FormalPara Key Summary Points

Current data encourage a broader application of autologous hematopoietic stem-cell transplantation (AHSCT) for treating patients with multiple sclerosis (MS).

Our analysis showed a significant decrease in the Expanded Disability Status Scale (EDSS) score and annualized relapse rate after treatment compared with the pretreatment period.

Our meta-analysis showed that 68% of patients with MS experience no evidence of disease activity (NEDA) after AHSCT.


Multiple sclerosis (MS) is characterized by chronic inflammation, neurodegeneration, and immune-mediated responses of the central nervous system (CNS), leading to demyelination, gliosis, and axonal damage [1, 2]. MS can cause permanent disability, reduce the quality of life, and shorten life span. Over the past two decades, disease-modifying therapies (DMTs) have been developed and approved, all of which have different efficacy and safety profiles. There have been considerable benefits for patients with relapsing–remitting MS (RRMS), as well as reduced clinical relapse. Although DMTs had a marginal effect on disability progression in RRMS, they failed to achieve an acceptable outcome in other subtypes of MS, such as progressive and treatment-refractory types [3,4,5].

In 1995, the use of autologous hematopoietic stem-cell transplantation (AHSCT), previously approved to treat hematological tumors, was introduced for severe autoimmune diseases [6, 7]. AHSCT is designed to remove the impaired immune system and then regenerate new immune cells to prevent the recurrence of neuroinflammatory symptoms [8, 9]. Previous studies have demonstrated the benefits of AHSCT in providing longer-term remission than conventional therapies. Also, the effectiveness and safety of this treatment approach were reported in autoimmune disease, especially in patients with MS who had not responded to DMTs [10, 11]. A retrospective cohort study on 120 patients with MS treated with AHSCT demonstrated a significantly decreased relapse rate at 2 and 4 years of follow-up, as well as a decrease in magnetic resonance imaging (MRI) T2 lesions. The study reported that 93% of patients were relapse-free at 2 years and 87% at 4 years. Based on the findings of this study, AHSCT was capable of preventing an increase in the Expanded Disability Status Scale (EDSS) scores [12]. Another study in patients with RRMS reported that five out of ten cases had complete remission after AHSCT at the end of the 10 years of follow-up. Also, three cases demonstrated improvement, so there is the possibility of complete remission after AHSCT [13]. Burt et al.’s study with a sample population of around 500 reported that AHSCT was a beneficial one-time treatment for RRMS. In contrast, their results showed less effectiveness of AHSCT in newly diagnosed secondary progressive MS [14]. Nowadays, AHSCT is recognized as a rapid treatment for relapsing or progressive multiple sclerosis. As a result, the National Multiple Sclerosis Society has acknowledged AHSCT as a feasible treatment option for patients with MS with high disease activity, as evidenced by relapse rates and new MRI lesions, despite the use of second-line DMTs, or in those with contraindications to conventional treatments. Indeed, patients under 50 years of age whose disease duration is less than 10 years are the best candidates for AHSCT [15]. A previous systematic review and meta-analysis demonstrated that progression-free survival after AHSCT in patients with MS was 75%, and estimated disease activity-free survival was 61% after 48 months [16].

As MS is generally not a life-threatening disease, concerns over mortality rates have previously restricted AHSCT application to treat MS. However, AHSCT has proven its safety over the past few years due to technical advances and careful patient selection in transplant centers. Thus, studies have reported that AHSCT led to decreased EDSS scores in most cases, although some patients had increased EDSS scores following the procedure [17,18,19]. In light of these contradictory results, we aimed to conduct a comprehensive systematic review and meta-analysis to investigate the efficacy and safety of AHSCT.


We conducted this systematic review and meta-analysis following the Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) checklist [20]. This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors.

Search Strategy

We performed a comprehensive literature search on PubMed, Scopus, and Web of Science in February 2022. The following terms were used in our search strategy: “Multiple Sclerosis” OR “Sclerosis, Multiple” OR “Sclerosis, Disseminated” OR “Disseminated Sclerosis” AND “autologous hematopoietic stem cell transplantation” OR “AHSCT” OR “stem cell.” A manual search of the reference lists of previous review studies was also performed to identify additional articles.

Eligibility Criteria

We included cohort studies, clinical trials, case–control studies, and case series that investigated the efficacy or safety of AHSCT in patients with MS. Conference abstracts that were indexed in PubMed, Scopus, or Web of Science were also screened. The studies that investigated other types of stem-cell therapy such as progenitor cells, embryonic stem cells, or programmed stem cells were excluded. Also, case reports and non-English studies were excluded. PICO in the present study was defined as follows. Problem or study population (P): patients with MS; intervention (I): AHSCT; comparison (C): none; outcome (O): efficacy and safety.

Study Selection

Two authors (N.R., F.A.) independently screened the titles and abstracts to identify relevant studies. The same investigators then reviewed the full text of the selected papers for final selection. Any disagreement was resolved by consultation with a third reviewer (F.N.).

Data Extraction

The same reviewers (N.R., F.A.) extracted the following data from the selected studies: study demographics, sample size, gender, mean disease duration, type of MS, regimen intensity, cell dosage, EDSS at baseline, annualized relapse rate (ARR) at baseline, and the endpoint results regarding the efficacy and safety of AHSCT. A combination of total body irradiation (TBI) plus anti-thymocyte globulin (ATG) (TBI/ATG) is considered a high-intensity conditioning regimen, while the intermediate-intensity regimen most commonly used is the BEAM (carmustine, etoposide, cytarabine, and melphalan) plus ATG (BEAM/ATG) according to the European Society for Blood and Marrow Transplantation (EBMT) classification. There was no considerable difference in the AHSCT procedure among studies.


EDSS after treatment, ARR after treatment, progression-free survival (PFS), relapse-free survival (RFS), event-free survival (EFS), MRI activity-free survival (MAFS), no evidence of disease activity (NEDA), incidence of new MRI lesions after treatment, overall survival (OS), and transplant-related mortality (TRM) were extracted as endpoint data. There was substantial heterogeneity in the follow-up duration among studies. We extracted the efficacy and safety outcomes by default 5 years after transplantation. In studies with a shorter follow-up duration, we extracted the data for the longest endpoint.

Quality Assessments

The quality of observational studies was assessed using the Newcastle–Ottawa scale (NOS) [21] and the Cochrane risk-of-bias assessment tool for clinical trials by two independent investigators (N.R., F.A.), and consulting the third investigator (F.N.).

Statistical Analysis

We used Stata 11.0 software (StataCorp LLC, College Station, TX, USA) for statistical analysis. The medians and interquartile range were converted to mean and standard deviation based on the Hozo et al. method [22]. A standardized mean difference (SMD) methodology was applied for EDSS and ARR. The other efficacy and safety outcomes were pooled with a random-effects model and a 95% confidence interval (CI). Also, subgroup analysis based on the type of study and regimen intensity was performed. The Cochrane Q test and I-squared (I2) statistic were used to evaluate the heterogeneity among included studies.


Search Results

Our comprehensive search and manual addition yielded 1008 articles after duplicate removal (Fig. 1). Our initial title and abstract screening excluded 894 studies. In the end, 50 studies entered our meta-analysis and systematic review after full-text screening [12, 13, 17, 23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70].

Fig. 1
figure 1

PRISMA flow diagram depicting the flow of information through the different phases of a systematic review

A total of 4831 patients with MS, aged 26–60 years, were included in our study (Table 1). Among studies, 41 were cohort studies, eight were clinical trials, and one was a case series. The average quality score was 7.36 for observational studies, which is acceptable. For clinical trials, there was low publication bias (Supplementary 1 and 2). The detailed features of included studies are presented in Table 1.

Table 1 Demographic and clinical characteristic of included studies

Efficacy of AHSCT

We measured the efficacy of AHSCT with several outcomes including EDSS score change, ARR change, PFS, RFS, EFS, MAFS, NEDA, and incidence of new MRI lesions after treatment.

Our analysis showed a significant decrease in EDSS score after treatment (SMD: −0.48, 95% CI −0.75, −0.22; Q = 239.52, P < 0.00, I2 = 91.34%) (Fig. 2). The ARR was also significantly reduced after AHSCT relative to the pretreatment period (SMD: −1.58, 95% CI −2.34, −0.78; Q = 133.36, P < 0.00, I2 = 95.77%) (Fig. 3).

Fig. 2
figure 2

Forest plot of EDSS score before and after treatment

Fig. 3
figure 3

Forest plot of ARR score before and after treatment

The pooled estimate of PFS after treatment was 73% (95% CI 69%, 77%; Q = 461.90, P < 0.00, I2 = 89.89%) (Supplementary 3). Furthermore, 81% of patients with MS who received AHSCT remained relapse-free (95% CI 76%, 86%; Q = 79.71, P < 0.00, I2 = 79.05%) (Supplementary 4). Investigation of EFS, which reflects the absence of any disease-related event, showed a pooled estimate of 63% (95% CI 54%, 73%; Q = 33.24, P < 0.00, I2 = 76.26%) (Supplementary 5). Also, the MAFS was 89% (95% CI 84%, 94%; Q = 3.25, P: 0.36, I2 = 26.66%) among included studies with low heterogeneity (Supplementary 6). New MRI lesions appeared in nearly 8% of patients who underwent AHSCT (95% CI 4%, 12%; Q = 5.31, P: 0.50, I2 = 0%) (Supplementary 7). Our meta-analysis showed that 68% of patients with MS experienced NEDA after AHSCT (95% CI 59%, 77%; Q = 37.93, P < 0.00, I2 = 75.97%) (Supplementary 8). The clinical outcomes are summarized in Fig. 4.

Fig. 4
figure 4

Clinical outcomes of AHSCT

Safety of AHSCT

The overall survival after transplantation was 94% (95% CI 91%, 96%; Q = 93.60, P < 0.00, I2 = 83.92%) (Supplementary 9). In addition, 4% of patients died from transplant-related causes (95% CI 2%, 6%; Q = 89.13, P < 0.00, I2 = 93.21%) (Supplementary 10).


Despite recent improvements in the application of AHSCT in MS, utilization of this treatment option is still limited. Many consider AHSCT among the final treatment strategies when other DMTs have failed [71]. In this systematic review, we aimed to address the lack of evidence supporting the confident application of AHSCT for patients with MS and to present a better view of the prospective benefits and potential risks.

The primary outcome measures for the efficacy of AHSCT were EDSS score change and ARR change. Regarding our analysis, both of these outcome measures showed reductions as a result of AHSCT. The decrease in the EDSS score is in line with previous meta-analyses, confirming the therapeutic application of AHSCT for halting the progression of MS [16, 72]. The reduction seen in ARR is also similar to the previous meta-analysis by Sormani et al., supporting the application of AHSCT in patients with MS with recurring relapses. It was previously shown that patients with RRMS are the most likely to benefit from AHSCT, besides having minimal transplant-related adverse effects compared with other MS subtypes [73, 74].

Based on our results, pooled estimates for PFS, RFS, and EFS showed promising results, confirming the effectiveness of AHSCT as a one-time and long-term treatment option for patients with MS. We also found slight but nonsignificant improvements in MAFS and incidence of new MRI lesions after treatment. Compared with other DMTs such as mitoxantrone (MTX), natalizumab, and alemtuzumab, AHSCT has shown better outcomes in controlling the progression and relapse of MS symptoms, in addition to achieving more extended periods of NEDA [70, 75, 76]. Currently the BEAT-MS (NCT04047628) trial is aiming to provide a comparison of the best available therapy versus AHSCT, though it is still in the patient recruitment stage. Further clinical trials are needed to elucidate a precise head-to-head comparison of these approaches.

We determined safety outcomes for AHSCT by overall survival and TRM. Contrary to previous findings, we found relatively high TRM. The initially high TRM of 3.6% decreased to 0.3% in studies post-2005 due to better patient selection, the use of proper regimens for immunoablation, and improved transplant techniques [73, 77, 78]. However, long-term outcomes measured by our analysis indicate higher TRM, raising a primary concern for AHSCT use in MS. We considered the endpoint of all TRM mainly at the end of 5-year follow-up duration; however, previous studies have considered a 100-day post-transplantation period for assessing TRM. This disparity in the definition of TRM may explain the observed difference in TRM between our research and previous meta-analyses.

As AHSCT targets the immune system, it can lead to several adverse events secondary to immune suppression. One study found that 79% of early non-neurological adverse effects, including neutropenic fever, sepsis, infections, and viral reactivation, were secondary to immunosuppression. Also, neurotoxicity occurred in 26 of 149 patients within 60 days of transplantation [60]. Late adverse events such as malignancies can be expected. Another study reported malignancies in nine of 281 patients [55]. Further studies with long follow-up duration are needed to determine the risk of potential adverse events after AHSCT in patients with MS.

AHSCT seems to hold better potential for treating patients with MS with different disease courses, as it is mainly considered among the final treatment options, and patient selection for AHSCT is usually made after many failed DMTs. The relatively high TRM of AHSCT versus other DMTs may be linked with patient characteristics. For instance, patients receiving AHSCT tend to have a more aggressive course of disease [55, 73]. Also, all AHSCT patients need to be protected from vaccine-preventable diseases, and the emergence of the COVID-19 pandemic has complicated this procedure in recent years [77]. Thus, the need for studies investigating the efficacy and safety of earlier AHSCT administration as mentioned in the EBMT criteria has increased.

EBMT recently issued guidelines with detailed patient characteristics appropriate for receipt of AHSCT, including highly active RRMS, disease duration less than 10 years, EDSS score equal to or less than 5.5, and age younger than 45 years [78]. By considering these in patient recruitment, achieving a better perspective on the efficacy and safety of AHSCT as a result of earlier administration is possible.

Although some guidelines have recently changed the position of AHSCT for RRMS from a “clinical option” to a “standard of care,” its use is still typically reserved for later in the disease course. As a result of growing evidence, equal footing of AHSCT with second-line DMTs for patients with RRMS is suggested [79]. Considering the superior efficacy of AHSCT in establishing long-term suppression of disease activity, it may be crucial to consider it before many of the second-generation DMTs to save time and prevent irreversible disease progression. However, this needs to be further investigated in large randomized controlled trials comparing the safety and efficacy of different DMTs with AHSCT in patients with distinct MS subtypes. There is a growing number of ongoing observational studies and clinical trials which can provide more evidence regarding the efficacy and safety of AHSCT in patients with MS and lead to optimization of this procedure (NCT numbers NCT03477500, NCT05029206, NCT04674280, NCT04047628).

Our study was limited in some aspects. First, due to the lack of studies focusing on specific subtypes of MS, we could not carry out a subgroup analysis. Also, there was relatively high heterogeneity between included studies, which led us to use random-effects analysis. Different patient characteristics, follow-up times, disease durations, subtypes of MS, conditioning regimens, and transplant techniques may have resulted in this heterogeneity.

Nevertheless, compared with a previous study by Ge et al. investigating the safety and efficacy of AHSCT in patients with MS [16], our study has several advantages. First, we investigated a greater number of efficacy and safety outcomes to give a comprehensive view of AHSCT in patients with MS. They excluded observational studies and only included 18 papers with a total of 731 patients, while we included 50 studies with a total of 4831 patients with MS. Furthermore, we used a more comprehensive search strategy in more medical databases to minimize missing papers and publication bias.


AHSCT is highly efficacious in treating patients with MS in multiple aspects, including preventing disease progression and relapse in addition to reducing inflammatory responses and associated CNS lesions. The few studies that have compared the efficacy of this treatment approach with currently available DMTs have reasonably indicated a better outcome. Although the patients enrolled in AHSCT trials are usually refractory to DMTs and develop a more aggressive disease course, comparisons with other DMT studies still show encouraging results. In addition, with the increasing knowledge and expertise in the field of stem-cell therapy, AHSCT has become a safer treatment approach for MS. Altogether, current data encourage a broader application of AHSCT for treating patients with MS while still considering proper patient selection and transplant methods.