TRANSLATIONAL AND CLINICAL RESEARCH

Neurotrophic Bone Marrow Cellular Nests Prevent Spinal

Motoneuron Degeneration in Amyotrophic Lateral Sclerosis Patients:

A Pilot Safety Study

MIGUEL BLANQUER,a JOSE M. MORALEDA,a FRANCISCA INIESTA,a JOAQUIN GOMEZ-ESPUCH,a,b JOSE MECA-LALLANA,c

RAMON VILLAVERDE,bMIGUEL ANGEL PEREZ-ESPEJO,

dFRANCISCO JOSE RUIZ-LOPEZ,

eJOSE MARIA GARCIA SANTOS,

f

PATRICIA BLEDA,a VIRGINIA IZURA,g MARIA SAEZ,g PEDRO DE MINGO,g LAURA VIVANCOS,h RAFAEL CARLES,h

JUDITH JIMENEZ,h JOAQUIN HERNANDEZ,i JULIA GUARDIOLA,e SILVIA TORRES DEL RIO,f CARMEN ANTUNEZ,b

PEDRO DE LA ROSA,dMARIA JULIANA MAJADO,

aANDRES SANCHEZ-SALINAS,

aJAVIER LOPEZ,

j

JUAN FRANCISCO MARTINEZ-LAGE,d SALVADOR MARTINEZk

aHematopoietic Progenitors Transplant and Cell Therapy Unit, cNeurology, dNeurosurgery, eNeumology,gNeurophysiology, hNeuropsychology, and iAnesthesiology, Hospital Virgen de la Arrixaca, Universidad de

Murcia, Murcia, Spain; bNeurology and fRadiology, Hospital Morales Meseguer, Universidad de Murcia, Murcia,

Spain; jStatistical Analysis, Fundaci�on para la Formaci�on e Investigaci�on Sanitarias de la Regi�on de Murcia,

Hospital Universitario Virgen de la Arrixaca, Murcia, Spain; kInstituto de Neurociencias, UMH-CSIC, Alicante,

Spain

Key Words. Amyotrophic lateral sclerosis • Bone marrow • Adult stem cells • Somatic cell therapy • Stem cell transplantation • Clinical

trials

ABSTRACT

The objective of this article is to assess the safety of intra-spinal infusion of autologous bone marrow mononuclearcells (BMNCs) and, ultimately, to look for histopathologi-

cal signs of cellular neurotrophism in amyotrophic lateralsclerosis (ALS) patients. We conducted an open single arm

phase I trial. After 6 months observation, autologousBMNCs were infused into the posterior spinal cord funicu-lus. Safety was the primary endpoint and was defined as

the absence of serious transplant-related adverse events.In addition, forced vital capacity (FVC), ALS-functionalrating scale (ALS-FRS), Medical Research Council scale

for assessment of muscle power (MRC), and Norris scaleswere assessed 6 and 3 months prior to the transplant and

quarterly afterward for 1 year. Pathological studies wereperformed in case of death. Eleven patients were included.We did not observe any severe transplant-related adverseevent, but there were 43 nonsevere events. Twenty-two

(51%) resolved in �2 weeks and only four were still pres-ent at the end of follow-up. All were common terminologycriteria for adverse events grade �2. No acceleration in

the rate of decline of FVC, ALS-FRS, Norris, or MRCscales was observed. Four patients died on days 359, 378,

808, and 1,058 post-transplant for reasons unrelated to theprocedure. Spinal cord pathological analysis showed agreater number of motoneurons in the treated segments

compared with the untreated segments (4.2 6 0.8 moto-neurons per section [mns per sect] and 0.9 6 0.3 mns persect, respectively). In the treated segments, motoneurons

were surrounded by CD901 cells and did not show degen-erative ubiquitin deposits. This clinical trial confirms not

only the safety of intraspinal infusion of autologousBMNC in ALS patients but also provides evidencestrongly suggesting their neurotrophic activity. STEM CELLS

2012;30:1277–1285

Disclosure of potential conflicts of interest is found at the end of this article.

Author contributions: M.B.: conception and design, collection and assembly of data, data analysis and interpretation, manuscript writing,and final approval of manuscript; J.M.J. and S.M.: conception and design, data analysis and interpretation, manuscript writing, and finalapproval of manuscript; F.I.: administrative support, collection and/or assembly of data, and data analysis and interpretation; J.G.E.,J.M., and R.V.: conception and design, provision of patients, collection and/or assembly of data, and data analysis and interpretation;M.A.P.E.: conception and design and provision of patients; F.J.R.L., J.M.G.S., V.I., and L.V.: conception and design, collection andassembly of data, and data analysis and interpretation; M.S., P.d.M., R.C., and J.J.: conception and design, collection of data, and dataanalysis and interpretation; C.A.: administrative support and provision of study material; M.J.M. and A.S.S.: provision of study materialand collection of data; J.L.: assembly of data, and data analysis and interpretation; J.F.M.-L.: conception and design.

Correspondence: Jose M. Moraleda, M.D., Ph.D., Hematology Department, Hematopoietic Transplant and Cell Therapy Unit, Virgen dela Arrixaca Hospital, University of Murcia, Ctra. Cartagena km 7, 30120 El Palmar, Murcia, Spain. Telephone: þ34-968369532; Fax:þ34-968369088; e-mail: jmoraled@um.es Received December 8, 2011; accepted for publication February 15, 2012; first publishedonline in STEM CELLS EXPRESS March 13, 2012. VC AlphaMed Press 1066-5099/2012/$30.00/0 doi: 10.1002/stem.1080

STEM CELLS 2012;30:1277–1285 www.StemCells.com

INTRODUCTION

Amyotrophic lateral sclerosis (ALS) is a neurodegenerativedisease with a largely unknown pathogenesis and has no cure.Its reported incidence is 2.16 per 100,000 person years with amean onset age of 64.4 6 12.3 years [1] and 90% of casesare sporadic [2, 3]. Patients develop progressive muscular at-rophy leading to death, usually through respiratory failure, atapproximately 3 years following diagnosis [4]. The pathologi-cal hallmarks of sporadic ALS are degeneration and loss ofmotoneurons with astrocytic gliosis and the presence of intra-cellular inclusions in degenerating neurons and glia [5, 6].

The management of ALS is primarily palliative and sup-portive [2]. Despite much research, there has been little successin the search for disease modifying or neuroprotective agents.In fact, Riluzole is the only drug that has shown a survival ben-efit in ALS patients although this is limited to 6 months and isnot associated with improvement in clinical symptoms [7–10].

Cell therapy has been proposed as a treatment for ALS[11–14]. After cell transplants, some patients showed a transientstabilization of disease progression, but the underlying biologi-cal mechanisms of these beneficial effects remain speculative.

Our group tested the potential benefits of hematopoieticstem cells infusion into the spinal cord parenchyma of adultmdf mice, a mouse model of motoneuron degeneration [15]with established clinical and pathological signs of progressivespinal motoneuron loss. The mice experienced functionalimprovement measured by open field, footprint, and rotarodperformance as well as an increase in the number of motoneur-ons in the anterior horn of the spinal cord when compared withsham operated mice. However, no neural differentiation of thetransplanted cells was observed. Further analysis of the experi-mental spinal cords showed that the grafted cells formed cellu-lar nests surrounding the motoneurons and that they expressedglial derived neurotrophic factor (GDNF) mRNA and protein.This graft derived GDNF acting as a neurotrophic signal in themicroenvironment of spinal motoneurons, increased neuronalsurvival, and improved the functional performance observed inthe mice after the transplant. Indeed, the specific distribution ofbone marrow-grafted cells and degenerating motoneuronsrevealed a new mechanism that increases the neurotrophic effi-ciency of secreted signals by the grafted cells [16, 17]. Mazziniet al. have previously published the first data showing thesafety of the infusion of mesenchymal stem cells into the spinalcord of seven ALS patients [12]. The ex vivo expanded mesen-chymal stem cells were transplanted into the spinal cord at T7-T9 levels, with minimal side effects and absence of detrimentaleffects on neurological function. Taking into account our pre-clinical results and those of Mazzini et al., we designed a clini-cal trial aiming to assess the feasibility and safety of intraspinalinfusion of autologous bone marrow mononuclear cells(BMNCs) and, ultimately, to look for histopathological signs ofcellular neurotrophism in ALS patients.

METHODS

Overview

We designed an open single arm phase I trial to evaluate the fea-sibility and safety of intraspinal infusion of autologous BMNC.Safety was defined primarily as the absence of treatment-relatedserious adverse events, as defined by the CONSORT group [18]and, secondarily, as a rate of decline in the forced vital capacity(FVC) as well as the ALS-functional rating scale (FRS), Norris,and MRC scales not significantly greater than prior to the surgicalintervention.

Our plan was to recruit 10 patients into the trial, and we there-fore preselected 32 of 122 patients according to our inclusion crite-ria. An intensive selection process, including a complete medical,respiratory, and neurophysiological assessment, was performed atthe initial visit. A psychological evaluation was carried out toensure that the patients fully understood the experimental nature ofthe trial and the risks associated with the procedure as well as toassess their psychological stability. Finally, 13 patients wererecruited and were evaluated quarterly to monitor disease progres-sion. During that time, two patients were excluded because of sub-sequent nonfulfillment of the inclusion criteria. The transplant wasperformed 6 months after enrollment in the trial of the 11 remain-ing patients. Thoracic 3-4 level was chosen for the laminectomyand the BMNC transplant, to ensure spinal column stability andsafety of the procedure. The fact that the intercostal muscles inner-vated by these spinal cord segments are the most equilibrated asregards respiratory function [19] was also considered.

After the infusion, all patients were assessed at the Virgen dela Arrixaca University Hospital and the Morales Meseguer Uni-versity Hospital quarterly for 1 year. In addition, in order to fur-ther monitor the adverse events, open telephone interviews wereperformed weekly for 3 months and monthly thereafter, until theend of the 1 year follow-up period. Adverse events were gradedaccording to the common terminology criteria for adverse eventsv3.0 (CTCAE) [20].

The clinical trial was approved by the Clinical Trials EthicsCommittee of the Virgen de la Arrixaca University Hospital andthe Morales Meseguer University Hospital as well as by theAgencia Espa~nola de Medicamentos y Productos Sanitarios. AAI-Pharma (Madrid, Spain) acted as the external monitor of thestudy. The trial was registered at www.clinicaltrials.gov (identifierNCT00855400) and at the European Clinical Trials Database(EudraCT number 2006-003096-12).

Selection Criteria

Patients between 20 and 65 years old were eligible if they haddefinite ALS according to the El Escorial criteria [21, 22], spinalonset, and duration of disease between 6 and 36 months. FVChad to be �50% of that predicted, and a below 90% fall in oxy-gen saturation (T90) occurring in �2% of sleep time.

Patients were excluded if they had evidence of concomitantneurological, psychiatric, or systemic disease, had received treat-ment with corticosteroids, immunoglobulins, or immunosuppres-sors in the last 12 months, had been included in another clinicaltrial, required enteral or parenteral nutrition, were pregnant orwere unable to provide informed consent.

Transplant Procedure

The transplant procedure has been described in detail elsewhere[23]. Briefly, 60 ml of autologous bone marrow was harvested,under sedation with propofol (0.5 mg/kg) and fentanyl (50–100lg), by multiple aspirations in the posterior iliac crests. A ficolldensity gradient separation was performed to isolate the mononu-clear fraction. The BMNCs were then resuspended in 2.1 ml nor-mal saline and loaded into two 1 ml syringes. The excess 0.1 mlwas used to assess the number of BMNC obtained, their viabilitymeasured by trypan blue exclusion, and the number of CD34þ,CD117þ, and CD133þ cells by flow cytometry. Meanwhile, withcontinuous monitoring of motor and sensory evoked potentials, aT3-T4 laminectomy was performed to expose the underlying spi-nal cord. Using a purpose-modified injector device, the cells wereslowly injected (approximately 3 minutes per syringe) through a22G lumbar puncture needle into the most avascular pial surfaceon the posterior spinal funiculus, approximately 1–2.5 mm fromthe midline, at a depth of 6 mm from the surface. The procedurewas repeated either in the ipsilateral or contralateral posteriorfuniculus, depending on the pial vasculature distribution. Thissurgical approach was chosen because of our previous animalmodel experience [16] and because it had been safely used in theMazzini et al. preliminary study [12].

1278 Neurotrophic BMNC Intraspinal Transplant in ALS

Assessments

As shown in Supporting Information Table S1, neurological, re-spiratory, radiological, neurophysiological, laboratory, psycholog-ical, and neuropsychological evaluations were performed quar-terly from day �180 to þ360, day 0 being the transplant date.ECG, chest x-ray, neurosurgery, and anesthetic assessment wereperformed before the intervention. Immediate postsurgery carewas provided in the Neurosurgery Department for 1 week. Beforedischarge, patients had hematological and blood chemistry analy-sis as well as spinal cord magnetic resonance imaging (MRI),somato-sensory evoked potentials (SSEP), spirometry, and psy-chological assessment.

Neurological assessment was performed by two neurologistsfrom different hospitals and included a clinical interview and ex-amination as well as ALS-FRS, Norris, and MRC scales scoring.The registered scores were the means of the values independentlymeasured by each neurologist.

Spirometry and respiratory muscle assessments were per-formed with a ZAN 100 pulmonary spirometer and a ZAN 500Plethysmograph (Waldfenster, Germany, www.zan.de). FVC wasmeasured in the sitting position and was expressed as a percent-age of predicted value. Static mouth pressure was measured usinga flanged mouthpiece. Patients performed three maximal inspira-tory and expiratory efforts from residual volume and total lungcapacity and then maximum expiratory pressure (MEP) and maxi-mum inspiratory pressure (MIP), sustained for 1 second, weredetermined as previously described [24].

SSEP and intraoperative monitoring were performed with afour-channel Neuropack M1 Nihon-Kohden (Nihon Kohden, To-kyo, Japan, www.nihonkohden.com). Inferior extremity SSEPwas studied with CPi stimulation following a standard protocol[25]. A two-channel Synergy Oxford Medelec (Oxford Instru-ments Medical, Oxfordshire, UK, www.oxford-instruments.com)was used to perform electromyography of genioglossus/masseter,biceps, extensor digitorum communis, first interosseous, paraver-tebralis dorsalis, quadriceps, and tibialis anterior. Polysomnogra-phy was performed following the AASM criteria [26] with a 20-channel Neurofax Nihon-Kohden (Nihon Kohden, Tokyo, Japan,www.nihonkohden.com).

Brain and spinal cord MRIs were obtained using a 1.5-Timaging system (MRI LX.1.5T; GE Medical Systems, Milwau-kee, WI, http://www.gehealthcare.com). Spinal cord images wereacquired with a four-channel phased-array surface antenna, withT1 and T2 enhanced Fast Spin-Echo sagittal sequences and a Gra-dient Echo axial sequence centered on the surgery area. Imageswere examined for dural and extradural changes as well as intrin-sic spinal cord changes.

Individual and family semistructured clinical, psychological,and neuropsychological interviews were performed. Psychopatho-logical assessment with special focus on depression and anxiety,

attitude toward the disease and the trial, and family support wasalso carried out using the MMPI-2 for personality evaluation,POMS for anxiety and mood status as well as Euro-QoL5D forquality of life evaluation.

Statistical Analysis

Continuous variables were described using central tendency (me-dian or mean) and dispersion measurements (standard deviation).Categorical measurements were described using absolute and rela-tive frequency tables. The conditions of application of statisticalanalysis were checked and normality was verified by the Kolmo-gorov-Smirnov test and homoscedasticity by the Levene test. Non-parametric tests were used for analysis in case of breach of any ofthe conditions. All variables showed a normal distribution. Com-parison between categorical variables was carried out using theChi-squared test and comparison of continuous variables using Stu-dent’s t test and the repeated measures ANOVA test (independentvariable: time at which subjects are tested). Significant results fromANOVA tests were followed up by post hoc pairwise comparisonsusing Tukey’s studentized range test. Analysis was performed bycomparing the increase of the variables in time constants (3 and 6months). The level of significance in all statistical tests was alpha¼ 0.05 and the SPSS statistical package version 18.0 was used.

Autopsy and Histology

Prior to the spinal cord extraction, in situ exploration was per-formed to detect possible anatomical malformations. Transverseserial segments of 1.0 cm cervical, thoracic, and lumbar spinalcord were identified in relation to the injection levels, embeddedin paraffin, cut at 7 lm, mounted in 10 parallel series, and proc-essed by structural staining (Cresyl Violet 0.5% [CV]) and immu-nohistochemistry against specific neural (Tuj1, choline acetyl-transferase (ChAT), glial fibrillary acidic protein, Nestin, andNG2), bone marrow cells (CD34, CD90, CD68, CD45, CD44,and MCP1), and ALS neurodegeneration (TDP-43 and SOD1)markers (Supporting Information Table S2). Immunostaining wasperformed using the standard biotin-avidin/horseradish peroxidaseprocedure. Slides were incubated with 0.9% oxygen peroxide inphosphate buffered saline (PBS) with 0.1% Triton X-100 (PBS-T) for 30 minutes and then blocked in 1% albumin bovine(Sigma) in PBS-T for 1 hour. Primary antibodies were incubatedfor 24 hours at room temperature and were then incubated withcorresponding biotinyladed secondary antibodies for 1 hour atroom temperature, followed by Vectastain ABC reagent for1 hour (both Vector Laboratories, Burlingame, CA) and immuno-labeling revealed with diaminobenzidine tetrahydrochloride(DAB; for light brown or, using with nickel sulfate [0.08%],black precipitates) and/or AminoEthyl Carbazole (AEC; for redprecipitate), following the manufacturer’s recommendations. Sec-tions were finally mounted on slides, dehydrated, and covered

Table 1. Demographics and baseline characteristics

Patient no. Sex Age TTI (months) FVC (%) ALS-FRS Norris MRC ALS treatment

1 F 46 20 113 24 54 45 R, E, C, L, B, T2 M 61 21 104 28 74 37 E, C, L3 F 54 16 94 28 69 38 R, T4 F 43 40 105 36 87 47 R5 M 45 38 79 27 74 42 R, E6 M 31 33 104 34 79 46 R, E, C, B7 F 49 29 115 28 71 35 R8 F 50 23 99 30 70 46 R, E9 F 52 24 121 32 81 50 R, E, B10 M 43 14 120 35 83 50 R, L11 M 41 15 116 38 95 54 RMedian 46 21 105 30 74 46

Abbreviations: ALS-FRS, amyotrophic lateral sclerosis functional rating scale; B, baclofen; C, creatine; E, vitamin E; FVC, forced vitalcapacity; L, lithium carbonate; MRC, medical research council scale for assessment of muscle power; R, rilutek; T, tizanidine; TTI, timefrom diagnosis to infusion.

Blanquer, Moraleda, Iniesta et al. 1279

www.StemCells.com

Table

2.Related

nonsevereadverse

events

Adverse

events

Patientno.

12

34

56

78

910

11

Constipation

22

22

12

21

12

þ4to

þ365

þ1to

þ365

þ1to

þ7þ1

toþ6

þ2to

þ4þ1

toþ5

þ2to

þ5þ1

toþ3

þ0to

þ2þ1

toþ6

Painfulwound

11

21

21

2þ2

toþ6

þ3to

þ5þ1

toþ6

0þ2

toþ2

6þ1

toþ5

2þ1

toþ1

04

þ1to

þ93

Hypoesthesia

11

2/1

11

11

þ6þ2

toþ8

7þ1

toþ2

3þ4

toþ4

6þ2

toþ1

18

þ2to

þ365

þ2to

þ31

Fingertips

Erratic

þ24to

þ365

Lower

limbs

Lower

limbs

Wound/foot

Rightleg

Intercostal

pain

11

12

2þ5

toþ1

1þ1

toþ4

þ2to

þ11

þ1to

þ53

þ4to

þ27

Paresthesia

11

11

þ6to

þ11

þ2to

þ11

þ4to

þ46

þ4to

þ6Lips

Legs

Low

costal

Legs

Intracranialhypotension

22

2þ1

toþ3

þ9to

þ28

þ1to

þ6Dysesthesia

21

þ32to

þ118

þ32to

þ274

Headache

12

þ18to

þ22

þ4Vertigo

2þ1

toþ5

Transplantsite

hyperesthesia

1þ2

toþ4

3Insomnia

1þ2

toþ7

Commonterm

inologycriteria

foradverse

eventsgradeisshownin

thefirstlineofeach

adverse

eventanddurationin

daysafterthetransplantin

thesecondline.

Whereappropriate,

thelocationof

theadverse

effect

has

beenadded.

1280 Neurotrophic BMNC Intraspinal Transplant in ALS

with Eukitt. After immunostaining, spinal cord sections were pho-tographed microscopically (Leica Microsystems, Wetzlar, Ger-many). Motoneurons in the anterior spinal horn showing largesize, characteristic polygonal shape, and large nuclei were countedin 10 sections of each CV series and spinal segment (T1, T2; T4-T5; T6; T8; T9); with a separation of 280 lm to ensure a singlecounting per cell. Control T1–T9 spinal samples from two males66 and 68 years old with no history of neurological diseasesdonated by the Brain Bank of the Regi�on of Murcia (BCRM) werestudied following the same protocol.

Role of the Funding Source

The study funding sources were neither involved in the studydesign, data collection, analysis, interpretation, or the writing of thereport nor involved in the decision to submit the paper for publica-tion. The views expressed in this article do not necessarily reflectthose of the funding bodies. All authors had full access to the studydata, read and approved the final version of the paper, and were re-sponsible for the decision to submit the paper for publication.

RESULTS

Patient characteristics are described in Table 1. Eleven (fivemales, six females) were included in the trial with a medianage of 46 years (range 32–61). The median duration of the

disease from diagnosis to the BMNC infusion was 21 months(range 11–40). The median FVC at entry was 105% (range79–121), whereas the median ALS-FRS, Norris, and MRCscores were 30 (range 24–38), 74 (range 54–95), and 46(range 35–54), respectively. All patients received their usualmedical treatment, including Riluzole.

BMNC Harvest and Transplant

A median of 462 � 106 BMNC (range 138.00–602.87) wasinfused, which included a median of 2.77 � 106 CD34þ(range 0.55–23.39), 2.31 � 106 CD117þ (range 0.41–10.25),and 1.30 � 106 CD133þ (range 0.41–5.43) cells (SupportingInformation Table S3). The median of viable cells analyzedin nine out of 11 patients was 80% (range 70–95). Therewere no anesthetic complications and no immediate adversesurgical events were recorded, apart from a 10-minute 50%reduction in SSEP in one patient.

Nonsevere Treatment-Related Adverse Events

Table 2 shows the 43 definitely, probably, or possibly relatednonsevere adverse events. Most of them (51%) occurred dur-ing the first 2 weeks after surgery. Twenty-six percentresolved in the first 2 months, 14% lasted less than 9 months,and only 9% was still present at the end of follow-up. Allwere CTCAE grade �2 with constipation the most frequentand long lasting (10 patients), surgical wound pain up to 3

Figure 1. Progress of neurological scales and functional respiratory indexes. No significant differences in the rate of decline of the neurologicalscales or the functional respiratory measurements before and after D0 were found. Abbreviations: ALS-FRS, amyotrophic lateral sclerosis-func-tional rating scale; D0, day of the infusion; FVC, forced vital capacity; MEP, maximum expiratory pressure; MIP, maximum inspiratory pressure;MRC, Medical Research Council scale for assessment of muscle power.

Blanquer, Moraleda, Iniesta et al. 1281

www.StemCells.com

months in seven patients, and temporary intercostal pain (2months) in five patients. Small areas of hypoesthesia (sevenpatients), paresthesia (four patients), and dysesthesia (twopatients), located mainly in the lower limbs and of little clini-cal relevance, resolved in 3–4 months, except for a smallhypoesthesic zone in one foot in two patients, still present atthe end of follow-up. Three patients experienced symptoms oftransient intracranial hypotension. Other less common andpossibly related transient adverse events were headache, ver-tigo, transplant site hyperesthesia, and insomnia.

Severe Adverse Events

None of the six severe adverse events recorded in the studywere considered to be related to the procedure. Patient 2experienced syncope due to paroxysmal atrial flutter on dayþ137, and this required admission for 1 day but resolvedspontaneously. However, he died on day þ378 following amassive pulmonary embolism. Patient 3 needed hospitaladmission for 10 days because of a respiratory tract infectionon day þ236 and she died on day þ359 due to ALS progres-sion and respiratory insufficiency. Finally, patient 8 developedacute cholecystitis that required cholecystectomy in the lastquarter of follow-up. Patients 1 and 5 died 14 and 23 monthsrespectively after the end of follow-up, patient 1 due to Influ-enza A (H1N1) infection, and patient 5 due to pulmonaryinfection and respiratory failure. Patients 1, 2, and 3 donatedtheir central nervous system to the BCRM.

Patient Follow-up

No deterioration in the rate of decline of spirometric variables(FVC p ¼ .901, MIP p ¼ .85, and MEP p ¼ .27) or neurolog-ical scales (ALS-FRS p ¼ .515, Norris p ¼ .839, and MRC p¼ .167) was observed (Fig. 1). During follow-up, eightpatients experienced progressive deterioration and eventualcomplete loss of the SSEP, although no clinical correlationwas observed. However, patient 9, who had experienced atransient reduction in SSEP during the transplant, had normalSSEP throughout the trial. During follow-up, a drop of the

T90 under 2% was observed in patients 8 and 10 and, overall,patients’ sleep efficiency progressively decreased throughoutthe study.

Nine patients showed some grade of psychological adapta-tive reaction but none experienced depression, and mood, asassessed by the POMS scale, remained stable throughout the

Figure 2. Progress of POMS and EuroQoL scales. POMS scale for mood remained stable throughout the clinical trial, its values not reachingsignificant differences at any time point. Quality of life as measured by the EuroQoL scale decreased significantly before the transplant (p < .01)but remained stable afterward (p ¼ .649).

Figure 3. Magnetic resonance imaging (MRI) scans. (A): T1-weighted spin-echo sequence (immediate postsurgical MRI) at thelevel of the laminectomy. A fluid extradural collection (*) exerts amild mass effect on the spinal cord (arrows) that also shows sometiny hyperintense dots (arrowhead) suggesting a subacute bleed. (B):T2-weighted fast spin-echo sequence (immediate postsurgical MRI) atthe level of the laminectomy. The fluid collection (*) exerts a masseffect on the dural sac but the spinal cord is spared. However, itshows a mild hyperintensity in the surgical area (arrow). (C): T2-weighted fast spin-echo sequence (MRI at the end of follow-up) atthe level of the laminectomy. The previous collection is now a scar(*). The spinal cord shows a substantial deformity with an increasedanterior-posterior diameter (arrowhead).

1282 Neurotrophic BMNC Intraspinal Transplant in ALS

trial (POMS score p ¼ .45) (Fig. 2). The EuroQol scaleshowed a reduction in quality of life up to the time of infu-sion (p < .01), but it remained stable after the intervention (p¼ .649) (Fig. 2).

Radiological Evolution

All patients showed extradural-extraspinal postsurgical collec-tions. In patients 1, 3, 5, and 10, this even exerted a transientdiscrete mass effect on the spinal cord. Nine patients had an

immediate self-limiting postsurgical T2 hyperintensity in thespinal cord, suggesting intraparenchymal edema. A persistentT2 high signal was observed in patients 3 and 6 throughoutthe study. Although some degree of persistent spinal cord de-formity was detected in all patients after transplantation, onlyin two of them was the antero-posterior diameter greater thanthe transverse. Patients 3, 6, and 8 showed a small subacutehemorrhage on the postsurgical MRI, and this appearance per-sisted until the end of the study (Fig. 3).

Figure 4. Pathological features in the spinal cord. (A): Macroscopic picture showing a dorsal view of the thoracic spinal cord. Red arrows labelthe remaining pial scar where the needle was introduced into the left PF. (B): T4 transverse section showing the needle entry point (red arrow), tra-jectory (black arrows), and cell infusion area (surrounded by a red line). Microphotographs of Nissl staining on paraffin-embedded T4 spinal cordtransverse sections showing: (C) basophilic packed cells filling the needle track from the subpial region (red arrow); (D) spinal cord gray matter atthe level of the cell graft (black arrows show the trajectory through the dorsal funiculus to the grafted area [g] in the central gray matter). ChAt(red) and CD90 (black) double immunohistochemistry showing: (E) at rostral thoracic levels, T1-T2, to the injected segments, some ChAT-positiveneurons can be detected (arrows), but very few CD90þ cells were visible (arrowheads); (F) in sections at T4 level, both ChAT-positive motoneurons(arrow) and CD90þ cells (arrowheads) were detectable; (G, H) high-power micrographs of ChAT-positive motoneurons (MNs) surrounded byCD90þ cells (arrowheads) at this thoracic spinal level; (I) high-power micrograph showing a MN in a grafted anterior horn, T5, surrounded by anest of CD90þ cells (arrowheads). Processed sections to detect ChAT and CD45 (J) or CD68 (K) show that these bone marrow cell populations donot form peri-motoneuron cellular nests. Contiguous paraffin-embedded spinal sections processed by ChAT/CD90 and TDP-43 immunohistochemis-try showing: (L, M) a degenerating motoneuron with collapsed cytoplasm and vacuolization (L) and somatic skein-like TDP-43-positive inclusions(M), at distant thoracic segments, T1, from the grafted ones; (O, P) when we move closer to the grafted segments, at T2, some CD90 cells (arrow-heads) can be detected close to the motoneurons (O, arrow) and some granular TDP-43 immunoreaction is present in motoneurons cytoplasm andnucleus (P); finally (Q, R)at the grafted segments, T3-T4-T5, peri-motoneurons nests of CD90þ cells are detected (R, arrows) in motoneurons thatdo not show any TDP-43 immunoreactivity (Q, arrow,), while motoneurons devoid of cellular nests show intracytoplasmic globulous inclusions (Q,arrowhead). Scale bars ¼ 1 mm (A, B), 100 lm (E, F), and 50 lm (G--R). Abbreviations: AF, anterior funiculus; ALF, anterolateral funiculus; amf,anterior midline fissure; AR, anterior root; cc, central channel; dh, dorsal horn; DL, dentate ligament; g, grafted area; lh, lateral horn; PF, posteriorfuniculus; PLF, posterolateral funiculus; pms, posterior midline sulcus; vh, ventral horn.

Blanquer, Moraleda, Iniesta et al. 1283

www.StemCells.com

Pathology and Histology

Macroscopic examination of grafted spinal cord segments didnot show any anatomical anomaly or damage due to the surgi-cal procedure (Fig. 4A, 4B). Nissl-stained paraffin sectionsshowed local accumulation of packed basophilic spheroid cellsfilling the needle track (Fig. 4C, 4D). An accumulation of vas-cular profiles was also evident in the grafted gray matter at thebase of the anterior horns (Fig. 4D). The mean number of spi-nal motoneurons in the anterior horn of thoracic segmentsincreased as we approached the infused segments, from 0.9 60.3 motoneurons per section (mns per sect) in superior (T1)and inferior (T8 and T9) segments to 4.2 6 0.8 mns per sect atthe grafted levels (T4-T5) (Fig. 5). The differences betweensegments were significant (T1 to T2 p < .01; T2 to T4-T5 p <.01; T4-T5 to T8-T9 p < .01) but not between the three patientsanalyzed. Control spinal cord had a mean of 14.4 6 1.2 mnsper sect in T4-T6 anterior horns. Cells expressing hematopoi-etic markers as CD45, CD90, and CD68 could be identified inthe spinal gray matter (Fig. 4E–4K). Interestingly, small spher-ical cells surrounding the motoneurons, forming a cellular nest,were observed exclusively in the grafted segments (Fig. 4I,4R). These nest cells did not express any neural marker, con-sistent with our earlier experimental data [16], they were nega-tive for CD45 and CD68 (Fig. 4J, 4K) and showed immunore-activity for CD90 (Fig. 4I, 4R). We also analyzed theneurodegenerative signs in the spinal motoneurons. In seg-ments caudal to the grafted level, a total of 41 ChAT-positivemotoneurons were counted. None of them had cellular nests,36 showed cytoplasmic collapse and ubiquitin intracellulardeposits (TDP-43 immunoreactive) (Fig. 4L, 4M), while fivedid not show these deposits. Conversely, no SOD1 immunore-action was detected (data not shown), confirming the sporadicALS form in our patients [27, 28]. Approaching the graftedsegments, the spinal motoneurons showed fewer degenerativesigns and intracytoplasmic deposits (Fig. 4O, 4P). Interest-ingly, motoneurons surrounded by CD90þ cellular nests didnot show ubiquitin deposits (76 out of 88 motoneurons[86.4%] counted at the T4 and T5 level in the three necropsies)(Fig. 4R, 4Q), while 11 out of 12 [91.6%] neighboring moto-neurons devoid of CD90 cellular nests showed globulous intra-cytoplasmic deposits (Fig. 4Q).

DISCUSSION

In the last 2 decades, almost 50 pharmacological agents havebeen tested in ALS patients but only Riluzole has demon-strated some therapeutic activity [9]. Thus, new therapeuticapproaches are urgently needed and stem cell transplants mayproduce a positive disease-modifying effect through multifac-torial mechanisms [29]. We have tested this hypothesis in ani-mal models of motoneuron degeneration and demonstratedthat cellular therapy using BMNC produces a neurotrophiceffect [16]. Since no clinical studies using intraspinal infusionof BMNC had been performed, we designed the present safetyclinical trial as a first step to explore a potential neurotrophiceffect in ALS patients.

The primary endpoint of the study was achieved becausenone of our 11 patients experienced any severe, procedure-related adverse event and, indeed, all the nonsevere adverseevents (definitely, probably, or possibly related) were mild(�grade 2) and most of them were transient (�2 months).No acceleration in the rate of decline of FVC, ALS-FRS,Norris, or MRC scales was observed. In addition, the MRIabnormalities observed were either transient or not clinicallyrelevant.

A thorough understanding of the biological mechanismsof cellular therapy is necessary before such an approach canbe advocated for ALS patients. The design of the trial wasprimarily based on our experimental results that demonstratedthat, in an animal ALS model, grafted BMNC migrated intothe anterior horn of the spinal cord, formed cellular nestsaround the motoneurons, and produced GDNF [16]. Our hy-pothesis was that neurotrophic mechanisms mediated byBMNC would modify the motoneuron microenvironment andthus favor their survival and, consistent with this, the autopsysamples showed some well-defined histological features asregards number, distribution, and lack of signs of motoneurondegeneration. There was a higher motoneuron count in the an-terior horns of the transplanted spinal levels (T4-T5) than inrostral and caudal thoracic levels. A significant number ofsmall basophilic CD90þ spherical cells adjacent to and/oraround the motoneurons were only observed at the graftedlevels and the morphology and arrangement of those cellswere identical to that observed in our animal model in whichwe demonstrated a GDNF-mediated neurotrophic effect onthe host motoneurons [16]. Moreover, these motoneurons sur-rounded by CD90þ cells had no ubiquitin deposits andneither did they show any other morphological signs ofdegeneration, strongly suggesting that the cellular-mediatedneurotrophic mechanism observed in the animal model wasalso reproduced in our patients. This suggests that autologousBMNC grafted into the anterior horns of ALS spinal cord sur-vive a long time as perineuronal nests and that they protectmotoneurons from neurodegeneration.

In summary, our data confirm the feasibility and safety ofinjecting BMNC into the thoracic spinal cord in ALS patients.Our results are in agreement with Deda et al. [13] whoinfused BMNC into C1 and C2 medullary levels of 10 ALSpatients without significant adverse events and with Mazziniet al. [11, 12] who injected bone marrow mesenchymal stemcells into T7-T9 levels with similar safety results. In contrastto Mazzini’s trial, we used BMNC because our basic researchstudies demonstrated that these cells have neurotrophic activ-ity [16] and higher graft viability than selected bone marrowmesenchymal stem cells [30]. We also wanted to infuse allthe cellular components that might deliver positive effects andmaintain technical simplicity to make the procedure morewidely available.

Figure 5. Mean motoneurons per spinal segment. The mean numberof motoneurons, although inferior to that of a control spinal cord,increased progressively from the distal segments of the spinal cord ofthe patients to the infused T4-T5 level, where the highest number ofconserved motoneurons was observed.

1284 Neurotrophic BMNC Intraspinal Transplant in ALS

CONCLUSIONS

In conclusion, this study confirms the feasibility and safety ofthe intraspinal infusion of BMNC and provides the first pub-lished evidence strongly suggesting their neurotrophic activityin ALS patients, showing histopathological signs of the reduc-tion of TDP-43 deposits in the spinal motoneurons.

ACKNOWLEDGMENTS

This work was supported by the Carlos III Institute (FIS EC07/90762), Advanced Therapies and Transplant General Direction(Health Ministry, Spain) (TRA-137), ISCIII Spanish Cell Ther-

apy Network (Tercel; RD06/0010/0023), DIGESIC-MECBFU2008-00588, Ingenio 2010 MEC-CONSOLIDERCSD2007-00023, GVA Prometeo Grant 2009/028, Rotary ClubElche-Illice and by the Fundaci�on Di�ogenes. We are grateful toDr. M.V. Tobin for checking the English wording of themanuscript.

DISCLOSURE OF POTENTIAL

CONFLICTS OF INTEREST

The authors have no conflicts of interest to report. The authorsdeclare that data reported here have been partially published inthe Spanish communication by Blanquer et al. [31].

REFERENCES

1 Logroscino G, Traynor BJ, Hardiman O et al. Incidence of amyotro-phic lateral sclerosis in Europe. J Neurol Neurosurg Psychiatry 2010;81:385–390.

2 Wijesekera LC, Leigh PN. Amyotrophic lateral sclerosis. Orphanet JRare Dis 2009;4:3–25.

3 Kiernan MC, Vucic S, Cheah BC et al. Amyotrophic lateral sclerosis.Lancet 2011;377:942–955.

4 Talbot K. Motor neuron disease: The bare essentials. Pract Neurol2009;9:303–309.

5 Piao YS, Wakabayashi K, Kakita A et al. Neuropathology with clini-cal correlations of sporadic amyotrophic lateral sclerosis: 102 autopsycases examined between 1962 and 2000. Brain Pathol 2003;13:10–22.

6 Neumann M, Sampathu DM, Kwong LK et al. Ubiquitinated TDP-43in frontotemporal lobar degeneration and amyotrophic lateral sclerosis.Science 2006;314:130–133.

7 Pastula DM, Moore DH, Bedlack RS. Creatine for amyotrophic lateralsclerosis/motor neuron disease. Cochrane Database Syst Rev 2010;6:CD005225.

8 Aggarwal S, Cudkowicz M. ALS drug development: Reflections fromthe past and a way forward. Neurotherapeutics 2008;5:516–527.

9 Zoccolella S, Santamato A, Lamberti P. Current and emerging treat-ments for amyotrophic lateral sclerosis. Neuropsychiatr Dis Treat2009;5:577–595.

10 Miller RG, Mitchell JD, Lyon M et al. Riluzole for amyotrophic lat-eral sclerosis (ALS)/motor neuron disease (MND). Cochrane DatabaseSyst Rev 2007;1:CD001447.

11 Mazzini L, Ferrero I, Luparello V et al. Mesenchymal stem cell trans-plantation in amyotrophic lateral sclerosis: A Phase I clinical trial.Exp Neurol 2010;223:229–237.

12 Mazzini L, Fagioli F, Boccaletti R et al. Stem cell therapy in amyotro-phic lateral sclerosis: A methodological approach in humans. Amyo-troph Lateral Scler Other Motor Neuron Disord 2003;4:158–161.

13 Deda H, Inci MC, Kurekci AE et al. Treatment of amyotrophic lateralsclerosis patients by autologous bone marrow-derived hematopoietic stemcell transplantation: A 1-year follow-up. Cytotherapy 2009;11:18–25.

14 Martinez HR, Gonzalez-Garza MT, Moreno-Cuevas JE et al. Stem-cell transplantation into the frontal motor cortex in amyotrophic lateralsclerosis patients. Cytotherapy 2009;11:26–34.

15 Blot S, Poirier C, Dreyfus PA. The mouse mutation muscle deficient(mdf) is characterized by a progressive motoneuron disease. J Neuro-pathol Exp Neurol 1995;54:812–825.

16 Cabanes C, Bonilla S, Tabares L et al. Neuroprotective effect of adulthematopoietic stem cells in a mouse model of motoneuron degenera-tion. Neurobiol Dis 2007;26:408–418.

17 Moraleda JM, Blanquer M, Bleda P et al. Adult stem cell therapy:Dream or reality? Transpl Immunol 2006;17:74–77.

18 Ioannidis JP, Evans SJ, Gotzsche PC et al. Better reporting of harmsin randomized trials: An extension of the CONSORT statement. AnnIntern Med 2004;141:781–788.

19 De Troyer A, Kirkwood PA, Wilson TA. Respiratory action of the in-tercostal muscles. Physiol Rev 2005;85:717–756.

20 Trotti A, Colevas AD, Setser A et al. CTCAE v3.0: Development of acomprehensive grading system for the adverse effects of cancer treat-ment. Semin Radiat Oncol 2003;13:176–181.

21 Brooks BR. El Escorial World Federation of Neurology criteria forthe diagnosis of amyotrophic lateral sclerosis. Subcommittee on MotorNeuron Diseases/Amyotrophic Lateral Sclerosis of the World Federa-tion of Neurology Research Group on Neuromuscular Diseases andthe El Escorial ‘‘Clinical limits of amyotrophic lateral sclerosis’’ work-shop contributors. J Neurol Sci 1994;124(suppl):96–107.

22 Brooks BR, Miller RG, Swash M et al. El Escorial revisited: Revisedcriteria for the diagnosis of amyotrophic lateral sclerosis. AmyotrophLateral Scler Other Motor Neuron Disord 2000;1:293–299.

23 Blanquer M, Perez-Espejo MA, Martinez-Lage JF et al. A surgicaltechnique of spinal cord cell transplantation in amyotrophic lateralsclerosis. J Neurosci Methods 2010;191:255–257.

24 Polkey MI, Green M, Moxham J. Measurement of respiratory musclestrength. Thorax 1995;50:1131–1135.

25 Cruccu G, Aminoff MJ, Curio G et al. Recommendations for the clini-cal use of somatosensory-evoked potentials. Clin Neurophysiol 2008;119:1705–1719.

26 Iber C, Ancoli-Israel S, Cheeson A et al. The AASM Manual for theScoring of Sleep and Associated Events-Rules, Terminology andTechnical Specifications. Westchester, IL: American Academy ofSleep Medicine, 2007.

27 Mackenzie IR, Bigio EH, Ince PG et al. Pathological TDP-43distinguishes sporadic amyotrophic lateral sclerosis from amyotrophiclateral sclerosis with SOD1 mutations. Ann Neurol 2007;61:427–434.

28 Mackenzie IR, Rademakers R, Neumann M. TDP-43 and FUS inamyotrophic lateral sclerosis and frontotemporal dementia. LancetNeurol 2010;9:995–1007.

29 Suzuki M, Svendsen CN. Combining growth factor and stem cell ther-apy for amyotrophic lateral sclerosis. Trends Neurosci 2008;31:192–198.

30 Pastor D, Viso-Le�on M, Jones J et al. Comparative effects betweenbone marrow and mesenchymal stem cell transplantation in GDNFexpression and motor function recovery in a motorneuron degenerativemouse model. Stem Cell Rev 2011; [Epub ahead of print].

31 Blanquer M, Perez Espejo MA, Iniesta F et al. [Bone marrow stemcell transplantation in amyotrophic lateral sclerosis: Technical aspectsand preliminary results from a clinical trial]. Methods Find Exp ClinPharmacol 2010;32(suppl A):31–37.

See www.StemCells.com for supporting information available online.

Blanquer, Moraleda, Iniesta et al. 1285