South Korea Renews Approval for ALS Stem Cell Therapy Neuronata-R After Phase 3 Data Reveal a Responder Subgroup

WHAT THIS ARTICLE COVERS

In May 2026, South Korea's Ministry of Food and Drug Safety (MFDS) renewed and updated the conditional approval of Neuronata-R (lenzumestrocel), an autologous bone-marrow-derived mesenchymal stem cell therapy for amyotrophic lateral sclerosis (ALS), narrowing its label to reflect a "slow progressor" patient subgroup that showed a statistically significant benefit in a Phase 3 trial that otherwise missed its primary endpoint. The decision — confirmed in a July 19, 2026 announcement from developer CorestemChemon — keeps one of the world's only licensed cell therapies for ALS on the market, gives regulators and clinicians a clearer biomarker-driven use case, and sets up a push toward a U.S. FDA filing later this year. It is a genuinely mixed but meaningfully positive story: the headline trial failed, but a pre-specified-enough subgroup analysis found real signal, and regulators responded by refining rather than revoking access.

A Trial That Missed, Then a Signal That Held Up

Neuronata-R has been conditionally approved in South Korea since 2014, making it one of the first licensed stem cell therapies for ALS anywhere. To convert that conditional status into full approval, CorestemChemon ran ALSummit, a randomized, controlled Phase 3 trial (NCT04745299) enrolling roughly 115 adults with ALS diagnosed within the prior two years, split across a two-dose treatment arm, a five-dose treatment arm, and placebo. When topline results were announced in late 2024, the trial did not meet its primary endpoint — the Combined Assessment of Function and Survival (CAFS) — across the full study population. That is an important, unglamorous fact, and it is the reason this therapy remains conditionally rather than fully approved even in Korea.

What changed the picture was a post-hoc analysis restricted to "slow progressors," a subgroup identified using a validated ALS progression-rate measure. In that subgroup, both dosing regimens showed statistically significant CAFS benefit at month 6 versus placebo (least-squares mean 20.95 vs. 13.66, p<0.024 for the two-dose group; 24.78 vs. 17.92, p<0.041 for the five-dose group), along with later but significant improvements on the ALS Functional Rating Scale-Revised (ALSFRS-R) and, in the five-dose group, slower decline in slow vital capacity — a measure of respiratory function that often predicts survival in ALS. These findings were presented at the 2025 PACTALS scientific meeting.

Why Regulators Acted on a Subgroup Analysis

Post-hoc subgroup findings are ordinarily treated with real caution, and rightly so — they were not the trial's pre-specified success criterion, and cherry-picking a subgroup after a trial fails is a well-known way to manufacture false hope. What makes this case more credible than most is convergence: the slow-progressor signal appeared consistently across three related measures (CAFS, ALSFRS-R, and vital capacity), tracked with dose (the five-dose arm generally outperformed the two-dose arm), and echoed an earlier Phase 2 trial in 64 patients that had also shown slowed progression at four and six months, even though that smaller trial found no long-term survival difference. The MFDS's response was proportionate to that evidence: rather than granting a blanket full approval or revoking the drug outright, it updated Neuronata-R's label to the slow-progressor population and kept the approval conditional, meaning further evidence generation is still expected.

Safety and Real-World Experience

One thing working in Neuronata-R's favor is a genuinely long safety track record. Because it uses each patient's own bone marrow-derived cells (avoiding donor-matching and graft-versus-host concerns), CorestemChemon reports no treatment-related serious adverse events across more than 400 commercially treated patients in South Korea since 2014, plus roughly 190 additional clinical trial participants. That is a meaningfully large, multi-year real-world safety base for a cell therapy — though it does not substitute for controlled efficacy evidence, and adverse-event reporting from a single company's registry has inherent limits compared with independent post-marketing surveillance.

The Road to a U.S. Filing — and Real Caveats

CorestemChemon holds FDA (2018) and EMA (2019) orphan drug designations for the therapy and says it intends to submit a Biologics License Application to the FDA in the third quarter of 2026, pursuing an accelerated-approval pathway built around biomarker data (reductions in neurofilament light chain) rather than waiting for a confirmatory survival trial — a strategy the company has explicitly compared to the precedent set by tofersen's accelerated approval for SOD1-ALS. The company also announced it was selected for an NSF-backed regenerative-medicine manufacturing initiative, the Piedmont Triad Regenerative Medicine Innovation Engine, and plans to relocate its U.S. subsidiary to North Carolina, alongside a roughly $17.8 million capital raise to fund manufacturing scale-up.

The caveats here matter and should not be glossed over. The PACTALS subgroup data, as of this writing, has not been published in a peer-reviewed journal, which independent ALS advocacy organizations have flagged as a real limit on how confidently the results can be interpreted. Subgroup analyses, however consistent, are not the same evidentiary standard as a met primary endpoint, and the FDA has generally required more from cell and gene therapy sponsors than Korea's conditional-approval framework does — the agency's own officials have recently pushed for more rigorous, longer-term follow-up standards for cell therapies generally. Neuronata-R remains unapproved and unavailable outside South Korea, a BLA has not yet been filed, and the timeline to any possible U.S. availability is likely years away even in the optimistic accelerated-approval scenario the company is pursuing.

Bottom Line

This is not a clean, unambiguous "cure" story, and it shouldn't be reported as one — the pivotal trial for Neuronata-R failed its primary goal. But the underlying signal in slow-progressing ALS patients was consistent across multiple measures, tracked with dose, echoed earlier trial data, and was judged credible enough by an independent national regulator to justify keeping a real therapy on the market for the patients most likely to benefit, with a decade-long, large-scale safety record behind it. For a disease with almost no effective treatments, a regulator narrowing rather than pulling authorization based on real (if unpublished) subgroup data is a legitimately encouraging, if measured, development — one worth watching as CorestemChemon moves toward an FDA filing.