Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Autologous mesenchymal stem cells for the treatment of secondary progressive multiple sclerosis: an open-label phase 2a proof-of-concept study

Autologous mesenchymal stem cells for the treatment of secondary progressive multiple sclerosis:... Articles Autologous mesenchymal stem cells for the treatment of secondary progressive multiple sclerosis: an open-label phase 2a proof-of-concept study Peter Connick,* Madhan Kolappan,* Charles Crawley, Daniel J Webber, Rickie Patani, Andrew W Michell, Ming-Qing Du, Shi-Lu Luan, Daniel R Altmann, Alan J Thompson, Alastair Compston, Michael A Scott, David H Miller, Siddharthan Chandran Summary Lancet Neurol 2012; 11: 150–56 Background More than half of patients with multiple sclerosis have progressive disease characterised by accumulating Published Online disability. The absence of treatments for progressive multiple sclerosis represents a major unmet clinical need. On January 10, 2012 the basis of evidence that mesenchymal stem cells have a benefi cial eff ect in acute and chronic animal models of DOI:10.1016/S1474- multiple sclerosis, we aimed to assess the safety and effi cacy of these cells as a potential neuroprotective treatment for 4422(11)70305-2 secondary progressive multiple sclerosis. See Comment page 123 *These authors contributed Methods Patients with secondary progressive multiple sclerosis involving the visual pathways (expanded disability equally status score 5·5–6·5) were recruited from the East Anglia and north London regions of the UK. Participants received Department of Clinical intravenous infusion of autologous bone-marrow-derived mesenchymal stem cells in this open-label study. Our Neurosciences (P Connick MRCP), Anne MacLaren Laboratory for primary objective was to assess feasibility and safety; we compared adverse events from up to 20 months before Regenerative Medicine treatment until up to 10 months after the infusion. As a secondary objective, we chose effi cacy outcomes to assess the (D J Webber PhD, R Patani MRCP), anterior visual pathway as a model of wider disease. Masked endpoint analyses was used for electrophysiological and Department of Clinical selected imaging outcomes. We used piecewise linear mixed models to assess the change in gradients over time at the Neurosciences (A W Michell PhD), Department of Pathology point of intervention. This trial is registered with ClinicalTrials.gov, number NCT00395200. (Prof M-Q Du PhD), Department of Veterinary Medicine Findings We isolated, expanded, characterised, and administered mesenchymal stem cells in ten patients. The mean (S-L Luan PhD), and Department dose was 1·6×10⁶ cells per kg bodyweight (range 1·1–2·0). One patient developed a transient rash shortly after of Clinical Neurosciences (Prof A Compston FMedSci), treatment; two patients had self-limiting bacterial infections 3–4 weeks after treatment. We did not identify any University of Cambridge, serious adverse events. We noted improvement after treatment in visual acuity (diff erence in monthly rates of change Cambridge, UK; NMR Research –0·02 logMAR units, 95% CI –0·03 to –0·01; p=0·003) and visual evoked response latency (–1·33 ms, –2·44 to –0·21; Unit, Institute of Neurology, p=0·020), with an increase in optic nerve area (diff erence in monthly rates of change 0·13 mm², 0·04 to 0·22; Department of Neuroinfl ammation, University p=0·006). We did not identify any signifi cant eff ects on colour vision, visual fi elds, macular volume, retinal nerve College London, London, UK fi bre layer thickness, or optic nerve magnetisation transfer ratio. (M Kolappan MRCP, Prof D H Miller FMedSci); Blood and Marrow Transplant Unit, Interpretation Autologous mesenchymal stem cells were safely given to patients with secondary progressive multiple Addenbrooke’s Hospital, sclerosis in our study. The evidence of structural, functional, and physiological improvement after treatment in some Cambridge, UK (C Crawley FRCP, visual endpoints is suggestive of neuroprotection. M A Scott PhD); Medical Statistics Department, London School of Hygiene and Tropical Funding Medical Research Council, Multiple Sclerosis Society of Great Britain and Northern Ireland, Evelyn Trust, Medicine, London, UK NHS National Institute for Health Research, Cambridge and UCLH Biomedical Research Centres, Wellcome Trust, (D R Altmann PhD); Department Raymond and Beverly Sackler Foundation, and Sir David and Isobel Walker Trust. of Brain Repair and Rehabilitation, University College London, London, UK Introduction Multipotent mesenchymal stromal cells are bone- (Prof A J Thompson FRCP); and Multiple sclerosis (MS) aff ects more than 2 million people marrow cells that can be expanded ex vivo and will readily Centre for Clinical Brain Sciences, worldwide and is the most common non-traumatic cause diff erentiate into mesodermal cell derivatives. In addition MRC Centre for Regenerative, of disability in young (<50 years) European adults. It is a to tissue engineering applications that target the repair of Medicine University of Edinburgh, Edinburgh, UK multifocal CNS disorder that has two distinct clinical phases skeletal tissue defects, biological properties independent (Prof S Chandran PhD) corresponding to inter-related pathological processes: focal of diff erentiation suggest that mesenchymal stem cells Correspondence to: infl ammation that drives activity during the relapse-remitting could have a therapeutic role through strategies other Prof Siddharthan Chandran, 8 stage and neuro degeneration that underlies progressive than tissue replacement in diseases such as MS. These Centre for Clinical Brain Sciences, disease char acterised by accumulating fi xed disability. strategies include neuroprotection through paracrine MRC Centre for Regenerative Medicine, Centre for Although important advances in treatment to reduce eff ects on the CNS microenvironment, augmentation of 3,4 Neuroregeneration, University of relapse rate have been made in the past two decades, no endogenous axonal and myelin repair processes, and Edinburgh, Chancellor’s Building, 8,9 treatments are available for the roughly half of patients immune regulatory activity. Increasing evidence shows Edinburgh EH16 4SB, UK with MS who have progressive disease. There is therefore both neuroprotection and functional improvement after siddharthan.chandran@ed. ac.uk a great and unmet clinical need for the development of infusion with mesenchymal stem cells in mouse models 10–14 neuroprotective treatments. of relapsing-remitting and chronic MS. 150 www.thelancet.com/neurology Vol 11 February 2012 Articles Clinically, mesenchymal stem cells have been used in (Hyclone; Perbio Science, Northumberland, UK), plated at the treatment of immune-mediated human diseases a density of 1×10⁸ cells per cell factory (Nunc, Thermo including steroid-resistant graft-versus-host disease and Scientifi c, Northumberland, UK). Near confl uent cultures 15–17 systemic lupus erythematosus. Three recent reports (>80%) were treated with 0·25% trypsin-EDTA (Invitrogen) have also described the use of intrathecally delivered and replated at 3·5×10⁶ cells per cell factory. Mesenchymal autologous mesenchymal stem cells in MS without adverse stem cells were harvested and cryopreserved in 4·5% events or signifi cant changes in general clinical outcomes human albumin solution (Bio Products Laboratory, 18–20 (webappendix). However, assessment of neuroprotection Hertfordshire, UK) with di methyl sulphoxide (Origen See Online for webappendix in the context of MS is challenging because of clinical and Biomedical Inc, Helsingborg, Sweden) at a fi nal pathological heterogeneity. To increase sensitivity for concentration of 10%. Mesenchymal stem cells were then structural and functional treat ment eff ects, the use of characterised in accordance with International Society of eligibility criteria that select cohorts with specifi c and Cellular Therapy recommen dations. Briefl y, this included clinically eloquent lesions, such as those of the anterior evidence of trilineage diff erentiation potential (adipocyte, visual pathway, enables assess ment of tailored and detailed chondrocyte, osteocyte) and fl ow cytometry to confi rm 22,23 outcomes. By use of this approach, we aimed to compare expression of CD73, CD90, and CD105 surface molecules safety and effi cacy outcomes for patients with secondary (>95%) and absence of CD34, CD45, CD14, and CD3 progressive MS before and after intravenous treatment (<2%). Release criteria for clinical use included absence of with autologous mesenchymal stem cells. contamination by pathogens (as documented by aerobic and anaerobic cultures and mycoplasma testing), and lack Methods of any genomic copy number changes as assessed with Participants 1-Mb-resolution bacterial artifi cial chromosome array Between November, 2007, and August, 2010, we did an comparative genomic hybridisation. open-label phase 2a proof-of-concept study involving We administered autologous mesenchymal stem cells participants recruited from the East Anglia and north intravenously to patients with secondary progressive MS. London regions of the UK (identifi ed from MS and Administration of the cells was done as a day-case general neurology clinics). We screened patients for procedure. To reduce type I hypersensitivity reactions, eligibility between November, 2007, and June, 2009. premedication with 10 mg chlor pheniramine, 100 mg Eligible participants were those aged 18–65 years with hydrocortisone, and 10 mg metoclopramide was given clinically defi nite MS according to the Poser criteria, 30 min before administration of the cells. Cryo preserved an expanded disability status scale (EDSS) score of cells were thawed (≤4 min) and im mediately infused over 2·0–6·5, clinical evidence of optic nerve involvement 15 min through a peripheral venous cannula. After (defi ned as a history of optic neuritis, Uhthoff ’s administration of cell suspensions, we infused normal phenomenon, or optic atrophy on examination), saline (500 mL) over 4 h. abnormal visual evoked potentials from one or both Our primary objective was to assess feasibility and safety; eyes consistent with de myelination, a retinal nerve fi bre our secondary objectives were to assess effi cacy on clinical, layer thickness of at least 45 μm in one eye, a T2 lesion electrophysiological, and structural outcomes, in addition on MRI of the optic nerve (webappendix), and the to providing information on the mechanism of any capacity to give consent. Patients were excluded if they recorded eff ect. We used a sentinel lesion approach based had a bleeding disorder, had received interferon beta or on the diseased anterior visual pathway to increase power glatiramer acetate within 6 months of trial entry, or had to detect treatment eff ects, and we used a pretest–post- previously used other disease modifying therapies at test design to compare adverse events and effi cacy any point. All patients gave written informed consent measures before and after the intervention. We assessed before study entry and approval was obtained from the participants at 3–6 month intervals for at least 12 months local ethics committee (Cambridgeshire 2 regional before and 6 months after treatment (webappendix). ethics committee). Assessment at each timepoint was split into two visits with a gap of less than 2 weeks; clinical assessment and visual Procedures evoked responses were done in Cambridge, UK, and MRI, We generated clinical-grade mesenchymal stem cells optical coherence tomography, and neuro-ophthal- under good manufacturing practice conditions with mological assessments were done at the University College standard operating procedures. Briefl y, we separated London Institute of Neurology (London, UK). Clinical bone-marrow mononuclear cells by density gradient assessment involved neurological and medical history with centrifugation in Ficoll-Paque Premium (GE Healthcare recording of adverse events and scores on the EDSS, MS UK Ltd, Buckinghamshire, UK). We resuspended the functional composite (MSFC), Addenbrooke’s cognitive washed cells in phosphate-buff ered saline/EDTA (Miltenyi examination revised, 29-item MS impact scale, and Beck Biotec Ltd, Surrey, UK) and cultured them in Dulbecco’s depression inventory II. Whole and central fi eld modifi ed Eagle’s medium (low glucose; Invitrogen, checkerboard pattern-reversal visual evoked responses Paisley, UK) supplemented with 10% fetal bovine serum were recorded with reversal achromatic checks subtending www.thelancet.com/neurology Vol 11 February 2012 151 Articles 60´ at the eye. Neuro-opthalmological assessment included Visual evoked responses, optic nerve area, optic nerve visual acuity with a retroilluminated early treatment magnetisation transfer ratio, and optic nerve diff usion diabetic retinopathy study chart, contrast acuity with tensor imaging based outcomes were assessed by a retroilluminated Sloan charts, colour vision with the single observer (MK) from whom participant status Farnsworth–Munsell 100-hue test, and visual fi eld (before or after treatment) was masked. Lesional analysis assessment by automated static perimetry (Humphrey was done after image acquisition at each visit. Brain fi eld analyser, 30-2 protocol). Optical coherence tomography volume and whole-brain magnetisation transfer ratio images were acquired by a single operator (MK) with a were done with automated methods with minimal time domain optical coherence tomograph (Stratus OCT manual corrections. Model 3000; Carl Zeiss Meditec, Dublin, CA, USA). MRI images were acquired with a Magnetom 3·0 T Statistical analysis Tim Trio scanner (Siemens, Erlangen, Germany) with a We used piecewise linear mixed models to assess, for a 12-element receiver head coil. Several MRI-based given measure, the change in gradient over time at the measures were assessed: optic nerve cross-sectional area; point of intervention; the given measure was the response optic nerve diff usion tensor imaging measures of variable, with the time from intervention and the time fractional anisotropy, mean diff usivity, axial diff usivity, multiplied by an after-intervention interaction term as and radial diff usivity; optic nerve magnetisation the two predictors. Such models allow estimation of the transfer ratio; whole-brain T2 lesion volume; whole-brain gradients before and after intervention, and of the T1 hypointense lesion volume; whole-brain magnetisa- gradient change with its statistical signifi cance. For tion transfer ratio; brain T1 hypointense lesion analyses of data involving separate values for each eye magnetisation transfer ratio; and brain T2 lesion over time, we added an additional level to the model with magnetisation transfer ratio. Intersessional stability of individual eyes as levels within participants. For the imaging measures was confi rmed by contemporary EDSS score, although the before and after gradients were assessment of ten locally recruited healthy volunteers. estimated as above, the test of gradient change used the non-parametric Wilcoxon sign rank test to compare the two ratios: change in EDSS score before or time interval before versus change in score after or time interval after. 98 patients assessed for eligibility There was no evidence of deviation from model assumptions. In particular there was no evidence for 84 ineligible non-normality or heteroscedasticity of residuals, or evidence against linearity assumptions. All of the 14 eligible reported mixed models achieved convergence with estimates for both the variance components and the fi xed 3 declined to participate eff ects. Unrecordable visual evoked responses, related to severe dysfunction due to disease, were represented by 11 participants recruited amplitude values of 0 μV and latency values of 180 ms (the maximum recorded during our study). Analyses were done with Stata SE (versions 9.2 and 11). Power 1 withdrew consent calculations could not be done before the study because of the lack of information from previous studies on 10 underwent bone-marrow aspiration, had mesenchymal stem potential eff ect sizes. This trial is registered with cells cultured to target dose and cryopreserved, and were treated with the cells ClinicalTrials.gov, number NCT00395200. Figure 1: Study profi le Role of the funding source The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or Measure at recruitment writing of the report. The corresponding author had full Number of participants 10 access to all the data in the study and had fi nal Sex ratio (men:women) 7:3 responsibility for the decision to submit for publication. Age (years) 48·8 (4·1; 40–53) Duration of multiple sclerosis (years) 14·4 (7·9; 5–26) Results Expanded disability status score 6·1 (0·3; 5·5–6·5) Figure 1 shows the study profi le and table 1 the Time since last clinical episode of optic neuritis 11·9 (8·2; 2–26) participants’ characteristics. All participants had (years)* secondary progressive MS, with clinical and electro- Data are n or mean (SD; range). *Nine participants. physiological evidence of optic nerve involvement. In the 2–26 years before recruitment, nine patients had Table 1: Participants’ characteristics clinical optic neuritis (three bilateral) and one had 152 www.thelancet.com/neurology Vol 11 February 2012 Articles treatment discontinued owing to disease progression p=0·003 2 years before recruitment into our trial). We successfully isolated and cultured mesenchymal −0·05 stem cells to the target dose from all bone-marrow aspirates (mean total cultured dose 2·0×10⁶ cells per kg, range 1·1×10⁶–3·7×10⁶). Mean culture duration was 24 days (20–30). Patients received a single infusion of autologous cells after monitoring for a mean of 17·3 months (14·1–20·9) during the pretreatment phase. The mean administered dose was 1·6×10⁶ cells per kg bodyweight 0·05 (1·1×10⁶–2·0×10⁶); mean volume of cell suspensions was 167·2 mL (range 89–246). We did not record any adverse events during infusion. One patient developed a macular rash over the anterior chest at about 3 h after the start of 0·10 infusion that resolved spontaneously over 12 h; a further B patient described scalp pruritus beginning 1 week after p=0·016 treatment and resolving spontaneously 2 weeks later. Two patients had infections: a self-limiting upper-respiratory tract infection 3 weeks after infusion (not requiring treatment) and an Escherichia coli urinary-tract infection 4 weeks after infusion (treated with oral antibiotics). Results of weekly blood testing of clinical chemistry, haematology, and immunology during the 4 weeks after infusion was unremarkable. Compared with pre treatment titres, no changes were evident in the post-treatment −2 period for T-cell subset counts (CD3, CD4, CD8, CD19, and CD56) or humoral immunity assessed by titres to common −4 antigens (mumps, measles, rubella, varicella zoster, tetanus, Haemophilus infl uenzae type B, and pneumo coccal p=0·006 antigens 1, 3, 4, 5, 6B, 7F, 8, 14, 18C, 19A, 19F, and 23F). We 0·4 did not identify any delayed adverse events during the post- treatment phase (mean 7·0 months, 5·8–10·2). 0·2 After treatment, there was an improvement in log of minimum angle of resolution (logMAR) visual acuity (fi gure 2, table 2) and low contrast visual acuity (table 2, webappendix). No signifi cant changes were evident in colour vision or visual fi elds. Physiological measures −0·2 showed a post-treatment reduction in visual evoked response latency and an increase in visual evoked response −0·4 amplitude; imaging measures showed an increase in optic nerve area after treatment (fi gure 2, table 2). No change was evident in macular volume, retinal nerve fi bre layer −0·6 thickness, or optic nerve magnetisation transfer ratio. Before After There was reduction after treatment in general disability Figure 2: Change in visual function, visual evoked response amplitude, and progression measured by EDSS (table 2). We did not optic nerve area identify a change in the MSFC or in measures of Paired monthly estimated rates of change in log of minimum angle of depression, cognition, and self-reported eff ect of MS on resolution (logMAR) visual acuity, whole-fi eld visual evoked response latency, daily living. T1 hypointense lesion volume decreased after and optic nerve area are shown for individual patients before and after treatment connected by solid lines. Pretreatment and post-treatment mean treatment and magnetisation transfer ratio increased, but rates of change are also shown connected with a dashed line. Signifi cance these changes were not statistically signifi cant. We did tests are shown for a diff erence between mean rates of change before and not identify any changes in the rate of T2 lesion after treatment. accumulation or general brain atrophy after treatment. Uhthoff ’s phenomenon. Two patients described a single clinical relapse event in the pretreatment phase, neither Discussion of which involved the anterior visual pathway. One Our proof-of-concept study provides evidence that an patient had been previously treated with disease intervention might aff ect the disease course in pro- modifying therapy (interferon beta for 1 year, with gressive MS. Specifi cally, we show that after intravenous www.thelancet.com/neurology Vol 11 February 2012 153 Optic nerve area (mm per month) Visual evoked response latency (ms per month) Visual acuity (logMAR units per month) Articles Further more, assessing neuroprotective therapies in Rate of change Diff erence in rate of change p value MS presents a substantial challenge because of the after treatment (95% CI) variability in disease features and course, combined Before After with insensitivity of generic clinical outcomes. We treatment treatment therefore adopted a sentinel lesion approach based on Vision a detailed assessment of the anterior visual pathway as Visual acuity (logMAR) 0·0050 –0·0207 –0·0205 (–0·0325 to –0·0085) 0·003 a model of wider processes. We chose the anterior 25% contrast acuity (logMAR) 0·0022 –0·0207 –0·0202 (–0·0330 to –0·0073) 0·011 visual pathway because of convergence of reliable and 5% contrast acuity (logMAR) 0·0083 –0·0372 –0·0371 (–0·0560 to –0·0181) 0·001 validated outcomes for clinical function, physiology, 1·25% contrast acuity (logMAR) 0·0063 –0·0370 –0·0369 (–0·0552 to –0·0185) 0·001 and structure. Nevertheless, because of wide variation Colour vision (Farnsworth–Munsell 0·1017 –0·0975 –0·1011 (–0·2567 to 0·0544) 0·070 between individuals in the rate of disease progression, 100-hue test √total error score) a further challenge in testing advanced therapies such Visual fi eld (mean deviance) 0·0395 0·00311 0·0192 (–0·1062 to 0·1445) 0·893 as cell-based interventions is to design early stage trials Full fi eld visual evoked response 0·4843 –0·8438 –1·3280 (–2·4447 to –0·2114) 0·020 that achieve adequate power. On this basis, we used a latency (ms) pretest–post-test design to increase eff ect size and Full fi eld visual evoked response –0·1084 0·1503 0·2587 (0·0705 to 0·4469) 0·007 amplitude (μV) therefore increase statistical power by 40–80%. The Macular volume (mm³) 0·0002 0·0041 0·0040 (–0·0135 to 0·0214) 0·654 limitations of this approach are that change evident Retinal nerve fi bre layer thickness –0·0052 0·0474 0·0527 (–0·3533 to 0·4586) 0·799 after treatment cannot be attributed exclusively to the (μm) eff ects of treatment since factors we did not record Optic nerve area (mm²) –0·0216 0·1046 0·1262 (0·0368 to 0·2155) 0·006 might also contribute. Changes identifi ed after Optic nerve magnetisation transfer 0·0656 0·0529 0·0529 (–0·1271 to 0·2328) 0·565 treatment therefore need to be confi rmed as treatment ratio (pu) eff ects by replication in trials with random allocation General between comparator groups. Such trials require Expanded disability status scale 0·0257 –0·0012 –0·0269 (–0·0431 to –0·0107) 0·028 substantial investment, and feasibility, safety, and Multiple sclerosis functional –0·0217 0·0141 0·0359 (–0·0275 to 0·0992) 0·267 eff ect-size-defi ning studies such as ours therefore have composite (Z score) a key role in informing decisions about whether further Addenbrooke’s cognitive 0·0492 0·2690 0·2198 (–0·1343 to 0·5739) 0·224 studies are justifi able and how they should be designed. examination (revised) Further limitations of our study include the small 29-item multiple sclerosis impact –0·3710 –0·5152 –0·1443 (–2·0865 to 1·7979) 0·884 cohort size and lack of masking for clinical outcomes. scale There is also risk of type I error due to multiple Beck depression inventory II 0·0965 –0·2663 –0·3628 (–0·9378 to 0·2121) 0·216 statistical comparisons; our results should therefore be T1 lesion volume (mm³) 204·35 –60·73 –265·08 (–574·85 to 44·69) 0·094 regarded as hypothesis generating and will need T1 lesion magnetisation transfer –0·1867 0·5791 0·7659 (–0·1389 to 1·6706) 0·097 ratio (pu) confi rmation in future studies. Nevertheless, inter- T2 lesion volume (mm³) 155·89 20·90 –134·98 (–579·64 to 309·67) 0·552 pretation of post-treatment changes is aided by masked T2 lesion magnetisation transfer –0·1738 0·3859 0·5597 (–0·2703 to 1·3896) 0·186 electrophysiological and imaging outcomes that are ratio (pu) probably resistant to observer bias or placebo eff ects. Total brain volume (%) –0·0880 –0·1470 –0·0590 (–0·1434 to 0·0254) 0·171 Moreover, unlike designs in which treatment is started immediately after recruitment, post-treatment Data are units per month unless otherwise stated. MAR=minimum angle of resolution. pu=percent units. change in our study is robust to regression to the mean Table 2: Effi cacy outcomes because of our prolonged pre treatment assessment phase. administration of autologous mesenchymal stem cells, We do not know the precise mechanism by which patients with secondary progressive disease improved on mesenchymal stem cells might act in our study. measures of visual function, physiology, and structure However, the fi ndings from our masked analyses without evidence of signifi cant adverse events. Improve- showed an increase in optic nerve area and reduction in ments in visual acuity and contrast sensitivity after visual evoked response latency that are consistent with treatment were accompanied by changes in masked a neuroprotective eff ect because of the promotion of outcome measures, as a reduction in visual evoked myelin repair. Our fi ndings of a possible reduction in response latency, increase in visual evoked response brain T1 hypointense lesion volume and an increase in amplitude, and an increase in optic nerve area. General brain T1 lesion magnetisation transfer ratio provide 29,30 disability progression measured by EDSS was also indirect support for this idea. Remyelination is the reduced after treatment. regenerative process by which myelin sheaths are Despite recent major advances in immunomodulatory restored to demyelinated axons and the failure of therapies, there are no treatments to slow, stop, or remyelination is implicated in the neuronal and axonal reverse the accumulation of fi xed disability in secondary loss that underlies progressive disability. Although our progressive MS. This relates in part to the complex and study was not designed specifi cally to address the eff ects incompletely understood biology of progression. of intervention on infl ammatory MRI metrics, we did 154 www.thelancet.com/neurology Vol 11 February 2012 Articles preparation. RP was involved in participant assessments. S-LL and Panel: Research in context M-QD developed and undertook comparative genomic hybridisation analysis techniques. Systematic review Confl icts of interest We searched Medline (1950 to August, 2011), Embase We declare that we have no confl icts of interest. (1980 to August, 2011), and the Cochrane Central Register of Acknowledgments Controlled Trials (The Cochrane Library issue 4, 2011) with Interpretation of optic nerve imaging was done by Katherine Miszkiel. the terms “multiple sclerosis” and “mesenchymal stem cells” Gordon Plant was also involved in initial planning. Andrew Henderson was involved in the optical coherence tomography training. for clinical trials published up to August, 2011, that report the Michael Sharpe, Peter Sandercock, and Gordon Smith provided eff ect of mesenchymal stem cells on the rate of disease comments on the report during preparation. progression in secondary progressive multiple sclerosis. We References did not limit our search by language. We identifi ed three 1 Pugliatti M, Rosati G, Carton H, et al. The epidemiology of multiple published trials in which autologous ex-vivo expanded sclerosis in Europe. Eur J Neurol 2006; 13: 700–22. mesenchymal stem cells were administered intrathecally, 2 Compston A, Coles A. Multiple sclerosis. Lancet 2008; 372: 1502–17. 3 Coles AJ, Compston DAS, Selmaj KW, et al. Alemtuzumab vs intrathecally and intracisternally, or intrathecally and interferon beta-1a in early multiple sclerosis. N Engl J Med 2008; intravenously (webappendix). 359: 1786–801. 4 Kappos L, Freedman MS, Polman CH, et al. Eff ect of early versus Interpretation delayed interferon beta-1b treatment on disability after a fi rst Consistent with previous studies, our fi ndings show that clinical event suggestive of multiple sclerosis: a 3-year follow-up analysis of the BENEFIT study. Lancet 2007; 370: 389–97. intravenous administration of autologous mesenchymal 5 Coles A. The curious incident of disability in multiple sclerosis stem cells to patients with secondary progressive multiple trials. Lancet Neurol 2006; 5: 899–900. sclerosis is feasible and safe. Our fi ndings also suggest 6 Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of structural, functional, and physiological improvement after adult human mesenchymal stem cells. Science 1999; 284: 143–47. 7 Arthur A, Zannettino A, Gronthos S. The therapeutic applications treatment consistent with neuroprotection. of multipotential mesenchymal/stromal stem cells in skeletal tissue repair. J Cell Physiol 2009; 218: 237–45. 8 Uccelli A, Laroni A, Freedman MS. Mesenchymal stem cells for the not identify any change in the rate of T2 lesion treatment of multiple sclerosis and other neurological diseases. accumulation after treatment. Similarly, the lack of Lancet Neurol 2011; 10: 649–56. eff ect on optical coherence tomography measures after 9 Payne N, Siatskas C, Barnard A, Bernard CCA. The prospect of stem cells as multi-faceted purveyors of immune modulation, repair treatment supports a view that structural change in and regeneration in multiple sclerosis. Curr Stem Cell Res Ther 2011; unmyelinated axons was not a signifi cant factor. The 6: 50–62. lack of signifi cant change in optic nerve magnetisation 10 Zappia E, Casazza S, Pedemonte E, et al. Mesenchymal stem cells ameliorate experimental autoimmune encephalomyelitis inducing transfer ratio measures suggests that the clinical and T-cell anergy. Blood 2005; 106: 1755–61. neuro physiological improvements might relate to 11 Zhang J, Li Y, Lu M, et al. Bone marrow stromal cells reduce axonal diff use tissue repair in the diseased anterior visual loss in experimental autoimmune encephalomyelitis mice. pathways or to our small study size and technical J Neurosci Res 2006; 84: 587–95. 12 Bai L, Lennon DP, Eaton V, et al. Human bone marrow-derived limitations, in view of the restricted previous experience mesenchymal stem cells induce Th2-polarized immune response of optic nerve magnetisation transfer ratio at 3 T. Taken and promote endogenous repair in animal models of multiple together our fi ndings are consistent with a growing sclerosis. Glia 2009; 57: 1192–203. 13 Gerdoni E, Gallo B, Casazza S, et al. Mesenchymal stem cells body of published work in acute and chronic models of eff ectively modulate pathogenic immune response in experimental MS showing neuroprotective eff ects of mesenchymal autoimmune encephalomyelitis. Ann Neurol 2007; 61: 219–27. stem cells independent of directed diff erentiation or 14 Kassis I, Grigoriadis N, Gowda-Kurkalli B, et al. Neuroprotection 8,9 and immunomodulation with mesenchymal stem cells in chronic cell replacement (panel). Central or peripheral experimental autoimmune encephalomyelitis. Arch Neurol 2008; mechanisms that might explain these results include 65: 753–61. immuno regulation, and modifi cation of the cellular 15 Le Blanc K, Frassoni F, Ball L, et al. Mesenchymal stem cells for treatment of steroid-resistant, severe, acute graft-versus-host environment causing trophic or anti-infl ammatory disease: a phase II study. Lancet 2008; 371: 1579–86. eff ects. Further more, recent studies suggest that 16 Sun L, Akiyama K, Zhang H, et al. Mesenchymal stem cell mesenchymal stem cells can promote the endogenous transplantation reverses multiorgan dysfunction in systemic lupus CNS repair processes of oligo dendrogenesis and erythematosus mice and humans. Stem Cells 2009; 27: 1421–32. 32,33 17 Carrion F, Nova E, Ruiz C, et al. Autologous mesenchymal stem cell remyelination, raising the possibility that re- treatment increased T regulatory cells with no eff ect on disease activity myelination underlies the evidence for neuroprotection in two systemic lupus erythematosus patients. Lupus 2010; 19: 317–22. in our study. 18 Mohyeddin Bonab M, Yazdanbakhsh S, Lotfi J, et al. Does mesenchymal stem cell therapy help multiple sclerosis patients? Contributors Report of a pilot study. Iran J Immunol 2007; 4: 50–57. SC, DHM, AC, and AJT were involved in the overall design of the 19 Yamout B, Hourani R, Salti H, et al. Bone marrow mesenchymal study. DRA was involved in statistical aspects of trial design and stem cell transplantation in patients with multiple sclerosis: a pilot analysis. MAS and CC were involved in design and execution of study. J Neuroimmunol 2010; 227: 185–89. mesenchymal stem cell isolation, expansion, storage, and 20 Karussis D, Karageorgiou C, Vaknin-Dembinsky A, et al. Safety and administration. DJW assisted in characterisation of the mesenchymal immunological eff ects of mesenchymal stem cell transplantation in stem cells. AWM did masked assessments of visual evoked potentials. patients with multiple sclerosis and amyotrophic lateral sclerosis. PC and MK were involved in all aspects of trial execution and report Arch Neurol 2010; 67: 1187–94. www.thelancet.com/neurology Vol 11 February 2012 155 Articles 21 Mehta LR, Schwid SR, Arnold DL, et al. Proof of concept studies for 27 Oberg AL, Mahoney DW. Linear mixed eff ects models. tissue-protective agents in multiple sclerosis. Mult Scler 2009; Methods Mol Biol 2007; 404: 213–34. 15: 542–46. 28 Lublin FD, Reingold SC. Defi ning the clinical course of multiple 22 Kolappan M, Henderson APD, Jenkins TM, et al. Assessing sclerosis: results of an international survey. National Multiple structure and function of the aff erent visual pathway in multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of sclerosis and associated optic neuritis. J Neurol 2009; 256: 305–19. New Agents in Multiple Sclerosis. Neurology 1996; 46: 907–11. 23 Connick P, Kolappan M, Patani R, et al. The mesenchymal stem 29 Barkhof F, Bruck W, De Groot CJA, et al. Remyelinated lesions in cells in multiple sclerosis (MSCIMS) trial protocol and baseline multiple sclerosis: magnetic resonance image appearance. cohort characteristics: an open-label pre-test: post-test study with Arch Neurol 2003; 60: 1073–81. blinded outcome assessments. Trials 2011; 12: 62. 30 Schmierer K, Scaravilli F, Altmann DR, Barker GJ, Miller DH. 24 Poser CM, Paty DW, Scheinberg L, et al. New diagnostic criteria for Magnetization transfer ratio and myelin in postmortem multiple multiple sclerosis: guidelines for research protocols. Ann Neurol sclerosis brain. Ann Neurol 2004; 56: 407–15. 1983; 13: 227–31. 31 Franklin RJM, Ffrench-Constant C. Remyelination in the CNS: 25 Dominici M, Le Blanc K, Mueller I, et al. Minimal criteria for from biology to therapy. Nat Rev Neurosci 2008; 9: 839–55. defi ning multipotent mesenchymal stromal cells. The International 32 Rivera FJ, Couillard-Despres S, Pedre X, et al. Mesenchymal stem Society for Cellular Therapy position statement. Cytotherapy 2006; cells instruct oligodendrogenic fate decision on adult neural stem 8: 315–17. cells. Stem Cells 2006; 24: 2209–19. 26 Johnson NA, Hamoudi RA, Ichimura K, et al. Application of array 33 Bai L, Caplan A, Lennon D, Miller RH. Human mesenchymal stem CGH on archival formalin-fi xed paraffi n-embedded tissues cells signals regulate neural stem cell fate. Neurochem Res 2007; including small numbers of microdissected cells. Lab Invest 2006; 32: 353–62. 86: 968–78. 156 www.thelancet.com/neurology Vol 11 February 2012 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png The Lancet Neurology Unpaywall

Autologous mesenchymal stem cells for the treatment of secondary progressive multiple sclerosis: an open-label phase 2a proof-of-concept study

Download PDF
7 pages

Loading next page...
 
/lp/unpaywall/autologous-mesenchymal-stem-cells-for-the-treatment-of-secondary-aCIIWEeHre

References (38)

Publisher
Unpaywall
ISSN
1474-4422
DOI
10.1016/s1474-4422(11)70305-2
Publisher site
See Article on Publisher Site

Abstract

Articles Autologous mesenchymal stem cells for the treatment of secondary progressive multiple sclerosis: an open-label phase 2a proof-of-concept study Peter Connick,* Madhan Kolappan,* Charles Crawley, Daniel J Webber, Rickie Patani, Andrew W Michell, Ming-Qing Du, Shi-Lu Luan, Daniel R Altmann, Alan J Thompson, Alastair Compston, Michael A Scott, David H Miller, Siddharthan Chandran Summary Lancet Neurol 2012; 11: 150–56 Background More than half of patients with multiple sclerosis have progressive disease characterised by accumulating Published Online disability. The absence of treatments for progressive multiple sclerosis represents a major unmet clinical need. On January 10, 2012 the basis of evidence that mesenchymal stem cells have a benefi cial eff ect in acute and chronic animal models of DOI:10.1016/S1474- multiple sclerosis, we aimed to assess the safety and effi cacy of these cells as a potential neuroprotective treatment for 4422(11)70305-2 secondary progressive multiple sclerosis. See Comment page 123 *These authors contributed Methods Patients with secondary progressive multiple sclerosis involving the visual pathways (expanded disability equally status score 5·5–6·5) were recruited from the East Anglia and north London regions of the UK. Participants received Department of Clinical intravenous infusion of autologous bone-marrow-derived mesenchymal stem cells in this open-label study. Our Neurosciences (P Connick MRCP), Anne MacLaren Laboratory for primary objective was to assess feasibility and safety; we compared adverse events from up to 20 months before Regenerative Medicine treatment until up to 10 months after the infusion. As a secondary objective, we chose effi cacy outcomes to assess the (D J Webber PhD, R Patani MRCP), anterior visual pathway as a model of wider disease. Masked endpoint analyses was used for electrophysiological and Department of Clinical selected imaging outcomes. We used piecewise linear mixed models to assess the change in gradients over time at the Neurosciences (A W Michell PhD), Department of Pathology point of intervention. This trial is registered with ClinicalTrials.gov, number NCT00395200. (Prof M-Q Du PhD), Department of Veterinary Medicine Findings We isolated, expanded, characterised, and administered mesenchymal stem cells in ten patients. The mean (S-L Luan PhD), and Department dose was 1·6×10⁶ cells per kg bodyweight (range 1·1–2·0). One patient developed a transient rash shortly after of Clinical Neurosciences (Prof A Compston FMedSci), treatment; two patients had self-limiting bacterial infections 3–4 weeks after treatment. We did not identify any University of Cambridge, serious adverse events. We noted improvement after treatment in visual acuity (diff erence in monthly rates of change Cambridge, UK; NMR Research –0·02 logMAR units, 95% CI –0·03 to –0·01; p=0·003) and visual evoked response latency (–1·33 ms, –2·44 to –0·21; Unit, Institute of Neurology, p=0·020), with an increase in optic nerve area (diff erence in monthly rates of change 0·13 mm², 0·04 to 0·22; Department of Neuroinfl ammation, University p=0·006). We did not identify any signifi cant eff ects on colour vision, visual fi elds, macular volume, retinal nerve College London, London, UK fi bre layer thickness, or optic nerve magnetisation transfer ratio. (M Kolappan MRCP, Prof D H Miller FMedSci); Blood and Marrow Transplant Unit, Interpretation Autologous mesenchymal stem cells were safely given to patients with secondary progressive multiple Addenbrooke’s Hospital, sclerosis in our study. The evidence of structural, functional, and physiological improvement after treatment in some Cambridge, UK (C Crawley FRCP, visual endpoints is suggestive of neuroprotection. M A Scott PhD); Medical Statistics Department, London School of Hygiene and Tropical Funding Medical Research Council, Multiple Sclerosis Society of Great Britain and Northern Ireland, Evelyn Trust, Medicine, London, UK NHS National Institute for Health Research, Cambridge and UCLH Biomedical Research Centres, Wellcome Trust, (D R Altmann PhD); Department Raymond and Beverly Sackler Foundation, and Sir David and Isobel Walker Trust. of Brain Repair and Rehabilitation, University College London, London, UK Introduction Multipotent mesenchymal stromal cells are bone- (Prof A J Thompson FRCP); and Multiple sclerosis (MS) aff ects more than 2 million people marrow cells that can be expanded ex vivo and will readily Centre for Clinical Brain Sciences, worldwide and is the most common non-traumatic cause diff erentiate into mesodermal cell derivatives. In addition MRC Centre for Regenerative, of disability in young (<50 years) European adults. It is a to tissue engineering applications that target the repair of Medicine University of Edinburgh, Edinburgh, UK multifocal CNS disorder that has two distinct clinical phases skeletal tissue defects, biological properties independent (Prof S Chandran PhD) corresponding to inter-related pathological processes: focal of diff erentiation suggest that mesenchymal stem cells Correspondence to: infl ammation that drives activity during the relapse-remitting could have a therapeutic role through strategies other Prof Siddharthan Chandran, 8 stage and neuro degeneration that underlies progressive than tissue replacement in diseases such as MS. These Centre for Clinical Brain Sciences, disease char acterised by accumulating fi xed disability. strategies include neuroprotection through paracrine MRC Centre for Regenerative Medicine, Centre for Although important advances in treatment to reduce eff ects on the CNS microenvironment, augmentation of 3,4 Neuroregeneration, University of relapse rate have been made in the past two decades, no endogenous axonal and myelin repair processes, and Edinburgh, Chancellor’s Building, 8,9 treatments are available for the roughly half of patients immune regulatory activity. Increasing evidence shows Edinburgh EH16 4SB, UK with MS who have progressive disease. There is therefore both neuroprotection and functional improvement after siddharthan.chandran@ed. ac.uk a great and unmet clinical need for the development of infusion with mesenchymal stem cells in mouse models 10–14 neuroprotective treatments. of relapsing-remitting and chronic MS. 150 www.thelancet.com/neurology Vol 11 February 2012 Articles Clinically, mesenchymal stem cells have been used in (Hyclone; Perbio Science, Northumberland, UK), plated at the treatment of immune-mediated human diseases a density of 1×10⁸ cells per cell factory (Nunc, Thermo including steroid-resistant graft-versus-host disease and Scientifi c, Northumberland, UK). Near confl uent cultures 15–17 systemic lupus erythematosus. Three recent reports (>80%) were treated with 0·25% trypsin-EDTA (Invitrogen) have also described the use of intrathecally delivered and replated at 3·5×10⁶ cells per cell factory. Mesenchymal autologous mesenchymal stem cells in MS without adverse stem cells were harvested and cryopreserved in 4·5% events or signifi cant changes in general clinical outcomes human albumin solution (Bio Products Laboratory, 18–20 (webappendix). However, assessment of neuroprotection Hertfordshire, UK) with di methyl sulphoxide (Origen See Online for webappendix in the context of MS is challenging because of clinical and Biomedical Inc, Helsingborg, Sweden) at a fi nal pathological heterogeneity. To increase sensitivity for concentration of 10%. Mesenchymal stem cells were then structural and functional treat ment eff ects, the use of characterised in accordance with International Society of eligibility criteria that select cohorts with specifi c and Cellular Therapy recommen dations. Briefl y, this included clinically eloquent lesions, such as those of the anterior evidence of trilineage diff erentiation potential (adipocyte, visual pathway, enables assess ment of tailored and detailed chondrocyte, osteocyte) and fl ow cytometry to confi rm 22,23 outcomes. By use of this approach, we aimed to compare expression of CD73, CD90, and CD105 surface molecules safety and effi cacy outcomes for patients with secondary (>95%) and absence of CD34, CD45, CD14, and CD3 progressive MS before and after intravenous treatment (<2%). Release criteria for clinical use included absence of with autologous mesenchymal stem cells. contamination by pathogens (as documented by aerobic and anaerobic cultures and mycoplasma testing), and lack Methods of any genomic copy number changes as assessed with Participants 1-Mb-resolution bacterial artifi cial chromosome array Between November, 2007, and August, 2010, we did an comparative genomic hybridisation. open-label phase 2a proof-of-concept study involving We administered autologous mesenchymal stem cells participants recruited from the East Anglia and north intravenously to patients with secondary progressive MS. London regions of the UK (identifi ed from MS and Administration of the cells was done as a day-case general neurology clinics). We screened patients for procedure. To reduce type I hypersensitivity reactions, eligibility between November, 2007, and June, 2009. premedication with 10 mg chlor pheniramine, 100 mg Eligible participants were those aged 18–65 years with hydrocortisone, and 10 mg metoclopramide was given clinically defi nite MS according to the Poser criteria, 30 min before administration of the cells. Cryo preserved an expanded disability status scale (EDSS) score of cells were thawed (≤4 min) and im mediately infused over 2·0–6·5, clinical evidence of optic nerve involvement 15 min through a peripheral venous cannula. After (defi ned as a history of optic neuritis, Uhthoff ’s administration of cell suspensions, we infused normal phenomenon, or optic atrophy on examination), saline (500 mL) over 4 h. abnormal visual evoked potentials from one or both Our primary objective was to assess feasibility and safety; eyes consistent with de myelination, a retinal nerve fi bre our secondary objectives were to assess effi cacy on clinical, layer thickness of at least 45 μm in one eye, a T2 lesion electrophysiological, and structural outcomes, in addition on MRI of the optic nerve (webappendix), and the to providing information on the mechanism of any capacity to give consent. Patients were excluded if they recorded eff ect. We used a sentinel lesion approach based had a bleeding disorder, had received interferon beta or on the diseased anterior visual pathway to increase power glatiramer acetate within 6 months of trial entry, or had to detect treatment eff ects, and we used a pretest–post- previously used other disease modifying therapies at test design to compare adverse events and effi cacy any point. All patients gave written informed consent measures before and after the intervention. We assessed before study entry and approval was obtained from the participants at 3–6 month intervals for at least 12 months local ethics committee (Cambridgeshire 2 regional before and 6 months after treatment (webappendix). ethics committee). Assessment at each timepoint was split into two visits with a gap of less than 2 weeks; clinical assessment and visual Procedures evoked responses were done in Cambridge, UK, and MRI, We generated clinical-grade mesenchymal stem cells optical coherence tomography, and neuro-ophthal- under good manufacturing practice conditions with mological assessments were done at the University College standard operating procedures. Briefl y, we separated London Institute of Neurology (London, UK). Clinical bone-marrow mononuclear cells by density gradient assessment involved neurological and medical history with centrifugation in Ficoll-Paque Premium (GE Healthcare recording of adverse events and scores on the EDSS, MS UK Ltd, Buckinghamshire, UK). We resuspended the functional composite (MSFC), Addenbrooke’s cognitive washed cells in phosphate-buff ered saline/EDTA (Miltenyi examination revised, 29-item MS impact scale, and Beck Biotec Ltd, Surrey, UK) and cultured them in Dulbecco’s depression inventory II. Whole and central fi eld modifi ed Eagle’s medium (low glucose; Invitrogen, checkerboard pattern-reversal visual evoked responses Paisley, UK) supplemented with 10% fetal bovine serum were recorded with reversal achromatic checks subtending www.thelancet.com/neurology Vol 11 February 2012 151 Articles 60´ at the eye. Neuro-opthalmological assessment included Visual evoked responses, optic nerve area, optic nerve visual acuity with a retroilluminated early treatment magnetisation transfer ratio, and optic nerve diff usion diabetic retinopathy study chart, contrast acuity with tensor imaging based outcomes were assessed by a retroilluminated Sloan charts, colour vision with the single observer (MK) from whom participant status Farnsworth–Munsell 100-hue test, and visual fi eld (before or after treatment) was masked. Lesional analysis assessment by automated static perimetry (Humphrey was done after image acquisition at each visit. Brain fi eld analyser, 30-2 protocol). Optical coherence tomography volume and whole-brain magnetisation transfer ratio images were acquired by a single operator (MK) with a were done with automated methods with minimal time domain optical coherence tomograph (Stratus OCT manual corrections. Model 3000; Carl Zeiss Meditec, Dublin, CA, USA). MRI images were acquired with a Magnetom 3·0 T Statistical analysis Tim Trio scanner (Siemens, Erlangen, Germany) with a We used piecewise linear mixed models to assess, for a 12-element receiver head coil. Several MRI-based given measure, the change in gradient over time at the measures were assessed: optic nerve cross-sectional area; point of intervention; the given measure was the response optic nerve diff usion tensor imaging measures of variable, with the time from intervention and the time fractional anisotropy, mean diff usivity, axial diff usivity, multiplied by an after-intervention interaction term as and radial diff usivity; optic nerve magnetisation the two predictors. Such models allow estimation of the transfer ratio; whole-brain T2 lesion volume; whole-brain gradients before and after intervention, and of the T1 hypointense lesion volume; whole-brain magnetisa- gradient change with its statistical signifi cance. For tion transfer ratio; brain T1 hypointense lesion analyses of data involving separate values for each eye magnetisation transfer ratio; and brain T2 lesion over time, we added an additional level to the model with magnetisation transfer ratio. Intersessional stability of individual eyes as levels within participants. For the imaging measures was confi rmed by contemporary EDSS score, although the before and after gradients were assessment of ten locally recruited healthy volunteers. estimated as above, the test of gradient change used the non-parametric Wilcoxon sign rank test to compare the two ratios: change in EDSS score before or time interval before versus change in score after or time interval after. 98 patients assessed for eligibility There was no evidence of deviation from model assumptions. In particular there was no evidence for 84 ineligible non-normality or heteroscedasticity of residuals, or evidence against linearity assumptions. All of the 14 eligible reported mixed models achieved convergence with estimates for both the variance components and the fi xed 3 declined to participate eff ects. Unrecordable visual evoked responses, related to severe dysfunction due to disease, were represented by 11 participants recruited amplitude values of 0 μV and latency values of 180 ms (the maximum recorded during our study). Analyses were done with Stata SE (versions 9.2 and 11). Power 1 withdrew consent calculations could not be done before the study because of the lack of information from previous studies on 10 underwent bone-marrow aspiration, had mesenchymal stem potential eff ect sizes. This trial is registered with cells cultured to target dose and cryopreserved, and were treated with the cells ClinicalTrials.gov, number NCT00395200. Figure 1: Study profi le Role of the funding source The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or Measure at recruitment writing of the report. The corresponding author had full Number of participants 10 access to all the data in the study and had fi nal Sex ratio (men:women) 7:3 responsibility for the decision to submit for publication. Age (years) 48·8 (4·1; 40–53) Duration of multiple sclerosis (years) 14·4 (7·9; 5–26) Results Expanded disability status score 6·1 (0·3; 5·5–6·5) Figure 1 shows the study profi le and table 1 the Time since last clinical episode of optic neuritis 11·9 (8·2; 2–26) participants’ characteristics. All participants had (years)* secondary progressive MS, with clinical and electro- Data are n or mean (SD; range). *Nine participants. physiological evidence of optic nerve involvement. In the 2–26 years before recruitment, nine patients had Table 1: Participants’ characteristics clinical optic neuritis (three bilateral) and one had 152 www.thelancet.com/neurology Vol 11 February 2012 Articles treatment discontinued owing to disease progression p=0·003 2 years before recruitment into our trial). We successfully isolated and cultured mesenchymal −0·05 stem cells to the target dose from all bone-marrow aspirates (mean total cultured dose 2·0×10⁶ cells per kg, range 1·1×10⁶–3·7×10⁶). Mean culture duration was 24 days (20–30). Patients received a single infusion of autologous cells after monitoring for a mean of 17·3 months (14·1–20·9) during the pretreatment phase. The mean administered dose was 1·6×10⁶ cells per kg bodyweight 0·05 (1·1×10⁶–2·0×10⁶); mean volume of cell suspensions was 167·2 mL (range 89–246). We did not record any adverse events during infusion. One patient developed a macular rash over the anterior chest at about 3 h after the start of 0·10 infusion that resolved spontaneously over 12 h; a further B patient described scalp pruritus beginning 1 week after p=0·016 treatment and resolving spontaneously 2 weeks later. Two patients had infections: a self-limiting upper-respiratory tract infection 3 weeks after infusion (not requiring treatment) and an Escherichia coli urinary-tract infection 4 weeks after infusion (treated with oral antibiotics). Results of weekly blood testing of clinical chemistry, haematology, and immunology during the 4 weeks after infusion was unremarkable. Compared with pre treatment titres, no changes were evident in the post-treatment −2 period for T-cell subset counts (CD3, CD4, CD8, CD19, and CD56) or humoral immunity assessed by titres to common −4 antigens (mumps, measles, rubella, varicella zoster, tetanus, Haemophilus infl uenzae type B, and pneumo coccal p=0·006 antigens 1, 3, 4, 5, 6B, 7F, 8, 14, 18C, 19A, 19F, and 23F). We 0·4 did not identify any delayed adverse events during the post- treatment phase (mean 7·0 months, 5·8–10·2). 0·2 After treatment, there was an improvement in log of minimum angle of resolution (logMAR) visual acuity (fi gure 2, table 2) and low contrast visual acuity (table 2, webappendix). No signifi cant changes were evident in colour vision or visual fi elds. Physiological measures −0·2 showed a post-treatment reduction in visual evoked response latency and an increase in visual evoked response −0·4 amplitude; imaging measures showed an increase in optic nerve area after treatment (fi gure 2, table 2). No change was evident in macular volume, retinal nerve fi bre layer −0·6 thickness, or optic nerve magnetisation transfer ratio. Before After There was reduction after treatment in general disability Figure 2: Change in visual function, visual evoked response amplitude, and progression measured by EDSS (table 2). We did not optic nerve area identify a change in the MSFC or in measures of Paired monthly estimated rates of change in log of minimum angle of depression, cognition, and self-reported eff ect of MS on resolution (logMAR) visual acuity, whole-fi eld visual evoked response latency, daily living. T1 hypointense lesion volume decreased after and optic nerve area are shown for individual patients before and after treatment connected by solid lines. Pretreatment and post-treatment mean treatment and magnetisation transfer ratio increased, but rates of change are also shown connected with a dashed line. Signifi cance these changes were not statistically signifi cant. We did tests are shown for a diff erence between mean rates of change before and not identify any changes in the rate of T2 lesion after treatment. accumulation or general brain atrophy after treatment. Uhthoff ’s phenomenon. Two patients described a single clinical relapse event in the pretreatment phase, neither Discussion of which involved the anterior visual pathway. One Our proof-of-concept study provides evidence that an patient had been previously treated with disease intervention might aff ect the disease course in pro- modifying therapy (interferon beta for 1 year, with gressive MS. Specifi cally, we show that after intravenous www.thelancet.com/neurology Vol 11 February 2012 153 Optic nerve area (mm per month) Visual evoked response latency (ms per month) Visual acuity (logMAR units per month) Articles Further more, assessing neuroprotective therapies in Rate of change Diff erence in rate of change p value MS presents a substantial challenge because of the after treatment (95% CI) variability in disease features and course, combined Before After with insensitivity of generic clinical outcomes. We treatment treatment therefore adopted a sentinel lesion approach based on Vision a detailed assessment of the anterior visual pathway as Visual acuity (logMAR) 0·0050 –0·0207 –0·0205 (–0·0325 to –0·0085) 0·003 a model of wider processes. We chose the anterior 25% contrast acuity (logMAR) 0·0022 –0·0207 –0·0202 (–0·0330 to –0·0073) 0·011 visual pathway because of convergence of reliable and 5% contrast acuity (logMAR) 0·0083 –0·0372 –0·0371 (–0·0560 to –0·0181) 0·001 validated outcomes for clinical function, physiology, 1·25% contrast acuity (logMAR) 0·0063 –0·0370 –0·0369 (–0·0552 to –0·0185) 0·001 and structure. Nevertheless, because of wide variation Colour vision (Farnsworth–Munsell 0·1017 –0·0975 –0·1011 (–0·2567 to 0·0544) 0·070 between individuals in the rate of disease progression, 100-hue test √total error score) a further challenge in testing advanced therapies such Visual fi eld (mean deviance) 0·0395 0·00311 0·0192 (–0·1062 to 0·1445) 0·893 as cell-based interventions is to design early stage trials Full fi eld visual evoked response 0·4843 –0·8438 –1·3280 (–2·4447 to –0·2114) 0·020 that achieve adequate power. On this basis, we used a latency (ms) pretest–post-test design to increase eff ect size and Full fi eld visual evoked response –0·1084 0·1503 0·2587 (0·0705 to 0·4469) 0·007 amplitude (μV) therefore increase statistical power by 40–80%. The Macular volume (mm³) 0·0002 0·0041 0·0040 (–0·0135 to 0·0214) 0·654 limitations of this approach are that change evident Retinal nerve fi bre layer thickness –0·0052 0·0474 0·0527 (–0·3533 to 0·4586) 0·799 after treatment cannot be attributed exclusively to the (μm) eff ects of treatment since factors we did not record Optic nerve area (mm²) –0·0216 0·1046 0·1262 (0·0368 to 0·2155) 0·006 might also contribute. Changes identifi ed after Optic nerve magnetisation transfer 0·0656 0·0529 0·0529 (–0·1271 to 0·2328) 0·565 treatment therefore need to be confi rmed as treatment ratio (pu) eff ects by replication in trials with random allocation General between comparator groups. Such trials require Expanded disability status scale 0·0257 –0·0012 –0·0269 (–0·0431 to –0·0107) 0·028 substantial investment, and feasibility, safety, and Multiple sclerosis functional –0·0217 0·0141 0·0359 (–0·0275 to 0·0992) 0·267 eff ect-size-defi ning studies such as ours therefore have composite (Z score) a key role in informing decisions about whether further Addenbrooke’s cognitive 0·0492 0·2690 0·2198 (–0·1343 to 0·5739) 0·224 studies are justifi able and how they should be designed. examination (revised) Further limitations of our study include the small 29-item multiple sclerosis impact –0·3710 –0·5152 –0·1443 (–2·0865 to 1·7979) 0·884 cohort size and lack of masking for clinical outcomes. scale There is also risk of type I error due to multiple Beck depression inventory II 0·0965 –0·2663 –0·3628 (–0·9378 to 0·2121) 0·216 statistical comparisons; our results should therefore be T1 lesion volume (mm³) 204·35 –60·73 –265·08 (–574·85 to 44·69) 0·094 regarded as hypothesis generating and will need T1 lesion magnetisation transfer –0·1867 0·5791 0·7659 (–0·1389 to 1·6706) 0·097 ratio (pu) confi rmation in future studies. Nevertheless, inter- T2 lesion volume (mm³) 155·89 20·90 –134·98 (–579·64 to 309·67) 0·552 pretation of post-treatment changes is aided by masked T2 lesion magnetisation transfer –0·1738 0·3859 0·5597 (–0·2703 to 1·3896) 0·186 electrophysiological and imaging outcomes that are ratio (pu) probably resistant to observer bias or placebo eff ects. Total brain volume (%) –0·0880 –0·1470 –0·0590 (–0·1434 to 0·0254) 0·171 Moreover, unlike designs in which treatment is started immediately after recruitment, post-treatment Data are units per month unless otherwise stated. MAR=minimum angle of resolution. pu=percent units. change in our study is robust to regression to the mean Table 2: Effi cacy outcomes because of our prolonged pre treatment assessment phase. administration of autologous mesenchymal stem cells, We do not know the precise mechanism by which patients with secondary progressive disease improved on mesenchymal stem cells might act in our study. measures of visual function, physiology, and structure However, the fi ndings from our masked analyses without evidence of signifi cant adverse events. Improve- showed an increase in optic nerve area and reduction in ments in visual acuity and contrast sensitivity after visual evoked response latency that are consistent with treatment were accompanied by changes in masked a neuroprotective eff ect because of the promotion of outcome measures, as a reduction in visual evoked myelin repair. Our fi ndings of a possible reduction in response latency, increase in visual evoked response brain T1 hypointense lesion volume and an increase in amplitude, and an increase in optic nerve area. General brain T1 lesion magnetisation transfer ratio provide 29,30 disability progression measured by EDSS was also indirect support for this idea. Remyelination is the reduced after treatment. regenerative process by which myelin sheaths are Despite recent major advances in immunomodulatory restored to demyelinated axons and the failure of therapies, there are no treatments to slow, stop, or remyelination is implicated in the neuronal and axonal reverse the accumulation of fi xed disability in secondary loss that underlies progressive disability. Although our progressive MS. This relates in part to the complex and study was not designed specifi cally to address the eff ects incompletely understood biology of progression. of intervention on infl ammatory MRI metrics, we did 154 www.thelancet.com/neurology Vol 11 February 2012 Articles preparation. RP was involved in participant assessments. S-LL and Panel: Research in context M-QD developed and undertook comparative genomic hybridisation analysis techniques. Systematic review Confl icts of interest We searched Medline (1950 to August, 2011), Embase We declare that we have no confl icts of interest. (1980 to August, 2011), and the Cochrane Central Register of Acknowledgments Controlled Trials (The Cochrane Library issue 4, 2011) with Interpretation of optic nerve imaging was done by Katherine Miszkiel. the terms “multiple sclerosis” and “mesenchymal stem cells” Gordon Plant was also involved in initial planning. Andrew Henderson was involved in the optical coherence tomography training. for clinical trials published up to August, 2011, that report the Michael Sharpe, Peter Sandercock, and Gordon Smith provided eff ect of mesenchymal stem cells on the rate of disease comments on the report during preparation. progression in secondary progressive multiple sclerosis. We References did not limit our search by language. We identifi ed three 1 Pugliatti M, Rosati G, Carton H, et al. The epidemiology of multiple published trials in which autologous ex-vivo expanded sclerosis in Europe. Eur J Neurol 2006; 13: 700–22. mesenchymal stem cells were administered intrathecally, 2 Compston A, Coles A. Multiple sclerosis. Lancet 2008; 372: 1502–17. 3 Coles AJ, Compston DAS, Selmaj KW, et al. Alemtuzumab vs intrathecally and intracisternally, or intrathecally and interferon beta-1a in early multiple sclerosis. N Engl J Med 2008; intravenously (webappendix). 359: 1786–801. 4 Kappos L, Freedman MS, Polman CH, et al. Eff ect of early versus Interpretation delayed interferon beta-1b treatment on disability after a fi rst Consistent with previous studies, our fi ndings show that clinical event suggestive of multiple sclerosis: a 3-year follow-up analysis of the BENEFIT study. Lancet 2007; 370: 389–97. intravenous administration of autologous mesenchymal 5 Coles A. The curious incident of disability in multiple sclerosis stem cells to patients with secondary progressive multiple trials. Lancet Neurol 2006; 5: 899–900. sclerosis is feasible and safe. Our fi ndings also suggest 6 Pittenger MF, Mackay AM, Beck SC, et al. Multilineage potential of structural, functional, and physiological improvement after adult human mesenchymal stem cells. Science 1999; 284: 143–47. 7 Arthur A, Zannettino A, Gronthos S. The therapeutic applications treatment consistent with neuroprotection. of multipotential mesenchymal/stromal stem cells in skeletal tissue repair. J Cell Physiol 2009; 218: 237–45. 8 Uccelli A, Laroni A, Freedman MS. Mesenchymal stem cells for the not identify any change in the rate of T2 lesion treatment of multiple sclerosis and other neurological diseases. accumulation after treatment. Similarly, the lack of Lancet Neurol 2011; 10: 649–56. eff ect on optical coherence tomography measures after 9 Payne N, Siatskas C, Barnard A, Bernard CCA. The prospect of stem cells as multi-faceted purveyors of immune modulation, repair treatment supports a view that structural change in and regeneration in multiple sclerosis. Curr Stem Cell Res Ther 2011; unmyelinated axons was not a signifi cant factor. The 6: 50–62. lack of signifi cant change in optic nerve magnetisation 10 Zappia E, Casazza S, Pedemonte E, et al. Mesenchymal stem cells ameliorate experimental autoimmune encephalomyelitis inducing transfer ratio measures suggests that the clinical and T-cell anergy. Blood 2005; 106: 1755–61. neuro physiological improvements might relate to 11 Zhang J, Li Y, Lu M, et al. Bone marrow stromal cells reduce axonal diff use tissue repair in the diseased anterior visual loss in experimental autoimmune encephalomyelitis mice. pathways or to our small study size and technical J Neurosci Res 2006; 84: 587–95. 12 Bai L, Lennon DP, Eaton V, et al. Human bone marrow-derived limitations, in view of the restricted previous experience mesenchymal stem cells induce Th2-polarized immune response of optic nerve magnetisation transfer ratio at 3 T. Taken and promote endogenous repair in animal models of multiple together our fi ndings are consistent with a growing sclerosis. Glia 2009; 57: 1192–203. 13 Gerdoni E, Gallo B, Casazza S, et al. Mesenchymal stem cells body of published work in acute and chronic models of eff ectively modulate pathogenic immune response in experimental MS showing neuroprotective eff ects of mesenchymal autoimmune encephalomyelitis. Ann Neurol 2007; 61: 219–27. stem cells independent of directed diff erentiation or 14 Kassis I, Grigoriadis N, Gowda-Kurkalli B, et al. Neuroprotection 8,9 and immunomodulation with mesenchymal stem cells in chronic cell replacement (panel). Central or peripheral experimental autoimmune encephalomyelitis. Arch Neurol 2008; mechanisms that might explain these results include 65: 753–61. immuno regulation, and modifi cation of the cellular 15 Le Blanc K, Frassoni F, Ball L, et al. Mesenchymal stem cells for treatment of steroid-resistant, severe, acute graft-versus-host environment causing trophic or anti-infl ammatory disease: a phase II study. Lancet 2008; 371: 1579–86. eff ects. Further more, recent studies suggest that 16 Sun L, Akiyama K, Zhang H, et al. Mesenchymal stem cell mesenchymal stem cells can promote the endogenous transplantation reverses multiorgan dysfunction in systemic lupus CNS repair processes of oligo dendrogenesis and erythematosus mice and humans. Stem Cells 2009; 27: 1421–32. 32,33 17 Carrion F, Nova E, Ruiz C, et al. Autologous mesenchymal stem cell remyelination, raising the possibility that re- treatment increased T regulatory cells with no eff ect on disease activity myelination underlies the evidence for neuroprotection in two systemic lupus erythematosus patients. Lupus 2010; 19: 317–22. in our study. 18 Mohyeddin Bonab M, Yazdanbakhsh S, Lotfi J, et al. Does mesenchymal stem cell therapy help multiple sclerosis patients? Contributors Report of a pilot study. Iran J Immunol 2007; 4: 50–57. SC, DHM, AC, and AJT were involved in the overall design of the 19 Yamout B, Hourani R, Salti H, et al. Bone marrow mesenchymal study. DRA was involved in statistical aspects of trial design and stem cell transplantation in patients with multiple sclerosis: a pilot analysis. MAS and CC were involved in design and execution of study. J Neuroimmunol 2010; 227: 185–89. mesenchymal stem cell isolation, expansion, storage, and 20 Karussis D, Karageorgiou C, Vaknin-Dembinsky A, et al. Safety and administration. DJW assisted in characterisation of the mesenchymal immunological eff ects of mesenchymal stem cell transplantation in stem cells. AWM did masked assessments of visual evoked potentials. patients with multiple sclerosis and amyotrophic lateral sclerosis. PC and MK were involved in all aspects of trial execution and report Arch Neurol 2010; 67: 1187–94. www.thelancet.com/neurology Vol 11 February 2012 155 Articles 21 Mehta LR, Schwid SR, Arnold DL, et al. Proof of concept studies for 27 Oberg AL, Mahoney DW. Linear mixed eff ects models. tissue-protective agents in multiple sclerosis. Mult Scler 2009; Methods Mol Biol 2007; 404: 213–34. 15: 542–46. 28 Lublin FD, Reingold SC. Defi ning the clinical course of multiple 22 Kolappan M, Henderson APD, Jenkins TM, et al. Assessing sclerosis: results of an international survey. National Multiple structure and function of the aff erent visual pathway in multiple Sclerosis Society (USA) Advisory Committee on Clinical Trials of sclerosis and associated optic neuritis. J Neurol 2009; 256: 305–19. New Agents in Multiple Sclerosis. Neurology 1996; 46: 907–11. 23 Connick P, Kolappan M, Patani R, et al. The mesenchymal stem 29 Barkhof F, Bruck W, De Groot CJA, et al. Remyelinated lesions in cells in multiple sclerosis (MSCIMS) trial protocol and baseline multiple sclerosis: magnetic resonance image appearance. cohort characteristics: an open-label pre-test: post-test study with Arch Neurol 2003; 60: 1073–81. blinded outcome assessments. Trials 2011; 12: 62. 30 Schmierer K, Scaravilli F, Altmann DR, Barker GJ, Miller DH. 24 Poser CM, Paty DW, Scheinberg L, et al. New diagnostic criteria for Magnetization transfer ratio and myelin in postmortem multiple multiple sclerosis: guidelines for research protocols. Ann Neurol sclerosis brain. Ann Neurol 2004; 56: 407–15. 1983; 13: 227–31. 31 Franklin RJM, Ffrench-Constant C. Remyelination in the CNS: 25 Dominici M, Le Blanc K, Mueller I, et al. Minimal criteria for from biology to therapy. Nat Rev Neurosci 2008; 9: 839–55. defi ning multipotent mesenchymal stromal cells. The International 32 Rivera FJ, Couillard-Despres S, Pedre X, et al. Mesenchymal stem Society for Cellular Therapy position statement. Cytotherapy 2006; cells instruct oligodendrogenic fate decision on adult neural stem 8: 315–17. cells. Stem Cells 2006; 24: 2209–19. 26 Johnson NA, Hamoudi RA, Ichimura K, et al. Application of array 33 Bai L, Caplan A, Lennon D, Miller RH. Human mesenchymal stem CGH on archival formalin-fi xed paraffi n-embedded tissues cells signals regulate neural stem cell fate. Neurochem Res 2007; including small numbers of microdissected cells. Lab Invest 2006; 32: 353–62. 86: 968–78. 156 www.thelancet.com/neurology Vol 11 February 2012

Journal

The Lancet Neurology – Unpaywall

Published: Feb 1, 2012

There are no references for this article.