A recent study has presented a new approach that researchers believe could enhance the ability of genetics to predict how effectively a patient will respond to commonly used medications, as well as the severity of potential side effects.
A UCLA study has presented a new approach that researchers assert could enhance the predictive power of genetics regarding how well a patient responds to widely prescribed medications and the intensity of any side effects they might experience.
Published in the journal Cell Genomics, this research
revealed that information from extensive collections of sequenced human genomes and biological data, referred to as biobanks, can lead to fresh understanding of the genetic foundation of responses to commonly used drugs.
Study lead author and UCLA Bioinformatics Ph.D. candidate, Michal Sadowski, noted that the typical approach for analyzing the genetics of medication responses relies on pharmacogenomic research involving genotyped participants from randomized controlled trials. However, these studies often involve a limited number of participants, can be expensive, and may not even be practical for certain medications, Sadowski explained.
The genetic information in biobanks offers numerous advantages. It includes sequenced genetic data from large groups of individuals, both those taking and not taking specific medications, and can be analyzed at a lower cost. Though biobank data cannot replace randomized controlled trials, it can reveal crucial insights that enhance future research and contribute to the growing field of genetic-based treatment outcome predictions, Sadowski stated.
“We aspire that, in the future, this will empower doctors and patients to more accurately assess the benefits and risks of treatments in a personalized manner, allowing for more informed and timely decisions regarding treatment initiation,” Sadowski said. “We anticipate that the analysis of biobank data will be particularly beneficial for widely prescribed medications.”
This study, guided by UCLA Neurology, Computational Medicine, and Human Genetics professor Noah Zaitlen and UChicago Genetic Medicine assistant professor Andy Dahl, utilized genetic information from over 342,000 individuals in the UK Biobank.
The researchers investigated how genetic variations influenced responses to four of the most commonly prescribed medications globally: statins (for high cholesterol), metformin (for type 2 diabetes), warfarin (for blood clots), and methotrexate (for autoimmune diseases and cancer).
Sadowski and his team aimed to assess the extent to which genetic variation contributed to the variability in drug responses and identify specific genes involved.
“If genetics can explain a significant portion, then it serves as a reliable predictor for medication responses,” Sadowski noted. “For instance, if you’re considering statins for cholesterol management, your doctor can examine your genetic information and provide insights, including potential side effects. If your genetic indicators suggest a favorable response and a low risk of side effects, it could be a wise choice to proceed with treatment.”
As an illustration, the research identified 156 genes that may influence the effectiveness of statins on LDL cholesterol levels. Overall, genetic differences among individuals accounted for approximately 9% of the variability in drug response.
The study also discovered that gene-drug interactions can impact the effectiveness of a genetic risk assessment tool known as a polygenic score. Polygenic scores summarize the cumulative impact of numerous genetic variants to predict an individual’s risk of developing certain traits or diseases. The generation of these scores requires training on genetic information from large groups and has notable limitations, including being predominantly based on data from individuals of European descent.
Sadowski’s findings indicated that standard polygenic scores may not perform well in clinical situations because they contain data from both statin users and non-users.
“We were taken aback by the significant differences in performance of polygenic predictors between individuals on and off medications,” Sadowski remarked. “We were also surprised by the extent of drug-specific heritability for some outcomes. These findings imply that more genetic associations and other elements of unexplained heritability might be unveiled through future studies focused on specific contexts of complex diseases.”
The study acknowledges several limitations, highlighting the need for further research to enhance the reliability of insights drawn from observational data in biobanks and to better comprehend the constraints of genetic risk predictions.
