The dual-pathway in silico screen improved high-risk detection specificity from 0.923 to 1.000 while maintaining precision in identifying trafficking-mediated cardiotoxicity.
Does a dual-pathway in silico screening framework improve the detection of trafficking-mediated cardiotoxicity compared to acute-only hERG screens?
A novel dual-pathway in silico screening tool successfully detects 'hidden' trafficking-mediated cardiotoxicity, improving the specificity of proarrhythmia risk prediction in early drug discovery.
Absolute Event Rate: 0% vs 0%
Background: The Comprehensive in vitro Proarrhythmia Assay (CiPA) initiative successfully shifted safety pharmacology from binary hERG screening toward mechanistic multichannel integration. However, a critical blind spot remains: most in silico workflows assume ion-channel density (Gmax) is static. Consequently, these models fail to detect "hidden" cardiotoxicity from compounds that prolong QT primarily by reducing hERG (KCNH2) surface expression (trafficking inhibition) rather than acutely blocking the pore. Methodology: We present Biosentry In Silico, a dual-pathway screening framework designed to close this gap. The architecture couples: Pathway 1 (Acute): Standard pore block modeling using Hill-equation multichannel inhibition. Pathway 2 (Chronic): A novel steady-state trafficking scalar—a closed-form analytical evaluation of the equilibrium reduction in functional hERG density. Unlike computationally expensive time-domain protein lifecycle simulations, the steady-state scalar allows for high-throughput evaluation while preserving mechanistic sensitivity to chronic liabilities. Key Findings: Detection of Hidden Risk: The dual-pathway screen correctly flags trafficking-mediated risk in exemplars such as Probucol (Strong trafficking signal) and Pentamidine (Onset signal), both of which are labeled as "Low Risk" by standard acute-only hERG screens. Preservation of Specificity: Balanced blockers like Verapamil remain correctly classified as low risk, demonstrating that the addition of the trafficking pathway does not introduce false positives. CiPA Benchmark: On a label-level benchmark of the CiPA reference subset (n=21), the dual-pathway logic improved high-risk detection specificity from 0.923 (acute-only) to 1.000, while maintaining precision. Conclusion: This study validates a scalable, mechanistic solution for detecting trafficking-mediated proarrhythmia in early drug discovery, offering a robust safeguard against late-stage attrition due to non-block mechanisms.
Mustafa Al Yagoub (Sun,) reported a other. The dual-pathway in silico screen improved high-risk detection specificity from 0.923 to 1.000 while maintaining precision in identifying trafficking-mediated cardiotoxicity.