Stem Cell Shot OSCA Shows Early Knee Cartilage Repair

For people living with knee osteoarthritis, today’s treatments mostly manage pain until the joint is worn out enough to replace. A new cell therapy presented at the American College of Rheumatology’s ACR Convergence 2026 in Orlando suggests that may not always have to be the endgame. In a 113-patient phase 2a trial, a single knee injection called OSCA roughly halved pain over six months and, on MRI, showed early signs of cartilage regrowth (MedPage Today).

MedPage Today’s headline put it boldly: the therapy “reverses damage.” The data are genuinely encouraging. They are also early, small and industry-funded, so it’s worth looking closely at what was shown and what wasn’t.

What OSCA is

OSCA, developed by South Korea’s Kangstem Biotech, combines two components in one injection (ACR abstract LB19):

Osiramestrocel (“OS”): donor mesenchymal stem cells derived from umbilical cord blood.
Cartilage acellular matrix (“CA”): a cartilage scaffold stripped of its cells, which MedPage Today reports is taken from pigs.

The idea is to pair fast symptom relief with tissue repair. Presenter Jae-Young Park, MD, of CHA University in Korea, described a proposed mechanism that targets not just the joint lining (synovium) but also the bone beneath the cartilage and pain-sensing nerves, aiming for less inflammation, less pain and structural improvement (MedPage Today).

How the trial worked

The multicenter, randomized, placebo-controlled study enrolled 113 Korean patients with moderate knee OA. Each received a single injection of placebo (37 patients), a mid dose of 50 million cells (38) or a high dose of 100 million cells (38). Researchers tracked pain and function monthly for six months, scored MRI scans at month 6 and measured 11 inflammation and cartilage-turnover biomarkers (ACR abstract).

Two details matter for interpretation. Safety was technically the primary focus, with efficacy outcomes exploratory. And patients averaged about 62 years old with a body mass index around 25, leaner than many US patients with knee OA, which MedPage Today notes may limit how well the results generalize (MedPage Today).

The results: pain, function and the MRI signal

Pain and function. At six months, pain on a 100-mm scale fell by 44.4 points (about 61%) with the mid dose and 37.7 points (53%) with the high dose, versus 14.9 points (23%) with placebo, all highly significant. Knee function on the IKDC scale improved by 22.0 and 19.3 points versus 4.1 for placebo. WOMAC and KOOS scores also improved significantly with both doses (ACR abstract).

Responders. About 78% of mid-dose and 77% of high-dose patients met OMERACT-OARSI responder criteria, versus 39% on placebo, and Park said the gap was visible just four weeks after the injection (MedPage Today).

Structure. This is the headline-grabbing part, and the most nuanced. The MOCART cartilage score rose 2.7 points with the mid dose and 1.6 with the high dose, while it fell 0.9 with placebo. Only the mid-dose gain was statistically significant. The overall WORMS whole-joint score did not differ from placebo, though a subregion analysis found significantly improved bone-marrow abnormalities in one compartment with the high dose (Healio).

Biomarkers. The high dose significantly lowered inflammatory markers TNF-α and IL-1β and cartilage-breakdown markers COMP and urinary CTX-II compared with placebo (Rheumatology Advisor).

Safety

Adverse drug reactions occurred in 3 mid-dose patients (7.9%) and 8 high-dose patients (21.1%), versus none on placebo, a statistically significant difference. MedPage Today reports these were injection-site reactions. No one discontinued because of them, and there were no treatment-related serious adverse events (ACR abstract; MedPage Today).

Why it matters, and what to watch

No approved therapy changes the course of knee OA. “Patients with moderate knee osteoarthritis have few options,” Park told Healio, noting they are often too early for joint replacement while current treatments only ease symptoms (Healio). A one-shot treatment that relieves pain and nudges tissue toward repair would be a real shift.

Key open questions:

Durability. Patients will be followed for another six months plus a five-year extension, which should show whether the benefits last (MedPage Today).
Repeat dosing and patient range. It’s unknown whether repeat injections are needed, or how the therapy performs in milder or more severe disease.
Dose puzzle. The lower dose did slightly better on pain and cartilage scores, while the higher dose did better on bone marrow and biomarkers.
Independence and scale. All authors report ties to Kangstem Biotech, and larger phase 2b and 3 trials are still needed. Kangstem has said it is seeking global pharma partners, and MedPage Today notes the therapy “won’t be cheap.”

Key takeaways

OSCA pairs umbilical cord blood stem cells with an acellular cartilage scaffold in a single knee injection.
In 113 patients, it roughly halved pain at six months, with responder rates near 78% versus 39% on placebo.
MRI showed a modest, significant cartilage-score gain with the mid dose, an early structural signal rather than proof of reversal.
Side effects were mostly injection-site reactions, with no treatment-related serious events.
Longer follow-up and larger, independent trials will decide whether this becomes the first disease-modifying knee OA therapy.

Conclusion

OSCA’s phase 2a results are among the most promising regenerative signals knee osteoarthritis has seen. They don’t yet prove that damaged knees can be rebuilt, but they give researchers a credible reason, and a clearer design, to test that idea at scale.

This article reports on early-stage research and is not medical advice. OSCA is an investigational therapy and is not approved for use.