May 2018 Edition Vol.11, Issue 5

Discussing Next Generation Value-based Payment Designs

John McCleery

At the recently held 2018 Community Oncology Conference: Keeping Patients at the Center, sponsored by the Community Oncology Alliance (COA), Kavita Patel, MD, MSHS, Co-Founder, Tuple Health moderated a discussion titled, “Moving the OCM 2.0 from Concept to Universal Payment Model.” The panel included Bruce Gould, MD, Medical Director, Northwest Georgia Oncology Centers and Mark Fendrick, MD, Director, Value-Based Insurance Design, University of Michigan.

Patel opened the discussion with a summary of concepts that are positive about the current Oncology Care Model (OCM), such as open to all payers, patient engagement, and Monthly Enhanced Oncology Service (MEOS) payments; and key issues affecting OCM practices such as practice transformation requirements, attribution, reporting requirements, risk adjustment, and financial incentives.

Moving into the discussion on next generation payment models or OCMs.v2 and what they would look like, Dr. Gould opened his dialogue by asking the audience to consider “what is value in drug therapy?” which is typically a part of any value discussion. He strayed away from the oft-used definition that value = quality / cost and instead offered a simple definition that “value is what something is worth to you.”

Putting that definition into more clinical terms as it relates to drug costs and drug therapies, he thought a more appropriate equation to apply to future payment design models is value = clinical benefit / drug price + toxicity + resource utilization. While drug utilization may not be a big component at this time, when “we start talking about CAR T therapies, it will become a factor,” he said.

To make his point on value as it relates to drug price and therapies and how that would relate to future payment designs, he presented what he referred to as 2 “extreme examples”. The first was a study on tamoxifen use after adjuvant chemotherapy in premenopausal women with node positive breast cancer. He showed a graphic of a Kaplan Meier survival curve from a clinical trial that sought to answer the question whether there was any clinical benefit of adding tamoxifen after patients with node positive breast cancer had received chemotherapy.

He pointed out that the trial results were modest, with “a 2% to 3% absolute difference between those who got tamoxifen and those who didn’t” (Figure 1). And that was after 8 years of study. He focused on the cost of tamoxifen, which he noted was only $17.75/month at Walgreen’s.

When is comes to a newer drug, neratinib, for example, used in HER2 positive breast cancer in patients who are receiving adjuvant therapy, he considered its cost juxtaposed with the cost of tamoxifen.

In Figure 2, he notes that at 60 months in the neratinib trial data, the difference in disease-free survival (DFS) “is less than about 3 percentage points.” Adding, “and that’s not overall survival—where patients are living longer—it’s living cancer free. And that’s a major benefit, but at what price?” he asked.

The monthly cost of neratinib is $11,200, and for one year it’s $134,000. The problem, he said, is that “we have no way of selecting the 2 or 3 patients that are going to benefit.” That means, “we have to spend about $13 million on 100 patients to find the 2 or 3 patients who would receive the clinical benefit of adding neratinib as adjuvant therapy. “If there was a biomarker, that would say this drug worked in a certain number of patients, that would be very helpful, but other than Herceptin, there’s no other biomarker that’s specific to this drug.”

Moving on to his second example, Dr. Gould offered that no oncologist would argue that adding Perjeta to Herceptin and/or chemotherapy hasn’t been a major breakthrough in treatment of patients who have metastatic breast cancer that overexpress HER2 neu. He stated, “Perjeta works great in metastatic breast cancer”; and in combination with Herceptin, he thought a lot of oncologists may logically think it would work just as well in the HER2 neu adjuvant setting.

However, in looking at the data from the APHINITY trial, which compared patients with HER2 neu positive breast cancer who received chemotherapy and a year of Herceptin with or without Perjeta, “we see the benefits are much more modest, to say the least. It’s a less than 2% difference,” stated Dr. Gould (Figure 3).

When it comes to cost, one year of adjuvant Perjeta costs $76,000; one year of adjuvant Herceptin and Perjeta combo costs $174,000. Dr. Gould pointed out that to find the 2 patients who would benefit in terms of DFS, 100 patients would need to be treated with the combination, “because we don’t have a marker to pick those 2 who are going to benefit. And the cost of trying to help those 2 patients skyrockets to over $7,000,000.”

“Obviously there are some strategies,” he said to think about in the context of second generation payment models, such as outcomes-based treatment in which the payer is reimbursed if the drug doesn’t work after a certain time. Or indication-based pricing, where, for example, in the adjuvant setting of Perjeta, a lower price would be paid to the manufacturer; and in the clinical scenario, if the drug that is being used shows a high benefit then the manufacturer would be paid a higher price.

A third strategy would be value-based insurance designs. Mark Fendrick took over the session at this point to discuss value-based insurance designs. What keeps Dr. Fendrick up at night is the tension between “what we can do and what we actually do do.” We are heading into what he termed “Star Wars science,” but out delivery system is more like “Flintstones delivery.”

“We need to bring oncology care from the stone age to the space age,” he said. And that it’s very important when discussing next generation payment models to shift the conversation from “how much we spend to how well we spend.” He referred to Dr. Gould’s presentation and offered that sometimes a million dollars spent is “an amazing investment and sometimes $17 spent is completely wasteful.”

He further added that benefit designs that favor the patient need to be addressed by all stakeholders. Patients have gone “from paying flat copays to percentage coinsurance to stifling deductibles that half the patients enrolled in these health plans know they don’t have the capital needed to pay.” He noted that overall, “Americans don’t care about the cost of cancer care; they care about what it’s going to cost them.”

Patients should not need to have a bake sale to afford a drug that is going to save their life, he said. “This is not the sign of a civilized country,” he said, adding “We have got to come together for some really meaningful changes in these [payment] designs to ensure that the patient is considered” so they can afford their treatment.

According to Dr. Fendrick, there is no clinical service that is high value or low value. For Dr. Fendrick, the clinical value of a service is determined by who gets it, when it the course of their disease, who provides it, and where. Where oncology care is being provided “is a topic that would make most men’s hair go gray.”

For new payment designs, what needs to be taken into consideration, for Dr. Fendrick, entails benefit plan designs that instead of setting cost sharing based on what things cost, that patients pay out of pocket for how important it is for the health that’s produced. How that would play out is that patients would pay more for drugs that produced higher results and lower for drugs that produced lower results, instead of the 20% coinsurance most Americans pay for all specialty drugs, regardless if that drug produces high differences in survival curves or low differences in survival curves.

He notes that this healthcare reform idea has very little support among policy stakeholders on Capitol Hill. But also noted that the Medicare Advantage VBID demo, currently piloted, allows Medicare Advantage plans to lower cost sharing for high-value, high-performing providers, and services.

Post a Comment