Just before Christmas, the FDA approved Spark Therapeutics’ Luxturna, a one-time treatment for a rare form of inherited blindness. The first true gene therapy in the U.S. to be given a green light by the agency, it represented a legitimate medical breakthrough. Patients in a clinical trial described seeing snowflakes and stars for the first time.

However, with the innovation came questions about cost. Predictions varied, but “really expensive” was the overwhelming consensus. In that regard, Spark didn’t disappoint. In early January, it unveiled Luxturna’s price tag: $425,000 per eye, or $850,000 for both.

Remarkably, many saw it as a relative bargain. Yes, calling anything that costs $850,000 a bargain — such as a yacht or an apartment — sounds ridiculous on its face, but the therapy came in under Wall Street’s prediction — a cool $1 million.

More surprisingly, in addition to all those zeros, was Spark’s novel pricing model. The company announced it would tie Luxturna’s price to outcomes, as well as launch an installment-payment program with the Centers for Medicare and Medicaid Services.

The details of the former are what captured the industry’s attention: If a patient doesn’t experience gains in vision within 30 or 90 days, or if those gains aren’t maintained 30 months after the treatment, the company would refund some of the cost to insurers. Spark has already announced a deal with the nonprofit Harvard Pilgrim and is in discussions with other payers, CEO Jeffrey Marrazzo said during a presentation to investors at JP Morgan’s annual healthcare conference in January.

See also: Nearly every $100,000 drug approved in the last three years was for a rare disease

Immediate response to the announcement was largely positive. Bloomberg dubbed it “the right move,” one that “highlights a recent trend toward (relative) pricing moderation with big implications for the drug industry.”

Executives from Eli Lilly, GlaxoSmithKline, Roche, Merck, and Bristol-Myers Squibb all told Reuters they are paying close attention to how Spark’s pricing negotiations play out. With a handful of gene therapies in late-stage trials, another gene-based therapy could soon follow in Luxturna’s footsteps.

Will the company’s outcomes-based model become a blueprint for the category? And more importantly, could it lead to a seemingly fairer pricing system?

INNOVATIVE AND OVERPRICED

“We are on the cusp,” says John Schneider, CEO and founder of Avalon Health Economics, a consulting firm specializing in health economics and outcomes research. Payers want to make novel treatments available to patients, he notes, but they also want additional assurances, given the lofty upfront costs.

Gerard Anderson, a health policy professor at Johns Hopkins University, agrees outcomes-based pricing will become a popular model for high-priced gene therapies. He believes the model originated across the pond, in countries that have adopted a single-payer model.

See also: Havas Health, Vencore form JV focused on rare diseases

For instance, in the U.K., the government relies on the National Institute for Health and Care Excellence (NICE) to determine what a drug is worth. The agency makes recommendations by evaluating the extent to which a treatment prolongs or improves a life. There’s an established cap: The government typically doesn’t pay more than $39,000 per quality adjusted life-year.

Drawing such a line is, by definition, a thorny and contentious exercise, but doing so creates leverage. To get their drugs approved, pharma companies will offer NICE significant discounts and other concessions, such as “not paying as much if a product is a failure for a patient,” Anderson says. “This system is being exported to the U.S.”

However, he is skeptical the framework will actually lower prices for Americans. Stateside, drug pricing is opaque and fragmented, with developers offering a range of rebates to a bevy of individual insurers and pharmacy benefit managers. There are no state-sanctioned reviews of a drug’s monetary value and the government largely stays out of price negotiations.

As a result, drug costs continue to rise at a rate far greater than inflation, vastly outstripping increases in other developed countries. Of course, there are arguments to be made in favor of this system, notably that high prices motivate drug companies to invest in research and development for novel treatments.

Anderson predicts that as outcomes-based payments become more popular in the U.S., rebates will simply be baked into a drug’s initial list price. “You’re just moving shells around,” he says.

See also: Top 25 rare disease brands, by sales

While Spark hasn’t said how much it will return to payers if the treatment fails to hit its benchmarks, Marrazzo indicated rebates won’t exceed 20% — which Schneider describes as a glorified discount. Nonetheless, he believes the model could still have a positive impact on the industry. Even if the financial stakes aren’t high, a risk-sharing agreement provides explicit incentives for pharma companies to develop drugs that work long-term, which could lead to improvements in clinical-trial design and other validation metrics.

THE BREAKING POINT

Peter Bach, director of Memorial Sloan Kettering’s Center for Health Policy and Outcomes, has a more cynical view. Outcomes-based pricing is solving a problem, he says, but neither affordability nor accountability has anything to do with it. As more six-figure therapies come to market, he explains, “This will become the industry’s latest ‘pay no attention to the man behind the curtain’ strategy. These deals will appear to address very high drug prices, but will have minimum financial impact.”

Luxturna isn’t the only gene-based treatment with a lofty payment: Gilead Sciences and Novartis’ CAR-T drugs are priced at $373,000 and $475,000, respectively. Their costs aren’t generally perceived as alarming because they treat rare diseases. But what happens when a revolutionary drug is developed for a common condition?

The U.S. healthcare system isn’t designed for such a breakthrough. The system could crack under financial pressure — and outcomes-based pricing won’t be enough to resuscitate it.

Let’s say a dementia drug that keeps patients sharp and out of nursing homes were to receive FDA approval, and let’s speculate that its developer priced it at $100,000 a year. On the surface, the cost isn’t unreasonable given the drug’s tangible value.

See also: Trump administration raises questions for rare-disease drugmakers

The situation could still pose nightmarish financial problems. An estimated 5.5 million Americans suffer from Alzheimer’s. Multiply that figure by $100,000, and the healthcare system is staring down an annual bill of $550 billion. Total healthcare spending last year was $3.3 trillion, or around 18% of GDP.

If we’re lucky and a suite of novel, curative drugs are just around the corner, there are two problems that need to be addressed. The industry needs an objective and comprehensive value-assessment framework that accurately weighs cost against innovation, says Bach. At the moment, the only organization that comes close to this description is the Institute for Clinical and Economic Review (ICER), a nonprofit that weighs in on how much a drug should cost based on its value. In January, it pronounced that for Luxturna to be cost-effective, it would need to be discounted between 74% and 82%. While there are signs payers are reading ICER’s evaluations, it’s unclear whether its reports are having any real-world impact.

Addressing the second problem requires a seismic shift. When a novel, patent-protected drug saves the healthcare system money, pharma companies pocket the majority of the surplus. Gilead’s hepatitis C cures, while pricey, could save tens of thousands down the road as patients avoid expensive procedures, such as liver transplants. The bulk of these savings is absorbed by the company, not passed down to consumers.

If the world were governed by clean economic principles, this would be a good setup: Drug companies that create the lion’s share of the savings should keep them. But it’s not and so it isn’t, says Anderson. In the U.S., innovative, life-changing drugs are so expensive that they are inaccessible to many patients. We can’t afford a system where all the cost savings go to the developer.

“It might be efficient to charge people the max value of a product, but it’s not equitable,” says Peter Hilsenrath, a healthcare economist at University of the Pacific.

“Drugs don’t work if people can’t afford them,” Anderson adds. “The key is access. I’m hoping there will be a way to measure a drug’s value and its worth to society.”