We evaluated the safety and efficacy of a single allogenic intravenous ASC infusion and daily oral VIT D in recent onset patients with T1D during 12 months. The procedure was safe and we observed a potential benefit in glycemic control, insulin dose and a temporary stability of pancreatic function until 6 months. This pilot study was the first to evaluate long term results of a single ASC infusion + VIT D in patients with recent onset T1D. Allogenic source of cells was chosen due to the possibility of impairment of mesenchymal stromal/stem cells immune properties in individuals with T1D (32, 33).
Acute safety of ASC infusion and 3 months adverse events have been published previously (31). Nine months after the intervention, one patient had a recurrence of a benign ovarian teratoma with a complete surgical removal. ASC has a differentiation potential capacity to mesodermal and other embryonic lineages, including adipocytes, osteocytes, chondrocytes, hepatocytes, neurons, muscle cells and epithelial cells, depending on the surrounding microenvironment (34, 35), however previous in vitro and in vivo studies showed the safety of intravenous infusion of MSC, including ASC (30, 36, 37). No tumor differentiation occurred in those studies (32–35). Mature benign teratoma is a germ line tumor and is the most frequent non-epithelial ovarian tumor with a chance of recurrence of 3–15% (38). This may be due to incomplete previous resection or missed small mature ovarian cysts at the time of removal rather than a recurrence (38)
Patients that received the ASC infusion + VIT D had a better glycemic control after 12 months. The treatment was also associated with a better β cell function during the first 6 months after infusion. Beneficial effects from MSC infusion in recent diagnosed T1D patients have been published previously. Carlsson et al and Hu et al both observed a better CP, lower HbA1c and lower insulin dose/kg 12 months after autologous bone marrow MSC infusion and 21 months after Wharton Jelly umbilical cord mesenchymal stroma cell (WJ-UC MSC) infusion, respectively, in patients with recent onset T1D (39, 40). Even though HbA1c and insulin dose/kg were lower in the intervention group for 12 months, better pancreatic function was observed only until 6 months, which may indicate a temporary effect of the intervention. Therefore, additional ASC infusions might be necessary to maintain that benefit, as seen with other autoimmune disease, such as multiple sclerosis (41). Moreover, other immunotherapies with ciclosporin, rituximab and humanized Fc-mutated anti-CD3 monoclonal antibody hOKT3 also had temporary effects (42–46). In previous studies with MSC, beneficial effects were sustained until 12–24 months, but none of those used ASC. Different mesenchymal stem cell sources could have distinct immunomodulatory and regenerative properties, contributing to different results in clinical trial (47, 48). In vitro, ASC present a more pronounced cytokine secretion than MSC from other sources, however this was not confirmed in vivo (47–49). The number of cells infused also might have contributed to the transient effect. In a recent metanalysis, the infusion of > 1 x 107 cells were associated to better outcomes (17), and we used 67.71x106. It is also possible that the slight difference is due to the lack of immunosuppression (50). Couri and cols evaluated de pancreatic function after autologous nonmyeloablative hematopoietic stem cell transplantation in 23 patients, after a median follow up of 30 months, 12 patients remained insulin-free with an increase of the CP AUC (50). Other authors confirmed the potential benefit of this therapy in inducing partial CR in patients with T1D, however these better results might have been secondary to the immunosuppressants that were used rather than the stem cell therapy itself (50, 51).
A higher pancreatic function is associated with less glycemic variability and severe hypoglycemia (2, 3). This difference could not be observed in our study since none of the patients had severe hypoglycemic episodes.
A significant number of patients were in partial CR in the intervention group, 85.7% and 57.17% in 6 and 12 months respectively. In other observational studies, this frequency varied from 0 to 20% in 6 months to 0 to 10% in 12 months (30). Patients that underwent partial CR presented a lower frequency of chronic microvascular complications after 7 years of follow up (52).
Some potential temporary immunomodulatory effect was observed after ASC infusion + VIT D in FOX P3 expression in CD8+ lymphocytes 1 month after intervention, as published previously by our group (31). However, this was not sustained after 3, 6 and 12 months. Immunoregulatory CD8+ T cells have been reported before (53) and were also associated with clinical response in a T1D trial with humanized Fc-mutated anti-CD3 monoclonal antibody hOKT3 (54). CD4+CD25+FOX P3 T cells are also critical for controlling autoimmunity and tolerance, this defective Treg cells has been associated to many autoimmune diseases, including T1D (55). Patients that were in partial CR after 6 and 12 months had a higher FOX P3 expression in CD4+ lymphocytes, suggesting that the immune activation might have had a role in the ASC + VIT D therapy in patients with a better response. This immune activation has been published previously by Haller et cols after umbilical cord MSC infusion in children with T1D, but was not associated with a better pancreatic function (56). In addition, after immunotherapy with teplizumab and otelixizumab, T1D patients had a transient modification of CD4+ lymphocyte expression profile (57, 58). Since our intervention group was treated with VIT D + ASC, we cannot confirm that this immunologic effect was solely by the ASC cells, by VIT D supplementation or both. We have previously reported a comparison between our patients and individuals that used only VIT D supplementation for 6 months and took part in a previous study. The CP AUC could not be compared since this group had only basal e peak CP, but ASC + VIT D had an increase in basal CP and a better HbA1c after 6 months when compared to VIT D supplementation group and controls (59). VIT D can act in lymphocyte proliferation, cellular autoimmune pathways and stimulated T regulatory response (19, 20). In this study, VIT D levels had a direct correlation with CP at T6, and, in those with CR after 6 months, vitamin D levels were higher. Previous clinical trials with vitamin D supplementation were inconsistent and a metanalysis has shown a small effect of VIT D on glycemic control in patients with recent-onset T1D (21–25).
Our study has some limitations. Firstly, as in most pilot studies, there was a limited size sample, which could explain the lack of statistical significance observed in some comparisons. Even though CP AUC was statistically similar between groups, at T0 2 patients in group 1 had a significant higher CP, which occurred by chance. Individual characteristics could had influenced this better pancreatic function. Moreover, there was not a group of patients that received intervention solely with VIT D, without ASC. Therefore, it is not possible to determine the exact beneficial effect of each intervention. An additional limitation is the location/route of administration of the ASC. In vivo studies have shown that after peripheral venous infusion, MSC tends to migrate to inflamed tissues (60, 61, 62), however some authors reported that only a small proportion of MSC reach the main target, while most migrate to the lungs and later distributed throughout the body (61, 62). Direct infusion in the peripancreatic region would be more effective and with longer lasting results (63). Finally, a longer follow up is still necessary to understand the sustained benefits in glycemic control and CP secretion. On the other hand, this pilot study has shown safety from a single ASC infusion with VIT D supplementation after 12 months, with a transient benefit in pancreatic function and glycemic control. A larger sample and a longer follow up is still necessary to understand the real benefit of this combined therapy. Other infusions could be necessary to maintain CP secretion after 6 months, since better results were observed util that time point.