Discovery of Covalent Ligands with AlphaFold3

Yoav Shamir, Ronen Gabizon, Adi Rogel, David Yin-wei Lin, Amy H. Andreotti, and Nir London

Journal of the American Chemical Society 2026

DOI: 10.1021/jacs.5c22222

Covalent inhibitors are a prominent modality for research and therapeutic tools. However, a scarcity of computational methods for their discovery slows progress in this field. AI models such as AlphaFold3 (AF3) have shown accuracy in ligand pose prediction, but their applicability for virtual screening campaigns was not assessed. We show that AF3 cofolding predictions and an associated predicted confidence metric ranks true covalent binders with near-optimal classification over property-matched decoys, significantly outperforming state-of-the-art covalent docking tools for a set of protein kinases. In a prospective virtual screening campaign against the model kinase BTK, we discovered a chemically distinct, novel, covalent small molecule that displays potent inhibition in vitro and in cells while maintaining marked kinome and proteomic selectivity. Co-crystallography validated the subangstrom accuracy of the predicted AF3 binding mode. These results demonstrate that AF3 can be practically used to discover novel chemical matter for kinases, one of the most prolific families of drug targets.