Page 342 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Chapter 14
of severe fluoropyrimidine-induced toxicity. No treatment-related deaths occurred in DPYD variant allele carriers who were treated with a reduced dose. Despite the low frequency of DPYD variant allele carriers, executing prospective DPYD genotyping did not increase costs, but reduced average costs slightly with €50 per patient, as was shown in the cost analysis of the trial (chapter 6).
Current PGx guidelines do not distinguish fluoropyrimidine dosing recommendations between treatment regimens. Fluoropyrimidine dosages in chemoradiation therapy are substantially lower compared to fluoropyrimidine dosages in other treatment regimens. Therefore, it was unclear if further fluoropyrimidine dose reductions could result in underdosing in DPYD variant allele carriers treated with chemoradiation therapy. In chapter 7 we compared severe toxicity between wild-type patients and DPYD variant allele carriers who received chemoradiation therapy, the latter group either treated with standard or reduced fluoropyrimidine dosages. DPYD variant allele carriers treated with regular fluoropyrimidine doses in chemoradiation therapy experienced severe toxicity more often compared to DPYD variant allele carriers treated with reduced fluoropyrimidine doses in chemoradiation therapy, showing dose reductions are required as well in this treatment regimen.
The feasibility of implementing prospective DPYD genotyping in daily clinical care was shown in chapter 8 of this thesis. The first 21 months of DPYD genotyping at the Leiden University Medical Center (LUMC) were investigated, starting with the introduction as routine care in April 2013 until the end of the observation period in December 2014. This study showed that the implementation of DPYD genotyping was first characterised by a learning or acceptance curve, but was feasible in a real world clinical setting with 90─100% of the patients treated with fluoropyrimidines being genotyped. This study also showed 90% dose adherence.
Another aspect of (DPYD) genotyping is the certainty of a test result, and the consequences of an erroneous result. In chapter 9 we describe the dilemma of confirmation practice as a quality control aspect of PGx testing. We discuss if it should be required to have two independent genotyping assays to correctly determine a genotype. In this study we discovered that, even after extensive validation, erroneous results can still occur due to misclassification of a genotype, e.g. caused by allele dropout. Despite the increase in costs and labour, a confirmation method is useful for genetic tests with high clinical impact, such as DPYD testing. Clear guidelines will help to align confirmatory laboratory practices for pharmacogenetics, which may need to be specified per gene and test.
Beyond current DPYD pharmacogenetics
In the first part of this thesis we describe how to reduce severe fluoropyrimidine-induced toxicity by DPYD genotyping of DPYD*2A, DPYD*13, c.2846A>T and c.1236G>A. Yet, it is known not all severe fluoropyrimidine-induced toxicity can be predicted by these four variants alone. Therefore, we investigated other options, beyond genotyping of the current four DPYD variants, to reduce severe fluoropyrimidine-induced toxicity. This is shown in the second part of this thesis, entitled “beyond current DPYD pharmacogenetics”.
In chapter 10, we present a first-time head-to-head comparison study of four DPD 340