Advancing Adverse Drug Reaction Prediction with Deep Chemical Language Model for Drug Safety Evaluation.

The accurate prediction of adverse drug reactions (ADRs) is essential for comprehensive drug safety evaluation. Pre-trained deep chemical language models have emerged as powerful tools capable of automatically learning molecular structural features from large-scale datasets, showing promising capabilities for the downstream prediction of molecular properties. However, the performance of pre-trained chemical language models in predicting ADRs, especially idiosyncratic ADRs induced by marketed drugs, remains largely unexplored. In this study, we propose MoLFormer-XL, a pre-trained model for encoding molecular features from canonical SMILES, in conjunction with a CNN-based model to predict drug-induced QT interval prolongation (DIQT), drug-induced teratogenicity (DIT), and drug-induced rhabdomyolysis (DIR). Our results demonstrate that the proposed model outperforms conventional models applied in previous studies for predicting DIQT, DIT, and DIR. Notably, an analysis of the learned linear attention maps highlights amines, alcohol, ethers, and aromatic halogen compounds as strongly associated with the three types of ADRs. These findings hold promise for enhancing drug discovery pipelines and reducing the drug attrition rate due to safety concerns.

Improved performance of non-preloaded and high flip-angle dynamic susceptibility contrast perfusion-weighted imaging sequences in the presurgical differentiation of brain lymphoma and glioblastoma.

OBJECTIVE: This study aimed to compare the accuracy of relative cerebral blood volume (rCBV) and percentage signal recovery (PSR) obtained from high flip-angle dynamic susceptibility contrast perfusion-weighted imaging (DSC-PWI) sequences with and without contrast agent (CA) preload for presurgical discrimination of brain glioblastoma and lymphoma. METHODS: Consecutive 336 patients (glioblastoma, 236; PCNSL, 100) were included. All the patients underwent DSC-PWI on 3.0-T magnetic resonance units before surgery. The rCBV and PSR with preloaded and non-preloaded CA were measured. The means of the continuous variables were compared using Welch's t-test. The diagnostic accuracies of the individual parameters were compared using the receiver operating characteristic curve analysis. RESULTS: The rCBV was higher with preloaded CA than with non-preloaded CA (glioblastoma, 10.20 vs. 8.90, p = 0.020; PCNSL, 3.88 vs. 3.27, p = 0.020). The PSR was lower with preloaded CA than with non-preloaded CA (glioblastoma, 0.59 vs. 0.90; PCNSL, 0.70 vs. 1.63; all p < 0.001). Regarding the differentiation of glioblastoma and PCNSL, the AUC of rCBV with preloaded CA was indistinguishable from that of non-preloaded CA (0.940 vs. 0.949, p = 0.703), whereas the area under the curve of PSR with preloaded CA was lower than non-preloaded CA (0.529 vs. 0.884, p < 0.001). CONCLUSION: With preloaded CA, diagnostic performance in differentiating glioblastoma and PCNSL did not improve for rCBV and it was decreased for PSR. Therefore, high flip-angle non-preload DSC-PWI sequences offer excellent accuracy and may be of choice sequence for presurgical discrimination of brain lymphoma and glioblastoma. CLINICAL RELEVANCE STATEMENT: High flip-angle DSC-PWI using non-preloaded CA may be an excellent diagnostic method for distinguishing glioblastoma from PCNSL. KEY POINTS: • Differentiating primary central nervous system lymphoma and glioblastoma accurately is critical for their management. • DSC-PWI sequences optimised for the most accurate CBV calculations may not be the optimal sequences for presurgical brain tumour diagnosis as they could be masquerading leakage phenomena that may provide interesting information in terms of differential diagnosis. • High flip-angle non-preloaded DSC-PWI sequences render the best accuracy in the presurgical differentiation of brain lymphoma and glioblastoma.

Unraveling Structural Alerts in Marketed Drugs for Improving Adverse Outcome Pathway Framework of Drug-Induced QT Prolongation.

In pharmaceutical treatment, many non-cardiac drugs carry the risk of prolonging the QT interval, which can lead to fatal cardiac complications such as torsades de points (TdP). Although the unexpected blockade of ion channels has been widely considered to be one of the main reasons for affecting the repolarization phase of the cardiac action potential and leading to QT interval prolongation, the lack of knowledge regarding chemical structures in drugs that may induce the prolongation of the QT interval remains a barrier to further understanding the underlying mechanism and developing an effective prediction strategy. In this study, we thoroughly investigated the differences in chemical structures between QT-prolonging drugs and drugs with no drug-induced QT prolongation (DIQT) concerns, based on the Drug-Induced QT Prolongation Atlas (DIQTA) dataset. Three categories of structural alerts (SAs), namely amines, ethers, and aromatic compounds, appeared in large quantities in QT-prolonging drugs, but rarely in drugs with no DIQT concerns, indicating a close association between SAs and the risk of DIQT. Moreover, using the molecular descriptors associated with these three categories of SAs as features, the structure-activity relationship (SAR) model for predicting the high risk of inducing QT interval prolongation of marketed drugs achieved recall rates of 72.5% and 80.0% for the DIQTA dataset and the FDA Adverse Event Reporting System (FAERS) dataset, respectively. Our findings may promote a better understanding of the mechanism of DIQT and facilitate research on cardiac adverse drug reactions in drug development.