Kaohsiung Medical University / Prof. Shu-Pin Huang

 Pain Points Solved 

• Clinical responses to second-generation hormonal therapies vary substantially among patients, and reliable biomarkers for predicting treatment response are lacking.
  
• Drug resistance arises through diverse mechanisms, which cannot be effectively distinguished using conventional gene analysis approaches.
• By integrating transcriptomic data with clinical information, this approach provides evidence to support precision medicine and reduces the risk associated with trial-and-error drug selection.

 Technology Introduction 

We adopted the Affymetrix transcriptome gene chip and analyzed the RNA expression from patients’ peripheral blood prior to, 3-month later, and after drug resistance developed. Using weighted gene correlation network analysis, we identified several gene modules correlated to specific drug responses. Then, we combined several machine learning algorithms and constructed classification models for different drug mechanisms. The overall accuracy for androgen synthesis inhibitors and androgen receptor blockers were 84.4% and 90%, respectively.

 Application Examples 

• The model can be used to predict patient responses to androgen synthesis inhibitors (ASI) or androgen receptor blockers (ARB) prior to treatment, assisting clinicians in selecting the most appropriate therapy.
  
• It can be applied in clinical trials of novel therapeutics to support patient stratification and treatment efficacy prediction.
• This approach can also be extended to gene response profiling studies in other cancer therapies.

 Related Links 

None

 Patent Name and Number 

TW I808838
US 18/578,721
CN 202280049348.2
EP 22845343.7

 Industry-Academia / Tech Transfer Partner 

None

 Honors and Awards  

The 21st National Innovation Award – Clinical Innovation Award

 Technical Contact  

Mr. Hung, Assistant Manager

Kaohsiung Medical University 
Tel: +886 7-3121101 ext. 2360
Email: R121084@kmu.edu.tw