JCO Article Insights: DYPD-Guided Dose Individualization and Survival Outcomes

In this episode of JCO Article Insights, host Dr. Soldato discussed with Dr. Knikman and Dr. Cats the findings of a study that assesses the influence of fluoropyrimidine dosing based on DYPD genotype on both progression-free and overall survival. The article, featured in the December edition of JCO, presents groundbreaking and reassuring data. Furthermore, it highlights emerging research challenges aimed at refining the prescription practices of one of the most widely utilized chemotherapy agents, striking a delicate balance between safety and efficacy.

TRANSCRIPT

The guest on this podcast episode has no disclosures to declare.

Davide Soldato: Welcome to this JCO Article Insights episode for the December issues of the Journal of Clinical Oncology. This is Davide Soldato, and today I will have the pleasure of interviewing Dr. Knikman and Dr. Cats, respectively first and corresponding authors of the manuscript titled "Survival of Patients with Cancer with the DPYD Variant Alleles and Dose Individualized Fluoropyrimidine Therapy: A Matched-Pair Analysis." 

Dr. Knikman is a clinical pharmacologist and assistant professor at the UMC Utrecht, while Dr. Cats is a gastroenterologist specializing in gastrointestinal oncology at the NKI in the Netherlands. 

Welcome, Dr. Knikman and Dr. Cats, and thank you for accepting our invitation today. 

Dr. Annemieke Cats: Thank you so much for the invitation.

Davide Soldato: So I just wanted to start by discussing the manuscript that you published. But before delving into the results of the manuscript that was published in the JCO, I just wanted to ask if you could give a brief overview of the DPD polymorphism and explain a little its relevance in the clinical practice. 

Dr. Annemieke Cats: The DPD polymorphism is very important in the metabolism of fluoropyrimidines. Fluoropyrimidines have been in practice for over seven decades now in the world and more than 2 million people received fluoropyrimidines in the beginning of this millennium. The indications for fluoropyrimidines have only been extended since then, so a lot of people are receiving this fluoropyrimidines. But with the good side of that there’s also another side and that is that there are a lot of side effects encountered by this chemotherapeutic drug. In the 1990s, it became clear that DPD was a key enzyme in the metabolism of fluoropyrimidines.

Dr. Jonathan Knikman: To better understand the toxicity associated which fluoropyrimidines are accompanied by, we have to take a closer look at the metabolism of fluoropyrimidines, and more specifically at the key metabolic enzyme which is dehydropyrimidin dehydrogenase, DPD in short. This enzyme breaks down the main active metabolite into inactive metabolites because 5-fluorouracil is the main active metabolite which is metabolized into inactive metabolites. However, if this enzyme does not function properly, this could lead to higher exposures of the active metabolites and subsequently more toxicity. This can be caused by mutations in the gene encoding for the DPD enzyme, which is the DPYD gene, and single nucleotide polymorphisms, so mutations in this gene can lead to less functional DPYD enzymes, subsequently can lead to more toxicity. 

Davide Soldato: So basically, patients that are harboring these SNPs in the gene encoding for the enzyme have a higher risk of toxicities. I think what is really important about the manuscript you published is that, apart from looking at the toxicities, side effects and pseudo profile among these patients who harbor these SNPs you also wanted to check whether this was associated with some reduction or at least with inferior clinical outcomes. The endpoints you selected were progression-free survival and overall survival. But I was really interested, and I think our readers and listeners would be interested in understanding a little bit the methods of the study. What was the cohort of patients that was selected? Was this a cohort composed only of patients with gastrointestinal malignancy or also different types of malignancies? And in this second case, if you included the patients with different types of malignancies, did you have any methods to be sure that there was not any differences among these patients at the very beginning? So basically, how you handled all the confounding factors that could potentially impact the analysis of clinical outcomes.

Dr. Annemieke Cats: To start your question, we have to go back to a previous study we performed, which was a prospective, multicenter study we performed in the Netherlands in 17 centers in which 1100 patients that had an indication for fluoropyrimidine therapy were included. In these 1100 patients, there were about 85 patients that were heterozygous carriers of a DPYD variant. What we did, we compared these two groups with each other, but before the DPYD carriers started, they had a reduced dose. The *2A variant carriers and the *13 carriers, they received a 50% dose, and the 1236 and 2846 they received a 75% dose. Those patients, a large amount of them consisted of patient with colorectal cancer, about 60%. And also breast cancer patients consist for a large amount of this group, it was about 20%. And then the rest were esophageal and gastric cancer mostly, and there was a group consisting of rare tumors as well like anal cancer. 

The comparisons showed that toxicity in the variant carriers was higher in the wild-type patients that received 100% dose. So despite those reductions, toxicity was higher in the carriers. But when we compared the variant carriers with historical controls, we saw that, especially for *2A, there was a large reduction of toxicity to a lesser amount. This was also the case for the 2846 variant carrier. But to a much lesser extent, there was a reduction in the 1236 carriers which worried us and this was the basis for a Clinical Pharmacogenetics Implementation Consortium (CPIC) to give a recommendation for starting for all these variant carriers with a 50% dose when treatment with fluoropyrimidine was indicated.

Davide Soldato: Thank you very much for the clarification. So basically, you started from one previous prospective study that was composed of a mix of different cancer types. And then in the study that you now published in the Journal of Clinical Oncology, you also added an additional part of patients who were carriers of this variants. I imagine that there was some variability between patients who were carrying the variants and those were not carrying a variant. So I wanted to understand a little bit, did you perform a matched analysis or did you try to be sure that there was not any confounding factors that could impact the results?

Dr. Jonathan Knikman: Yes, we performed an exploratory retrospective matched-pair analysis and we used a matched-pair analysis to select patients which are comparable between the DPYD variant carriers and the DPYD wild-type patients to ensure that on the most important factors, they were comparable and outcome was also comparable. In this matched-pair analysis, we compared progression-free survival and overall survival between DPYD variant carriers treated with the reduced dose and DPYD wild-type patients treated with the full dose. We matched these patients on five matching variables which were primary tumor type, stage of cancer, age within a range of plus or minus 10 years, sex, and treatment regimen to ensure the these patients as comparable as possible.  

Davide Soldato: And so now, moving on a little bit to the results, what were the results in brief regarding progression-free survival and overall survival for patients who were harboring a variant in the DPYD gene and those who were wild-type for this gene?

Dr. Annemieke Cats:The results of our study showed that there was no significant difference in progression-free survival and overall survival between the DPYD variant carriers pooled as one group, treated with the reduced dose, compared to wild-type patients treated with the full dose. This suggests that DPYD-guided dosing can be performed safely without compromising treatment effectiveness. However, when we take a closer look at the individual variants, our study showed that progression-free survival and overall survival were not negatively impacted by DPYD-guided dosing in the 2846 and the DPYD *2A variant groups. However, in the group of 1236 variant carriers, we did find a significant difference in progression-free survival with a hazard ratio of 1.43, indicating that progression-free survival  was shorter compared to the matched wild-type patients treated with the full dose. However, no difference was found in overall survival.  

So based on results on this study, we can conclude that DPYD-guided dosing can be used while treating patients with fluoropyrimidines without compromising effectiveness and improving safety. However, when patients do not experience toxicity or experience minimal toxicity, it is important to escalate the dose guided by toxicity to ensure maximum exposure to the treatment. 

Davide Soldato: Thank you very much. That was a very comprehensive overview of the results. I was really wondering regarding the variants in the 1236. So the one that you said was associated with a shorter progression-free survival. At the very beginning, Dr. Cats very well explained that in the first prospective trial that was done, among these patients, toxicity was still higher or they experienced more severe toxicities, even with a 25% dose reduction compared to what was planned. So I was wondering if you could speculate a little bit. Do you think that this reduction in the progression-free survival might potentially be associated with these higher rates of severe adverse events, and that maybe this has created a little bit of a gap in the adherence to chemotherapy?

Dr. Annemieke Cats: We looked into the mean dose that the 1236 variant carriers received, and this was about 75% of the dosage. So the dosage was what the protocol prescribed for the study. And also the number of cycles did not differ from the wild-type patients. So, I think the patients did not stop earlier with their treatment than was intended to. What we did see however is that in the whole group, only 10% of the variant carriers had some kind of dose modification, which could be both an increase of the dose or a reduction of the dose. So, there were only a few patients where the dose was modified. We know that the 1236 variant carriers are a very heterogeneous group. DPD enzyme activity also shows a wide range, ranging from normal to very low dose even we described a patient with a homozygous 1236 mutation. And we would expect that there would be no DPD enzyme activity, but there was still some activity in this patient. 

Davide Soldato: That's a very good insight on this particular topic. So, just related to the conversation that we were just having, do you think that in general, for all DPD variants, and in particular, as we discussed, for the 1236, do you think that genotyping is sufficient for now to understand which is the right dose for these patients, to balance toxicities and to obtain the best clinical outcomes? Or do you think that we need more sophisticated or more integrated types of monitoring? Should we, for example, in the 1236, look more carefully at pharmacokinetics and so understand if these patients can receive higher doses because maybe, as you were saying, the activity of the enzyme is really higher than what we are expecting based on the genotype? Do you envision something like this happening in the future, or do you consider it as a potential line of research on this topic? 

Dr. Jonathan Knikman: It's a very interesting question, and it's something we've thought about a lot. At the moment, I think genotyping is the way to go and is the most robust and most validated method currently available with fluoropyrimidines. And I think it's a bit of both. Currently, as mentioned, I would recommend DPYD genotyping, and also in the 1236 variant carriers, as we have seen in previous studies, that toxicity is still increased even while administering 75% dose instead of a full dose. So we see that there's still more toxicity. However, our study also shows that progression-free survival is shorter compared to wild-type patients. So, it's quite a complex situation as we still have more toxicity, but the progression-free survival is also shorter, and that's where we would advise the dose escalation part. So, I think it's a combination of genotyping and escalating the dose when possible, if there is no toxicity or limited toxicity, to ensure maximum exposure and to minimize the effect on progression-free survival while still trying to reduce toxicity.

Dr. Annemieke Cats: This is a very important question because we do not completely understand why toxicity is higher in the 1236 with a 25% dose reduction that may compromise progression-free survival. So we have to look in closer detail, and currently we are looking in closer detail to what do we have for a pharmacokinetic analysis, that you mentioned. What about DPD activity, enzyme activity? Can you titrate on that? And I think for now, it still stands that we should start with a 50% dose, but with the possible effectiveness of the dose in mind, you should go with an early titration, and I would say something about 25%. Although having said that, I have no data to underline this. And if you have the possibility to go for DPD enzyme activity in addition to genotyping, that would also help you to titrate doses on that. And that's where we stand now. But we need to know why these 1236 variant carriers have such a large range in activity and toxicity.

Davide Soldato: I also think, and I can ask you if you agree with me, that this analysis is really very important because I think it's one of the first reports regarding the analysis, specifically of progression-free survival and overall survival based on variants of DPD. But at the same time, I think that we also have to underline for our listeners that this was still a retrospective and exploratory analysis. So it's true that you observed this association with shorter progression-free survival. But still, I think that we will need also more core studies to validate these findings and to be really sure and also to perform additional analysis as to what the mechanism is as you were saying. I don't know if you agree on this limitation of the study, despite its importance in regarding clinical outcomes.

Dr. Annemieke Cats: We certainly agree with you that we have to keep in mind that it is an exploratory, retrospective analysis. Having said that, a randomized controlled trial certainly has some difficulties in it as well. Nowadays, it is not feasible to give patients with a DPYD variant a full dose. In 2018, there was a lawsuit in Oregon, and now recently in Ontario, Canada. There's also a lot of rumor in the news because of patients getting a full dose while having a DPYD variant. So the lethality of the toxicity in some variant carriers makes it not ethical to perform a randomized controlled trial. So we have to look for different designs that reflect real-world data and learn more about the genotype and also in different ethnicities. It is also very important because what we have studied considers the white population mostly and that's also a direction that future research should go to. 

Davide Soldato: Thank you very much. That was very insightful, especially the last part regarding ethnicity and the possibility that also these variants might be different according to that. 

So I think that the main message that we should pass is that when prescribing fluoropyrimidines, genotyping of DPD is fundamental because toxicities among patients harboring these variants can be severe and can also be lethal. You performed this retrospective exploratory analysis that provided us with overall reassuring data. There is still more research to be done, especially for the 1236 variant. So I think that that probably is our bottom line and main message for our listeners. 

And I just had one final question because in the manuscript that you published, you had localized, locally advanced, and also metastatic cancer that were all grouped together. Do you think that an additional line of research would be to specifically look at how these impactful outcomes only in the locally advanced and localized cancer compared, for example, with the metastatic?

Dr. Annemieke Cats: You raised a very good point because there is a lot of variety within this study, and now you're mentioning locally advanced, local, or metastatic cancer. It's also within the cancer types. We studied different cancer types as well to reach a large amount of patients. It would be good to have a more homogeneous population that you can derive your conclusions from. So, I think that would certainly help us in the future, and we should look into whether we could do this together because before we had such a large population, we would need several countries to work together. 

Davide Soldato: Yeah, you're totally right. Thank you very much for underlining that last point. 

Thank you very much, Dr. Knikman and Dr. Cats, for being here with us today and for explaining to us and our listeners the results of your research. That concludes this episode of JCO Article Insights. We discussed the manuscript titled "Survival of Patients with Cancer with DPD Variant Alleles and Dose of Individualized Fluoropyrimidine Therapy: A Matched Pair Analysis." This is Davide Soldato, your host. Thank you for your attention, and stay tuned for the next episode. 

The purpose of this podcast is to educate and inform. This is not a substitute for professional medical care and is not intended for use in the diagnosis or treatment of individual conditions. 

Guests on this podcast express their own opinions, experiences, and conclusions. Guest statements on the podcast do not express the opinions of ASCO. The mention of any product, service, organization, activity, or therapy should not be construed as an ASCO endorsement.