From QSAR to dynamics: A data-driven design of drug-like lactam-based HDAC inhibitors for diffuse large B-cell lymphoma

• A validated QSAR model (R² = 0.79, Q² = 0.69, CCC_ext = 0.9395) was developed to predict HDAC inhibitory activity. • Compound 14c identified as a potent lead with strong binding (–11.5 kcal/mol) and favourable ADMET profile. • Molecular dynamics confirmed the structural stability of 14c over 100 ns with lower RMSD than Tucidinostat. • MM/GBSA binding free energy for 14c (–59.70 kcal/mol) outperformed the reference drug (–38.26 kcal/mol). • Study offers a fully integrated computational workflow for HDAC inhibitor optimization in DLBCL therapy. Diffuse large B-cell lymphoma (DLBCL) continues to require more selective and effective therapeutic options beyond currently available HDAC inhibitors such as Tucidinostat. This study employed an integrated computational strategy combining QSAR modelling, molecular docking, ADMET profiling, and molecular dynamics (MD) simulations to identify novel HDAC2 inhibitors with promising therapeutic potential. A validated three-descriptor QSAR model (R² = 0.7869, Q²_LOO = 0.6870, CCC_ext = 0.9395) guided scaffold selection, identifying compound 14 with favourable predicted activity (pIC₅₀ = 6.57) and strong binding affinity (–11.1 kcal/mol). Tucidinostat showed comparatively lower predicted activity (pIC₅₀ = 5.69) and binding affinity (–10.5 kcal/mol) under the same evaluation. Structure-guided optimization yielded compound 14c, which demonstrated improved docking scores (–11.5 kcal/mol), high predicted intestinal absorption (93.6%), compliance with Lipinski’s rule of five, and a non-mutagenic ADMET profile. MD simulations over 100 ns indicated stable binding of 14c within the HDAC2 pocket, with reduced RMSD (∼1.1 Å) and lower protein backbone fluctuations relative to the Tucidinostat complex. MM/GBSA calculations suggested a more favourable binding free energy for 14c (–59.70 kcal/mol) compared with Tucidinostat (–38.26 kcal/mol), driven by van der Waals and lipophilic interactions. While these computational results highlight 14c as a priority candidate for synthesis and experimental evaluation, confirmation of its potential advantages over Tucidinostat will require further in vitro and in vivo validation.