Autologous haematopoietic stem cell transplantation for multiple sclerosis: a position paper and registry outline

Background: While substantial progress has been made in the development of disease-modifying medications for multiple sclerosis (MS), a high percentage of treated patients still show progression and persistent inflammatory activity. Autologous haematopoietic stem cell transplantation (AHSCT) aims at eliminating a pathogenic immune repertoire through intense short-term immunosuppression that enables subsequent regeneration of a new and healthy immune system to re-establish immune tolerance for a long period of time. A number of mostly open-label, uncontrolled studies conducted over the past 20 years collected about 4000 cases. They uniformly reported high efficacy of AHSCT in controlling MS inflammatory disease activity, more markedly beneficial in relapsing-remitting MS. Immunological studies provided evidence for qualitative immune resetting following AHSCT. These data and improved safety profiles of transplantation procedures spurred interest in using AHSCT as a treatment option for MS. Objective: To develop expert consensus recommendations on AHSCT in Germany and outline a registry study project. Methods: An open call among MS neurologists as well as among experts in stem cell transplantation in Germany started in December 2021 to join a series of virtual meetings. Results: We provide a consensus-based opinion paper authored by 25 experts on the up-to-date optimal use of AHSCT in managing MS based on the Swiss criteria. Current data indicate that patients who are most likely to benefit from AHSCT have relapsing-remitting MS and are young, ambulatory and have high disease activity. Treatment data with AHSCT will be collected within the German REgistry Cohort of autologous haematopoietic stem CeLl trAnsplantation In MS (RECLAIM). Conclusion: Further clinical trials, including registry-based analyses, are urgently needed to better define the patient characteristics, efficacy and safety profile of AHSCT compared with other high-efficacy therapies and to optimally position it as a treatment option in different MS disease stages.


Introduction
Autologous haematopoietic stem cell transplantation (AHSCT) is becoming increasingly important internationally in treating multiple sclerosis (MS). 1,26][7][8] However, many questions about using AHSCT in MS are still unsolved.The three ongoing RCT studies, STAR-MS in the United Kingdom, BEAT-MS in the United States and RAM-MS in Norway, 9 together with the upcoming NET-MS study in Italy, aim to confirm effects in highly active relapsing-remitting MS in comparison with high-efficacy treatments such as B cell depleting and other monoclonal antibodies.Data should be available within 3-4 years.Being aware that MS is a progressive disease from disease onset, 10 the possibly most relevant questions are which is the optimal window of opportunity for AHSCT as an alternative to standard therapy and what extent of clinical progression in relation to disease duration and age justifies AHSCT as a rescue therapy. 11Other major issues are the choice of the conditioning regimen (intensity of lymphoablation and extent of myeloablation required) and establishing the long-term safety in relation to the conditioning, in respect of the risks of malignancies and secondary autoimmune diseases.
In Germany, AHSCT for MS in general is not paid for by insurance companies and hence is only performed on an individual basis and expert opinion.Therefore, coverage needs to be negotiated from case-to-case and after rejection of compensation from insurance companies, patients need to pay on their own.Some try to reach out for a social court verdict.Neurologists have been mostly reluctant, which has led to only 57 patients being documented in the European Society for Blood and Marrow Transplantation (EBMT) database of 1.721 cases between 1995 and 2021. 9n addition, insurance companies in most cases do not cover treatment costs.Therefore, a relevant number of persons with MS (pwMS) opted for treatment to be performed in other countries in the past.Knowledge about the benefits and risks of AHSCT among German pwMS is sparse.To overcome the treatment and knowledge gap in Germany, MS neurologists and stem cell transplantation experts have joined to develop recommendations for AHSCT treatment in MS and strategies to gather treatment experience.This article represents recommendations developed as an overarching expert consensus by a task force on stem cell transplantation in MS established under the umbrella of the Clinical Competence Network Multiple Sclerosis (KKNMS), the German Working Group for Haematopoietic Stem Cell Transplantation and Cellular Therapy e.V. (DAG-HSZT), and the German MS Self-Help Association (DMSG).They are intended to help identify patients who are most suitable for AHSCT, to facilitate the coverage of costs by the health insurances, and to carry out and follow up the transplantations according to a common standard and document them in the German MS registry and the German Registry for Stem Cell Transplantation (DRST).The goal is to systematically document all experience with AHSCT in MS in Germany, including cases beyond the consensus criteria mentioned below, to specify clear indication criteria based on long-term data.

Criteria for the use of AHSCT
We will distinguish between narrower and broader criteria aligned with the Swiss standard for AHSCT in MS. 12 The narrower criteria correspond to the core criteria of EBMT 9 and the criteria of the American MS Society (National MS Society, NMSS) 13 based on recommendations of the American Society for Blood and Marrow Transplantation (ASBMT). 14though AHSCT is probably the most effective immunotherapy for MS and can reset the immune system, it is, at least based on our current knowledge, not a regenerative therapy.Current evidence indicates that young persons with a short disease duration course of MS, and high inflammatory activity on magnetic resonance imaging (MRI) (new T2 lesions and contrast agent lesions) and a high relapse frequency benefit most from this approach.Data are not conclusive in progressive disease courses and in the absence of MRI activity.For advanced disease stages with a long duration, older age and greater impairment, data argue against a benefit that would justify the risks of transplantation.Recent inflammatory activity on MRI is included in all current criteria.First introduced in a short section on evidence, the criteria are discussed individually and then summarised in tabular form for relapsing and secondary progressive MS (SPMS).Primary progressive MS (PPMS) is discussed separately at the end.
Disease course.AHSCT is primarily used to treat relapsing-remitting MS.The data on SPMS suggest a significantly lower efficacy.In the meta-analysis by Sormani et al., 15 relapsing and non-relapsing patients were compared.Here, a clear difference was found in favour of RRMS.Treatment-related mortality (TRM) was also significantly lower in RRMS.In the metaanalysis by Zhang and Liu, 5 RRMS showed 81% progression-free survival (PFS), SPMS 78% and PPMS 60%.In a long-term observed cohort of 281 evaluable patients, with median follow-up of 6.6 years, Muraro et al. 16 reported significantly worse PFS in progressive versus relapsing form of MS [hazard ratio (HR), 2.33; 95% confidence interval (CI), 1.27-4.28].
In the real-world case series by Burt et al., 17 PFS at 4 years was 95% in RRMS and 66% in SPMS.The Italian experience in n = 210 pwMS from 20 Italian centres was reported by Boffa et al. 18 In all, 86 patients were diagnosed SPMS and 2 patients PPMS.PFS was 85.5% in RRMS at 5 years compared with 71% in progressive group.Another analysis of the subgroup of SPMS with n = 79 from the Italian cohort applied a matching analysis with established MS treatments and confirmed a longer time to disability progression in active SPMS after AHSCT (HR = 0.50; 95% CI = 0.31-0.81;p = 0.005). 11In the series by Nicholas et al., 19 no differences in the effect on PFS between RRMS and progressive MS could be detected.Neither demographics nor progression nor type of prior therapy were predictive. 11,17commendation: Active RRMS despite highly effective immunotherapy is the key indication for AHSCT.In younger patients with active SPMS and a conversion from RRMS no longer than within the preceding 3 years without severe impairment (Expanded Disabilty Status Scale Score, EDSS ⩽ 6.5), the use of AHSCT can be considered.
Age.Beyond the age of 50, AHSCT, like all other highly effective MS immunotherapies, should be discussed critically, though very few data on the effect of age on AHSCT are available. 20In the meta-analysis by Sormani et al., 15 patients > and < 36 years were compared.Here, no convincing difference in therapeutic effect was found.Age did not affect mortality.The long-term follow-up case series by Muraro et al., 16 however, reported age as one significant variable affecting PFS comparing ages 18-31, 32-37, 38-44 and beyond 45.
Recommendation: A lower regenerative capacity of the nervous system, ageing of the immune and nervous systems, comorbidities, and the risk of other comorbidities argue against AHSCT in pwMS beyond 50 years.In older age, a short disease duration and high inflammatory activity are a pre-requisite to consider AHSCT.

Impairment and progression of impairment.
In the meta-analysis by Sormani et al., 15 patients with EDSS > 5.5 and ⩽ 5.5 were compared.Descriptively, there was an advantage for less affected patients, but this was not significant.However, TRM was significantly higher in EDSS > 5.5.In the real-world case series by Burt et al., 17 all groups (EDSS 2-4, EDSS 4.5-6, EDSS 6.5, EDSS 7-8) benefitted from AHSCT.Muraro et al., 1 formulates that patients with an EDSS ⩾ 7 are at high risk of treatment failure.Mariottini et al. 21recently reported a comparison of AHSCT (n = 31) and cyclophosphamide (CYC) therapy (monthly in first year, then every 2 months in second year, n = 62) in SPMS (mean 40 years, 13 years MS, EDSS 6.0).After 5 years, PFS was very similar between AHSCT (70%) and CYC (81%) treated patients, while AHSCT significantly suppressed more pronounced relapse activity.In an earlier case series of n = 26 patients with SPMS, Mariottini et al. 22 found PFS for 42% after 5 and 30% after 10 years.Recent disability accrual has a favourable prospect over impairments that exist already for many years.
Recommendation: A free walking distance of at least 100 m with assistance (EDSS 6.0) is generally considered the upper limit of mobility impairment for AHSCT treatment, based on the idea that the degeneration in the nervous system should not be too advanced.In individual cases, AHSCT can be considered beyond EDSS 6.0 if the patient is very young, has a short, highly aggressive course and has a high level of inflammatory activity.In the same way, a patient who is currently only mildly affected can be treated if the inflammatory activity (many relapses, high MRI activity) and neurological impairment are high during relapses.The dynamics of progression must also be taken into account.Thus, according to the EBMT criteria based on Menon et al., 23 an EDSS 6.0 after a maximum of 5 years of disease progression or an EDSS 6.0 before age 40 are also accepted as criteria.
Disease duration.Detailed analyses are not available.However, the effects of disease duration often closely resemble those of age.
Recommendation: In general, based on the available studies, MS should not be present for longer than 10 years (first certain manifestation, not time of diagnosis), which also conforms to the idea that degeneration plays an increasing role after a longer disease course.Here, too, differentiation must be made for individual cases.If MS starts in childhood or adolescence, AHSCT can also be considered after 15 years with MS.In older patients with such a long disease period, AHSCT treatment should only be considered in cases with very high inflammatory activity.
Relapses.To our knowledge, data comparing the impact of high and low relapse rates before transplantation are not available.

Recommendation:
In principle, relapse activity should be present in the period before transplantation.Given the heterogeneity of relapses and the uncertainty in diagnosing relapses, ideally, only relapses with EDSS-relevant changes should be evaluated.However, in highly active disease courses, differentiation between relapse and progression can be difficult.In case of an uncertain relapse classification, other criteria become more important for determining the indication: MRI activity, age and disease duration.In the criteria according to Menon et al., 22 an SPMS conversion since less than 3 years is mentioned.Here, however, a distinction from PPMS with superimposed relapses remains blurred.

MRI activity.
In the real-world case series by Burt et al., 17 SPMS patients with Gd+ in the year before AHSCT were more likely to benefit.Bur-man et al. 24 and Mancardi et al., 25 reported similar evidence.
Recommendation: Basically, MRI activity, that is, new contrast enhancement(s) or new/size-progressive T2 lesions in the last year, is a pre-requisite for AHSCT, although there is no international consensus on the necessary number of lesions.However, highly effective therapy can potentially lead to suppression of MRI activity in the presence of clear clinical progression.Whether this progression is neurodegenerative or an expression of diffuse inflammation not showing up on MRI is unclear.There is no doubt that contrast enhancement and T2 lesions represent only parts of the MRI inflammatory activity in MS.Therefore, affected persons without MRI activity are also suitable in individual cases, provided they are young, have a short disease duration and show considerable progression dynamics.

Previous immunotherapy.
No data are available specifically analysing AHSCT outcomes after documented treatment failure of highly or moderately effective MS immunotherapies; however, many patients, who had received AHSCT in the past had already failed one or several approved therapies before being offered AHSCT.Indeed, one long-term observational cohort study documented the association of a higher number of previous immunotherapies with worse PFS. 16his observation led these authors and other experts to recommend against multiple treatment failures as a pre-requisite for consideration of AHSCT and that failure of no more than two prior disease modifying treatments (DMTs) would be preferred. 1commendation: In principle, treatment failure with one highly potent medication (ocrelizumab, ofatumumab, rituximab, natalizumab, alemtuzumab or a similar active substance) is required before selecting AHSCT, also in light of negotiating cost coverage with health insurance companies.In cases of aggressive disease and marked disease progression under an initial medication, AHSCT should already be considered in order not to miss the most favourable time of opportunity for this treatment.In individuals with clearly aggressive disease, AHSCT might be justified even as a first-line treatment on a case-by-case basis.
Based upon group consensus, the Task force proposes the following core and extended criteria adapting mostly the Swiss standard. 12The criteria listed below are not definite predictors for a response to AHSCT but are intended to describe a corridor for its use.Therefore, criteria have to be examined individually and will not necessarily be entirely fulfilled in each case.Currently, all AHSCT cases continue to be individual attempts, in German 'Heilversuch' to slow down the disease activity and progression (Table 1).

Use in primary progressive MS (PPMS).
There is no specific study on PPMS, although PPMS data are available in cohorts with progressive patients.Most immunotherapies do not work convincingly in PPMS; hence AHSCT should be considered with great caution.The prototypic PPMS patient, more likely to be male, around 50 years of age, often very slowly insidious in onset, with a leading spinal course and little inflammatory activity, does not appear to be a good candidate for AHSCT.On the other hand, primary progressive inflammatory disease courses are also present in young people.Combined with the pathophysiological concept that MS is basically a progressive disease with varying degrees of focal inflammatory activity, AHSCT should also be considered in individual cases of PPMS.Ideally, these patients fulfil activity criteria through MRI activity and imposed relapses.Without MRI activity, PPMS could only be considered in cases with an aggressive course, 23 that is, EDSS 6.0 after 5 years or EDSS 6.0 before age 40 and with enhanced consideration of the benefit versus risk balance.

Major treatment risks.
A major concern of neurologists is mortality due to AHSCT which has been about 7% in the earliest treatment times from 1995 to 2000. 1 However, TRM has substantially declined to 1% in the meta-analysis by Zhang and Liu. 5 Looking at the most recent cases from 2017 to 2021, mortality rate was 0.2% in EBMT data 26 as well as also 0.2% in the Burt real-world cohort. 17Other single-centre or regional case series reported higher rates and current estimates of risk of TRM is ~1%. 1 In addition, secondary autoimmune diseases have been reported in about 5% and secondary cancer in about 3%.These data need to be considered with caution as long-term follow-up is limited.

Practical management guidance of AHSCT in MS
A strong interaction of a neurological centre with profound MS expertise and a committed and experienced stem cell transplantation unit or clinic is crucial for a scientifically sound and consistent patient selection as well as for optimal treatment and follow-up including low-barrier clinical visits and any necessary readmissions.As expertise is correlated with the number of procedures performed, concentrating the work in a limited number of centres in each country is preferable.
Therapy decision.Decision of a neurological stem cell board (2 neuroimmunologists or MS experts, 1 haematologist stem cell transplanter, 1 neuroradiologist) is based on clinical information and MRI presentation.Written documentation of the decision is maintained in the hospital information system.To assure optimal assessment and counselling of patients, it is advisable to establish an interdisciplinary collaboration at the respective centre with fertility experts, pneumonologist, cardiologist, neuropsychologist/psychiatrist and often also gastroenterologist, and infectious disease experts.
Informed consent.Comprehensive patient information in oral and written form, ideally discussed in the presence of relatives and in several sessions, must address not only the opportunities but also the risks, especially the mortality risk, the risk to fertility, and the long-term risks of cancer and secondary autoimmune diseases.
Distance to previous therapies.General rules cannot be formulated.In view of an aggressive disease course, a critical assessment of each individual Once the graft is released by the professional cell manufacturer, conditioning and AHSCT can take place.
Conditioning.In the meta-analysis by Sormani et al., 15 high-medium and low-intensity regimens were compared.Here, only a descriptive difference in favour of high-intensity therapy was found.The conditioning regimen did not affect mortality.In the meta-analysis by Zhang and Liu, 5 there was no clear advantage of highintensity regimens on progression (PFS: medium intensity 73%, low intensity 85% and high-intensity 58%).However, mortality was significantly higher in high-intensity regimens (0% in low versus 6% in high).Analysis of EBMT data on different conditioning regimens is still pending but for conditioning both CYC and ATG (antithymocyte globulin) or ALG (anti-T lymphocyte globulin) or BEAM (BCNU, etoposide, cytarabine, melphalan) with ATG were mostly used with promising outcomes and manageable toxicities 9 with a more suitable safety profile for CYC and ATG according to the real-world case series by Burt et al. 17 Recommendation: The conditioning regimen should preferably use 200 mg/kg body weight CYC (4 × 50 mg/kg body weight).According to the safety and warning statements in the product information, treatment with CYC can lead to germline damage (oocytes and sperm).Therefore, fertile female patients should be advised about egg freezing prior to treatment, and male patients should be advised about sperm preservation.
As an alternative to CYC, the BEAM-ATG protocol can be used.To prevent engraftment of residual or reinfused autoreactive T cells, especially in case of an unselected AHSC graft, ATG (Thymoglobulin®) at a cumulative dose of 5-7.5 mg/kg body weight or ALG (Neovii®) at a cumulative dose of 60 mg/kg prior to AHSCT 17 should be added.In addition, ATG might also support proper immune reconstitution by the induction of regulatory T cells. 27Premedication to avoid allergic reactions, infusion and concomitant medication are performed according to local protocols.
ATG/ATLG also requires prophylactic medication according to local standards.Pneumocystis jirovecii infections with sulfamethoxazole/trimethoprim can be given for 6 months.In case of intolerance, atovaquone or dapsone can be administered alternatively, or a monthly pentamidine inhalation can be given.
All patients are monitored weekly for CMV and EBV by PCR in peripheral blood until day 30 and at every visit thereafter.Possible reactivations are monitored, and preventive treatment is given according to site guidelines.Additional routine safety visits, as recommended by the on-site haematologist/transplant specialist, are conducted according to local protocols (Table 2).
Other infection prophylaxis.Antifungal prophylaxis as a standard is not recommended.In protracted neutropenia, posaconazole, but also other drugs, such as fluconazole or itraconazole or micafungin, can be given according to local guidelines.Ofloxacin or ciprofloxacin for infection prophylaxis in neutropenia can be given until confirmed transplantation (engraftment) according to local guidelines.Pre-emptive CMV treatment with ganciclovir or valganciclovir can be administered if a replicative infection is proven (positive pp65 test or PCR).
Transfusions.All blood products for administration in the first year after transplantation must be irradiated.CMV-negative patients should be transfused only with CMV-negative blood products or with an appropriately adapted system.Platelet and red cell transfusions should be administered according to site guidelines and protocols for allogeneic stem cell transplantation.
Other • Meningococcal (meningococcal ACWY conjugate and meningococcal B vaccine) • Hepatitis B (titre determination recommended after completion of vaccinations) From 24 months after transplantation • Measles/mumps/rubella (live vaccine) after titre control Rehabilitation.MS-competent neurological rehabilitation from month 3 after transplantation should be sought for all patients, ideally in a clinic with expertise in AHSCT for MS or corresponding cooperation with a haemato-oncologist, as summarised in EBMT guidelines. 28uroimmunological follow-up.In the first year, follow-up should be quarterly, followed by halfyearly clinical controls.A cranial MRI and, ideally, a spinal MRI as 're-baseline MRI' and safety assessment should be carried out 3-6 months after transplantation, then annually or earlier if clinically indicated.Although there is little evidence for any ongoing focal inflammation after AHSCT, the database is not sufficient to exclude evolution of new inflammatory activity after AHSCT.

Study concept: German REgistry cohort of autoLogous haematopoietic stem cell transplantation in MS
German REgistry cohort of autoLogous haematopoietic stem cell transplantation in MS (RECLAIM) will combine two registry platforms, the German MS registry and the DRST, managed by DAG-HSZT, which is linked to the EBMT database.In addition, patients will be invited to contribute to the Observational Study in MS treated with aHSCT (OMST) of EBMT led by R. Saccardi.The aim is to document all pwMS living in Germany who have been treated with AHSCT, even if outside Germany.We will focus on the long-term effects of AHSCT on disability and MRI data, as well as safety, tolerability and toxicity, based on yearly clinical evaluations.To be included, patients are invited to approach 1 of the 160 local MS centres contributing to the MS registry.In addition, patients without access or unwilling to access an MS centre can provide their data to the central study centre in Hamburg and will be monitored remotely as well.The study start is planned for spring 2023.

TherapeuTic advances in neurological disorders
Volume 16

Table 1 .
Recommendations.must determine if AHSCT is justified under therapy that is still immunologically effective.
EDSS, Expanded Disbailyyt Status Scale Score; MRI, magnetic resonance imaging; MSFC, Multiple Sclerosis Functional Composite; RRMS, relapsing-remitting MS; SPMS, secondary progressive MS. a In the case of progression during relapse, higher EDSS values can be present that justify an AHSCT, just as values < 3.0 can be present in a relapse-free interval.bIncrease in EDSS: 1 point for patients with EDSS < 5.5, 0.5 points for EDSS ⩾ 5.5.case
therapeutics.Antiemetics are administered according to local practice procedures or institutional guidelines.Infusions are administered and managed according to local guidelines. journals.sagepub.com/home/tan9