The correlation between DPYD*9A (c.85T > C) genotype and dihydropyrimidine dehydrogenase deficiency phenotype in patients with gastrointestinal malignancies treated with fluoropyrimidines: Updated analysis

544 Background: The correlation between DPYD*9A (c.85T > C) genotype and dihydropyrimidine dehydrogenase (DPD) deficiency phenotype is controversial. In our cohort of 28 patients with gastrointestinal malignancies (GI) treated with fluoropyrimidines, DPYD*9A was the most commonly diagnosed variant (46%) and there was a noticeable genotype-phenotype correlation (Khushman et al). In this updated analysis, a larger cohort of a mixed racial background was genotyped for DPYD*9A variant to confirm the incidence and genotype-phenotype correlation. Methods: Between 2011 and 2018, in addition to genotyping for high risk DPYD variants (DPYD*2A, DPYD*13 and DPYD*9B), genotyping for DPYD*9A variant was performed on 113 patients. Fluoropyrimidines-associated toxicity was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (v 5.0). Results: DPYD variants were identified in 61 patients. DPYD*2A was identified in 3 patients and DPYD*9B was identified in 2 patients (one patient had double heterozygous *9A and *9B). Heterozygous DPYD*9A was identified in 46 patients (41%) and homozygous DPYD*9A was identified in 11 patients (10%). Among patients with DPYD*9A variant, Caucasians represented 51% and African Americans represented 46%. 27 patients (47%) were females. Grade 3-4 toxicities were experienced in 26 patients with mutant DPYD*9A (3 patients had homozygous DPYD*9A) and in 20 patients with no identified DPYD mutation (P = 0.7035). In patients who received full dose fluoropyrimidines (N = 85), grade 3-4 toxicities were experienced in 22 patients with mutant DPYD*9A (2 patients had homozygous DPYD*9A) and in 17 patients with no identified DPYD mutation (P = 0.8275). Conclusions: In our updated analysis, DPYD*9A variant was the most commonly diagnosed variant. The correlation between DPYD*9A variant and DPD clinical phenotype was not reproduced. The noticeable correlation that we previously reported is likely due to small sample size and patients’ selection and testing bias.