The objective of this study was to retrospectively validate a newly designed scoring system for TEAEs of liver-directed adherent stem cell therapy after liver transplantation (Figure 2). This score will be used in an approved phase-I study (MiSOT-I). The validation was conducted retrospectively in a cohort of 187 recipients of liver allografts who did not receive investigational cell therapy.
The majority of the study population showed no TEAEs (score 0) in accordance with the MiSOT-I score. This was expected, because the cut-off values of the score parameters were chosen to be a high barrier for clearly unacceptable clinical events in the further development of this and other stem cell therapies. For all modalities, the relative frequency of severe TEAEs (score 3) did not exceed 5%. Previous studies looking at comparable complications after organ transplantation have shown rates of pulmonary embolism or acute respiratory distress syndrome of 0.37%  and 5.5% , respectively. Also, portal venous occlusion, hepatic artery thrombosis, and hepatic vein stenosis, which accounted for most of the intraportal/infusional grade 3 TEAEs in our cohort, have reported rates of up to 2.6%, 3.2%, and 1.5%, respectively [30, 31].
By contrast, anaphylactic reactions (grade 3 parameter for systemic toxicity) are extremely rare in the clinical setting of solid organ transplantation, to our knowledge only one such case has been described in the published literature . When comparing the three modalities, the highest relative frequency of a score of 3 was most often pulmonary-related (day 4 = 3.5%). This is consistent with previous studies suggesting a high rate of pulmonary complications following orthotopic liver transplantation . Systemic TEAEs were the least frequent, which can be explained by the general low incidence of postoperative anaphylaxis . Thus, in view of the grade 3 events in our cohort, the results of this study confirm and further quantify the findings in the literature concerning pulmonary, hepatic, and systemic function after deceased-donor liver transplantation.
A further subgroup analysis for patients with grade 3 events failed to provide a valuable hypothesis on which descriptors are associated with severe TEAEs (Table 3). Previous investigations have shown that patients with alcoholic cirrhosis achieve the same postoperative survival and complication rates as non-alcohol-related transplantations . However, for all remaining parameters, numerous studies show that high age , male gender , non-A blood type of the recipient , low donor creatinine or bilirubine , a long cold ischemia time , and a high MELD score  all significantly correlate with an increased postoperative morbidity and mortality rate after liver transplantation. Hence, a correlation between grade 3 TEAEs and any of these parameters was expected but was not established in our cohort. A possible explanation for this discrepancy is the difference in size of the two compared subgroups (n=12 vs. n=175), although this was considered in the design of the statistical analysis. Another reason for this observation may be the choice of exclusion criteria (Figure 1), since most previous comparative studies included re-transplanted patients and patients above 65 years of age.
Biopsy-proven rejection-free survival after non-living related orthotopic liver transplantation was analyzed. Here, we grouped patients receiving CNI-free, sirolimus-free, bottom-up immunosuppression (as will be used in the MISOT-I study) against all other patients . Patients who received ciclosporin A, tacrolimus, or sirolimus as their primary immunosuppression presented with a rejection-free graft survival rate of 72.9% after 365 days of follow-up (mean follow-up = 206.59 days). The non-CNI, non-sirolimus, bottom-up group presented with rejection-free survival of 57.1% in comparison. Since time to biopsy-proven acute rejection will be a secondary endpoint of the MiSOT-I trial, the analysis of the rejection time in the present cohort can serve as a retrospective comparator, naturally with all bias and shortcomings of such an analysis.
Unavailability of data and low data consistency over the analyzed time period was the key limitation of this retrospective study. However, in the light of no other available data, this study will still be the most valuable comparator for MiSOT-I and other investigational phase I studies applying adherent stem cell therapies. No high-risk patients for the elected events could be identified from our present cohort and, therefore, we have no further means to exclude patient groups from MiSOT-I.