Thyroid cancer (TC) is a very common malignancy worldwide. Chief among the innovative molecular drug targets for TC are epigenetic modifications. Increased telomerase activity in cancer cells makes telomerase a novel target for epigenetic anticancer drug innovation. Recently, telomerase reverse transcriptase ( TERT ) gene promoter ( TERTp ) mutations (C228T and C250T) were reported at high frequency in TC cell lines and tumor biopsies. In this study, three representative TC cell lines, mutant TERTp (TPC1), mutant BRAF / TERTp (KTC2), and wild-type TERTp (WRO), were screened with a drug library composed of 51 epigenetic drugs: 14 Aurora kinase inhibitors; 23 histone deacetylase inhibitors; 5 sirtuin modifiers; 3 hypoxia-inducible factor inhibitors; 2 DNA methyltransferase inhibitors; 2 histone methyltransferase inhibitors, a histone demethylase inhibitor, and a bromodomain inhibitor. Effects of the drugs on cell growth at 48 and 72 h were compared. PF-03814735, a small-molecule inhibitor of Aurora kinase A (IC ₅₀ = 0.8 nM) and B (IC ₅₀ = 5 nM), was the most potent on KTC2 cells, whereas CUDC-101, a multitarget inhibitor, was effective on both WRO and KTC2 cells. Notably, PF-03814735 was found to be the most effective epigenetic drug on cell lines harboring the C228T mutation. In conclusion, these new findings offer specific guidance on dose and time course selection to design novel therapeutic interventions against TC using PF-03814735, and specifically target cells carrying the TERTp ^C228T mutation. In a larger context of drug discovery science, these findings inform new strategies to forecast optimal treatment regimens for TC, particularly with Aurora kinase inhibitors and in ways guided by epigenetic drug design.

HerCanPred, a machine-learning-based pathogenicity classifier specifically optimized for 63 cancer-predisposition genes, was developed to improve the interpretation of missense variants in hereditary cancer syndromes. This model integrates sequence conservation with structural features derived from AlphaFold2 (AF2) structures. HerCanPred achieved a strong performance, outperforming 23 established predictors. SHAP analysis identified AF2-derived structural features, specifically local pLDDT confidence scores and relative solvent accessible area, as the strongest predictors of variant impact. Benchmarking the strengths and limitations of HerCanPred against existing methods showed that misclassification of pathogenic variants was concentrated in disordered and surface-exposed regions, whereas benign failures were more broadly distributed. HerCanPred and three established predictors were also applied to over 57,000 variants of uncertain significance (VUS) from the same gene set. Notably, 166 VUS were reassigned as pathogenic and 75 as benign, with an enrichment of the NF1, FH, and MLH1 genes. By combining gene-specific training with 3D structural information, HerCanPred provides a robust framework for reducing diagnostic uncertainty. Our findings demonstrate that targeted, structure-aware tools can contribute to resolving VUS, providing a rational basis for systematic variant reinterpretation and more informed medical management in hereditary cancer care.