DPYD and 5-fluorouracil toxicity: Results of largest study to date support association

Published: December 1st, 2014

Category: Stories

Use of the antimetabolite 5-fluorouracil (5–FU) is associated with variability in both efficacy and toxicity. Dihydropyrimidine dehydrogenase (DPD) is the rate-limiting catabolic enzyme for 5–FU and eliminates approximately 85% of administered drug. Variability in DPD enzyme activity is affected by multiple factors, including three DPYD gene variants (DPYD*2A, D949V, and I560S). Although some studies have documented an association between these variants and 5–FU toxicity, additional data are needed to inform clinical practice.

Lee and colleagues genotyped patients with stage III colon cancer enrolled in a randomized phase III trial and tested associations between DPYD variants and 5–FU toxicity. Investigators assessed patients for adverse events (AEs) biweekly and used the National Cancer Institute (NCI) Common Toxicity Criteria. A total of 25 DPYD variants functionally characterized to cause decreased DPD activity were tested as part of a larger genetic biomarker screening panel. The primary study outcome was rate of AEs common in 5–FU treated patients (defined as proportion of patients with at least one grade ≥3 5–FU adverse event [5FU–AE] during treatment). Secondary outcomes included rate of any grade ≥3 AE and disease-free survival.

Of the 2,886 eligible patients, complete AE and genotype data were available for 2,594 patients. A total of 33.1% of patients (n = 859) reported any grade ≥3 5FU–AE. The incidence of grade ≥3 5FU–AEs in DPYD*2A, I560S, and D949V carriers was 80% (22/25), 50% (2/4), and 81.5% (22/27), respectively. Researchers found associations between grade ≥3 5FU–AEs and DPYD*2A (OR = 15.21; 95% CI 4.54–50.96, P < 0.001) and D494V (OR = 9.10; 95% CI 3.43–24.10, P < 0.001) variants, which remained significant after adjustment for numerous variables.

Clinical Implications

This is the largest study to date investigating the association between DPYD genetic variants and 5–FU AEs. Although previous meta-analyses of studies examining these associations have yielded conflicting results, investigators wrote that variability in the design of included studies, patient populations, and other factors have limited the power of these analyses. Rather, these conflicting results highlighted “the need for assessment in larger patient populations receiving comparable treatment regimens with uniform clinical data,” authors wrote.

This statement is echoed in an accompanying editorial, in which the editorialist wrote, “For *2A and D949V, all the tenets of a clinically applicable pharmacogenetic marker are met.” Specifically, the variants are functional, are involved in a major 5–FU metabolic pathway, and patients carrying these variants have a higher risk for AEs. “In light of the current results, clinicians are strongly encouraged to consider testing for *2A and D949V in patients treated with either 5–FU- or capecitabine-based regimens,” the editorialist concluded.

More information on DPYD testing, including available genetic test options, FDA-approved biomarker information, and recommendations of the Clinical Pharmacogenetics Implementation Consortium and the Dutch Pharmacogenetics Working Group guidelines are available through the Pharmacogenomics Knowledgebase.


Lee AM et al. DPYD variants as predictors of 5-fluorouracil toxicity in adjuvant colon cancer treatment (NCCTG N0147). J Natl Cancer Inst. 2014 December [Epub ahead of print].

Innocenti F. DPYD variants to predict 5-FU toxicity: the ultimate proof [editorial]. J Natl Cancer Inst. 2014 December [Epub ahead of print].