Clinical Safety and Effectiveness of Adipose-Derived Stromal Cell vs Stromal Vascular Fraction Injection for Treatment of Knee Osteoarthritis: 2-Year Results of Parallel Single-Arm Trials

Background: There are currently no disease-modifying treatments available for knee osteoarthritis (OA), although cultured adipose-derived stromal cells (ASCs) have shown promise in experimental models. However, given the regulatory limits on the use of cultured cells in humans, previous trials have focused primarily on the stromal vascular fraction (SVF) intra-articular injection. Therefore, the therapeutic value of ASCs for knee OA remains unknown. Purpose: To study ASC versus SVF intra-articular injection in patients with Kellgren-Lawrence (KL) knee OA grades 2 to 4 in parallel single-arm trials. Study Design: Cohort study; Level of evidence, 2. Methods: A total of 80 patients were enrolled, with 42 (72 knees) receiving ASC intra-articular injection and 38 (69 knees) receiving SVF. Patient-reported outcome measures were assessed at 1, 3, 6, 12, and 24 months using the Knee injury and Osteoarthritis Outcome Score 5 (KOOS5) and pain visual analog scale (VAS). The percentages of patients achieving the minimal clinically important difference (MCID) and Patient Acceptable Symptom State (PASS) were also calculated. Per protocol, a subset of the ASC group received an ASC booster injection after 6 months. A repeated-measures analysis of variance compared results between treatment arms and by KL grade over time. Results: Patient-reported outcome measures improved substantially after both treatments ( P < .05 at all time points), with the ASC group more likely to achieve the MCID (50% vs 24%; P = .01) and PASS (45% vs 24%; P = .04) for the pain VAS and the MCID (43% vs 16%; P = .02) for the KOOS5 at 12 months, although not at 24 months. Knees treated with ASC for KL grade 2/3 OA had significantly superior outcomes compared with those with KL grade 4 OA for the KOOS5 ( P = .01) and pain VAS ( P = .03), but no such difference was observed in knees treated with SVF. Three patients receiving ASCs (7%; all KL grade 3) sought additional nonoperative treatment by 24 months versus 9 patients receiving SVF (24%; all KL grade 3) ( P = .06). ASC booster injections conferred no additional benefit. Notably, patients in the ASC cohort reported more injection-site pain and swelling after the booster injection than after the initial injection ( P < .01). Conclusion: This represents the first head-to-head comparison of ASCs and SVF for the treatment of knee OA in humans. ASC and SVF injections both substantially improved knee pain and function at all follow-up time points, although ASC injections demonstrated significantly better improvements with regard to the MCID and PASS for the pain VAS and the MCID for the KOOS5 at 12 months. There appears to be no benefit to a booster ASC injection after initial treatment. Given less donor-site morbidity and equivalent superior outcomes at 2 years, the use of ASCs over SVF in the treatment of knee OA may be warranted.