UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Implementation of pharmacogenetic risk prediction models in pediatric oncology Bendyshe-Walton, Tessa


Adverse drug reactions (ADRs) are increasingly recognized as important and sometimes irreversible complications of cancer treatment¹,². Anthracyclines and cisplatin, two widely-used chemotherapeutic agents in the treatment of childhood malignancies, have contributed to the increased 5-year survival rates for childhood cancer to over 82% today³. Their use, however, is limited by the occurrence of anthracycline-induced cardiotoxicity in up to 57%⁴ of treated children and cisplatin-induced ototoxicity in 60-70%⁵-⁷ of treated children. Genetic associations for the susceptibility of these two ADRs have been discovered and replicated⁸-¹², and clinical practice guidelines have been published¹³,¹⁴ outlining which associations have sufficient evidence for their use in clinical practice. Based on these clinical practice guidelines, pharmacogenetic risk prediction models that combined several genetic variants into one predicted outcome for anthracycline-induced cardiotoxicity and cisplatin-induced ototoxicity were developed using logistic regression. In this study, pharmacogenetic risk prediction models for two common ADRs were implemented into clinical practice in pediatric oncology at BC Children’s Hospital. Between July 2013 and September 2018, 279 patients were enrolled in the study and have had their pharmacogenetic risk prediction results returned to their treating oncologists. Results have been incorporated into treatment decision-making and have resulted in treatment modifications such as the use of cardioprotective and otoprotective drugs, increased audiological and cardiac monitoring, and the use of results to decide between different treatment protocols. Prospective evaluation of the occurrence of cardiotoxicity and ototoxicity currently demonstrates that pharmacogenetic-tested patients have experienced significantly less cardiotoxicity than previously treated patients that did not receive pharmacogenetic results over the same follow-up period (3.4% versus 11.8%, p=0.0005). Rates of cisplatin-induced ototoxicity in patients that received pharmacogenetic testing were similar to previously-treated patients used to develop the risk prediction model (58.9% versus 66.7%, respectively), and none of the patients that have received treatment modifications as a result of pharmacogenetic testing have developed clinically relevant ototoxicity (≥ grade 2 ototoxicity). Interviews with patients/families (n=11) and oncologists (n=4) demonstrated that patients/families felt more involved in treatment decisions and were reassured by understanding their risk of toxicity. Oncologists indicated that testing helped ensure that treatment and long-term monitoring were appropriate for each patient.

Item Citations and Data


Attribution-NonCommercial-NoDerivatives 4.0 International