<i>DPYD</i> testing in GI cancers: A real-world analysis of costs and quality of care.

Abstract

821 Background: Fluoropyrimidines are foundational agents in GI oncology, however they can cause severe toxicity in patients with reduced dihydropyrimidine dehydrogenase activity due to mutations in the DPYD gene. While complete deficiency is rare, partial deficiency affects 3–8% of the U.S. population - a clinically meaningful minority. Pretreatment DPYD genotyping identifies at-risk patients, personalizing fluoropyrimidine dosing to reduce toxicity. Despite NCCN guidance to consider upfront DPYD testing, adoption in the US remains limited with a lack of real-world data on cost and quality outcomes. Methods: We retrospectively analyzed 233 consecutive patients who initiated fluoropyrimidine therapy for GI cancers at two Moffitt Ambulatory Centers, 1/2024-9/2025. Patients were categorized as receiving upfront DPYD genotyping/dose modification ( n =122) or as untested/standard dosing ( n =111). Outcomes included adverse events (AEs), hospitalizations for toxicity, dose reductions or treatment discontinuation within 60 days and cost estimates. Between-group differences were assessed using t-test, chi-square or Fisher’s exact tests, as appropriate. Results: Baseline characteristics were similar, with mean age of 60 in both and majority having stage III/IV disease and ECOG 1. Upfront testing identified 5% DPYD variant prevalence ( n =6), all of whom received genotype-guided dose reductions. The tested cohort had fewer clinically significant AEs than the untested group (17 vs 25; p =0.044). No hospitalizations for toxicity occurred in the tested cohort versus 6 in the untested group ( p =0.01). The cost of hospitalizations due to treatment-related toxicity was estimated at $303,528, whereas the cost of implementing upfront DPYD testing was $44,880. In the untested group, 8% discontinued fluoropyrimidines permanently due to toxicity ( n =9), while none did in the tested group, leading to significantly higher treatment adherence with upfront testing ( p =0.001). Pretreatment DPYD genotyping was projected to save $2356/patient starting fluoropyrimidines by mitigating risk of severe toxicity and hospitalization. Conclusions: In this real-world study, DPYD genotyping and personalized dosing was associated with significantly less toxicities, hospitalizations, and treatment discontinuations compared with standard dosing. These findings support broader adoption of upfront DPYD testing as a strategy to improve patient safety, treatment adherence, and reduce healthcare costs. Outcome, n (%) DPYD Tested (n=122) Untested (n=111) P-value Dose-limiting Treatment Related Adverse Effects 17 (14) 25 (23) 0.0443 Dose Reductions 18 (15) 25 (23) 0.0634 Discontinuation of therapy 0 9 (8) 0.0011 Hospitalizations 0 6 (5) 0.0109 Cost per patient ($)* 367.86 2724.49 *Calculated using estimated total cohort costs of hospitalization/upfront genotyping and averaged per patient. Diverse payer mix model.