Allogeneic Mesenchymal Stem Cells Can Mitigate Early, Ongoing, and Severe Graft Versus Host Disease but Require Delayed Administration for Optimal Effect.

Mesecnhymal stem cells have been observed to inhibit graft versus host disease clinically, however the timing of infusion of these cells has not been well characterized. In previous studies, we have observed MSC to rescue lethally irradiated hosts that had received sub-optimal numbers of stem cells, permit the reduction of host conditioning while establishing equal or better levels of engraftment than the combination of intensive host conditioning and untreated HSC grafts, and enable xenogeneic engraftment (rat→ mouse) suggesting that administration of MSC in combination with an allogeneic transplant significantly alters host immune responses to enhance engraftment.. These findings could only be observed if MSC were given on the same day as the bone marrow stem cells. The purpose of this study was to determine to what extent MSC might affect donor immune responses involved in GVHD and to determine the optimal timing of these effects, in order to optimize the maximal beneficial effects of allogeneic stem cell grafts engineered with MSC. Since GVHD, mediated by donor T cells, requires host antigen presentation for initiation, we tested whether the effect of MSC occurred before or after this interaction. We used an established GVHD model in which 20x 10^6 Balb/c bone marrow cells in combination with 15 x 10^6 Balb/c splenocytes were administered to lethally irradiated B6 recipients to test whether MSC (1.0 x 10⁵) could inhibit initiation of GVHD and to what extent these cells could mitigate or abrogate ongoing GVHD.

In control animals, we observed donor T cell expansion to occur in the absence of B6 host T cells with corresponding destructive effects resulting in 100% lethality by day 48. Four experimental groups (n=10 per group) were used to test MSC administration at 4 time points: 1) on day 0 following co-culture with the graft to test whether cell contact between MSC and GVHD-producing splenocytes is necessary, 2) on day 2 to test whether donor T cell exposure to host antigen is required, 3) on day 20, to test the magnitude of effect of MSC on ongoing GVHD, and 4) on day 30 in which GVHD is severe and usually irreversible. Mice were weighed twice weekly and monitored daily for survival and clinical evidence of GVHD (ruffled fur, cachexia, alopecia, and diarrhea). When compared to survival of control animals, no statistically significant effect was observed when MSC were given with the stem cell grafts on day 0. Strikingly, survival was significantly increased to 60% when given on day 2 (p=0.01, log rank test), to 50% when given on day 20 (p=0.005), and to 40% for day 30 treated animals (p=0.009). Following MSC infusion, those animals that developed signs of GVHD such as ruffled fur and alopecia had dramatic improvement of these physical findings with most surviving animals experiencing a complete reversal to normal appearing fur.

The observation that no effect occurred with MSC administered at the time of bone marrow transplantation suggests that the mechanism of effect requires host antigen presentation. We conclude that optimal timing for the infusion of donor specific MSC to abrogate GVHD begins after donor T cells have encountered host antigen and can be equally effective during early, late, and severe GVHD. Clinical strategies involving the use of allogeneic stem cell grafts engineered with MSC are likely to be more powerful in overcoming GVHD if the MSC infusion is administered in a delayed fashion.