Is uracil enough for effective pre-emptive DPD testing?
Niels Heersche,Maja Matic,Ron H.J. Mathijssen,Marieke J.H. Coenen,Ron H.N. van Schaik
DOI: https://doi.org/10.1515/cclm-2024-0742
2024-07-20
Clinical Chemistry and Laboratory Medicine (CCLM)
Abstract:To the Editor, With great interest, we read the article by Launay et al. [1], which compared results of pre-emptive dihydropyrimidine dehydrogenase (DPD) deficiency testing for fluoropyrimidine treatment using both DPYD genotyping and plasma uracil levels. The impressive sample size demonstrates that routine pre-treatment DPD deficiency testing is feasible and has been fully adapted into clinical practice in France. However, we have several comments and concerns regarding the data and its interpretation. Firstly, the authors opted to use a genotyping panel that omits haplotype B3 (HapB3; c.1129–5923C>G (rs75017182) and the commonly tested variant in linkage p.E412E; c.1236G>A (rs56038477)). Although indeed conflicting results regarding the adequate dose for HapB3 carriers exist, its association with increased toxicity rates, such as severe neutropenia and diarrhea, remains undisputed. In meta-analysis a relative risk rate for severe grade ≥3 fluoropyrimidine related toxicity of 1.72 (1.22–2.42) was found [2], and even among 51 patients treated at a 75 % dose intensity, an increased incidence of grade ≥3 adverse events of 39 % persisted [3]. The exclusion of HapB3 from the current analysis therefore hampers adequate comparison between both tests. Secondly, the authors argue that targeted genotyping is not sufficiently reliable in the detection of patients who have a complete and profound DPD deficiency. In support of this argument, details are provided of five patients who had a complete DPD deficiency according to French health authority recommendations, with only one of them having a DPYD variant genotype (see ref [1], table 1, patient C: homozygous *2A genotype). Interestingly, this is also the only patient who had unmeasurable dihydrouracil levels. Reliable measurement of uracil and dihydrouracil levels, as noted by the authors, is heavily dependent on pre-analytical handling. Lack of adherence to recommendations regarding sample handling can result in false positive DPD deficiency diagnoses and subsequent undertreatment due to unnecessary dose-reductions. Moreover, renal impairment and abnormal liver function may also result in false positive results. Notably, one of the five described patients (i.e. patient B) had a very high dihydrouracil level, resulting in legitimate concerns of the authors regarding the reliability of the completely deficient status in this specific case. Regrettably, data on sample handling as well as clinical data relating to kidney and liver function were unavailable to the authors, making a final conclusion regarding the patient's DPD status difficult. Nonetheless, we have concerns regarding the validity of the diagnosis of a completely deficient DPD status in the remaining three patients as well, suggesting that, in contrast to the authors' conclusion, both genotyping and phenotyping are able to identify completely deficient patients. Literature on patients with a complete DPD deficiency is sparse. However, case-reports of four completely DPD deficient patients (genotype confirmed) show that these patients not only share the inability to form 5-FU metabolites (DHFU, FUPA or FBAL being not quantifiable), but also have undetectable levels of dihydrouracil [4], [5], [6]. Since the remaining three patients who were identified as being completely DPD deficient (i.e. patients A, D, and E) had detectable dihydrouracil levels, with two of the patients even having dihydrouracil levels within the physiological range, we consider it unlikely that these patients are truly completely DPD deficient. Although the patients match the criteria for a complete DPD deficiency based on uracil testing guidelines, we believe they are more likely to be partially deficient patients, in which case a major treatment line (i.e. fluoropyrimidines) has been unnecessarily withheld from them. Still, the absence of common DPYD variants in these patients highlights an important limitation of targeted DPYD genotyping: its inability to detect rare or unknown genetic variants. In that regard, DPYD sequencing of these patients would greatly aid interpretation of the results, as presence of rare DPYD variants could support diagnosis of a DPD deficiency. Consequently, this would demonstrate the added value of combining DPYD genotyping with uracil-based phenotyping, especially for patients with extremely elevated uracil levels. For now however, hard conclusions cannot be drawn from these cases due to the absence of additional data. Finally, the lack of clinical outcomes or a corresponding DPD enzyme activity measured in peripheral blood mononuclear cells greatly hampers the interpretation of the results. Currently, prospective evidence for dose-reductions based on pre-emptive DPD testing is only available for genotyping [3, 7]. The evidence for uracil-based dosing is largely retrospective [7]. Unfortu -Abstract Truncated-
medical laboratory technology