FDA surprised Novartis, patient advocates and physicians by approving a label for the pharma's gene therapy that creates hope that newborns diagnosed with spinal muscular atrophy can be treated before symptoms appear -- in time to head off permanent, severe disability. The label also gives the company cover for a price that would be hard to justify for the narrower population that it studied in the trials to support its marketing application.
On May 24, FDA approved the gene therapy manufactured by AveXis, a unit of Novartis AG, for all patients with SMA under two years old with bi-allelic mutations in the SMN1 gene.
This is a much broader population than was included in the studies that supported approval. A completed trial and an ongoing trial enrolled only symptomatic Type 1 SMA patients with mean ages of 3.4 months and 3.9 months, respectively.
Although it's not unheard of for FDA to grant a label beyond the study population, this one took the company by surprise.
FDA's label sets up Zolgensma onasemnogene abeparvovec-xioi to compete with Spinraza nusinersen from Biogen Inc. to treat pre-symptomatic babies diagnosed with SMA. The two products provide hope that widespread screening of newborns could allow patients with SMA to lead lives that aren't cut tragically short and that some may avoid disability.
Effective treatment of SMA has profound benefits for children and their families. Prior to approval of Spinraza in December 2016, infants diagnosed with Type 1 SMA were unable to sit on their own and had a life expectancy of less than two years. Individuals with Types 2 and 3 SMA suffer debilitating neuromuscular damage.
The indication helps Novartis justify a $2.125 million wholesale acquisition cost, a price that was just within a range the Institute for Clinical and Economic Review (ICER) benchmarked as cost-effective based on the product's value. "We and I think Novartis were surprised that the label was broader than the study population."
Michael Sherman, Harvard Pilgrim
If the label had adhered strictly to the population studied in clinical trials, the upper limit of ICER's range would have been $1.5 million, according to a report it issued in April. ICER released an updated estimate on May 24 to reflect the broader indication, as well as data from an ongoing clinical trial.
Novartis had hoped to be able to mitigate the unprecedented price by offering to accept substantial financial risk if Zolgensma fails to perform as expected.
AveXis has been trying for a year to persuade CMS to find a way around the Medicaid best price rule that prevents it from entering into outcomes-based contracts that put most of Zolgensma's price at risk.
AveXis and Novartis aren't the first companies to have plans for outcomes-based contracts founder on best price. Spark Therapeutics Inc., for example, has been unable to get traction for a robust outcomes-based payment model for its gene therapy, Luxturna voretigene neparvovec-rzyl, approved to treat a rare form of blindness. (see "Breaking Ground").
HHS Secretary Alex Azar has expressed strong support for outcomes-based contracts, but CMS Administrator Seema Verma has sent mixed signals. At a recent press briefing she suggested that outcomes-based contracts could lead to higher prices.
Making payments for one-time therapies contingent on performance over time does not address concerns over prices, but it does provide assurances to payers that they are getting what they've paid for.
Questions about the appropriate prices, and the desire for assurances that one-time therapies provide durable benefits, will become more pressing as the number of gene therapies expands. Value subset
Novartis CEO Vas Narasimhan highlighted Zolgensma's pricing congruence with the revised ICER value assessment, telling reporters the "price of Zolgensma is expected to be within the range of traditional cost-effectiveness thresholds used by ICER when updated for its full labeled indications."
It sits at the upper bound of the value-based price estimated for pre-symptomatic patients - and well above the threshold ICER set for symptomatic SMA patients.
ICER's value-based price range of "$750,000-$1.5 million is still applicable to Type I patients with symptoms," ICER President Steven Pearson told BioCentury. "We decided not to try to create a hybrid price based on a guess on the mix that will be treated in year one, year two, etc., but $2.1 million truly is an upper bound under favorable assumptions about the durability of treatment benefits and the most cost-effective population -- pre-symptomatic patients."
Novartis said it will market Zolgensma for treatment of the estimated 400 infants who are newly diagnosed with SMA in the U.S. every year, as well the about 700 patients under two years old.
AveXis, which created the gene therapy, was acquired by Novartis in April 2018 when the drug was in Phase III.
After submitting the Zolgensma BLA, AveXis told SMA caregivers and physicians it expected FDA to approve a label for Zolgensma -- called AVXS-101 at the time -- that reflected the population in the product's clinical trials. The label raises questions for payers who will be asked to cover populations that haven't been studied.
"Based on the data included in the applications, our expectation is that the initial label will be for intravenous (IV) use of AVXS-101 for infants with SMA Type 1," AveXis stated in an October 2018 open letter to the SMA community.
The FDA approval not only makes Zolgensma available to pre-symptomatic patients and to patients who are older than those AveXis has studied, it also opens the door to treat Type 2 and Type 3 SMA.
Type 2 SMA typically has an onset between seven and 18 months, and Type 3 SMA has an onset after 18 months.
The label raises questions for payers who will be asked to cover populations that haven't been studied.
"We and I think Novartis were surprised that the label was broader than the study population," Michael Sherman, SVP and CMO of Harvard Pilgrim Health Care Inc., told BioCentury. "We're still negotiating, including the question of exactly what population we cover."
Sherman said based on discussions with AveXis and the clinical data submitted to FDA, Harvard Pilgrim "had anticipated that the approval would be for the more limited population, under eight and half kilos and under six months old, commensurate with the population study. We have internally agreed to cover that study population."
Harvard Pilgrim hasn't decided whether it will cover Zolgensma for the full spectrum of patients in the FDA label, Sherman said. Beaten by best price
Novartis had hoped to launch Zolgensma with a robust outcomes-based contract but, like other manufacturers, it has not been able to overcome the Medicaid best price rule.
Under the rule, if a manufacturer gives a rebate for even a single patient of more than 23.1% -- or 17.1% for a drug like Zolgensma that is approved solely for pediatric use - that rebated price becomes the highest price for all Medicaid purchases.
Because of the best price rule, the maximum risk Novartis will take in an outcomes-based contract is 17.1%, or about $363,000.
"We'd like to be able to provide payers with much bigger rebates for how the product does over time, but right now we're limited by the best price impact that would generate," AveXis President David Lennon told BioCentury. "The best price rule is the largest impediment to access that we face in this country today."
Michael Sherman, Harvard Pilgrim
Sherman agrees that Novartis should be liable for a much larger rebate. "You should really should have a robust outcomes-based agreement to make sure it delivers" on the promise demonstrated in clinical trials.
"The outcome measures we have in place, which were designed for Type 1 symptomatic patients, aren't necessarily relevant to a broader population," he added.
Rebates will be triggered by death or permanent ventilation, Lennon told reporters on May 22.
Sherman said he couldn't reveal the discount Novartis is offering Harvard Pilgrim, but he confirmed that it does not exceed 17.1%.
Harvard Pilgrim has tentatively agreed to pay the full price of Zolgensma up front and to be eligible for a rebate if the drug fails to meet performance goals any time during the first five years after treatment.
While this arrangement will work for a treatment of an extremely rare condition like SMA, a more robust outcomes-based agreement could be needed to enable payments for one-time treatments for more common diseases, Sherman said.
"The best price rule is the largest impediment to access that we face in this country today," Sherman told BioCentury. In the absence of outcomes-based agreements, payers could refuse to cover expensive treatments based on limited data, especially when there are unresolved concerns about the durability of treatment, he said.
AveXis has had discussions with CMS over the last year about best price, Lennon told BioCentury.
"We have engaged CMS in a discussion on a framework that could potentially allow that to happen for our product," Lennon said. "We're still waiting for their feedback and agreement about if it could be executed under current policies or not. If they say no, it requires legislative action.""We'd like to be able to provide payers with much bigger rebates for how the product does over time."
David Lennon, AveXis
CMS head Verma acknowledges that new payment models for curative therapies are needed, but she doesn't seem to be sold on outcomes-based contracts.
"Drugs that are coming out are one-time curative treatments and so these are very different from what we've had in the past," Verma told reporters on May 22. "The payment systems we have in place are really not set up to deal with that."
Verma said she is "especially concerned with state Medicaid programs where they are on a one-year or two-year budget cycle and they are having to pay for multi-million dollar drugs for lots of their beneficiaries."
Verma reported that "companies are coming to us and saying 'Can we pay for this depending on the outcomes. If someone is living for three years we'll pay this and if the person has passed away we'll give a rebate.'"
The CMS Administrator added that the she worries that outcomes-based contracts "actually encourage higher launch prices because they potentially are going to be giving rebates if the treatment isn't effective."
Verma's concern about outcomes-based contracts is "legitimate in the sense that we in this country don't have a mechanism for coming up with a starting price, and it goes beyond Zolgensma," Sherman said. "I've done a lot of outcomes-based agreements and it's not a panacea. If you have something where there is 50% risk and it's priced at three times what it should be, that doesn't really get you the fair value."
Mark Trusheim, strategic director at MIT's NEWDIGS (NEW Drug Development ParadIGmS) program, told BioCentury that he's heard state Medicaid directors speculate "that any new financing approach will simply allow higher pricing by making things seem affordable."
Determining the value or appropriate price for a therapy can be considered separately from designing performance-based contracts.
Making payments contingent on the outcomes achieved by an intervention should insulate Medicaid and other payers from some unjustified costs, Trusheim said. "A performance-based approach as opposed to straight-out financing allows for fair pricing and protects Medicaid from products that don't perform as well as hoped or expected."
Biogen Inc. (NASDAQ:BIIB), Cambridge, Mass.
Centers for Medicare and Medicaid Services (CMS), Silver Spring, Md.
Institute for Clinical and Economic Review (ICER), Boston, Mass.
Massachusetts Institute of Technology (MIT), Cambridge, Mass.
Novartis AG (NYSE:NVS; SIX:NOVN), Basel, Switzerland
SMN1- survival of motor neuron 1 telomeric