Background: Intravenous drug use (IVDU) represents the major factor of HCV transmission, but the treatment uptake among people who inject drugs (PWID) remains low owing to a false presumption of low efficacy. The aim of our study was to assess treatment efficacy in PWID and factors determining adherence to therapy. Methods: A total of 278 consecutive patients starting DAA (direct-acting antivirals) therapy were included, divided into two groups: individuals with a history of IVDU, PWID group (N = 101) and the control group (N = 177) without a history of IVDU. Results: Sustained virological response 12 weeks after the end of therapy (SVR12) was achieved by 99/101 (98%) and 172/177 (98%) patients in the PWID and control group, respectively; in PWID group, two patients were lost to follow-up, and in the control group, four patients relapsed and one was lost to follow-up. PWID patients postponed appointments significantly more often, 29 (28.7%) in PWID versus 7 (4%) in the control group, p = 0.001. Thirteen of 101 (12.9%) and six of 177 (3.4%) patients in the PWID and in the control group, respectively, missed at least one visit (p < 0.01). However, postponing visits led to a lack of medication in only one PWID. In the PWID group, older age (p < 0.05; OR 1.07, 95% CI 1.00–1.20) and stable housing (p < 0.01; OR 9.70, 95% CI 2.10–56.20) were factors positively contributing to adherence. Contrarily, a stable job was a factor negatively influencing adherence (p < 0.05; OR 0.24, 95% CI 0.06–0.81). In the control group, none of the analyzed social and demographic factors had an impact on adherence to therapy. Conclusions: In PWID, treatment efficacy was excellent and was comparable with SVR of the control group. Stable housing and older age contributed to a better adherence to therapy.

The HADDOCK team participated in CAPRI rounds 47-55 as both server, manual predictor, and scorers. Throughout these CAPRI rounds, we used a plethora of computational strategies to predict the structure of protein complexes. Of the 10 targets comprising 24 interfaces, we achieved acceptable or better models for 3 targets in the human category and 1 in the server category. Our performance in the scoring challenge was slightly better, with our simple scoring protocol being the only one capable of identifying an acceptable model for Target 234. This result highlights the robustness of the simple, fully physics-based HADDOCK scoring function, especially when applied to highly flexible antibody-antigen complexes. Inspired by the significant advances in machine learning for structural biology and the dramatic improvement in our success rates after the public release of Alphafold2, we identify the integration of classical approaches like HADDOCK with AI-driven structure prediction methods as a key strategy for improving the accuracy of model generation and scoring.

Adenosine receptors are a family of G protein-coupled receptors with increased attention as drug targets on different indications. We investigate the thermodynamics of ligand binding to the A₃ adenosine receptor subtype, focusing on a recently reported series of diarylacetamidopyridine inhibitors via molecular dynamics simulations. With a combined approach of thermodynamic integration and one-step perturbation, we characterize the impact of the charge distribution in a central heteroaromatic ring on the binding affinity prediction. Standard charge distributions according to the GROMOS force field yield values in good agreement with the experimental data and previous free energy calculations. Subsequently, we examine the thermodynamics of inhibitor binding in terms of the energetic and entropic contributions. The highest entropy penalties are found for inhibitors with methoxy substituents in meta position of the aryl groups. This bulky group restricts rotation of aromatic rings attached to the pyrimidine core which leads to two distinct poses of the ligand. Our predictions support the previously proposed binding pose for the o-methoxy ligand, yielding in this case a very good correlation with the experimentally measured affinities with deviations below 4 kJ/mol.

Molecular docking excels at creating a plethora of potential models of protein-protein complexes. To correctly distinguish the favourable, native-like models from the remaining ones remains, however, a challenge. We assessed here if a protocol based on molecular dynamics (MD) simulations would allow to distinguish native from non-native models to complement scoring functions used in docking. To this end, first models for 25 protein-protein complexes were generated using HADDOCK. Next, MD simulations complemented with machine learning were used to discriminate between native and non-native complexes based on a combination of metrics reporting on the stability of the initial models. Native models showed higher stability in almost all measured properties, including the key ones used for scoring in the CAPRI competition, namely the positional root mean square deviations and fraction of native contacts from the initial docked model. A Random Forest classifier was trained, reaching 0.85 accuracy in correctly distinguishing native from non-native complexes. Reasonably modest simulation lengths in the order of 50 to 100 ns are already sufficient to reach this accuracy, which makes this approach applicable in practice.

Large-scale cancer sequencing efforts coupled with deep functional genomic interrogation of cancer models provides us an increasingly detailed view of the critical cancer drivers required for persistent growth and survival of cancer cells. Targeting such drivers has provided significant therapeutic benefit to patients across a wide range of cancers, yet progress has been, in large part, confined to the targeting of catalytic domains in kinases. To expand the reach of therapeutics into new classes of therapeutic targets, we need to identify and validate specific mechanisms and pockets to enable drug discovery. To enable this, we are developing a method called Hit-finding by Cysteine-Scanning (HCS) for identifying druggable pockets in proteins of interest. Here, we seek to couple the ability to create systematic variant libraries across proteins of interest with increasingly diverse covalent small molecule/fragment libraries. Thus, HCS involves screening pooled cysteine (cys)-variants against libraries of covalent compounds. To demonstrate the utility of this approach, we applied HCS to systematically screen 189 KRAS-G12D variants. We successfully purified KRAS cys-variant recombinant proteins in pools of seven or eight achieving coverage of 184/189 variants. KRAS variant pools were then screened against a pilot library of 47 covalent compounds with diverse reactivity detecting reactive variant-ligand pairs by intact LC-MS. This allowed us to quantify the accessibility and reactivity across the entire KRAS-G12D protein. The data confirmed the known SII-pocket in KRAS and revealed potentially novel ligand-bound conformations of oncogenic KRAS mutant proteins. In parallel, the same KRAS cys-variant library was generated as a barcoded library in a mammalian expression system. Upon screening this library in intact cells we observed that Sotorasib functionally inhibit the relevant KRAS SII-pocket cys-variants, enabling enrichment of these variants in a negative selection screen. Importantly, binding of a panel of SII-pocket compounds to variant H95C was found to be independent of the KRAS mutant context. Our results indicate that HCS could be a useful tool for the development of inhibitors for currently undruggable oncogenic KRAS mutants as well as potentially other drug targets. Citation Format: Laurens Moore van Tienen, Shadwa Bayoumi, Khaja Muneeruddin, Nancy Leymarie, David Kornfilt, Yuemin Bian, Devishi Kesar, Ruitong Li, Thomas Atack, Alexandra-Mariela Popa, Zuzana Jandova, Philipp Trollmann, Andreas Bergner, Klaus Rumpel, Dirk Kessler, William R. Sellers. Hit-finding by Cysteine-Scanning (HCS): A method for finding druggable pockets [abstract]. In: Proceedings of the AACR Special Conference: Targeting RAS; 2023 Mar 5-8; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Res 2023;21(5_Suppl):Abstract nr B037.