Supervised by: Merissa Hickman BSc (Hons). Merissa studied Biomedical Science at the University of Hull. She is a current postgraduate student at the University of Cambridge studying Genomic Medicine. She was awarded a Cambridge Trust Scholarship to complete her MPhil.


Introduction: The DPYD gene plays a pivotal role in encoding the production of the DPD enzyme, which catalyses a group of chemotherapeutic drugs known as fluoropyrimidines, widely used in cancer treatment.

Aims and Objectives: This research paper highlights the essential features of DPYD testing and assesses its effectiveness through an extensive review of clinical trials, drawing insights into the feasibility of DPYD testing implementation in both France and the UK. Notably, countries that mandate DPD testing in cancer treatment also conduct clinical trials, the results of which indicate a significant correlation between increased DPYD testing adoption and reduced toxicity responses in cancer treatment. Clinical data from DPYD testing-implementing countries, such as France and the UK, support this assertion as they exhibit lower rates of decline and toxicity in chemotherapy treatment compared to countries that are less stringent in implementing DPYD testing. This research paper also delves into a critical evaluation of the efficacy of DPYD testing mandates, with a specific focus on France’s proactive implementation and the UK’s lack thereof.

Main Findings: Studies conducted within both England and France found that implementing individualised dosing in response to DPD testing results lowered toxicity rates due to fluoropyrimidines by 3-5%. This implies other European countries with similar gene pools, including the UK, should implement a DPD testing mandate as France has.

Conclusion: While the efficacy of DPD testing has been established as useful and effective in lowering rates of toxicity in response to fluoropyrimidines, there is also the matter of testing cost. This study did not cover cost effectiveness of the test, and as there is a substantial monetary requirement to take this test, this presents ethical and economic concerns for the healthcare system. 

1) Introduction

A comprehensive examination of current literature reveals that DPYD testing is already a standard practice across Europe, where the establishment of clear clinical guidelines, including reimbursement policies, has been instrumental in the efficient implementation of DPYD testing. Equally vital is the recognition amongst oncologists for the impact of toxicity and fatality rates in patients receiving fluoropyrimidines. Recent research indicates that toxicity is experienced by 10-40% of patients receiving fluoropyrimidines, with a fatality rate ranging from 0.2% to 0.8% (Martins et al., 2020). Moreover, Cancer Research UK has unveiled partial DPD deficiency in 2 to 8% of the human population and severe DPD deficiency in 1% of individuals. DPYD testing plays a pivotal role in identifying potential gene variants that may alter the shape of the DPD enzyme, consequently affecting drug metabolism. This, in turn, opens the door to the application of pharmacogenomics in fluoropyrimidine-based treatments, leading to reduced drug doses for patients with DPYD gene variants. The awareness that 8% of the European population and 3-8% of the Caucasian population have DPD deficiency has been brought to the forefront by the New Zealand Medicines and Medical Devices Safety Authority. The mandatory implementation of DPYD testing in hospitals and medical institutes worldwide ensures that individuals with lower DPD enzyme levels receive more efficient treatment, resulting in reduced toxicity. 

Clinics and hospitals have started to obligate a form of DPD testing to patients receiving chemotherapeutic drugs. These countries include European countries such as France and, to some extent, the UK. By collecting statistics and data from these countries and comparing these data, a correlation between an increase in frequency of DPYD testing implementation and decreased toxicity in cancer treatment and decreased fatality rates can be seen. Through the implementation of pharmacogenomics into cancer treatment and by using DPYD tests, patients will receive drugs  that are the most effective in treating them.

2) Role of Fluoropyrimidines in Cancer Treatment

Fluoropyrimidines are used to treat cancer because it can prevent the division of proliferating cancerous cells. Fluorouracil, a type of fluoropyrimidine, works as an inhibitor with its ability to prevent DNA or RNA repair and synthesis. Previous studies show it achieves this by interfering with the formation of the enzyme-substrate complex involving thymidylate synthase (TS) via enzyme inhibition (Longley et al 2003). This thoroughly disrupts DNA replication and limits the cell’s ability to replicate. 

Further research into fluoropyrimidine activity pathways reveal that the detailed analysis of this process involves the catabolism of fluoropyrimidine into its inactive metabolites: FdUMP, FdUTP, and FUTP (Thorn et al., 2011). Each of these metabolites has its own action mechanism. FdUMP prevents the binding of dUMP with TYMS, and this then prevents the conversion of dUMP into dTMP, which is involved in DNA synthesis. FdUMP also disrupts the folate cycle, which synthesises the start codon in DNA replication. TYMS inhibition causes imbalances in fDUMP and FUTP, causing the misincorporation of FUTP into DNA and FdUMP into RNA.

A study by Odotof Et. al in 2015 shows that cancerous cells cannot undergo mitosis if they lack the genes that code for enzymes and proteins, making fluoropyrimidines excellent anticancer drugs. However, fluoropyrimidine accumulation can cause overexposure of the thymidylate synthase to fluorouracil, which may result in the development of resistance to its mechanism. 

The overaccumulation of fluoropyrimidines often occurs due to a deficiency in dihydropyrimidine dehydrogenase (DPD), an enzyme responsible for breaking down these drugs. When DPD isn’t functioning properly, fluoropyrimidines can build up to toxic levels. This buildup can result in the formation of harmful substances that not only affect cancer cells but also harm healthy cells in the body. This toxicity poses a significant challenge in cancer treatment, as it can lead to side effects and limit the effectiveness of these drugs.

3) DPD Testing

The enzyme responsible for fluoropyrimidine breakdown is dihydropyrimidine dehydrogenase (DPD). This enzyme serves as the fluoropyrimidine rate-limiting enzyme, which regulates its anti-cancer effect. In regards to fluorouracil, a widely-used fluoropyrimidine, DPD metabolises 80% of fluorouracil into inactive metabolite, dihydrofluorouracil (DHFU), in the liver. This prevents fluorouracil accumulation and its overexposure to the TS enzyme (Dean, 2016.)

The DPD enzyme plays a vital role in regulating the effects of this wide scope of anti-cancer medication. However, approximately 0.5% of patients who have cancer have a severe DPD deficiency and 2-8% of adult cancer patients have a moderate DPD enzyme deficiency (Tutsi et al., 2018). For these patients, receiving fluorouracil at doses typically prescribed to individuals without DPD deficiency can lead to unwelcome side effects due to the toxic medication buildup. Such effects include nausea, vomiting, and hair loss (Saif et al., 2008).

DPD deficiency is caused by mutations in the DPYD gene which encodes for the DPD enzyme. However, it’s worth noting that not all DPYD gene variants exert the same level of reduction in DPD enzyme activity. Variations in this gene give rise to a spectrum of enzymatic capabilities, with certain variants posing a higher risk of fluoropyrimidine toxicity.

For instance, a comprehensive meta-analysis conducted by Deac found that specific DPYD gene variants, such as DPYD*13 and c.1236G>A/HapB3, are associated with a heightened risk of experiencing fluoropyrimidine-related toxicity. These genetic markers, through their influence on DPD enzyme activity, serve as valuable indicators for tailoring cancer treatment to an individual’s specific needs.

(Deac, 2020)


Effect on DPD enzyme DOSE 
C.2846A>T Variant  Homozygous received 50% reduction. Heterozygous receives  25% reduction 
c.1129-5923C>G Pre mRNA  splicing
c.1236G>A 50% decrease in enzyme activity Dose reduction 25%

Adverse drug reactions are a result of ineffective metabolism of a drug. Toxicity in cancer treatment occurs as a result of these adverse drug reactions due to variance in drug metabolism across patients (Dan M. Roden, MD 2020). In regards to fluoropyrimidine medications, variations in the DPYD gene lead to different rates of its metabolism in the body. Without adjusting the doses, low-metabolising patients can experience toxic buildup of this anti-cancer medication designed to hasten cell death. Fluoropyrimidine buildup due to DPD deficiency can cause severe inflammation, ulceration of gastrointestinal tract, and low white blood cell concentration (Yuya Hagiwara et.Al 2022). 

However, these side effects can be avoided through implementation of DPYD testing, which allows the personalisation of fluoropyrimidine drug prescription in cancer treatment. Personalised medicine means supplying patients with drug doses that are most effective in treating them. Through DPD testing, every patient will receive a drug dose based on their genotype. 

Therefore, to implement effective cancer treatment, the discovery of every patient’s DPD enzyme activity is required. If variants in the DPYD genes are identified and altered DPD metabolic activity is found, this implies that these patients require lower fluorouracil drug doses to avoid an adverse toxicity response. Through pharmacogenomics, patients will experience reduced toxic side effects and expensive chemotherapeutic drugs, like fluorouracil, are conserved. 


The application of pharmacogenomics in cancer treatment presents challenges due to the narrow therapeutic window required to determine the optimum treatment (Morawska et al., 2018). A drawback of DPYD testing is that not all DPD deficiencies result from genetic mutations in the DPYD gene; mutations in other genes such as TYSM, ABCB1, and MTHF can also lead to poor fluoropyrimidine metabolism (Pratt & Scott, 2017). This can result in a deficiency in DPD enzyme functionality going undetected through DPYD testing if no variations in DPYD have been acquired.

Furthermore, certain variations, such as p.D949V, exhibit low sensitivity values, leading to negative testing results for individuals at risk of toxicity. This discrepancy arises because carriers and non-carriers of the p.D949V variant can experience similar toxicity levels (Innocenti et al., 2020). However, it is important to note that the positive predictive values for p.D949V are significantly higher than its negative predictive values.Mutations in DPD at an exon splice site have been analysed and compared between patients with low and high DPD activity (Ridge et al., 1998). Variations observed at these DPYD codons were not consistently associated with low DPD enzyme deficiency, a rare occurrence compared to the high percentage of DPYD testing trials identifying DPD enzyme insufficiency (Innocenti et al., 2020).

Another study by Offer et al. (2014) defines DPD insufficiency as an inability of the DPD enzyme to break down fluoropyrimidines into inactive metabolites due to low DPD enzyme levels or an altered, non-complementary active site of DPD.

Additionally, certain variants such as DPYD c.1679T>G, c.1236G>A/HapB3, and c.1601G>A have been associated with DPD deficiency, but definitive clinical evidence for the validity of these variants is lacking (Meulendijks et al., 2015).


Genotype-based DPD detection usually searches for genetic mutations in the DPYD gene, including single nucleotide polymorphisms. In contrast, phenotype-based DPD detection measures plasma uracil concentration. DPD phenotyping provides more accurate results than DPYD genotyping because some rare DPYD variants cannot be detected (Paulsen et al., 2022).

However, DPYD genotyping has reduced the frequency of fluoropyrimidine-related hospitalisation from 19% to 0%. Moreover,  4.8% of DPYD variant carriers who didn’t undergo a DPYD test died, compared to 0% deaths in DPYD variant carriers who received DPYD genotype-guided dosing of fluoropyrimidine (Paulsen et al., 2023). DPYD testing has proven to be more cost-effective (Grimalde et al., 2022).

On the other hand, while DPD phenotyping is more accurate, its implementation can be challenging because the results are influenced by factors such as food intake, cardiac rhythm, and kidney function. Furthermore, DPD phenotyping results may vary between laboratories (Paulsen et al., 2022). Likewise, while DPD phenotyping offers more accurate DPD deficiency detection, DPYD genotype testing is more cost-effective and easier to implement (Paulsen et al., 2022).

In general, DPYD genotype testing is more clinically accepted and frequently used across European countries (Innocenti et al., 2020).

Literature Review Of DPD Genotype

A genotypic analysis study revealed that two single nucleotide polymorphisms (SNPs), p.D949V and p.V32I, correlated with grade 3 or higher fluorouracil (Boige et al. 2016). In related adverse events (AE), yielding p-values less than 0.01 (Boige et al., 2016). More than 36.7% of absolute AE rates were attributed to p.D949V carriers and 13% to p.V732I carriers. Both of these DPYD gene mutations were associated with hematologic AEs and p.V732I with neutropenia, weakening the immune system’s ability to mitigate infections (namely bacteria) (Boige et al., 2016).

DPYD genotyping for p.D949V has proven to yield a specificity of 100%, sensitivity of 2%, a negative predictive value of 51% and a positive predictive value of 86%. Furthermore, testing for p.V732I highlighted a specificity of 90%, sensitivity of 16%, negative predictive value of 52% and a positive predictive value of 61% (Boige et al., 2016). As determined by the results, the relatively high positive predictive value of 86% of the p.D949V variant suggests that testing for this genotypic mutation is associated with patients having DPD deficiency and are highly susceptible to grade 3 or greater fluorouracil AEs. Additional research validates this finding whereby the occurrence of grade 3 or higher fluorouracil AEs and hematologic AEs were associated with p.V732I with a confidence interval of 95% and p value of 0.002. This variant was located in 20% of patients who were confirmed with grade 3 or higher haematological AEs compared to the 4.92% control patients who did not carry the p.V732I mutation. However, the results of the p.D949V variant were not significant enough to show an association between the control group and patients with that SNP (Boige et al., 2016). 

Fang et al. (2016) noted similar findings. With genotype-guided dosing compared to the standard dosing of 5-FU, the occurrence of grade 3 or higher toxicity declined from 73% to 28% with a confidence rating of 95%. Moreover, the symptoms of the toxicity that did occur in the 28% were short term in contrast to the long term toxicity that is usually associated to those with the variant allele receiving a normal dosage of 28.03 ~ 38.94mg·h/L of 5-FU (Fang et al., 2016). Furthermore, findings by Deenen et al. (2016) highlighted a reduction in toxicity related deaths of patients who were receiving genotype guided dosing from 10.4% (5 out of 58 historical control patients) to 0% (Deenen et al., 2016).

A major concern associated with genotype-guided dosing is that the reduction in the dosage would result in lower efficacy of 5-FU. Like many other chemotherapeutic drugs, 5-FU is characterised by having a narrow therapeutic window (Hashimoto et al., 2020). In recent years, therapeutic drug monitoring (TDM) has been a considered better approach to the standard body surface area (BSA) approach as BSA traditionally has been an inefficient predictor of systemic drug exposure. Research indicates that response rates were greatly improved and toxicity related AEs were reduced for patients undergoing TDM based 5-FU dosages, advocating for the need to personalise dosage and medication (Hashimoto et al., 2020).

Linda M Henricks et al’s study validated this finding as even reduced dosages mitigated any reduction in the efficacy of 5-FU. For DPYD*2A carriers, a 50% reduction in initial dosage was sufficient. However, it was noted that further reduction in dosage required research to confirm its efficacy (Henricks et al., 2018).

The nature of these studies and their results highlight an evolutionary process of advancements in DPYD genotyping, preventing toxicity related to 5-FU. The relationship established between successful DPYD genotyping in patients and a reduction in toxicity related AEs due to reduced dosage of 5-FU have been promising as seen by the research of Deenen et al. (2016) and Hashimoto et al. (2020). Nonetheless, the rudimentary success of DPYD genotyping itself must be assessed and improved for it to have greater impact in reducing 5-FU related toxicity. As mentioned by Boige et al. (2016), the negative predictive value of 52% does indicate that the occurrence of false positives is relatively common, which may hinder the success of 5-FU’s efficacy as a chemotherapeutic drug for those who have been diagnosed as a false positive. Improvements to achieving a lower false positive value and further research to establish a correlation between the occurrence grade 3 or higher haematological AEs, and the p.D949V variant being present in patients, would allow for DPYD genotyping to overall be a successful method to reduce toxicity.

This conclusion has also been present in research done by Brooks et al. (2022). The research noted that whilst DPYD genotyping increased cost for patients by $78, it also improved survival rates by 0.0038 quality adjusted life years (QALYs). This ultimately led to an incremental cost effective ratio (ICER) of $20,506/QALY. Further with the price of $174 per DPYD screening and reduction in averted hospitalisations, the ICER amounted to $37,30/QALY. Through the use of a willingness to pay threshold of $50,000/QALY, researchers concluded that 96.2% of interaction preferred screening to no screening (Brooks et al., 2022). This clearly indicated a positive trajectory for DPYD genotyping to be incorporated in healthcare systems all around the world, given that consistent research and improvement are being made to the technology of DPYD genotyping. 

Analysis of the Efficacy of DPD Testing in France

In April 2019, France became the first country in the world to mandate DPD testing. This decision was based on various studies from the EMA, reporting that it was a highly recommended test prior to cancer treatment using fluoropyrimidines, particularly 5-fluorouracil (5-FU) and capecitabine. With an average of 80,000 people receiving fluoropyrimidines as chemotherapy for cancer in France alone, the toxicity dangers pushed the country into implementing mandatory DPD testing to limit sickness and fatality across the region (NHS, 2020). 

France’s preferred method of testing became the LC-MS/MS test, a phenotype test. This phenotypic test has the advantage that the ratio of Uracil and dihydrouracil (U/UH2) is measured directly as opposed to observing the gene. As dihydrouracil is the processed form of uracil after the DPD enzyme’s catalysation, the U/UH2 ratio can provide specific results on the exact efficacy of the DPYD gene and the amount of DPD enzyme in the body in an actionable scenario, unlike the genotypic test. However, a limitation to the implementation of this test on a nationwide scale was the various steps and complications in the phenotypic testing process. This was resolved through a partnership with Shimdu Corporation, which devised a fully-automated phenotypic testing process with a combination of an extraction machine and a mass spectrometer. The benefit to this process was that the combined prep and analysis time per test was cut to only 30 minutes or less with efficient standardisation (Zimmerman, 2020). 

To analyse toxicity and fatality rates due to chemotherapy using fluoropyrimidines in France prior to the DPD testing mandate, data collection was conducted in the Centre-Val de Loire region of France. This involved an exhaustive assessment of patients receiving FP-based chemotherapy during 2013-2014. This dataset formed the foundation for estimating the national annual patient population undergoing 5-FU or capecitabine treatments. Furthermore, a comprehensive cohort of 513 patients diagnosed with incident solid tumours and receiving first-line FP-based chemotherapy was examined to ascertain the true incidence of severe adverse effects (SAEs) associated with the treatment (Guellec et al.,2020).  

The extrapolation of regional data to a national level revealed that an estimated 76,200 patients receive annual treatment with 5-FU (53,100 patients) or capecitabine (23,100 patients) (Guellec et al., 2020). The incidence of SAEs during the initial two cycles of treatment was calculated to be 19.3%, with a 95% confidence interval (CI) of 16-23%. This figure included one toxic death, representing 0.2% of cases (95% CI 0-1%) (Guellec et al.,2020). Over the initial 6 months of treatment, the SAE incidence rate rose to 32.2% (95% CI 28-36%). Notably, the incidence of events categorised as death, life-threatening prognosis, or incapacity/disability stood at 1.4% (95% CI 0.4-2.4%) for the first two cycles and 1.6% (95% CI 0.5-2.6%) during the initial 6 months. These findings translate to an annual occurrence of approximately 1200 patients experiencing life-threatening prognosis or incapacity/disability in France, including 150 toxic deaths. In total, an estimated 14,700 people were recorded to have an SAE incident across France (Guellec et al., 2020). However, the limitations of this analysis and estimate include the fact that this study was conducted in only a certain region of France, and using this to estimate cancer and toxicity rates across the whole country leaves significant room for error in the estimates. On the other hand, due to the unique nature of cancer from person to person, no cohort of people would be enough to determine exactly the rates of cancer across the country, so this was the best option. 

In 2021, a study was applied to a randomly sampled cohort of 59 routine patients undergoing 5-FU-based therapy for digestive cancers. While none of these patients exhibited total DPD deficiency, 23% demonstrated the poor metabolizer phenotype, and one patient was identified as profoundly deficient. Consequently, the 5-FU doses in patients with the poor metabolizer phenotype were reduced by an average of 35% compared to non-deficient patients. Remarkably, despite the substantial reduction in 5-FU dosing, similar efficacies were achieved across both subsets of patients (Launay, 2021). Clinical benefit rates were 40% in both groups, stable disease rates were 40% (without individualised dosing) vs. 37% (with individualised dosing), and progressive disease rates were 20% in both subsets (Launay, 2021). Notably, no significant differences in toxicities were observed between the groups within the study whether they were receiving full dosage or not, and only 3% of early severe toxicities were recorded amongst the entire sample (Launay, 2021). This value notably contrasts with the typical range of 15-30% severe toxicities reported with standard 5-FU therapy. 

As observed, the fluoropyrimidine toxicity had dropped in the patient sample, however, the efficacy was nearly identical. By adjusting the dosage according to the analysis of the DPYD gene, the SAE incidence in patients had fallen by a significant amount, indicating a successful outcome of the new DPD healthcare policy. 

It must be noted that the data and results analysed lacked comprehensive patient samples as compared to other studies, especially when speaking on a nationwide scale. However, cancer’s nature is extremely individualised from person to person, with restrictions on the width and breadth of studies that can be performed, especially with limited resources. Therefore, by using data samples and extrapolating them to reach a national level, we can find great efficiency in data analysis and make the best hypotheses on the effect of DPD testing in France.

Therefore, it  can be concluded that due to the DPD testing mandate in France, the toxicity rates due to fluoropyrimidine buildup in DPD-deficient patients have been drastically reduced by the new outlook of individualised dosing. 

Analysis Of The Efficacy Of DPD Testing In The UK

Genomic testing in England was made routinely available (not compulsory) because of the seven national NHS Genomic Laboratory Hubs (GLHs) from August 2020. Approximately 38,000 patients are treated with fluoropyrimidine-based therapy each year in England,  and there is approximately 10-40% probability of facing severe or lethal toxicity and 1% probability of facing fatal toxicity (Public Health England, 2020; Amstutz 2018). Therefore, this policy is definitely a logical course of action. However, due to this not being compulsory, this policy may result in some different outcomes compared to France, which has fully implemented DPD testing. This section will explore the effects of DPD testing in England.

The policy states that patients who are commencing treatment that contains fluoropyrimidine-based therapy, whether it be 5- 5-fluorouracil, capecitabine, or tegafur, should be screened for the four DPYD variants c. 1905+ 1G>A (rs3918290) DPYD*2A, c. 2846A>T (rs67376798), c.1679T>G (rs55886062) DYPD*13, and c.1236G>A/HapB3DPYD (rs56038477) – all of which have been associated with being a cause of DPD deficiency. Via the clinical pathway, tests can be ordered upon consent for chemotherapy using fluoropyrimidine.

A study was conducted in Britain in 2010, the aim being to assess the contribution of harmful DPYD variants to fluoropyrimidine toxicity in British cancer patients as, at the time, toxicity was a major clinical problem. This was associated with DPYD variants that caused DPD deficiency, although this was previously deemed insignificant in Britain. This study was performed on 47 patients (27 female, 20 male with a mean age of 62 years), predominantly affected by GI malignancy who have been experiencing grade 3 or 4 fluoropyrimidine toxicity.

The most common toxicities experienced were myelotoxicity (37.5%) and diarrhoea (37.5%), with mucositis (19.5%), hand-foot syndrome (3.6%), and neurotoxicity (1.8%) also being experienced. A total of 8.5% of patients had the 1905+1G>A variant, all four being female and three of whom suffered from  severe diarrhoea. Five other cases had other sequence variants (2846A>T n = 4, 1679T>G n = 1). Overall, 9 of the 47 patients (19%) carried DPD deficiency caused by deleterious DPYD variants, which is significantly large considering that people believed this was an insignificant cause at the time. However, because of the relatively small sample size affiliated with only two hospitals, this may not be an accurate representation of the whole population. However, this still proves the DPYD mutation. The study concluded that prior testing and appropriate dosing can diminish the influence of DPD deficiency 

The pre-treatment was first enforced in April 2020 and became widespread thereafter, with England taking action in August of the same year. Following this, a study was carried out that explored the experience of treating patients with DPD deficiency with fluoropyrimidine chemotherapy following the onset of policies being placed in a large oncology centre in the United Kingdom. The study involved examining records of 23 patients with DPYD mutation, who had started chemotherapy between April and November of 2020 using the GAS (Gene Activity Score) as a guide to calibrating dosage. They were tested for mutations in a method that is considered clinically actionable by the Clinical Pharmacogenetics Implementation Consortium.

The majority of patients began with a 50% dose. One patient started with 100% dosage, which led them to experience mild diarrhea subsequent to the second cycle. They later underwent DPD testing and had their dosage reduced to 50% for their following cycles. Three out of 23 chemo-radiotherapy patients began with a dosage of 76%, as 50% seemed to be too low to be effective (subtherapeutic dosage). One out of these three patients had no toxicities, another had grade 2 nausea and non-neutropenic fever, and the last patient experienced 6 weeks of pancolitis. Overall, 7 of the 23 patients had no toxicities. In addition to all of this, 5 patients in total had further reduction of dosage and none had an increase of dosage.

This experiment showed that, unfortunately, serious adverse events (or SAEs) do still occur despite dose reduction according to GAS. This does mean that DPYD variants are not the only cause. However, with the help of DPD testing and curated dosage, it was shown to have an effect on patients by reducing toxicities in most of the patients in this experiment. Although this was performed in a small group, it shows that testing can definitely improve effectiveness.

Results / Conclusion

The many developments in genomics research and the push towards the incorporation of pharmacogenetics and personalised medicine within healthcare have engendered a substantial discourse challenging the efficacy of these policy measures. Certain proponents posit that the strides achieved in oncological research, coupled with the enhancement of individualised  healthcare encounters, warrant the adoption of such practices. Conversely, a counter argument asserts that the translation of these research advancements into broader disease applications remain to be determined, and that prioritising individual-centric approaches over public welfare could potentially impact the long-term sustainability of these programs, particularly within developing nations or third-world countries. One example of this is the reluctance to enact the deployment of DPD testing, notwithstanding its potential to confer advantages upon patients. The efficacy of DPD testing via its impact on toxicity levels and patient mortality, which could lead to contemplating universal integration, is determined through a comprehensive analysis of the compiled statistical data originating from both England and France. France conducted an empirical study that initially revealed the incidence rates of severe adverse events (SAEs) linked to FP-based chemotherapy to be calculated at 19.3% within a regional demographic. Subsequent to the intervention of testing and personalised dosing applied to a randomised subset in 2020, both patients with the poor metabolizer phenotype and those receiving conventional dosing exhibited conspicuously analogous efficacy levels. This observation is indicative of a noteworthy 12% reduction in toxicity cases among patients undergoing standard 5-FU therapy. 

Similarly, a parallel study undertaken in England featured a dosage adjustment of 35% for patients characterised by poor metabolizer phenotypes, mirroring the outcomes derived from France’s study conducted in 2020. The findings from England indicated a substantial decrease of 12% in patient toxicity related to 5-FU therapy. Remarkably, both within England and France, the incorporation of DPD testing followed by personalised dosing engendered a significant reduction in toxicity rates. The parallel studies conducted within these two European countries serve as accurate examples, unequivocally illustrating the incontrovertible affirmative impact directed by the implementation of DPD testing within their respective demographic frameworks. As previously mentioned, it poses a challenge to generalise regional data for national-level implications, as such an approach lacks comprehensiveness and accuracy in depicting population statistics. Nevertheless, the localised impact within a specific population highlights that DPD testing indeed yields more than a marginal reduction in patient toxicity. Thus, it is fair to say that the impact of DPD testing could justify implementation in the future.


Akutsu, D. et al. (2022). Severe Gastrointestinal Disorder Due to Capecitabine Associated with Dihydropyrimidine Dehydrogenase Deficiency: A Case Report and Literature Review, Reveiw, 61(16) 2449-2455) 

Andersen, S.E. et al. (2022). DPYD genotyping and dihydropyrimidine dehydrogenase (DPD) phenotyping in clinical oncology. A clinically focused minireview, Review, (5). 

Boige, v., et al.  (2023). Current diagnostic and clinical issues of screening for dihydropyrimidine dehydrogenase deficiency. European Journal of Cancer, (181).

Boige, V., et al.. (2016). DPYD Genotyping to Predict Adverse Events Following Treatment With Fluorouracil-Based Adjuvant Chemotherapy in Patients With Stage III Colon Cancer: A Secondary Analysis of the PETACC-8 Randomized Clinical Trial. JAMA Oncology, [online] 2(5), pp.655–662. doi:

Bocse, H.F., Bocsan,I.C. Buzoianu, A,D., Burz, C.C., Daec, A.L. ( 2020) A review on the importance of genotyping and phenotyping in fluoropyrimidine treatment. Med Pharm Rep. (93)(3). P. (223-230)

Brooks, G.A., Tapp, S., Daly, A.T., Busam, J.A. and Tosteson, A.N.A. (2022). Cost-effectiveness of DPYD Genotyping Prior to Fluoropyrimidine-based Adjuvant Chemotherapy for Colon Cancer. Clinical Colorectal Cancer, 21(3), pp.e189–e195. doi:

Chantal Barin-Le Guellec, et al. (2020). Toxicities associated with chemotherapy

Deenen, M.J. et al. (2016). Upfront Genotyping of DPYD*2A to Individualize Fluoropyrimidine Therapy: A Safety and Cost Analysis. Journal of Clinical Oncology, 34(3), pp.227–234. doi:

Fang, L. et al. (2016). Determining the optimal 5-FU therapeutic dosage in the treatment of colorectal cancer patients. Oncotarget, 7(49), pp.81880–81887. doi:

Hashimoto, Y. et al. (2020). Current Status of Therapeutic Drug Monitoring of 5-Fluorouracil Prodrugs. Anticancer Research, 40(8), pp.4655–4661. doi:

Henricks, L.M. et al. (2018). DPYD genotype-guided dose individualisation of fluoropyrimidine therapy in patients with cancer: a prospective safety analysis. The Lancet Oncology, 19(11), pp.1459–1467. doi:

Johnston, p., Harkin,D., Longley, D., Longley DB, Harkin DP, Johnston. (2003) PG5-Fluorouracil: mechanisms of action and clinical strategies. Nat Rev Cancer 3: 330-338, ResearchGate. 3(5), pp. 330-338

Iriart, J.A.B. (2019). Medicina de precisão/medicina personalizada: análise crítica dos movimentos de transformação da biomedicina no início do século XXI. Cadernos de Saúde Pública, 35(3). doi:×00153118.

Loganayagam, A., Arenas-Hernandez, M., Fairbanks, L., Ross, P., Sanderson, D.S., Marinaki, M.M . (2010). The contribution of deleterious DPYD gene sequence variants to fluoropyrimidine toxicity in British cancer patients. Cancer Chemotherapy and Pharmacology, 65(2):403-6.

Longley, D., Harkin,D., Johnston, p., Longley DB, Harkin DP, Johnston. (2003) PG5-Fluorouracil: mechanisms of action and clinical strategies. Nat Rev Cancer 3: 330-338, ResearchGate. 3(5), pp. 330-338

Martens, K., Huntjens, D., Rigter, T., Bet, P., Cornel, M. (2019). DPD Testing Before Treatment With Fluoropyrimidines in the Amsterdam UMCs : An Evaluation of Current Pharmacogenomics Practice. National Library of Medicine

Manon Launay et al. (2015). Beating the odds: efficacy and toxicity of dihydropyrimidine dehydrogenase-driven adaptive dosing of 5-FU in patients with digestive cancer. British Journal of Clinical Pharmacology, [online] 81(1), pp.124–130. Doi: .

NHS. (2020). Clinical Commissioning Urgent Policy Statement Pharmacogenomic testing for DPYD polymorphisms with fluoropyrimidine therapies [URN 1869] (200603P) Commissioning position Information about pharmacogenomic testing for DPYD polymorphisms with fluoropyrimidine therapies. (n.d.). Available at: f.

Pratt, V.M., Scott, S.A. (2017) Personalized Medicine in Cancer Treatment. DPYD,

Thorn, C. et al. (2011). PharmGKB summary: carbamazepine pathway. National Library of Medicine. (12), pp.906-10

Wang, L., Howlett, S., Essapen, S. (2021). Treating patients with dihydropyrimidine dehydrogenase (DPD) deficiency with fluoropyrimidine chemotherapy since the onset of routine prospective testing-The experience of a large oncology center in the United Kingdom. Seminars in oncology, 49(2), pp.170-177. 

Zimmerman. (2020). Fully-automated DPD deficiency testing. Available at: [Accessed 26 Aug. 2023].