Driving Innovation in Cancer Care: The Future of Value-Based Care in Oncology

In this interview, Gordon Kuntz discusses the Delphi Method, a Predictable Cost of Care Model developed by the Predictable Cost of Care Working Group that enhances oncology market access strategies by standardizing and broadening the economic modeling of cancer treatment costs beyond just drug prices, while aiming to inform pathway developers, payers, and practices.

Transcript:

Gordon Kuntz: Hi, I'm Gordon Kuntz. I am the moderator for the Clinical Pathways Congress for the last several years. It's one of the great honors of my year and of my career to be involved with the Pathways Congress. I'm a consultant with pharma around clinical pathways and the oncology ecosystem.

What is the Predictable Cost of Care Working Group?

Kuntz: The Predictable Cost of Care Working Group is a group of oncology pathway developers and pharma sponsors who have come together to create an industry standard solution for evaluating the cost of cancer care. Developing a model around the cost of cancer care is very, very complicated because the cost of cancer care is highly variable. Sometimes patients need more services, sometimes they need less services, sometimes they need supportive care, sometimes they need especially intense treatment. However, our belief is that there is a core of costs, outside of simply the cost of the drug, that are predictable, that are knowable, and that we can develop utilization models and find independent basis for that will help all of those users and the industry as a whole, to understand what the costs are beyond just the cost of the drug. Those users might be pathway developers, they might be payers, they might be oncology practices.

There are a lot of different potential use cases for this information. The challenge today is that pharma doesn't develop those kinds of models on a routine and normalized basis. Different pharma companies will come up with different models. It's very difficult for somebody, even a pathway developer, to evaluate those models. The notion is, if we come together as an industry, that we can create a standard that pharma can follow in developing these models. They might have other models as well, but they would develop this model. They would submit that to pathway developers. They might submit that to payers. They might submit that to practices as they're trying to make a case for their drug. First and foremost is always the clinical message, however: efficacy, safety, and then we always talk about cost. Historically, we've talked about the cost of the drug. We now want to talk about a broader set of costs with this predictable cost of care.

How might the new industry standard for economic models developed by the Predictable Cost of Care Working Group impact current market access strategies for oncology treatments?

Kuntz: The challenge with current economic modeling in market access strategies around cancer care is that generally we only look at the cost of drug. In a pathways program, the cost of drug at average sale price (ASP) is typically what's been used and, frankly, that's the standard that's been used for the last 20 years. That encompasses a significant portion of the cost, but honestly not all of the cost. There's about 30 to 50% of the cost of a cancer episode that isn't related to the drug directly. It's related to the side effect management, the premeds, and the other costs that might be associated, some of which are predictable, with that cancer care. For a pharma company to be able to tell their total economic story to a pathway developer or to a payer or a practice who is interested in some element of total cost of care, they need to be able to describe more of the elements.

But one of the challenges that we've had is they've sort of made up numbers along the way. There will be an application of a number related to, for example, a hospital stay, and without having a standard to go to, like we do with the cost of the drug itself—as I mentioned previously, the average sale price—we need some independently verifiable unit cost numbers as well as then where the number of units, the utilization side, has been derived. With those elements in place, there's an industry standard that we're trying to develop here, which will allow the pathway developers to ingest and analyze this data on a much more efficient basis. Then pharma can develop these models, they can submit their data and have that economic picture be considered alongside the clinical value that a drug might have.

How does the Predictable Cost of Care Model compare to traditional cost effectiveness models used in market access discussions?

Kuntz: Traditional cost-effectiveness modeling in market access will typically deal with some of the similar issues, but the Predictable Cost of Care modeling that we're trying to develop here is an industry standard that actually creates independent, verifiable standards for the cost side of the equation. The utilization side—we’re looking at deriving that initially from a clinical trial, although in the future we might be able modify that and refine that through real-world evidence. The Predictable Cost of Care Model is useful for anyone who's accepting risk. So that can be, obviously, a payer; it might be a practice, and therefore the pathway for either a payer or a practice that needs to reflect that total cost in the consideration of often very expensive drugs. Efficacy is still number one, but they need to be informed by this greater picture of what the total cost of that drug actually is.

What are the potential implications for pricing and reimbursement negotiations regarding what you just discussed?

Kuntz: The implications of the Predictable Cost of Care Model around things like price negotiations between pharma and either payer, practices, or their GPOs can definitely be influenced by this total cost of care. Today we really look at what's the cost of the drug, can I get a discount? Some pretty simple math, if you will. That's actually predicated on a model that is rapidly fading. Historically, practices would look at the difference between what they had to pay for a drug and what they could get reimbursed for a drug, and essentially a margin that they could make that could come from a number of sources—discounts, rebates, rebates, whatever it might be—compared with the negotiation with the payer. That's no longer how practices look at their economic opportunity. They're looking at a value-based arrangement in many cases, which, by necessity, is related to the total cost of care.

It's very difficult to estimate a total cost of care if all the variables are up in the air and the points don't matter. They matter a lot to practices. So having a Predictable Cost of Care Model that people accept as reasonable, accurate, independently verifiable, provides an additional basis for these kinds of negotiations. And if a pharma company has a strong economic argument alongside their clinical argument, of course, it's going to put them in a better position to negotiate with either payers or with practices who are trying to understand what the total cost of care is going to be for their drug.

When discussing potential implications for pricing and reimbursement negotiations, pharma often looks at ICER and quality adjust life years to inform their pricing. How does the Predictable Cost of Care Model influence ICER strategies?

Kuntz: So the Predictable Cost of Care Model, in its first iteration, is perhaps a little simple. It's a complex topic. I won't make any excuses for really trying to approach this with something that we could actually get done. We talked a lot in the group about things like quality of life and some issues like that. The ways that patients access the healthcare and the burden on patients. Honestly, for the first round, it's probably a little bit complicated to try and do that all at once. So we're really focused initially on that predictable cost of care. I believe that the model will continue to evolve.

One of the things about this model that I think is really quite important is I've been involved with clinical pathways for 20 years. And 20 years ago when clinical pathways were essentially developed initially about 2004, 2005, the way that we handle clinical pathways today is the way that we handled them 20 years ago. It has stood the test of time, it's obviously worked very well, but it hasn't evolved, and drugs cost multiples of what they did. We thought they were insanely expensive 20 years ago. They're now multiples of that today. Understanding this total cost picture is actually more relevant today than it's ever been. I hope that this Predictable Cost of Care Model will be adopted by pathway developers and by pharma on an industry standard so that we can evolve the pathways model to include these kinds of things. And as the model evolves itself, we can include more patient-relevant measures like quality of life and the number of touches to the healthcare system—those sorts of things to help a patient voice be involved as well.

Is there also a push to keep patients in mind or do you feel like the way things are set up now, everyone's trying to align on the larger goal and focus on cost?

Kuntz: The Working Group is set up to solve in some ways multiple problems at the same time. One is to really start to understand this cost picture. We've been able to obviously evaluate the cost of drug for a very long time. We simply look at ASP. You can look it up online, the government publishes it quarterly, and you can figure out what the average sale price for a drug is. In trying to understand this broader cost picture, it's really seen as an additive piece to the existing methodologies that people use to evaluate whether a drug is medically appropriate, the single best choice, the single best recommendation within a pathways program—that sort of thing—which are really clinically oriented and should be clinically oriented approaches. That's not going to change. What the group is trying to do is to add some substantial information to that around this predictable cost of care.

The group is especially mindful of the cost to patients, but it's very difficult to evaluate that. In a perfect world, we'd be able to know what each patient's financial responsibility would be when they first walk up to their oncologist's office and say, “I'm here for an appointment with the doctor.” But that's very difficult to understand because the benefit structures, because of where that patient is in their use of those benefits, all sorts of things make that especially difficult. So we can't really evaluate that. We've had a number of discussions and are very mindful of the fact that patients have nonfinancial costs associated with cancer care. There’s travel time. Maybe the person has a caregiver who comes with them, or they need childcare. Distance can be a huge issue. Obviously, the financial cost can be a huge issue. Time can be a big issue if the patient feels well enough to be at work, but has to be in the doctor's office 3, 4, or 5 days a week for infusions; this is especially disruptive for them in trying to maintain a normal life. I do think we'll come back to this. I think there are some measures that we've discussed. We just kind of decided that wasn't on our core mission, and we want to put a bow on this and reach a consensus that we can then publish and get that piece ready. I think in phase 2, phase 3, we'll come back to some of these issues and look deeper at that patient impact and couple that with the direct costs to payers and to practices of these other additional costs beyond the drug.