(263) Pharmacogenomics in Cystic Fibrosis Patients: A Personalized Medicine Approach

Background: Cystic Fibrosis (CF) is a multisystem genetic disorder requiring the use of many acute and chronic medications. The use of genetic information (pharmacogenomics, PGx) to tailor therapeutic approaches is increasing in many areas. We hypothesize that although genetic variants influencing medication response likely exist in the CF population similar to the general population, their increased use of medications to manage their disease increases opportunity for intervention to mitigate undesirable drug responses (UDR) and treatment failures.

Methods: Eighty-two people with CF (pwCF) donated DNA samples for PGx variant analysis and were genotyped using Illumina GSA microarray. Two years of medication encounter data, including UDR, was gathered from the electronic medical record for each participant. PharmGKB and CPIC databases were utilized in the annotation of medications and PGx variants. Clinical annotation of PGx variants was performed using an open-source software, PharmCAT.

Results: This cohort of pwCF had a mean age of 22.1 years (Range: 3-66 years old) and were 93.9% Caucasian. They were prescribed 286 unique medications over the 2-year observation and experienced a total of 3336 medication exposure events (MEEs). The median number of unique medication exposures was 28 medications for any individual and on average antibiotics comprised 33.1 % (SD=13.46, n=82) of these MEEs. There were no statistically significant differences in the total number of MEEs between adult and pediatric patients, but adults were prescribed medications from more varied classes (p=0.0107). Of the 286 medications, only 33 were associated with a PGx-based dosing guideline; of these 33 medications, only three were used by a majority of our cohort. Of the 16 medications taken by more than 50% of the cohort: 56.3% did not have any annotations, 6.3% had no dosing guidelines (NDG) but had variant annotations, and 18.8% had NDG but had clinical annotations. Approximately 61% of patients had an actionable variant in the CYP2C19 gene and 22% of these patients experienced an ADE correlated with their PGx-genotype for CYP2C19. Overall, 97.6% of patients had at least 1 clinically-actionable PGx variant; over 70% of patients had ≥3.

Conclusions: Similar to other population-based studies, almost every patient in this cohort had at least one actionable (highly-evidenced) PGx variant, and many were taking one or more medications with lower-level evidence, supporting a potential role for other PGx variants. The high medication burden of this population increases their likelihood of repeated prescription changes for UDR, and some experienced predictable adverse drug events, both of which could be mitigated by knowing patients’ PGx profiles at the time of prescribing. Implementation of PGx guidelines in CF care centers should be a priority, to aid in preventing unnecessary UDR as well as to promote further research in the medications prescribed to pwCF.

Acknowledgements: All authors completed work on this project while at Children's of Alabama and/or the University of Alabama at Birmingham. J.S.G. is supported by National Institutes of Health (NIH) Grant K23HL143167 and the Cystic Fibrosis Foundation (CFF) Grants GUIMBE18A0-Q and GUIMBE20A0-KB.

References: