Gene expression signatures predict response to therapy with growth hormone

Area of Science:

• Endocrinology
• Genomics
• Personalized Medicine

Background:

• Recombinant human growth hormone (r-hGH) treats growth disorders like growth hormone deficiency (GHD) and Turner syndrome (TS).
• Current methods for predicting patient response to r-hGH are imprecise, leading to suboptimal treatment and high costs.
• Accurate prediction of therapeutic response is crucial for effective clinical management.

Purpose of the Study:

• To identify predictive markers for r-hGH treatment response in GHD and TS patients.
• To evaluate the utility of genetic markers and baseline blood transcriptome for predicting growth outcomes.
• To develop a more accurate method for assessing individual patient response to r-hGH therapy.

Main Methods:

• Recruited 71 GHD and 43 TS patients for a 5-year study of r-hGH response.
• Analyzed 1219 genetic markers and baseline blood transcriptome data.
• Utilized random forest machine learning to assess the predictive power of transcriptomic data for growth response.

Main Results:

• No single genetic marker accurately predicted r-hGH response.
• Identified a consistent set of genes in both GHD and TS whose expression predicted therapeutic response with high accuracy (AUC > 0.9).
• Combining transcriptomic markers with clinical phenotype significantly reduced predictive error.

Conclusions:

• Baseline gene expression profiles can accurately predict therapeutic response to r-hGH in GHD and TS patients.
• A predictive model integrating transcriptomic data and clinical phenotype offers improved accuracy over current methods.
• This research supports the development of a genomic test for personalized r-hGH treatment strategies.