NEW YORK – Evidence is growing to support the value of preemptive, universal pharmacogenomic DYPD testing to help oncologists avoid severe side effects for patients receiving fluoropyrimidine chemotherapies, alongside new professional guidelines that aim to help standardize the expansion of this testing across labs and cancer centers.
Investigators from Atrium Health, a multisite cancer center in the Carolinas that has pioneered an electronic health record-based prompting system to drive DPYD testing amongst its oncologists, recently shared their own data from the pilot phase of this effort, which offered strong evidence for the value of preemptive versus reactive testing.
The publication, in JCO Precision Oncology, comes almost a year after Atrium expanded its in-house genotyping program to help oncologists use these chemotherapies more safely. Fluoropyrimidines are widely prescribed but known to cause severe and life-threatening toxicities in a subset of patients who harbor certain alterations in the DYPD gene.
These variants are associated with decreased production of dihydropyrimidine dehydrogenase (DPD), the protein that helps break down fluoropyrimidines. The resulting poor metabolization puts patients at risk of adverse reactions including infections, bleeding, and mouth sores that can quickly become severe.
Atrium reported in its study that among 757 patients who received DPYD genotyping beginning in 2020, 45 were heterozygous carriers of a detrimental variant. In 442 patients with three-month follow-up, grade-3 toxicities occurred in more than 60 percent of patients who were tested after beginning therapy, versus only 31 percent of those tested preemptively — nearly the same rate as wild-type patients, or those without a detected variant.
Patients tested reactively had tenfold higher odds of hospitalization compared with their preemptively tested counterparts.
"We didn't plan for this," said Jai Patel, chair of the department of cancer pharmacology and pharmacogenomics at Atrium's Levine Cancer Institute. "But, because we weren't mandating that doctors do the test or when they do the test, it offered a real-world look at what happens when we integrate this kind of program."
"We clearly saw that if we do the pretreatment testing and adjust the dose, we can significantly reduce severe toxicities and hospitalizations compared to the reactive testing group that were carriers," he said.
According to Patel, the field is already seeing an acceleration in the spread of DPYD testing as evidence like this emerges and as Atrium and other pioneers demonstrate practicality and feasibility. Payor coverage has also been expanding rapidly.
"I think we have seen a shift, starting with Medicare back in 2020 when the MolDx program issued coverage of targeted molecular diagnostic tests, which included DPYD," Patel said. With most Medicare administrative contractors now following MolDx's moves, most states do now cover DPYD, he added, alongside a majority of private or commercial payors, though there are certainly some that still do not.
Patel said that Atrium started billing for its test last year, and that the vast majority of its patients now get their tests reimbursed.
"We submit prior authorizations on every test we get, and more than 90 percent are being authorized," he said. "We submit a letter of medical necessity, and we cite our own internal data to show that we observe a reduction in severe toxicities and hospitalizations … and with that, we have actually seen much better authorization rates where most patients are paying little to nothing."
Patel is not alone in recognizing a shift in the field toward more universal implementation of DPYD genotyping. The Association for Molecular Pathology last week published a new set of consensus recommendations to aid in the design and validation of clinical DPYD genotyping assays and promote standardization of testing across different laboratories, citing the emergence of patient advocacy groups that "are working to raise awareness about fluoropyrimidine toxicity and the availability of DPYD testing, as well as advocating for universal testing."
The new DPYD report is the latest in a series of recommendations developed by the AMP PGx Working Group, and includes a two-tier categorization of variants recommended for inclusion in clinical tests.
The Tier 1 recommended variants all have a well-characterized effect on functional activity of protein and/or gene expression, have an appreciable minor allele frequency in a population or ancestral group, have available reference materials for assay validation, and are technically feasible for clinical laboratories to interrogate using standard molecular testing methods.
The Tier 2 list comprises optional variants that meet at least one, but not all, of the Tier 1 criteria.
What the AMP recommendations do not include is a push for preemptive testing. But Patel said that his hope is that by building on its recently published data with other partner institutions, Atrium will be able to further cement the value of pretreatment testing and inspire others to adopt its model.
"The majority of hospital systems are on either Epic or Cerner … so they all have the ability to integrate alerts. And even if the health system or clinic doesn't have the infrastructure to build a DPYD test in-house, there's obviously several commercial labs that offer testing." These include cancer genomics firms like Tempus, and lab giants like Labcorp.
He also noted that test turnaround time has improved over the last several years. "That always used to be an issue where we simply couldn't wait two weeks to get those results back, but now, I'd say that the average turnaround time is probably about a week or maybe even a little bit less for some commercial labs," he said.
Given these improvements, Patel said he thinks it is "perfectly feasible" for clinics and hospitals to be able to set this up.
The larger barrier, he said, is a lack of guidelines, either from the US Food and Drug Administration or the National Comprehensive Cancer Network, to compel oncologists to test for DPYD variants.
Although a recent label change by the FDA maintained that DPYD testing isn’t required for use of fluorouracil-based products, it did update its stance to state that testing needs to be considered and that there should be a discussion with the patient. "That language never existed before," Patel said. "So, because it now actually states the patient must be informed about the availability and the implications of testing, that is now a liability concern for providers."
Moving forward, Atrium is working on taking its data back to its own clinics and providers to try to increase its own rate of pretreatment testing.
This has prompted new conversations about how to make it as easy as possible for providers, leading to a new innovation, which the team is currently working on, that will integrate DPYD testing into the order set for chemotherapy.
"Now, every time they pull the order set for a patient starting a chemotherapy regimen that includes capecitabine [marketed by Genentech as Xeloda] or 5-fluorouracil, it will automatically order a DPYD test, unless they opt out," Patel said.
"If they do opt out, we are going to follow up with that," he added. "They'll need to provide a reason, and we're going to follow up with them to ensure that their patient knows that they've opted out."
He and his colleagues are also working on an expanded study in collaboration with the University of Michigan, pulling together "almost a dozen" sites that have implemented DPYD testing.
"We've applied for some grant funding to try to cover some of the logistical costs of doing this, and the idea is that we're hoping to pull together a real-world pharmacogenomics consortium," Patel said. With a larger cohort, the aim is to be able to study not just toxicities but also the impact of PGx-guided dosing on drug efficacy and overall outcomes, which could help persuade oncologists and guidelines bodies of the overall clinical utility of upfront DPYD genotyping.
Some hesitancy around DPYD testing has rested on worries about the impact of reducing the doses of these drugs, and although there has been some data demonstrating that dose reduction doesn't affect efficacy, it hasn't been convincing enough. With a volume of "probably over 5,000 tests per year" between Atrium and its proposed partners, that could change, Patel said.
"At this point, there's not going to be a randomized trial that's ever going to be done, for a number of reasons, one of which is that it's not ethical," Patel said. "And when those studies are not ethical, I think it's perfectly reasonable to rely on pragmatic real-world data, to then compare what happens when we actually implement testing."
The team is also working on health economic analyses, Patel added, highlighting the fact that the recent study showed not only a reduction in toxicities, but also in hospitalizations. "[That] was huge, so right now we're doing the cost evaluation of that shift," he said.