Page 32 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Chapter 2
and phenotyping included the DHU/U ratio. Two hundred forty seven patients with grade 3–5 toxicity were retrospectively tested. In total, 3% of all patients carried one or more mutations. Twenty seven out of 247 retrospectively tested patients died of whom 16 (59%) and 24 (89%) were identified with genotyping or phenotyping, respectively. The combined approach would have identified 98% of grade 3–5 toxicity patients and 100% of mortalities.63
(Cost) Effectiveness of DPD deficiency testing
A prospective, multicentre study was conducted by Deenen et al., in which 2,038 patients were screened for DPYD*2A prior to start with 5-FU or capecitabine.64 Twenty-two patients (1.1%) were heterozygous carriers of DPYD*2A and patients received an initial dose reduction of 50% when starting therapy, followed by dose titration based on clinical tolerance. Toxicity results showed that the risk of grade ≥3 toxicity was significantly reduced to 28% compared to 73% in historical controls (p<0.001). Drug-induced death reduced from 10% to 0%. This study convincingly shows that pre-treatment genotyping of DPYD*2A followed by dose adjustment in carrier patients improves patient safety. A cost analysis was executed using a decision analytic model from a health care payer perspective, including only direct medical costs. Genotyping costs were €75 per test. The average total treatment cost per patient was slightly lower for screening (€2,772) than for non-screening (€2,817). The approach was shown to be feasible in routine clinical practice.64 Ahmed et al. presented a cost analysis of a retrospective screening for four DPYD variants in 31 patients who experienced grade 3–5 toxicity. Five patients carried a variant and were admitted to the ICU due to toxicity. The costs of hospital admission (€155,083) were much higher than the screening costs of all patients starting with fluoropyrimidine therapy for CRC during the study period (€26,800).53 Another retrospective study of 48 patients shows cost effectiveness with DPYD screening costs for four variants being almost nine times lower than hospital admissions of four patients (£1,776 versus £15,525; approximately €2,500 versus €21,500).58 We must bear in mind that genotyping technology is developing fast and prices continue to decline.37 Phenotyping tests have been recently reviewed by van Staveren et al., and to our knowledge, to date no additional cost-effectiveness analysis for a phenotyping test has been published.29
Recommendations and guidelines of DPYD pharmacogenomics
Warnings or contraindications for using 5-FU/capecitabine in DPD deficient patients are stated by the FDA and EMA.65,66 This is meaningless without knowing, and thus testing a patient for DPD deficiency. No formal recommendations on pre-therapeutic (upfront) screening for DPD deficiency are given by health authorities, regulatory agencies or guideline committees from the National Comprehensive Cancer Network or American Society of Clinical Oncology. The European Society for Medical Oncology explicitly states that they do not recommend upfront routine testing for DPD deficiency despite the risk of severe and potential lethal toxicity.67 It is unknown to us what arguments underlie this recommendation. Only in cases of severe toxicity due to 5-FU treatment DPD deficiency screening is strongly recommended, and exposure to standard dose of 5-FU is contraindicated in proven DPD deficiency patients, according to guidelines published in 2012.67 The lack of official recommendations on pre- therapeutic genotyping is limiting the process of implementation. One of the reasons may
30