Newborn screening urged for fatal neurological disorder, spinal muscular atrophy (SMA)

Spinal muscular atrophy (SMA) is a rare degenerative disorder that typically causes death within a few years after birth. There is a genetic test to detect SMA. An ongoing clinical trial has had fantastic success where the 25 newborns with SMA have survived without dependence on a ventilator for two years. However, the trial is not yet complete nor published. With a July 8th deadline looming, should the FDA approve genetic screening for SMA at birth?

EDIT: Full text below. The article looks like it should be open access, but here it is in case it's not working:

Roughly once a day in the United States, a child is born with a fatal genetic disorder that destroys motor neurons in the brain stem and spinal cord. In its worst and most common form, spinal muscular atrophy (SMA) kills children when they are still toddlers, as their respiratory muscles fail.

But 18 months ago, the Food and Drug Administration approved a first, promising treatment: a drug that restores production of a key protein missing in SMA (Science, 16 December 2016, p. 1359). Now, SMA advocacy groups and members of Congress are urging Secretary of Health and Human Services (HHS) Alex Azar to recommend that all 4 million infants born in the United States each year be tested for SMA. They argue that affected children should be identified and treated when the new drug likely helps the most—before neurons die.

By law, Azar faces an 8 July deadline, but such deadlines have been missed in the past. And although an advisory panel voted in February in favor of screening all newborns, some of its experts dissented. They noted that key studies of the new treatment—a drug called nusinersen (marketed as Spinraza by Biogen of Cambridge, Massachusetts)—are still ongoing, involve small numbers of children, and are unpublished.

But delay “would be a tragedy for children born in the interim who may benefit from screening because they will miss the window for receiving treatment when it is most effective,” 14 members of the House of Representatives wrote to Azar last month, urging speedy approval. An HHS spokesperson says Azar is “still reviewing this important issue.”

Checking for the SMA mutation in a drop of blood would cost $1 to $5 per newborn (although the drug itself costs $750,000 for the first year and $350,000 annually after that). But there's a high bar for adding a disorder to the 34 conditions for which screening is recommended. Among other criteria, data must show that outcomes improve if treatment begins before symptoms appear. In this case, the key data come from an ongoing, Biogen-sponsored trial called NURTURE in which 25 newborns with confirmed SMA got the drug before symptoms developed. In July 2017, when the oldest baby had been followed for 25 months, all the children were still alive, none were ventilator-dependent, and those old enough sit up without support could do so—an achievement unheard of in babies that die early from SMA.

But those striking results have not been published in peer-reviewed journals because the patients are still young and the trial is ongoing. When the Advisory Committee on Heritable Disorders in Newborns and Children, convened by HHS, met in February—before the promising data were released—five of 13 voting members opposed routine SMA screening.

“It concerns me about not having published, peer-reviewed literature,” said one, Joan Scott, who directs services for children with special health needs at the Health Resources and Services Administration in Rockville, Maryland. Scott Shone, a senior public health analyst at RTI International in Research Triangle Park, North Carolina, added, “If all this robust data exists, why was it not presented?” He noted that no one knows whether the drug will continue to help kids as they age. “There's a huge unknown there,” Shone said.

There simply hasn't been time to accumulate long-term data, says Wildon Farwell, senior medical director in clinical development at Biogen. But, he says, “All the data we see shows that treating patients before symptom onset allows the potential for greater benefit than waiting until symptoms occur.”

A theoretical model developed for the advisory committee by outside experts estimated that screening would spot about 150 babies each year who otherwise would not be identified until symptoms set in; diagnosis can take months after that. Those months may be crucial: One study showed that in the sickest patients, 90% of motor neurons are destroyed by the age of 6 months.