CLN Article
Supported by
Pharmacogenomics (PGx) testing is rapidly expanding, particularly in cancer treatment, to ensure patients receive the right drug at the correct dose at the right time. However, as the field grows, the lack of standardized guidelines and the ensuing debates about testing criteria can seem overwhelming—to laboratorians and clinicians alike.
We asked Ryan Nelson, PharmD, medical director of precision medicine at ARUP, about what he’s seeing in the field. In addition to his work at ARUP, he also participated in the Standardizing Laboratory Practices in Pharmacogenomics (STRIPE) Annual Meeting and Consensus Workshop at US Pharmacopeia (USP) in October, and is part of MetaCensus, a blockchain-enabled platform that is enabling cross-discipline meta-analysis on complicated topics like this one.
One of the bigger challenges we have seen is that just because you are looking to implement the test does not mean that the clinicians will order it or be able to interpret it.
My advice is to get the appropriate department chair or clinical chair, the department of clinical pharmacology, and the case management group together, and talk about who they see as the groups of clinicians most likely to successfully implement PGx.
The ones that first come to mind are usually pancreatic, colon, gastrointestinal (GI) stromal, and, if a pediatric hospital, lymphoid malignancies and myeloid leukemias.
Another crucial area is supportive care for cancer patients receiving antidepressants that are often needed to manage the inherent emotional challenges, as well as anti-nausea medications to manage the emetogenic nature of many cancer treatments.
It also is important for selecting the appropriate antifungal agent for bone marrow transplant patients at risk of breakthrough fungal infections.
I do not recommend an institution do a carte blanche “we’re putting PGx everywhere” approach, because they are not likely to successfully implement the program or measure its impact if successfully deployed. Instead, pick one or two clinical sectors with the highest chance of success and go from there.
Let’s start with somatic testing. If I were a clinician at any cancer center, the basic answer would be the National Comprehensive Cancer Networks’ (NCCN) cancer-specific guidelines and the NCCN Biomarker Compendium. Clinicians can identify the cancer type, refer to the appropriate tables, and ensure they select the correct panel.
Things get particularly interesting when you need both germline and somatic testing. Pancreatic, colon, ovarian, prostate, and breast cancer all have germline and somatic ramifications for therapeutic decision-making. NCCN provides excellent guidance on heritable cancer-related germline testing. However, they do not provide guidance on germline tests for PGx.
For PGx guidance, I strongly advocate for the use of the Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Dutch Pharmacogenetics Working Group (DPWG) guidelines.
Concerning which laboratory tests are most appropriate for a given clinical circumstance, I recommend the test cover the Association for Molecular Pathology (AMP) PGx Working Group’s Tier 1 alleles and potentially Tier 2 alleles as well. I also recommend the lab test be certified by the College of American Pathologists (CAP).
For PGx testing, clinicians generally fall into four groups: those that are completely unaware; those that are aware but uncertain; and then those who have a level of certainty and either have or have not implemented it. All of them have relatively vindicated rationales for their current stances, though that will hopefully change.
In general, a small group of elite clinicians utilize germline PGx testing in oncology, though it’s slowly becoming more common. To me, it is quite encouraging how much more uptake there has been over the last decade—even the last five years.
As far as solid tumors go, lung cancer has been a leader in somatic PGx applications in terms of the number of therapeutically actionable targets with many indicated therapies.
For germline PGx, colorectal, pancreatic, and GI stromal cancers have seen a big impact on reducing toxicity from chemotherapy in clinics where it has been implemented (particularly in Europe where DPYD and UGT1A1 genotype-guided chemotherapy dosing is more common). There are many more, but these are the most frequently tested from my perspective.
But this landscape continues to change. Every year, I attend the American Society of Clinical Oncology (ASCO) and American Association for Cancer Research (AACR) conferences, and the discussion of oncology biomarkers becomes more ubiquitous. I see that trajectory continuing.
At ARUP, we offer both panel testing and individual gene testing. We try to ensure that even our panel tests offer genes with gene-drug associations tied to actionable recommendations backed by CPIC. Of course, it does depend on the clinical scenario.
All these choices are based on the clinicians' specific clinical applications and the individual patient decisions they are trying to make.
At the same time, we make sure we’re being judicious about reducing redundancies in the event there are downstream applications for that PGx testing, which is often the case.
But considering a test’s potential use in the future versus immediate utility is a tricky thing for insurance companies to know how to handle. They often don’t want to cover more than what is absolutely necessary at a given time point.Insurers would be wise not to let the immediate clinical decision take favor over the greater clinical context to develop over time. In many cases, selecting a test that benefits a patient now — and further down the clinical road — can save actual lives and money.
One of the most significant gaps is also an opportunity. Currently, there is not enough collaboration between the CPIC, NCCN, ASCO, and the FDA. I would like to see these groups sit down at the same table and embrace their differing perspectives, explaining their rationales and where they see gaps in the evidence.
All humans have innate biases that we think we control well but often fail to. We must embrace our risk of blind spots in our analyses and conclusions and rely on other minds with differing views to stratify for our blind spots. Only then will we come away with the least wrong, most informed assessment of the evidence.
One example: We held a session during the STRIPE Annual Meeting and Consensus Workshop where clinicians that author NCCN guidelines, others that treat patients guided by DPYD testing, the FDA, and ECRI participated on a panel to discuss their viewpoints around whether DPYD-genotype-guided chemotherapy should be standard of care or not. The FDA just recently reiterated a safety announcement we are encouraged by around DPYD testing.
Similarly, there is also not enough collaboration and communication among healthcare providers, namely, pharmacists, physicians, nurses, genetic counselors, and lab medicine professionals, all of whom are central to the topic of PGx and the different ways it can be applied. It’s one of the reasons we created MetaCensus.
There is a growing number of centers across the U.S. — both National Cancer Institute-designated and not — that have begun to understand the importance of PGx testing. And we hope that this year we’ll see a few collaborations here at the University of Utah’s Huntsman Cancer Institute lead to expanded PGx testing.
Supportive oncology care groups have also been expanding PGx testing for antidepressant selection. I commend and celebrate them for working to expand that. One of their bigger challenges is that they do not have enough support for prior authorization. It is both a logistical and policy problem.
Many gaps exist in the PGx space, and the approaches needed to resolve them look like most challenges humans face — unified minds curious to learn, eager to improve, and relentless at embracing being wrong to become more right. We need to remove ourselves from the comfort of our echo chambers, have constructive dialogue with bright minds that see things from different lenses than our own, and build tools that favor evidence over opinion.