Potential anti-inflammatory mechanism of action of mesenchymal stromal cells in osteoarthritis patients results in overall improvement in pain and symptoms

We initiated the first Canadian clinical trial using autologous bone marrowderived mesenchymal stromal cells (BM-MSCs) to treat 12 patients with Kellgren-Lawrence III or IV osteoarthritis (OA) of the knee. Patients received 1, 10 or 50 million MSCs in a single intra-articular injection in the affected knee. 12 patients (mean age of 56, range 46-65); 7/12 male) received BM-MSCs without any adverse events. All patients showed improvements in diseasespecific outcomes in KOOS and Western Ontario and McMaster Universities Osteoarthritis (WOMAC) score at 12 months. Paired t-test p-values of the means at 12 months vs. baseline are as follow for KOOS: ADL (p = 0.026), pain (p = 0.002), Symptoms (p = 0.012), QOL (p = 0.019), Sports and Recreation (p = 0.053) and WOMAC Function (p < 0.001), Pain (p = 0.026), and Stiffness (p < 0.001) subscales. Individual analyses show that 8/12 patients showed an improvement in Minimal Clinically Important (MCI) change of 10 points relative to baseline based on pain and symptom indices at 6 months. The MCI analyses appear to correlate with interim MRI analysis of synovitis based on gadolinium uptake suggestive of anti-inflammatory mechanism of action of MSCs. However, full 12-month MRI analyses are pending. Analysis of the synovial fluid microenvironment at baseline and 3 months postinjection showed decreased levels of pro-inflammatory monocytes/macrophages with concomitant increase in PGE2 and decrease in IL12p40 levels, suggestive of MSC-mediated inflammation resolution. Analyses of 12 patientspecific MSCs showed elevated gene and protein expression of anti-inflammatory markers that correlated with the stratification of patients based on MCI analyses suggesting that donor heterogeneity is partially responsible for the observed clinical efficacy differences. In conclusion, autologous BM-MSCs are safe and result in significant improvements in overall KOOS and WOMAC scores at 12 months, despite donor heterogeneity. A dose of 50 x 106 MSCs may result in better outcomes based on individual patient MCI analyses of KOOS scores. For the first time, we are able to show clinically that MSC injection resulted in less inflammatory synovial environment, which correlated with patient reported outcomes and potential MRI changes in synovial inflammation.