The Case of Spinal Muscular Atrophy: Getting to treatment

Part 2

A few years ago, a bright grade-school student, accompanied by relatives, was in my office in the Department of Pathology and Cell Biology at Columbia University. Across the corridor, a class of PhD students was taking a course that asked what molecular and cell biologists could contribute to the treatment of serious diseases. Most PhD students study basic problems and use convenient organisms — fruit flies, mice, worms, amoebae or bacteria, but usually not people. Dr. Steven Spitalnik, a pathologist from our department, along with a neurologist with long experience in motor neuron diseases led the discussion. The graduate students knew a lot about molecular and cellular biology, but not the great clinical problems in medicine, and so they had studied a number of diseases in depth over a year. One of them was Spinal Muscular Atrophy (SMA) and now they were going to meet the patient who had been chatting in my office. 

This child had a late onset form of SMA. Although the symptoms appeared late, they would progress; even now this patient got up off the floor in a peculiar way, a three-point stance, due to weakened muscles. This was six years ago and there was no treatment for such patients; there was a gene therapy for mice whose genes had been mutated to give them the weakened muscles of SMA.    

How might one attack the problem in humans? There were two immediate ways.

Genome sequencing discovered a closely related gene called SMA2. It makes a protein product SMN1 (Survival of Motor Neurons), except for one change, which renders it nearly inactive. Researchers created a small segment of DNA and injected it into the cerebrospinal fluid, which bypasses the blood-brain barrier. It corrected the mutation in mice, so that their muscles made effective SMN1 and did not deteriorate. Since an infant’s brain is more than a thousand times larger than that of a mouse, the fear was that it might not work in a newborn human.

The drug, now called Nusinersen, corrects the SMA2 gene to produce the missing protein. As of a year ago, 122 infants had been treated and 41 percent improved to the point of being able to breathe, sit, and in some cases, to walk. The results were so compelling that the clinical trial was stopped so that children in the control arm of the study could get the drug. One disadvantage is that it must be injected into the spinal fluid every four months. Still, it does not always work.

There is a second treatment. If you are going to attempt something difficult in biology or medicine, it is a good idea to let evolution take the first steps. Is there a relatively harmless virus that could deliver DNA to a neuron guarded by the blood-brain barrier? Decades of sometimes discouraging research had found one, called AAV-9 or adeno-associated virus 9. It delivers whatever DNA it is carrying — in this case, researchers have replaced much of its DNA with an SMA1 gene. The hope was that the virus would introduce the SMA1 DNA into motor neurons and replace the missing function. (How these recombinant viruses are made and how they work in detail is fascinating for PhD students and your correspondent, but is a subject for another day.)

Such a virus was tested extensively and finally injected into newborns showing early symptoms of SMA. This sounds simple, but is not. The paper that resulted, “Single-Dose Gene Replacement Therapy for Spinal Muscular Atrophy,” from an international consortium, appeared in the New England Journal of Medicine in November 2017. With one injection, a great number (trillions/kg) of AAV9 viruses carrying SMA1 DNA was injected. This caused apprehension among researchers, but there was only mild inflammation.  

Of the first 15 patients, all improved, and none died, as untreated children always had by 20 months. Some improved more than others, perhaps depending on how soon the virus was administered. This was a Phase I trial to establish safety and dose and fortunately the early results were compelling. Unlike Nusinersen, this treatment requires a single injection into a vein.  

That class was six years ago and I presume that the brave kid who talked to us started treatment some time ago. The New England Journal of Medicine now includes an animated summary in its major papers, such as the ones cited above.  It can be found online (go to the journal’s website, www.nejm.org, and plug the title of the paper into the “search” line), or if you email me, I will send it to you. 

This advance has stimulated comment that is sometimes overwrought, using clichés like “paradigm shift,” which always makes me queasy. The treatments provide improvement, not necessarily a cure. We are a long way from victory with this or other genetic diseases, but it is a start.


Richard Kessin, PhD is Professor Emeritus of Pathology and Cell Biology at the Vegelos College of Physicians and Surgeons of Columbia University. He lives in Norfolk and can be reached at Richard.Kessin@gmail.com.