Objective: To summarize the evidence on immunoablative therapy followed by autologous hematopoietic stem cell transplantation (aHSCT) to manage severe and treatment-refractory multiple sclerosis (MS).

Methods: We collected all the published studies of aHSCT in any form of MS from 1995 to 2016, carefully excluding reports that were updated in subsequent studies. Endpoints were transplant-related mortality (TRM), rate of disease progression, and no evidence of disease activity (NEDA) status. A weighted metaregression based on a Poisson model was run, assessing whether there were study-specific characteristics with an effect on TRM and progression.

Results: Fifteen studies including 764 transplanted patients were pooled in the meta-analysis. The pooled estimate of TRM was 2.1% (95% confidence interval [CI] 1.3%–3.4%). TRM was higher in older studies (p = 0.014) and in studies with a lower proportion of patients with relapsing-remitting MS (RRMS) (p = 0.028). A higher baseline Expanded Disability Status Scale (p = 0.013) was also significantly associated with a higher TRM. Pooled rate of progression was 17.1% at 2 years (95% CI 9.7%–24.5%) and 23.3% (95% CI 16.3%–31.8%) at 5 years. Lower 2-year progression rate was significantly associated with higher proportions of patients with RRMS (p = 0.004). The pooled proportion of NEDA patients at 2 years was 83% (range 70%–92%) and at 5 years was 67% (range 59%–70%).

Conclusions: The emerging evidence on this therapeutic approach in MS indicates that the largest benefit/risk profile form this therapeutic approach can be obtained in patients with aggressive MS with a relapsing-remitting course and who have not yet accumulated a high level of disability.