Researchers find that while individual genetic variants have modest effects, combining rare and common variants through polygenic risk scores enhances prediction of steroid-related risks, moving toward more personalized medicine in prescribing.

Practical implementation faces challenges in scalability and diversity; targeted use in high-risk individuals with longer-term steroid exposure is considered most feasible, and findings should be replicated in larger, ethnically diverse cohorts.

Integrating polygenic risk scores into routine prescribing is challenging, with emphasis on targeting high-risk patients and validating findings across more diverse populations to ensure broader applicability.

Key genetic variants linked to higher risk include CYP3A4 (osteoporosis) and CTLA4 (stroke and cataract); adding PRSs for osteoporosis improves risk prediction beyond age and sex, especially in younger patients at first prescription.

Incorporating bone health PRSs enhances prediction beyond demographic factors, notably for younger patients at first steroid exposure.

Specific variants such as CYP3A4 and CTLA4 indicate higher risk of side effects, with PRSs for osteoporosis providing additional predictive value.

The study confirms a dose–response relationship between steroid exposure and adverse effects, suggesting genomic data could enable earlier steroid-sparing or closer monitoring for high-risk patients.

Dose management remains crucial, with clear evidence that higher exposure increases risk of adverse outcomes.

A study presented at a genetics conference shows genetic data can improve prediction of side effects from oral corticosteroids used for chronic inflammatory conditions.

The conference presentation highlights that incorporating genetic information enhances forecasting of OCS-related adverse effects.

PRS-enhanced risk assessment could enable earlier steroid-sparing interventions or closer monitoring, pushing practice toward more personalized medicine.

Experts emphasize that integrating genomic data into routine prescribing could enable population-level genomics to support personalized care, contingent on broader access to genetic data.