BOSTON — The shelf life on pharma's time-tested pricing model is running out. It may not spoil today or tomorrow or in the next few years, but stakeholders agree it will eventually give way to a new system that better addresses cost conundrums like pricing a gene therapy.
In the meantime, drugmakers and payers have been experimenting with value-based contracts, agreements that tie how much the former group gets paid for its medicines with how well they work in patients once on the market. The contracts only recently started to gain traction, and have so far brought more questions than answers, according to industry experts attending the Drug Information Association's 2018 annual meeting.
Legal barriers persist
Food and Drug Administration guidance issued earlier this month offered some much-desired clarity about conversations between drug companies and payers. In the past, manufacturers were strictly limited to discussing the health benefits included in a drug's label. But now, they can present less rigorously tested economic analyses that are materially different from the FDA-approved label, so long as they make clear note of that difference.
"That's one barrier from last year where we made a lot of progress," Robert Duffield, counsel for market access at Novo Nordisk, said during a panel presentation at DIA.
Elsewhere, there has been less advancement. Duffield highlighted a couple of legal concerns that remain largely unresolved: one being value-based contracts may inadvertently violate the anti-kickback statute — which prohibits exchanging anything of value for a federal healthcare program referral — the other being they may further complicate price-reporting rules for government programs like Medicaid Best Price.
"There's no clear guidance on, if we give you the product for free and it doesn't work, is that a best price of zero? And what happens there?" he said.
Notably, these worries have weighed on manufacturers for months. Last March, a survey from industry trade group PhRMA found price-reporting metrics and the anti-kickback statute were barriers of high or very high importance for 64% and 46% of responders, respectively.
Traditionally, insurers and pharmacy benefit managers would weigh a drug's price against trial safety and efficacy results in deciding whether to cover it. Value-based contracts, however, have made them increasingly interested in findings from outside the clinic, such as patient-recorded outcomes or information collected via wearables.
"Going beyond claims data allows you develop a much more rich assessment of a specific drug's impact on the patient," Jim Clement, executive director of value-based care and supply chain strategy at Aetna, said during the panel.
High on manufacturers' to-do list is gathering more information at earlier stages of the clinical pathway, though that has presented challenges.
"You're really confronted with a number of variables that don't exist in traditional clinical research as it's performed today," Michelle Hoiseth, corporate vice president of real-world data services at Parexel, told BioPharma Dive in an interview.
"Is the data complete? Is it accurate? Are the data in fields that are extractable easily?" Those are a few of the worries, Hoiseth noted. Yet it would appear pharma has no choice but to tackle them, given the emphasis placed on real-world evidence in both the 21st Century Cures Act and PDUFA VI.
"The use of real world data for what we would call the emerging-use cases is still tricky business, and yet that's where this legislation is pushing us as an industry," she added.
Bottom lines are surely fueling the push as well. Amgen, Sanofi and Regeneron Pharmaceuticals have pursued real-world evidence to boost sales of their respective PCSK9 cholesterol drugs, which so far haven't lived up to the hype that surrounded them pre-approval. To that point, underperforming drug launches cost drugmakers around $12 billion in lost sales revenue over the last two years, according to an internal Parexel analysis.
Adding scale, metrics
The broader lens of real-world data often means monstrous volumes of information. To deal with it, drugmakers and payers are turning to an array of technological solutions, from blockchain and cloud-based data storage to robotic and probabilistic software. Integrating that tech too presents challenges.
"Where I've been taking the organization is to say: what is available to us in the information technology space to ease the pain, so to speak. Because I'm a national health plan and I still can't hire enough resources to make these truly scalable," Aetna's Clement said.
And scale aside, payers haven't come to any sort of consensus on how many data points or clinical outcomes should make their way into a value-based contract.
Clement, for instance, said the optimal contracts have three to five stacked metrics. He acknowledged, though, that even those based on an event with just two potential outcomes, like whether Novartis' Entresto treatment either leads to congestive heart failure (CHF) remission or not, aren't easy to parse out.
"When you're talking stacked metrics, you may be measuring five metrics for any one drug [and] the data requirements and the analytical horsepower explode exponentially."