TY - GEN
T1 - Clinical implementation of DPYD-genotypingand DPD-phenotyping in Denmark
AU - Paulsen, Niels Herluf
PY - 2022/12/22
Y1 - 2022/12/22
N2 - A group of chemotherapeutic drugs called fluoropyrimidines (FP) has been used in clinical oncology
since the beginning of the 1960s. FP acts via the metabolite 5-fluorouracil (5-FU) and includes the
drugs 5-fluorouracil, capecitabine, and tegafur. These different formulations of FP are the cornerstone
of treating several solid tumors related to the gastrointestinal tract, such as colorectal cancer, gastric
cancer, and pancreatic cancer, thereby making it one of the most used cancer drugs worldwide. FP
drugs are metabolized mainly by the rate-limiting enzyme dihydropyrimidine dehydrogenase (DPD). For decades it has been known that patients with decreased function of the DPD-enzyme are at elevated risk of severe adverse events, some deathly if treated with standard doses of FP. However,
everyday clinical use of DPD testing has been lacking. Several clinical studies have examined different testing methods to measure patient DPD-enzyme
activity before treatment and adjust the dose of FP accordingly. The two most commonly used methods in today's clinical practice are genotyping of the DPYD-gene, the encoding gene for the DPDenzyme, and measurement of the endogenous metabolite uracil (DPD-phenotyping).
Both testing methods have strengths and weaknesses, and the evidence supporting routine clinical use
is still relatively sparse. In this PhD project, we aimed to add more evidence supporting the use of DPYD-genotyping by
implementing a DPYD-genotype test for all patients treated with FP drugs at Odense University Hospital (OUH).The clinical implementation of the DPYD-genotype at OUH changed the treatment for a large group
of cancer patients. The project involved several departments at OUH, including the Department of
Clinical Biochemistry, the Department of Clinical Pharmacology, and the Department of Oncology.
Early in the process, we decided to provide the DPYD-genotype test to all patients, not just those
willing to participate in the clinical study.Before initiating the clinical study, the European Medicines Agency (EMA) recommended testing for
all patients before treatment with FP. This recommendation took many oncologic departments by
surprise in Denmark. In Odense, we could provide DPD testing to patients and fulfill the recommendation within a few months because we were already preparing the clinical study. Therefore OUH
became the first hospital in Denmark to offer routine DPYD-genotyping for all cancer patients intended for FP treatment.Patients were genotyped for four clinically relevant DPYD-variants: DPYD*2A, DPYD*13, D949V,
and HapB3. Patients identified as heterozygous for one of the tested DPYD-variants were treated with
a 50% dose of FP drugs to reduce their risk of FP-related toxicity.During this Ph.D. project, three studies were conducted, all focusing on the clinical use of DPYDgenotyping and DPD-phenotyping. Study I. This review article presents the evidence supporting the use of DPYD-genotyping and DPD-phenotyping in clinical oncology. The review is based on two systematic literature searches following the
PRISMA guidelines. We included clinical studies examining the frequency of toxicity in patients
tested with DPYD-genotyping or DPD-phenotyping. Nine and 12 studies met the inclusion criteria
for DPYD-genotyping and DPD-phenotyping, respectively. The data from the included studies are
presented and discussed in the review. We conclude in the review that the evidence supporting the
use of DPYD-genotyping and DPD-phenotyping in clinical oncology is sparse, and further evidence
is needed to guide clinicians in dosing FP.Study II.The following months after the EMAs recommendation, DPD-activity testing was implemented
across Denmark. Because no studies had examined the prevalence of DPD deficiency in the Danish
population, we decided to collect data from all hospital laboratories performing DPD testing across
Denmark from July 2020 to December 2021. In total, 4,228 Danish cancer patients were tested in the
data collection period. Of the 4,215 DPYD-genotyped patients, 92.4 % were DPYD wild type. The
distribution DPYD-variants was: DPYD*2A (1%), D949V (1.4%), DPYD*13 (0.2%), and HapB3
(4.9%). Four patients were compound heterozygous.Among 2,260 phenotyped patients, 2,132 (94.3%) had uracil concentrations[U] below 16 ng/ml indicating normal DPD activity. The remaining 128 (5.7%) patients had [U] ≥16 ng/mL and <150 ng/ml
indicating decreased DPD activity.The prevalence of clinically relevant DPYD-variants in Danish cancer patients was 7.6%, which is in
line with similar studies of other European populations. 5.7% of phenotyped patients had [U] ≥ 16
ng/ml. Patients carrying DPYD-variants had higher mean [U] than wild-type, but the uracil concentrations varied substantially in both groups. We found that the strength of the correlation between
genotype and phenotype was poor.Study III.The clinical benefit of DPYD-genotyping was examined in this clinical study. Cancer patients intended for their first treatment of FP were enrolled. Participants who consented to participate in the
study had their blood drawn for later DPD-phenotyping.In the study, 230 patients were included and treated with FP-dosings according to their DPYD-genotype. After treatment with FP, the occurrence of FP-related toxicity was compared with a historical
control cohort of patients also treated at OUH. The control group consisted of 492 cancer patients
treated with FP before the clinical implementation of the DPYD-genotype.The study examined the frequency of grade ≥ 3 FP-related toxicity and could not demonstrate a clinical benefit when comparing the two groups. Patients carrying DPYD-variants had, after the implementation, a lower frequency of FP-related hospitalization(19% vs. 0%) and death(4.8% vs. 0%).
Data from DPD-phenotyping demonstrated that wild-type patients with [U] ≥ 16 ng/ml had a higher
occurrence of overall grade ≥ 3 FP-related toxicity compared to patients with [U] < 16 ng/ml (28%
vs. 55%).
AB - A group of chemotherapeutic drugs called fluoropyrimidines (FP) has been used in clinical oncology
since the beginning of the 1960s. FP acts via the metabolite 5-fluorouracil (5-FU) and includes the
drugs 5-fluorouracil, capecitabine, and tegafur. These different formulations of FP are the cornerstone
of treating several solid tumors related to the gastrointestinal tract, such as colorectal cancer, gastric
cancer, and pancreatic cancer, thereby making it one of the most used cancer drugs worldwide. FP
drugs are metabolized mainly by the rate-limiting enzyme dihydropyrimidine dehydrogenase (DPD). For decades it has been known that patients with decreased function of the DPD-enzyme are at elevated risk of severe adverse events, some deathly if treated with standard doses of FP. However,
everyday clinical use of DPD testing has been lacking. Several clinical studies have examined different testing methods to measure patient DPD-enzyme
activity before treatment and adjust the dose of FP accordingly. The two most commonly used methods in today's clinical practice are genotyping of the DPYD-gene, the encoding gene for the DPDenzyme, and measurement of the endogenous metabolite uracil (DPD-phenotyping).
Both testing methods have strengths and weaknesses, and the evidence supporting routine clinical use
is still relatively sparse. In this PhD project, we aimed to add more evidence supporting the use of DPYD-genotyping by
implementing a DPYD-genotype test for all patients treated with FP drugs at Odense University Hospital (OUH).The clinical implementation of the DPYD-genotype at OUH changed the treatment for a large group
of cancer patients. The project involved several departments at OUH, including the Department of
Clinical Biochemistry, the Department of Clinical Pharmacology, and the Department of Oncology.
Early in the process, we decided to provide the DPYD-genotype test to all patients, not just those
willing to participate in the clinical study.Before initiating the clinical study, the European Medicines Agency (EMA) recommended testing for
all patients before treatment with FP. This recommendation took many oncologic departments by
surprise in Denmark. In Odense, we could provide DPD testing to patients and fulfill the recommendation within a few months because we were already preparing the clinical study. Therefore OUH
became the first hospital in Denmark to offer routine DPYD-genotyping for all cancer patients intended for FP treatment.Patients were genotyped for four clinically relevant DPYD-variants: DPYD*2A, DPYD*13, D949V,
and HapB3. Patients identified as heterozygous for one of the tested DPYD-variants were treated with
a 50% dose of FP drugs to reduce their risk of FP-related toxicity.During this Ph.D. project, three studies were conducted, all focusing on the clinical use of DPYDgenotyping and DPD-phenotyping. Study I. This review article presents the evidence supporting the use of DPYD-genotyping and DPD-phenotyping in clinical oncology. The review is based on two systematic literature searches following the
PRISMA guidelines. We included clinical studies examining the frequency of toxicity in patients
tested with DPYD-genotyping or DPD-phenotyping. Nine and 12 studies met the inclusion criteria
for DPYD-genotyping and DPD-phenotyping, respectively. The data from the included studies are
presented and discussed in the review. We conclude in the review that the evidence supporting the
use of DPYD-genotyping and DPD-phenotyping in clinical oncology is sparse, and further evidence
is needed to guide clinicians in dosing FP.Study II.The following months after the EMAs recommendation, DPD-activity testing was implemented
across Denmark. Because no studies had examined the prevalence of DPD deficiency in the Danish
population, we decided to collect data from all hospital laboratories performing DPD testing across
Denmark from July 2020 to December 2021. In total, 4,228 Danish cancer patients were tested in the
data collection period. Of the 4,215 DPYD-genotyped patients, 92.4 % were DPYD wild type. The
distribution DPYD-variants was: DPYD*2A (1%), D949V (1.4%), DPYD*13 (0.2%), and HapB3
(4.9%). Four patients were compound heterozygous.Among 2,260 phenotyped patients, 2,132 (94.3%) had uracil concentrations[U] below 16 ng/ml indicating normal DPD activity. The remaining 128 (5.7%) patients had [U] ≥16 ng/mL and <150 ng/ml
indicating decreased DPD activity.The prevalence of clinically relevant DPYD-variants in Danish cancer patients was 7.6%, which is in
line with similar studies of other European populations. 5.7% of phenotyped patients had [U] ≥ 16
ng/ml. Patients carrying DPYD-variants had higher mean [U] than wild-type, but the uracil concentrations varied substantially in both groups. We found that the strength of the correlation between
genotype and phenotype was poor.Study III.The clinical benefit of DPYD-genotyping was examined in this clinical study. Cancer patients intended for their first treatment of FP were enrolled. Participants who consented to participate in the
study had their blood drawn for later DPD-phenotyping.In the study, 230 patients were included and treated with FP-dosings according to their DPYD-genotype. After treatment with FP, the occurrence of FP-related toxicity was compared with a historical
control cohort of patients also treated at OUH. The control group consisted of 492 cancer patients
treated with FP before the clinical implementation of the DPYD-genotype.The study examined the frequency of grade ≥ 3 FP-related toxicity and could not demonstrate a clinical benefit when comparing the two groups. Patients carrying DPYD-variants had, after the implementation, a lower frequency of FP-related hospitalization(19% vs. 0%) and death(4.8% vs. 0%).
Data from DPD-phenotyping demonstrated that wild-type patients with [U] ≥ 16 ng/ml had a higher
occurrence of overall grade ≥ 3 FP-related toxicity compared to patients with [U] < 16 ng/ml (28%
vs. 55%).
U2 - 10.21996/1bfh-rf75
DO - 10.21996/1bfh-rf75
M3 - Ph.D. thesis
PB - Syddansk Universitet. Det Sundhedsvidenskabelige Fakultet
ER -