Risdiplam Raising SMN Levels in Older Patients in Ways That Seem Durable, Researcher Says of Early Trial Data

Treating teenagers and adults with oral risdiplam, a potential therapy for spinal muscular atrophy (SMA), led to sustained increases in blood levels of the key SMN protein missing in these patients, the lead investigator for a Phase 2 clinical trial said in an interview.

Claudia A. Chiriboga, MD, also said these early results in the first 12 SMA patients enrolled in the open-label JEWELFISH study (NCT03032172) showed risdiplam to be well-tolerated. JEWELFISH is now enrolling up to 180 infants through adults with types 1, 2, and 3 disease at sites in the U.S., U.K., Switzerland, and Italy; more information is available here.

In an interview with SMA News Today, Chiriboga, a pediatric neurologist and a professor at Columbia University Irving Medical Center (CUIMC), discussed these early findings, while also sharing her thoughts on possible future trial outcomes, what makes risdiplam stand apart, and who might benefit from combination treatment.

Data regarding risdiplam’s clinical effectiveness in these patients is not yet available.

SMA is caused by a faulty SMN1 gene, which reduces the levels of the SMN protein in motor nerve cells and leads to cell loss. Similar to Spinraza (nusinersen, by Biogen), the first approved SMA therapy, risdiplam is designed to boost the ability of an alternative gene — SMN2 — to produce full-length and functional SMN, unlike the shorter and unstable protein it normally produces.

JEWELFISH, as of Dec. 1, 2018, had six patients with type 2 and six with type 3 disease, two SMA forms whose symptoms typically become evident in the first years of life. Their median age was 20, ranging from 13 to 52 years old. “So many of those patients are quite advanced,” Chiriboga said.

After a median 14 months of treatment, data showed that SMN protein levels in the blood peaked within four weeks and were maintained at about two times the levels recorded at baseline or the study’s start.

Chiriboga, who is following JEWELFISH participants being treated at CUIMC, noted that a similar profile has been seen in other ongoing Phase 2/3 trials of risdiplam, namely FIREFISH (NCT02913482) in infants with type 1 SMA and SUNFISH (NCT02908685) in patients ages 2–25 with types 2 or 3.

Whether these SMN increases lead to significant improvements has not yet been analyzed, but Chiriboga mentioned preclinical animal data supporting such a link, with protein levels in the blood correlating with those in the brain, and risdiplam efficiently reaching the central nervous system (spinal cord and brain).

“I’m dying to see those data because I’m sure it’s going to show an association,” she said.

Safety, like efficacy, is also yet to be assessed, and enrolling more patients may enable a comparison between SMA types, Chiriboga added.

Existing data show that risdiplam, a liquid given orally once in the morning, is well-tolerated and “the children don’t seem to mind it.” JEWELFISH’s dose was based on determinations made in the first part of SUNFISH, which showed motor improvements regardless of disease severity or age at treatment start.

Its oral delivery means that risdiplam is distributed throughout the body, setting it apart from the specific effect on motor neurons of Zolgensma (onasemnogene abeparvovec-xioi, by Novartis and AveXis), and from Spinraza’s intrathecal (spinal canal) infusion. The benefits of systemic distribution are “probably most obvious in the younger patients … who have a lower level of SMN protein to the point that it can affect other organs much more strongly,” Chiriboga said.

Recent results from FIREFISH appear to support this, with most of its 21 infants showing meaningful motor benefits at 12 months of treatment, including seven (41.2%) in the proposed therapeutic dose group able to sit without support for at least five seconds.

Considering the rapid progression in babies with this most severe form of SMA and that treatment started at a relatively late age (a median close to 7 months old), these benefits are even more impressive, Chiriboga said.

“To see the magnitude of response was really surprising to me because we know that the later we treat them, the worse the response is,” she said. “I am persuaded myself … that [systemic delivery] is one of the reasons why [risdiplam] works so well in these infants.”

Preferences and potential combinations

Risdiplam — developed  by Roche and Genentech in collaboration with PTC Therapeutics and the SMA Foundation — is an optimized version of RG7800, tested in the Phase 1 MOONFISH trial (NCT02240355) that was stopped early. Compared with RG7800, risdiplam is “more specific, with better physical chemical properties and less off-target effects … possibly a better molecule,” Chiriboga said.

When first designed, JEWELFISH mainly intended to treat the nine MOONFISH patients and those given Spinraza before its U.S. approval, all between 12 and 60 years of age. The study’s goal, Chiriboga said, was to test if risdiplam worked in a different way and to see if having received another SMN2-targeting therapy would make any difference in terms of safety and tolerability.

Researchers later decided to broaden its target population to include younger patients (starting at 6 months old) and people given Spinraza commercially. Participants in the OLEOS Phase 2 trial (NCT02628743) of Roche’s olexosime — stopped after disappointing 18-month results — were also invited to enroll.

With Zolgensma’s successful development — and its May approval in the U.S. — the trial expanded again to include people on this gene therapy. Of note, none of these patients had enrolled in JEWELFISH as of July 9.

Zolgensma is intended to be a one-time treatment, providing a normal SMN1 gene and a lifelong production of SMN in motor neurons. Including these patients in JEWELFISH will enable investigators to judge whether a combination with risdiplam may be beneficial.

Compared with the rapid improvements seen in babies treated with Zolgensma presymptomatically in the SPR1NT Phase 3 trial (NCT03505099, still enrolling) or early in their disease process in its pivotal study (NCT02122952), Chiriboga thinks the gene therapy’s impact on people with more advanced SMA — being evaluated in the STRONG Phase 1 trial (NCT03381729) — may not be as clear. Disease progression brings extensive motor neuron loss.

So two therapies that raise protein levels by different routes “makes a lot of sense” in these patients, she said. This expectation might soon be tested. “A lot of patients who have received [Zolgensma] either through the [clinical] studies or through expanded access programs are asking to receive risdiplam. And we’re asking for more slots so that we can dose more patients.”

But greater benefits with earlier treatment are widely demonstrated, with most babies given Spinraza presymptomatically in the NURTURE Phase 2 study (NCT02386553) able to walk unassisted within a normal time frame. Likewise, symptomatic patients in the SHINE Phase 3 trial of Spinraza (NCT02594124) and in the SUNFISH study of risdiplam also show more significant improvements at younger ages.

“I think it goes with all SMA treatments,” Chiriboga said. “The earlier you are in your disease, the better the efficacy.”

JEWELFISH will not test risdiplam in combination with Spinraza as it will with Zolgensma, largely because of the lasting nature of a gene therapy. Patients using Spinraza need to stop the therapy and remain untreated for 90 days before starting a risdiplam screening process, Chiriboga said. The two treatments’ similar mechanism of action, including their work “in a very close location” of SMN2, is the reason for not testing the combination.

“I see it [risdiplam] more as a replacement therapy for nusinersen [Spinraza], because they’re just too similar in mode of action [and] it’s unclear what the interaction with it would be,” she said. Having alternative SMN2-targeting medicines may be particularly appealing to patients not eligible for Zolgensma, she added. “That’s a strong choice in preference, and oftentimes the mode of delivery plays a role there.”

Possible treatment choices in SMA, she believes, likely will depend greatly on when they are started, because “there’s maybe not a big difference” across therapies in terms of effectiveness “if you treat very, very early, before disease onset.” Of note, the global RAINBOWFISH trial (NCT03779334, now recruiting) is testing risdiplam in presymptomatic SMA babies up to 6 weeks old.

In contrast, combinations in more advanced SMA — given the lower likelihood of a gene therapy enabling full recovery — “may be advisable.” Such combinations may also extent to investigative treatments intending to boost muscle function, Chiriboga said.

“I think it’s a very individualized approach,” she said, adding that patient and family expectations are also key. “I think people should be free to decide if something isn’t working for them, to try another medication if it works better.”