Up until three years ago, Vamshi Rao didn't have any treatments to offer the families of babies born with a genetic disease that meant they likely wouldn't live past the age of two.
No medicines existed for the degenerative condition called spinal muscular atrophy, or SMA. So Rao, a neurologist at the Lurie Children's Hospital of Chicago, could only help manage expectations and day-to-day care.
His role, and that of other physicians, changed dramatically in 2016, when Biogen's Spinraza became the first drug to win U.S. approval for the condition, and then again this May, when the Food and Drug Administration cleared a gene therapy called Zolgensma.
Rao can now give families options that represent a realistic shot at a longer, better life for babies with SMA.
"We've never been in this situation before," he said in a September interview.
Between May and October, about 100 families facing the diagnosis opted for Zolgensma, just the second gene therapy for an inherited disease to reach commercial markets in the U.S. Unlike Spinraza, Zolgensma is designed to be a one-time treatment and, possibly, a cure.
Several of those infants were treated at Lurie Children's, where Rao is also helping run clinical studies sponsored by Novartis, the Swiss pharmaceutical company that sells Zolgensma. About 50 other U.S. institutions have administered Zolgensma commercially, outside of clinical testing.
Approval was a major accomplishment in a field that had for decades held the potential to develop curative fixes for devastating genetic disorders like SMA.
But its arrival — clouded by a scandal over manipulated animal testing data submitted by Novartis — made real once theoretical concerns over how patients will receive and pay for expensive, one-time therapies.
Infused via inactivated, hollowed-out viruses, Zolgensma replaces the missing or defective gene that causes SMA to develop. Infants in the Phase 1 study used to win approval could sit up independently, stand and, in two cases, even walk in the months and years after receiving Zolgensma — functional milestones that wouldn't have been reached otherwise.
Whether Zolgensma can actually be a cure won't be known definitively for years. But in theory it could be if given to infants early enough, before their affected neurons start to die, says Jerry Mendell of Nationwide Children's Hospital in Ohio, who led the first study of what would become Zolgensma.
The promise of lifelong benefit was central to Novartis' justification for pricing Zolgensma at $2.1 million per patient, more than any other drug before it.
While insurers initially balked at covering such a pricey, one-time treatment, Novartis says coverage is now in place for 90% of eligible patients on commercial plans and 30% of those on Medicaid.
"It's clear that payers are not happy here, but they have largely relented," said Ronny Gal, an analyst at Bernstein, in an interview.
"It's hard to push against a drug that takes a kid who would die and allows him to live. But for products that are less efficacious, I'd expect pushback."
Hoping to ease insurer concerns, Novartis offered to spread Zolgensma's cost over five years, and is working to set up agreements linking reimbursement to patient outcomes. While the latter option has proved popular, none have yet taken up the drugmaker on its pay-over-time proposal, said Dave Lennon, CEO of AveXis, the biotech developer of Zolgensma that Novartis bought for $8.7 billion last year, in an interview.
Also helping the pharma's case to insurers was an estimate from the Institute for Clinical and Economic Review that judged Zolgensma, if given very early, could be cost-effective at a price between $1.2 million and $2.1 million. (A more conservative finding by ICER, however, put Zolgensma's cost-effective price at just under $900,000.)
"Certainly you see now the methodology that ICER is using and the concept of value-based pricing is an extremely important part of the discussions with payers," said Lennon.
Yet early treatment, before symptoms arise, is dependent on genetic screening for SMA. While the federal government has recommended the condition be tested for in newborns since July 2018, only 15 states are currently doing so. Two more, Michigan and Colorado, are expected to soon begin screening.
"When states puts these programs in place, we do see much higher utilization in newborns of therapies, at rates that are sometimes three times as high as states that don't have newborn screening," said Lennon.
While Zolgensma's early market success suggests a one-time gene therapy can be commercialized, Novartis has also benefited from both SMA's rarity and the cost of existing treatment in winning over insurers.
In the U.S., only about 500 babies are born each year with SMA, 60% of whom are diagnosed with the most severe form, called Type 1. That's a much smaller population than other rare genetic conditions like hemophilia and sickle-cell disease, the target of other gene therapy programs in Phase 3 trials.
And Spinraza costs $750,000 for the first year of treatment, and $375,000 thereafter — potentially making $2.1 million more palatable if Zolgensma truly is a one-and-done treatment.
Novartis' achievement, however, is marred by a damaging data scandal that's put both the drug and company under scrutiny.
When submitting the application for FDA approval, Novartis included testing data from preclinical mice studies that was manipulated, it turned out, by (or at the behest of) top AveXis officials. Novartis management knew of the falsified data in March, but filed Zolgensma anyway.
In a show of its displeasure and a signal to other developers, the FDA warned Novartis of potential civil or criminal penalties, pending an investigation that has yet to complete. Zolgensma, however, could stay on the market, and the FDA affirmed the therapy's positive benefit-risk profile.
Novartis ousted the AveXis executives it says were to blame, and has sped up the integration of the biotech into its own quality control organization.
"Quality, in the grandest scheme, is always a journey," said AveXis' Lennon. "While we've accelerated this, we don't see an end to our commitment to quality improvement."
Months after addressing that controversy, the FDA placed partially on hold a study of Zolgensma due to concerns over toxicity seen in animals, possibly tied to how the therapeutic gene was delivered. The regulatory action revived previously raised warnings about the risk of inflammation in certain neurons exposed to a virally delivered gene.
In both respects, Zolgensma could be a case study in the risks as well as rewards involved for large pharmaceutical companies looking to get into gene therapy.
"This is one of the first gene therapies," said Gal. "We don't have a lot of experience infusing viral particles into the brain."