At the Community Oncology Alliance (COA) Annual Meeting (April 24, 2020), Lee Schwartzberg, MD, FACP, West Cancer Center and Research Institute (a member of OneOncology), moderated a stakeholder panel discussion on utility and access issues of comprehensive genomic profiling in cancer care.
Journal of Clinical Pathways sat down with Dr Schwartzberg to further examine the potential of genomic testing in clinical pathways and how built-in prior authorization for such testing could be a realistic possibility in the near future.
The panel at the COA Annual Meeting comprehensively broke down the power and the potential of genomic profiling in cancer care. Towards the end of the panel, Dr Alan Balch highlighted the potential of clinical pathways to include built-in prior authorization for genomic profiling. Where in a typical pathway would this genomic profiling exist, and why is this idea a good one?
Dr Schwartzberg: Comprehensive genomic profiling has come of age to the point where it truly belongs in pathways. The first place in pathways that profiling belongs for many cancers is at the diagnosis of metastatic disease, especially metastatic non-small cell lung cancer (NSCLC). We know that in this setting, there are multiple driver mutations that are actionable and that have approved therapies targeting them.
It is guideline adherent—in fact guideline necessary and therefore pathway necessary—to do multiplex testing at the time of metastatic disease. This practice has been the paradigm now for several years in this disease as well as other cancers.
It is not considered controversial at all that comprehensive genomic profiling should be done at that point for that disease. In general, almost every cancer should have comprehensive genomic profiling at the diagnosis of metastatic disease. I am a big proponent of understanding the landscape of a patient's disease in order to determine the course over multiple lines of therapy.
What is the largest challenge(s) for making this “built-in” prior authorization a reality or commonality across pathways?
Dr Schwartzberg: The largest challenges depend on the perspective of the stakeholder in the cancer care chain. During the COA panel, we heard that pathologists are adamant on doing comprehensive genomic profiling early and getting that information. Patients are as well. Physicians tend to be the same way, with the caveat that they want to be able to understand the information. Often times, there is this wealth of data delivered back, which can be difficult to navigate and use to make decisions.
At the other end of the spectrum, some payers have yet to be convinced that there is value to doing comprehensive genomic profiling in patients on a routine basis. This is really the crux of the matter.
The debate lies in when to do it, who should pay for it, and whether it requires prior authorization. I am of the opinion that we should be past this debate because the debate only exists on the basis of cost. Payers are trying to control cost, which is appropriate, but this really is not the right area to control cost.
There are two aspects of the cost equation in this discussion. The first is the actual cost of the test, which has gotten less expensive over time while providing more information. The average test now costs in the thousand dollar range. But given the fact that targeted therapies typically cost around $10,000 per month, the relative value of the cost of the test vs the cost of the therapy is disproportionate.
Payers are concerned that having a target on a comprehensive genomic profile report and a therapy that potentially could be used without a clear FDA indication will open them up to having to pay for drugs with uncertain benefit.
Is there an issue of payers not meeting providers halfway at the bargaining table to making built-in prior authorization happen?
Dr Schwartzberg: There is varying understanding at the payer level in regard to value. For those that have good insight into the potential of these tests, they are willing to meet the provider halfway. The momentum is trending toward doing so. Other payers have taken a more restrictive approach and have yet to engage with the provider.
There is quite a bit of variability in the third party payer market. Interestingly, the variability does not necessarily correlate with the size of the payer. Some of the larger payers remain steadfast on not paying for the tests.
Additionally, there is the added network piece; payers may contract with an individual testing company. If a payer has a contract with one laboratory and a provider has familiarity and comfort with another, there is a conflict set up about who should conduct the test. Providers may find interpretation of an unfamiliar report more difficult. Since there are an increasing number of laboratories conducting genomic profiles, this problem is likely to increase.
I believe there is value in creating relationships with a testing company because these are complex tests. Having access to expertise at a given laboratory is really important. For instance, the reports themselves typically have short explanations of the alterations found, particularly for the less common ones. Having a representative at the laboratory who can direct queries to experts in molecular pathology or molecular genetics can truly help put the results into a perspective that is clinically useful. More knowledge of esoteric and rare events will help the oncology provider make the best clinical decisions for his or her patients.
How do you envision genomic profiling inclusion on pathways impacting the structure of the pathway? Will it require a total physical overhaul to allow for separate treatment arms based on the results of genomic testing – whether that be on- or off-label use of therapies?
Dr Schwartzberg: The short answer is “yes.” We already have genomic profiling in pathways, particularly for on-label use.
Let's use NSCLC as an example. There are at least six or seven different driver alterations that have therapies that are the preferred regimen for the first- or second-line of care. These are built into pathways. Every pathway that I am familiar with has built in the genomic alteration just like it would the stage or other clinical factors that impact the decision-making for a particular treatment. The algorithm goes on in the second- or third-line as well as to where those genomic alterations and associated therapies fit.
The same is also true today in colorectal cancer. We make decisions in the first-line setting based on clinical factors, but also genomic profiling. There are multiple alterations that have to be taken into consideration in order to make the right therapy choice.
When it comes to non-FDA indications, the conversation is a bit more nuanced. There are two aspects to consider, the first of which is clinical trials. We all acknowledge that clinical trials should always be at the highest level of a pathway, since none of these patients are being treated in a curative setting. Having those alterations in a comprehensive genomic profile report at the time of diagnosis of metastatic disease will allow eligible patients to participate in novel and innovative clinical trials – even in the first-line setting.
For patients that have exhausted clinical trial options or not eligible for clinical trials and have exhausted standard therapy options, utilizing the comprehensive genomic profile to look for alterations that have a therapy indicated in another disease is, in my opinion, certainly appropriate.
How is OneOncology working to ensure genomic profiling prior authorization within its own pathways? What is the level of negotiating you are doing with payers and other stakeholders?
Dr Schwartzberg: Genomic profiling is a centerpiece in the clinical arena for OneOncology. A major focus of mine has been to begin to develop a precision oncology program that will dovetail with our pathways program.
We have been working towards a preferred relationship with a national laboratory that does comprehensive genomic profiling. As I stated before, having the ability to understand and interact with individuals that have expertise in molecular genetics, molecular pathology, and the complexity of these tests is really important.
We have also launched our pathways program, which in a joint effort with Flatiron, uses a tool called Flatiron Assist to build our pathway tool. We launched the beta testing 2 weeks ago. Thus far, it is functioning very well. The novel thing about this particular project is that it is all based on NCCN preferred regimens. The templates for NCCN have been ingested by Flatiron, whereas OneOncology is building the content. We have an authoring tool that allows us to write our own pathways based on the clinical features. The tool provides these features and allows us to pick the preferred regimen. These pathways are embedded with the genomic element. From the start, we have been creating a pathways system that is reliant on comprehensive genomic profiling results.
Additionally, in my own practice, we are having discussions with our major and local payers about making sure that they will place the companies we are partnering with to do comprehensive genomic profiling into their networks.
I envision different levels of complexity of molecular profiling as we go forward, some of which would be done at large practices. However, others like comprehensive genomic profiling may be beyond the scope of any practice. In fact, it may even be beyond the scope of a hospital system. Many academic groups have turned to partnering with a national vendor because it requires a lot of investment and it is a continually changing area.
Our focus at OneOncology is to use the result of comprehensive genomic profiling—provided by our partnership with a national laboratory—to influence our pathway design.
