BACKGROUND: We recently demonstrated that intracellular xenogen-contaminated autologous MSCs (FBS) and non-xenogen-contaminated allogeneic (ALLO) MSCs caused an adverse clinical response after repeated intra-articular injection in horses, whereas autologous (AUTO) MSCs did not. Our current objective was to use clinical data from the previous study to compare MSC stemness against adverse response indicated by synovial total nucleated cell count (TNCC) following intra-articular MSC injection. METHODS: Stemness, quantified by a trilineage differentiation (TLD) score; immunomodulation, quantified by mixed lymphocyte reactions (MLRs); and degree of MHCI expression, quantified by mean fluorescent intensity (MFI); were correlated to the synovial TNCC 24 h after naïve and primed injection. RESULTS: There was a trend of a negative correlation (p = 0.21, r = - 0.44) between TLD score and TNCC after primed injection in the ALLO group. Within the ALLO group only, there was a significant positive correlation (p = 0.05, r = 0.77) between MHCI MFI and TNCC after naïve injection and a trend (p = 0.16, r = 0.49) of a positive association of MHCI MFI to TNCC after primed injection. Within the FBS group only, there was a positive correlation (p = 0.04, r = 1) between TNCC and lymphocyte proliferation after both injections. CONCLUSIONS: The trend of a negative correlation of TLD score and TNCC in the ALLO, but not the FBS group, together with the association of MHCI expression and TNCC in the ALLO group, indicates that improved stemness is associated with reduced MSC immunogenicity. When inflammation was incited by xenogen, there was a strong correlation of lymphocyte activation in vitro to adverse response in vivo, confirming that MLRs in vitro reflect MSC immunomodulatory activity in vivo. The relationship of stemness in vitro, suppression of lymphocyte activation in vitro, MHCI expression in vitro, and clinical response in vivo should be further investigated.