Adipose-derived mesenchymal stem cells (AdMSC) for the treatment of secondary-progressive multiple sclerosis: A triple blinded, placebo controlled, randomized phase I/II safety and feasibility study

Abstract

BackgroundCurrently available treatments for secondary progressive multiple sclerosis(SPMS) have limited efficacy and/or safety concerns. Adipose-mesenchymal derived stem cells(AdMSCs) represent a promising option and can be readily obtained using minimally invasive procedures.Patients and methodsIn this triple-blind, placebo-controlled study, cell samples were obtained from consenting patients by lipectomy and subsequently expanded. Patients were randomized to a single infusion of placebo, low-dose(1x106cells/kg) or high-dose(4x106cells/kg) autologous AdMSC product and followed for 12 months. Safety was monitored recording adverse events, laboratory parameters, vital signs and spirometry. Expanded disability status score (EDSS), magnetic-resonance-imaging, and other measures of possible treatment effects were also recorded.ResultsThirty-four patients underwent lipectomy for AdMSCs collection, were randomized and thirty were infused (11 placebo, 10 low-dose and 9 high-dose); 4 randomized patients were not infused because of karyotype abnormalities in the cell product. Only one serious adverse event was observed in the treatment arms (urinary infection, considered not related to study treatment). No other safety parameters showed changes. Measures of treatment effect showed an inconclusive trend of efficacy.ConclusionInfusion of autologous AdMSCs is safe and feasible in patients with SPMS. Larger studies and probably treatment at earlier phases would be needed to investigate the potential therapeutic benefit of this technique.

Sample size re‐estimation for clinical trials with longitudinal negative binomial counts including time trends

“…Because the longitudinal gamma frailty model, which the sample size re‐estimation procedure depends on, cannot be fitted using standard software, the presented inference, sample size estimation, and BSSR methods have been coded in R and made available on CRAN within the package spass . A detailed example using data from Fernández et al is given in the Appendix . Both variance estimators considered are implemented, namely, using the observed sandwich estimator as well as the restricted sandwich estimator calculated under the null hypothesis.…”

“…For illustration, we consider data from Fernández et al, 33 available online as supplementary material to their publication. In this MS trial, active T1 lesions were measured at three time points (at baseline, 6 months after randomization, 12 months after randomization) on patients who were administered different doses of autologous Adipose-derived mesenchymal stem cells (AdMSC).…”

“…We now demonstrate the full procedure of sample size estimation and blinded sample size reestimation using the same data from Fernández et al 33 as in the application from Section 5.4, ie, comparing the low-dose treatment group and the placebo group.…”

“…Let us assume that an MS trial is being planned and initial data, in this case, consisting of the placebo and low-dose group from Fernández et al 33 has been collected in a pilot study. To assess the required parameters for sample size estimation, assuming an underlying exponential trend, the data can be fitted using the function fit.nb.gf() as follows.…”

“…At the end of the clinical trial, when the groups are unblinded, the function test.nb.gf() can be used for testing treatment effects. Again, we refer to the data set from Fernández et al 33 and test for treatment differences between the placebo and the low dose strata. Then, the null hypothesis H 0 ∶ 3 ≥ 0 can be tested as follows.…”