PII-112 - GENETIC PREDICTORS OF RESPONSE TO RISPERIDONE IN CHILDREN AND ADOLESCENTS: INSIGHTS FROM A PSYCHIATRIC HOSPITAL SERVING RURAL AND UNDERSERVED YOUTH

S. Killam¹, J. Staples², K. Brown³, R. Dalton³, E. Sather³, Q. Chen⁴, J. Loveland⁵, C. Schwanke⁵, A. Elias⁵, K. Bigos⁴, E. Woodahl³; ¹L.S. Skaggs Institute for Health Innovation - University of Montana, Missoula, MT, USA, ²University of Montana, Missoula, Montana, USA, ³University of Montana, Missoula, MT, USA, ⁴Johns Hopkins School of Medicine, Baltimore, MD, USA, ⁵Shodair Children's Hospital, Helena, MT, USA.

Background: Risperidone is a commonly used antipsychotic for treating psychiatric illness in children and adolescents, however, there is large variability in risperidone response and discontinuation rates remain high. Pharmacogenomics offers the opportunity to improve risperidone outcomes, yet studies in pediatric populations are limited. We conducted a genome-wide association study (GWAS) to investigate genetic and non-genetic predictors of risperidone response in 161 pediatric patients who received inpatient psychiatric care in a pediatric hospital in a rural and underserved setting.

Methods: Clinical and demographic data and risperidone treatment outcomes were collected retrospectively. Genomic DNA was genotyped using the Infinium Global Screening Array. All patients were of self-identified European ancestry. Participants were admitted on risperidone or prescribed risperidone during inpatient care. We also collected all concomitant medications, included other antipsychotics and CYP2D6 inhibitor use. Patient discontinuation of antipsychotics was evaluated with Kaplan-Meier and Cox proportional hazard models. In the GWAS analysis, genotyped and imputed variants were tested individually for statistical associations with outcomes using linear or logistic regression.

Results: While 41.0% of patients discontinued risperidone, patients remained on risperidone longer than other antipsychotics, with the exception of quetiapine. We also found female patients discontinued more quickly, as did patients in the acute treatment program compared to residential. We identified nine genetic variants significantly associated with risperidone outcomes: duration of risperidone treatment, frequency of risperidone discontinuation, maximum risperidone dose prescribed, time to readmission after discharge, and duration of hospital stay.

Conclusion: Our study is the first GWAS of risperidone response in pediatric populations, which provides valuable insights into the biological complexity of risperidone response, as well as moving towards precision antipsychotic treatment. Importantly, our study demonstrates the high value of conducting research in a community-based setting and demonstrates the need to expand research studies beyond academic medical centers.