Adoption and practice variation of DPYD genotyping prior to fluoropyrimidine therapy at two Canadian cancer centres.

Abstract

528 Background: Fluoropyrimidines are widely used in the treatment of solid malignancies. Variants in the dihydropyrimidine dehydrogenase ( DPYD ) gene increase the risk of severe fluoropyrimidine-associated toxicity. Following guidance from Ontario Health (Cancer Care Ontario) recommending DPYD genotyping prior to fluoropyrimidine therapy in 2023, we evaluated factors associated with testing, variant prevalence, and treatment initiation practices at two comprehensive cancer centres. Methods: We conducted a retrospective multicentre study of adult patients with solid tumors treated at Princess Margaret Cancer Centre (PMCC) and Cancer Centre of Southeastern Ontario (CCSEO) who received at least one dose of fluoropyrimidine chemotherapy for the first time between Mar 27 2023 and Oct 29 2025. Patients who initiated fluoropyrimidine treatment prior to DPYD genotyping availability were excluded. Data on DPYD genotyping for the four guideline-recommended variants, variant prevalence, treatment initiation before result availability, and treatment delays related to genotyping were collected. Chi-square and multivariable logistic regression analyses identified factors associated with DPYD genotyping. Results: Among 616 eligible patients, 447 (72.6%) underwent DPYD genotyping. Testing rates differed significantly by treatment centre and tumour site, but not by treatment intent (Table 1). In multivariable analysis, treatment at PMCC (aOR 0.61, 95% CI 0.40–0.93, p=0.021) and a diagnosis of head and neck cancer (aOR 0.21, 95% CI 0.08–0.51, p=0.001) were independently associated with lower odds of DPYD genotyping. Among tested patients, 73 (16.3%) initiated treatment before DPYD results were available and 42 (9.4%) experienced treatment delays while awaiting results. Twenty-six patients (5.8%) carried a clinically relevant DPYD variant. HapB3 was most common (n=19), followed by DPYD*2A (n=4) and DPYD*9B (n=4); with one patient compound heterozygote for HapB3 and DPYD*9B. Conclusions: Following implementation of DPYD genotyping at two cancer centres, there were gaps in uptake and timing of testing. Variation according to treatment centre and tumour site, together with treatment initiation before result availability, identifies opportunities to improve integration of pharmacogenomic testing into routine oncology practice and maximize fluoropyrimidine safety. DPYD genotyping uptake according to patient and treatment characteristics. Characteristic Tested n/N (%) p-value Centre 0.016 PMCC 287/413 (69.5) CCSEO 160/203 (78.8) Intent 0.075 Curative 261/346 (75.4) Palliative 186/270 (68.9) Tumor Site 0.003 Luminal GI 220/298 (73.8) HPB 109/149 (73.2) Breast 88/113 (77.9) Head + Neck 11/27 (40.7) Other 19/29 (65.5)