Cell-based versus corticosteroid injections for knee pain in osteoarthritis: a randomized phase 3 trial

Mautner K, Gottschalk M, Boden SD, et al. · Nature Medicine

At 12 months no cell injection beat the corticosteroid control or the other two on either co-primary endpoint. VAS pain change from baseline was -24.3 (BMAC), -19.4 (SVF), -20.1 (umbilical cord) and -20.9 (corticosteroid); against the control, BMAC -3.4 (P=0.19), SVF 1.5 (P=0.56), umbilical cord 0.8 (P=0.76), with KOOS pain giving the same answer and every sensitivity analysis agreeing. No group's MRI osteoarthritis score moved from baseline. No procedure-related serious adverse events occurred, though related non-serious events separated by arm: joint swelling 24.1% on umbilical cord tissue against 7.4% on corticosteroid (P=0.01), post-procedural contusion 38.6% on SVF against 12.2% on BMAC and none on the other two (P<0.0001), and hematoma 12.4% on SVF against 2.9% on BMAC (P=0.02).

Population

480 randomized (440 analyzed) adults with Kellgren-Lawrence II-IV knee OA, multicenter; titled a phase 3 trial and described in its own abstract as phase 2/3

Intervention

Single injection of autologous BMAC, autologous adipose stromal vascular fraction, or allogeneic umbilical cord tissue-derived MSCs

Comparator

Corticosteroid injection, randomized 3:1 within each of the three arms so every cell product met its own 40-patient control

Limitations

Single-blind, with an active comparator rather than placebo. Sponsored by the Marcus Foundation, which separately supports six of the authors; several authors work for Sanford Health, which holds a financial interest in InGeneron, the company whose stromal vascular fraction product this trial tested, and the first author consults for a micronized-fat company not used here.