Peripheral blood stem cell transplantation in the treatment of ... - Nature

Bone Marrow Transplantation, (1997) 20, 631–638  1997 Stockton Press All rights reserved 0268–3369/97 $12.00

Peripheral blood stem cell transplantation in the treatment of progressive multiple sclerosis: first results of a pilot study A Fassas1, A Anagnostopoulos1 , A Kazis2, K Kapinas2, I Sakellari1, V Kimiskidis2 and A Tsompanakou 1 Departments of 1Haematology and 2Neurology, The George Papanicolaou Hospital, Exokhi/Thessaloniki, Greece

Summary: Several experimental autoimmune diseases (AID), including allergic encephalomyelitis, ie the multiple sclerosis (MS) model, respond to TBI and chemotherapy followed by BMT. Remissions of AID may also occur in patients with concomitant malignancies treated with allogeneic or autologous BMT. These observations have emphasized the possibility of treating AID with highdose therapy and haematopoietic stem cell transplantation (HSCT). In a phase I/II pilot study, 15 patients with progressive MS were treated with BEAM followed by autologous blood SCT and antithymocyte globulin (ATG). Patients were severely disabled, with median EDSS and SNRS scores of 6 (5–7.5) and 42 (33–62), respectively. Cyclophosphamide (4 g/m2) and G/GMCSF (5 mg/kg/day) were used for stem cell mobilization, which caused no neurotoxicity. On days +1 and +2, ATG (2.5–5 mg/kg) was given for in vivo T cell-depletion. Allergy (93%) and infections (87%) were the principal toxic complications. Mild, transient, neurotoxicity was observed in six patients in the immediate post-transplant period. The median follow-up time is 6 months (6–18). Durable neurologic improvements have been detected on both the EDSS (7/15) and SNRS (15/15) systems. One patient worsened at 3 months and two have relapsed. Autologous HSCT appears feasible in MS; it does not aggravate disability and seems to offer a clinical benefit. However, these observations need confirmation and long-term outcomes will show if benefits counterbalance toxicity and cost. Keywords: haematopoietic stem cell transplantation; multiple sclerosis

Multiple sclerosis (MS) is a disease caused by selective destruction of myelin in the central nervous system. Genetic susceptibility, environmental triggers (possibly viral), and an abnormal immunological reaction lead to focal myelin destruction and secondary damage to axons. Myelin breakdown results from infiltration of the affected areas with lymphocytes and macrophages and is associated with focal neurological dysfunction.1 The disease is incurable and Correspondence: Dr A Fassas, Haematology Department, The George Papanicolaou Hospital, 570 10 Exokhi, Thessaloniki, Greece Received 5 March 1997; accepted 8 June 1997

seriously disabling, running a relapsing/remitting course which eventually becomes progressive (secondary progressive MS). In about 10–20% of cases it has a continuously progressive course from onset (primary progressive MS), which is slow or rapid, or even fulminant with shortened survival.2 There is strong evidence supporting an immunopathogenic basis for the disease which could, therefore, be regarded as an autoimmune process directed at one or other of the structural components of myelin. 3 Treatment involves anti-inflammatory, immunosuppressive, and immunomodulating agents,4,5 some of which are active in managing relapses or preventing them, but none is curative. Moreover, for patients with the progressive form of the disease, therapies are either unable to halt deterioration or have modest effects on clinical improvement.6–11 The rationale for using haematopoietic stem cell transplantation (HSCT) to treat MS is based on the principle of complete ablation of an aberrant immune system followed by reconstitution of a new immune system deriving from either an allogeneic donor or a T cell-depleted autologous transplant, as has been proposed for other autoimmune diseases (AID).12–16 Recapitulation of lymphocyte ontogeny may result in immune tolerance, especially in the allogeneic setting in which a graft-versus-host reaction could contribute to complete elimination of the patient’s residual myelinreactive cells.17 In the autologous setting it seems practically impossible to eliminate all autoreactive cells, but the patient may still benefit from a strong conditioning regimen followed by a graft containing very few lymphocytes. Experimental investigations have shown that inherited or induced AID in animals may be cured by BMT,12,18,19 and this is also true for the animal model of MS, experimental allergic encephalomyelitis (EAE), in which effective prevention or remission of the disease manifestations can be attained by TBI and cytoreductive drugs followed by allogeneic, syngeneic, or autologous BMT.17,20–22 In man, there is anecdotal evidence of resolution of AID in patients undergoing allogeneic BMT for concomitant malignancies.23–25 Autologous transplants, using either T celldepleted or unmanipulated grafts, have also been reported to effect clinical and immunological remissions in patients with AID. These remissions, however, were mostly of short duration, although occasional long remissions have been seen.25–28 The number of cases of patients with MS and concomitant leukaemia or lymphoma who have undergone allogeneic or autologous HSCT is very limited. In these cases, stabilization of disease activity, neurological

Stem cell transplantation in multiple sclerosis A Fassas et al

632

improvement, but also exacerbation of MS have been reported. 25 Allogeneic HSCT for AID would reasonably appear to be more efficacious than autologous transplants in maintaining immunological remission. However, this yet unknown advantage may not outweigh the increased systemic complication rate of GVHD and related morbidity. Therefore, if transplantation is used to treat AID, autologous HSCT would possibly be, at present, a safer alternative.

Patients and methods Study design The trial was approved by two institutional review committees. It was designed as a phase I/II pilot study, testing the protocol in 15 patients with advanced and active disease, aged below 55, who would be followed-up for at least 2 years. Target criteria were treatment-related toxicity, possible aggravation of neurologic status, improvement in disability, and relapse after improvement. Enrollment of patients took place from April 1995 to April 1996. For ethical reasons only two patients were treated at first and it was only after 4 months that more patients were entered into the study. By that time, the results in these two patients appeared very encouraging in terms of acceptable toxicity and improvement in disability. Patients The 15 patients (eight male, seven female) had definite MS according to the Poser criteria29 for a time period of 2–28 years (median, 10). The median age at presentation was 37 years (range, 24–54). They were all judged to be in a progressive phase of MS (pMS) for a median of 6 years (range, 2–17) and had evidence of clinical deterioration of at least 1.0 point on the Kurtzke Expanded Disability Status Scale (EDSS)30 over the preceding 12 months. They were also severely disabled, with a median EDSS of 6.0 (range, 5.0–7.5) at entry. Median score on the Scripps Neurologic Rating Scale (SNRS)31 was accordingly low, 42 (range, 33– 62). The patients had characteristic abnormalities on magnetic resonance imaging (MRI) scans and in one, brain lesions enhanced with gadolinium-DTPA at enrollment. The progressive phase of MS was primary in eight patients and secondary in seven, with median durations of 8 (range, 2–17) and 6 years (range, 3–8) respectively. Patients had previously received a variety of conventional therapies with little or no effect on disease progression and disability. These included soluble methylprednisolone pulses (14 cases), occasional low-dose prednisolone orally (six), azathioprine (five), mitoxantrone (one), beta-IFN (two), and i.v. immunoglobulin (IVIG, two cases). To be eligible for the trial, patients were required to have had no medical illness precluding transplantation and no mental dysfunction precluding the ability to give informed consent. All patients were negative for hepatitis B virus (HBV) antigens and for antibodies to the hepatitis C virus. Five patients who had evidence of past HBV infection (antibodies to HBV) were included in the study. Detailed

explanation of the protocol was given to the eligible patients and to their family members. Stem cell mobilization Cyclophosphamide (CY) 4 g/m2 was given for stem cell mobilization, followed by daily s.c. injections of G-CSF (lenograstim, eight patients) or GM-CSF (seven patients) at 5 mg/kg bw, commencing 1 day after CY infusion. Patients were planned to undergo at least two leukaphereses by means of a Fenwall CS3000-Plus cell separator (Baxter, Fenwal Division, Deerfield, IL, USA), starting when neutrophils reached 109/l. The minimum number of CD34+ cells to be collected for transplantation was set at 4 × 106/kg bw. As no monoclonal antibodies were available, ex vivo T cell depletion was not performed. Transplantation procedure The time interval from stem cell harvest to transplantation was 16–52 days (median, 33). The BEAM regimen was chosen for conditioning and was used at full dose, as in lymphomas: BCNU 300 mg/m2 on day −6, etoposide 200 mg/m2 and cytosine-arabinoside 200 mg/m2 on days −5 to −2, and melphalan 140 mg/m2 on day −1. On day 0, the stem cells were thawed and infused. On days +1 and +2, rabbit ATG (thymoglobulin, Merieux, Lyon, France) at 2.5 mg/kg bw/day was given to the first eight patients, as a sort of in vivo lymphocyte depletion, but the dose was considered low and was doubled (5 mg/kg/day) in the next seven patients. G-CSF (lenograstim) was administered at 10 mg/kg bw i.v. to all patients from day 0 until absolute neutrophil counts exceeded 1.5 × 109/l for 3 consecutive days. IVIG (Sandoglobulin; Sandoz, Basle, Switzerland) at 0.5 g/kg bw was given as supportive treatment on days −7, +8, +23, and +38. Oral ciprofloxacin (1 g) fluconazole (400 mg) and acyclovir (15 mg/kg bw) were given daily as infection prophylaxis. Patients were isolated, but not in laminar air flow systems, from the day of admission, ie 1 day before starting the conditioning, to the day of discharge. Nursing prophylactic measures were similar to those taken for patients undergoing transplantation for lymphomas. Toxicity and haematologic evaluation Adverse experiences were recorded according to the WHO grading system. Special attention was given to the nervous system during stem cell mobilization and in the immediate post-transplant period, in order to detect toxicities in the form of aggravation of existing symptoms or the appearance of new ones. After discharge from hospital, followup was continued on an outpatient basis until days +40 to +50. Patients were then re-examined at the end of the 3rd month post-HSCT and every 3 months thereafter, unless an earlier incident required management or hospitalization. Blood lymphocyte phenotype analyses were performed before stem cell mobilization, before conditioning for transplantation, and every 3 months after HSCT.


Two scoring systems were used to assess neurologic status, the Kurtzke EDSS30 and the Scripps NRS.31 On the EDSS, normal examination is scored as 0 and death as 10. Patient scores at presentation ranged from 5 to 7.5, indicating impairment of daily activities and degrees of disability varying between inability to walk (without aid) further than 200 metres and restriction to a wheelchair. On the Scripps scale, a score of 100 points represents maximum efficiency. Assessments were made by two neurologists working independently whose rating scores were averaged. Evaluations were done at entry, after mobilization of peripheral blood stem cells, ie before conditioning for HSCT, 1 and 3 months after stem cell infusion, and at 3-month intervals thereafter. Worsening and relapse after improvement were defined as a gain of 1.0 or more EDSS points or a loss of 10 or more SNRS points when scores were compared to the baseline or to the latest assessment scores, respectively. Smaller or no changes on the scales were regarded as indicating stabilization of disease status. Improvement in disability was indicated by reduction of EDSS of 1.0 or more points, or a gain of at least 10 SNRS points. Serial MRIs were performed along with neurologic clinical assessments. Other criteria included recording of evoked potentials, analysis of cerebrospinal fluid with detection of oligoclonal bands by immunoelectrophoresis, measurement of circulating cytokines, and determination of cytokine messenger RNA expression. The results of these investigations are not included in this paper.

633

100

60

50

Percentage

Neurologic evaluation

40

30 CD8+

20 CD4+ 10

At entry After CY (baseline) (before HSCT)

+3

+6

+9

+12

+15

+18

Months after HSCT

Figure 1 Kinetics of CD4+ and CD8+ lymphoid subsets following HSCT. Values are reported as percentages ± s.d.

7

(15) (15)

(13) (15) (13) (6)

(3)

(2)

(1)

Results Stem cell harvest and haematologic recovery after infusion The median number of CD34+ cells collected for reinfusion was 10 × 106/kg bw (range, 4.6–19 × 106/kg) and the median number of CD3+ cells was 60 × 106/kg bw (range, 16–270 × 106/kg). GM-CSF and G-CSF were equally effective in mobilizing CD34+ cells. ANC reached 0.5 × 109/l between days +8 and +11 (median, day +10) and platelet counts reached 50 × 109/l between days +8 and +15 (median, day +13). The median number of days spent in hospital for transplantation was 26 (range, 19–36 days). As expected, cell phenotype analysis showed a profound fall of CD4+ cells following HSCT in all patients without exception. The mean value of CD4+ cells dropped from 46 ± 5.6% (s.d.) at entry to a nadir of 15.1 ± 8.4% at 3 months post-HSCT (P , 0.001, paired t-test) and remained very low over the next 12 months. In contrast, the mean value of CD8+ cells rose from 23.6 ± 5.9% at entry to 50.5 ± 13.2% at 3 months (P , 0.001); from this peak, values declined gradually in the course of time but remained above the corresponding mean CD4+ values (Figure 1). Small, but significant, changes in CD4+ and CD8+ cell counts were also seen after stem cell mobilization, apparently caused by the high dose of CY.

Mean EDSS score ± s.e.

6

5

4

3

At entry After CY +1 (baseline) (before HSCT)

+3

+6

+9

+12

+15

+18

Months after HSCT

Figure 2 Changes in the Kurtzke (EDSS) rating scores. Figures in parentheses represent numbers of evaluated patients.

Toxicity of stem cell mobilization No major complications occurred during the cytopenic period following high-dose CY except for polymicrobial bacteremia (grade 3). Eight patients had no adverse manifestations; seven had 9 episodes of infection: 7 cases of fever (FUO) (grade 2), a urinary tract infection (UTI) due to Chromobacterium violaceum (grade 1, minor), and a case


634

90

(15) (15)

(13) (15)

(13)

(6)

(3)

(2)

(1)

Mean SNRS score ± s.e.

80

70

60

50

However, these were transient (1–2 days) and mild. On day +30, when the first post-transplant assessment was made, no patient was found to have increased his EDSS score by 1 point, or to have decreased his SNRS score. An increase of 0.5 EDSS points was noted in three cases, while the rest of the patients had stable or improved scores. On SNRS, all scores were improved. After discharge from hospital, three patients needed brief re-hospitalization (2–10 days) between days +20 and +80 for treatment of FUO or documented infections (bacteremia, sinusitis). Despite persistent low blood CD4+ cell counts, late infections (beyond day +100) did not occur, apart from one case of cystitis in a patient with known recurrent UTIS dating from the pre-transplant period. No patient has to date been re-admitted to hospital for a late disorder subsequent to the transplant procedure.

40

Improvement in disability At entry After CY +1 (baseline) (before HSCT)

+3

+6

+9

+12

+15

+18

Months after HSCT

Figure 3 Changes in the Scripps (SNRS) rating scores. Figures in parentheses represent numbers of evaluated patients.

of polymicrobial bacteremia, with Streptococcus viridans, Enterococcus faecalis, and Candida albicans recovered from blood culture. Grade 1 oral mucositis was seen in 3 patients; transient, grade 1, liver toxicity was detected in 1 patient (elevation of transaminases). One patient developed an allergic reaction soon after CY infusion (bronchospasm, grade 3). Moderate (grade 2) allergic reactions the ‘fever with drug’ type occurred in two patients receiving GMCSF. One other patient on GM-CSF developed fever with hypotension (grade 4 toxicity) necessitating a switch to GCSF. Apart from these allergic manifestations with GMCSF, there were no other differences between GM-CSF and G-CSF, which was very well tolerated. Regarding neurotoxicity, no toxic signs were noted. Moreover, one patient improved his EDSS score considerably, from 5 to 3.5, and five patients improved their SNRS scores by 11 to 18 points. Transplant-related toxicity No patient died as a result of protocol therapy. Table 1 shows all complications observed during the immediate post-transplant period. Allergy and infections were problematic. Fourteen of the 15 patients developed allergic reactions, mostly to ATG, severe enough (anaphylaxis) in two patients to preclude administration of the entire dose of ATG. To prevent anaphylactic phenomena, soluble methylprednisolone was subsequently administered on a regular basis, at 0.25 g twice daily for 2 days, along with the ATG. Infectious episodes occurred in 13 patients, bacteremia being the most frequent (Table 1). Liver toxicity was minor with no case of veno-occlusive disease (VOD). Within 11 days following stem cell infusion, six patients developed some adverse neurologic events or reported worsening of already existing ones, as shown in Table 1.

The first patient underwent HSCT 18 months ago and three patients have completed 1 year of follow-up. Minimum and median follow-up time of the whole group is 6 months. Mean EDSS and SNRS rating scores (±s.e.m.) are shown in Tables 2 and 3 and are also depicted in Figures 2 and 3. EDSS scores declined gradually with time while SNRS scores increased, indicating neurological improvement. A two-way analysis of variance performed at 6 months showed that the mean values of patient scores given at the scheduled assessments showed significant changes during this time period on both scales (P , 0.01 and P , 0.001, respectively). Paired t-test analyses of the mean differences of individual scores from corresponding baseline scores also showed significant changes (Table 2 and 3), which were more prominent on the SNRS as it is more sensitive and not based so much on walking ability. On the EDSS, one patient showed improvement after CY. After HSCT, six improvements were detected at the 1st-month assessment (40%, Table 2). Two of these six patients were found to have further improved at 3 months and one patient continued to improve at 9 and 15 months post-HSCT. This patient had a spectacular course, with an EDSS score of 5 at presentation and only 2 at 18 months. One late improvement was also seen in a patient who lost 1 EDSS point at 6 months. In all, seven patients improved. More meaningful changes, ie loss of 1.5 or more EDSS points, were detected in four of the seven patients: one improved by 3, one by 2.5, one by 2, and one by 1.5 points during the total follow-up period. The patient with the gadolinium-enhancing brain lesions on MRI improved by 1.5 points at 3 months. Patients who did not respond on the EDSS (8 of 15) have remained stable to date, 6 to 12 months (median, 6) post-HSCT, except for one, who gained 1 EDSS point at 6 months. The mean EDSS change (increase) in non-responding cases was 0.11 ± 0.19 at 6 months. No factors were found to be significantly related to favourable responses. Patients with primary pMS responded as well as patients with secondary pMS. There was an impression of better responses in patients below 40 years of age, with an EDSS score 5, and with a short dis-


Table 1

635

Transplant-related toxicity

Toxicity

Grade (WHO)

Allergy erythema fever and erythema fever and hypotension bronchospasm anaphylaxis Oral soreness, erythema ulcers GI bleeding Liver enzymes Infection bacteremia

No. of patients 14/15 4 9 5 2 2 7/15 6 1 1/15 3/15 13/15 9

1 2; 1 4 3 4 1 3 3 1 3

fungemia pneumonia UTI enteritis FUO

3 3 2 2 2

1 3 2 3 1

Neurologic aggravation of visual impairment headache confusion-disorientation deterioration of ataxia vertigo

Table 2

Cause

BCNU, stem cell inf., ATG ATG Stem cell inf., ATG Stem cell inf., ATG ATG

Steroids Staph. sp., E. cloacae, Kl. oxytoca C. albicans E. faecium, S. haemolyticus E. clocae, Citrobacter sp.

6/15 2 2 1 1 1

on on on on on

days +3, +4 day +3 day +5 day +10 day +11

Changes in the Kurtzke EDSS rating scores

Time

At entry (baseline) Before HSCT (after CY) After HSCT, months: +1 +3 +6 +9 +12 +15 +18

No.a

Mean EDSS score (s.e.)

Mean changeb

15 15

6.1 (0.2) 5.9 (0.3)

−0.2 (0.1)

15 15 13c 6 3 2 1

5.7 5.6 5.4 4.5 5.2 4.0 2.0

(0.3) (0.4) (0.4) (0.7) (1.0) (2.0)

−0.5 −0.5 −0.6 −1.3 −0.8 −2.0 −3

d

(0.2) (0.2)d (0.3)d (0.4)d (0.7) (1.0)

No. of patients improved by >1 point >1.5 points

1 (7%)

1 (7%)

6 (40%) 6 (40%) 6 (46%) 4 2 2 1

3 (20%) 4 (27%) 3 (23%) 3 1 1 1

a

Number of patients evaluated. Mean of the differences from baseline scores. c Data from two patients are missing. d P , 0.05; paired t-test. Compared to baseline, one patient got worse (= gained 1 EDSS point) at 6 months. b

ease duration (primary pMS) before HSCT. None of the differences, however, were statistically significant. On the SNRS, more obvious changes were seen (Table 3). Scores were found to have already improved after stem cell mobilization in five patients (due to CY) and also later, at 1, 3, and 6 months, in five, four, and one more patient, respectively. By 6 months, all patients gained 10 points. Five patients increased their already improved scores further at 3 months and late improvements were also seen, at 6 and 12 months (two and one patient, respectively). Considerable (around 30–40 SNRS points)

increases were noted in four patients. There was no worsening on this scale. Relapses No relapses have been detected on the EDSS and, improvements have been durable for 1+ to 18+ months (median, 8). However, on the SNRS, two patients relapsed at 3 and 9 months post-HSCT, after having improved for 2 and 5 months, respectively. One of these had received a minimal dose of ATG because he developed anaphylactic shock at


636

Table 3

Changes in the Scripps NRS rating scores

Time

At entry (baseline) Before HSCT (after CY) After HSCT, months: +1 +3 +6 +9 +12 +15 +18

No.a

Mean SNRS score (s.e.)

Mean changeb

15 15

49.5 (2.6) 57.4 (2.9)

+ 7.9 (1.7) d

13c 15 13c 6 3 2 1

65.0 68.1 72.0 73.7 69.0 75.0 78.5

(3.7) (2.8) (3.2) (3.9) (8.1) (4.0)

+13.7 +18.6 +22.6 +27.5 +31.7 +37.0 +35.5

(2.1) d (2.1) d (2.9) d (3.6) d (6.6)e (1.0)e

No. of patients improved by >10 points >20 points

5 (33%)

0 (0%)

9 (69%) 14 (93%) 12 (92%) 6 3 2 1

3 (23%) 8 (53%) 8 (62%) 5 2 2 1

a

Number of patients evaluated. Mean of the differences from baseline scores. Data from two patients are missing. d P , 0.001; paired t-test. e P , 0.05; paired t-test. After having improved, two patients relapsed at 3 and 9 months. Compared to baseline, there is no worsening. b c

infusion. Nevertheless, the SNRS scores of these patients have remained above (better than) their respective baseline scores. The rest of the patients (13 of 15) have maintained their improved SNRS scores for 1+ to 18+ months (median, 5). Moreover, we found no difference, in terms of response or relapse, between the two ATG doses administered (5 mg/kg bw and 10 mg/kg bw). Discussion The main objective of this study was to assess the feasibility and toxicity of HSCT in the treatment of MS. We also attempted to obtain some indications of the clinical efficacy of HSCT on this inexorably crippling disease in which, sometimes, even halting progression could be regarded as a positive treatment result (stabilization). Our results show that peripheral blood HSCT can be used with relative safety to patients with pMS without causing exacerbations of the disease. Some evidence is also provided that, in a number of patients, this kind of therapy can suppress disease progression and reduce disability, even in those with longstanding disease. Moreover, it seems that some ‘remissions’ may be sustainable, as in the two of our patients who have completed a follow-up of at least 15 months. When designing the protocol, important issues came into question, namely the choice of patients, the choice of highdose therapy, the kind of graft, and the means for an objective evaluation of results. For ethical reasons, only patients with advanced disease were eligible, somewhat unrewarding material, as these patients may be more susceptible to the debilitating consequences of HSCT32 or they may have much gliosis in the CNS. Also, since the pattern of deterioration cannot be easily identified in pMS,2,33 before HSCT, one has to consider, whether the disease is currently active or whether it is in a phase of slow progression or of spontaneous stabilization. Progression of disability over the previous year5,7–9 and serial gadolinium–enhanced brain MRI scans34 have been used as criteria for patient selection in MS trials. Serial MRI scans were not performed in our

patients before HSCT but they had all declined on the EDSS by at least 1 point over the preceding year, with 11 of them having evidence of very recent deterioration before HSCT. In experimental BMT, non marrow-ablative doses of TBI are less effective than high doses in preventing spontaneous relapses of EAE.17 In patients with AID, the most appropriate conditioning regimen is unknown. Relatively less myelotoxic regimens, eg high-dose CY alone or 2-chlorodeoxyadenosine, are less immunotoxic than TBI or high-dose busulphan (Bu)35 which, additionally, can reach and destroy effector lymphocytes residing in the CNS, a possibly important event for disease remission.36 However, TBI has been reported to aggravate neurological symptoms in rats17,21 and high-dose Bu may be associated with VOD and neurotoxicity.35 BEAM, which seems to have worked in the trial so far, can be argued to be less immunosuppressive or lymphotoxic than BuCy and may, therefore, not be sufficient to prevent relapse of autoimmunity. However, it is much less toxic. It is very myelotoxic and lymphotoxic, and designed for use especially in autologous HSCT in lymphomas. It has rarely been used in allogeneic transplants.37 It incorporates drugs that are not neurotoxic at the doses used,38 it can cross the blood–brain barrier (BCNU, high-dose melphalan, high-dose cytosine-arabinoside) and is immunosuppressive (BCNU and high-dose melphalan).37,39–41 To avoid anesthesia and for faster recovery we chose to use peripheral grafts, although the high numbers of re-infused lymphocytes may cause disease recurrence, just as allogeneic grafts from donors with AID can transfer the disease to recipients.42–44 To deplete lymphocytes in vivo, ATG was administered after stem cell infusion but it proved to be relatively problematic because of allergic reactions. Of the two patients who could not receive ATG, one relapsed (on the SNRS) but the other has maintained an improved SNRS score for 6+ months. In contrast to all five cases described by Euler et al,26 who relapsed after receiving unmanipulated stem cell grafts for AID, we had only two relapses. ATG may, therefore, be of benefit in the setting of autologous HSCT with unmanipulated peripheral blood grafts. However, our patients’ median follow-up time


is still short and longer surveillance is needed. Nevertheless, the question of the usefulness of other, possibly more effective, methods of lymphocyte depletion (positive CD34 cellselection, negative T cell-depletion) has not yet been settled. It would suffice to use only CSF for stem cell mobilization. However, by adding CY at 4 g/m2 one may gain a first impression of disease sensitivity to high-dose therapy, while lower doses of CY (eg 2 g/m2) may not yield improvements.7 The two CSFs used were administered in a randomized manner and yielded similar CD34+ cell counts. However, G-CSF was better than GM-CSF, which caused allergic phenomena, and is, therefore, preferable for mobilization and acceleration of neutrophil recovery. Infections – some significant (bacteremia, fungemia) – were frequent in the early post-infusion period, but late ones, despite persistent immunosuppression, have not as yet occurred. Some mild, transient neurotoxicity did occur after stem cell infusion. Probable causes were the drugs used for the conditioning, infection,32 and possibly the re-infused lymphocytes. The worsening detected in one patient at 6 months can easily be ascribed to the disease, but the possibility that HSCT had a detrimental effect on his neurologic status should be considered. Nevertheless, the procedure was, in general, well tolerated by the patients, without the need for long hospitalization. MS is one of the most difficult diseases in which to judge the effect of a therapy because of subjective influences. The ratings of our independently working neurologists were consistent with each other and did not differ statistically (data not shown). Despite reservations about objectivity, an efficacy analysis was also attempted. All patients showed improvement when they were evaluated on the SNRS; by 6 months they gradually gained 10–41 points. Even the two relapsed patients keep their scores above the corresponding baseline ones. Using the more strict EDSS, seven of 15 patients were found to have improved over the first 6 months, coming down the scale by 1–2.5 points. MRI analyses showed less disease volume in the brain at 3 and 6 months, but the differences were not statistically significant (data not shown). Interestingly, some patients continued to improve beyond 6 months and some improvements seem too great to occur in a chronic, crippling, disease like pMS. It seems that, despite clinically advanced disease, some patients still have reversible injuries to the CNS. Such patients are those with active disease on MRI11,34 but even patients without gadolinium-enhancing lesions may also benefit from HSCT. It is improbable that the improvements occurred spontaneously and to such a degree, since patients with pMS usually deteriorate steadily with time or remain stable, and only exceptionally can improve.2,33 However, for non-responsive patients, it is impossible to say that their stabilization is due to the therapy. The reason for the patients improvement is apparently the profound immunosuppression caused by HSCT which invariably induced a long-lasting depletion of CD4+ lymphocytes. Disease activity seems to depend upon the availability of circulating lymphocytes and certain monoclonal antibodies to T cells (OKT3, anti-CD4, CAMPATH-1H) have been reported to reduce disease activity on MRI and also to have some clinical effect.45–47 In this trial, CD4+ cell counts had invariably fallen but not all patients responded

on the EDSS. Also, there was no correlation between reinfused CD3+, CD4+, CD8+ cells and response (data not shown), either because the EDSS is not sufficiently sensitive to detect minor changes or because other mechanisms are also implicated in disease improvement. It will be interesting to see whether these improvements will pertain when CD4+ lymphocytes have resumed their normal blood levels. The impact of autologous HSCT on long-term outcome will determine its utility. If improvements are sustained in the future, it will be tempting to conclude that autoreactive cells have been eradicated and that tolerance has been acquired through induction of unresponsiveness to tissue antigens presented to newly emerging T lymphocytes.18 At present it seems that HSCT may be clinically worthwhile in pMS. The procedure, however, does have a certain mortality risk, probably below 5%, and the results of this study require confirmation especially as other therapies also claim to be of benefit in pMS.10,11,48,49 Issues of cost-effectiveness, early mortality, and, also, of late effects including secondary cancer50 will have to be carefully considered, before (and if) prospective, comparative trials are to be undertaken.

References 1 Poser CM. Notes on the pathogenesis of multiple sclerosis. Clin Neurosci 1994; 2: 258–265. 2 Matthews WB. Course and prognosis. In: Matthews WB, Compston A, Allen IV, Martyn CN (eds). McAlpine’s Multiple Sclerosis, 2nd edn. Churchill Livingstone: Singapore 1991, pp 139– 163. 3 Gulcher JR, Vartanian T, Stefansson K. Is multiple sclerosis an autoimmune disease? Clin Neurosci 1994; 2: 246–252. 4 van Oosten BW, Truyen L, Barkhof F, Polman CH. Multiple sclerosis therapy: a practical guide. Drugs 1995; 49: 200–212. 5 Cook SD, Devereux C, Troiano R et al. Effect of total lymphoid irradiation in chronic progressive multiple sclerosis. Lancet 1996; i: 1405–1409. 6 Hauser SL, Dawson DM, Lehrich JR et al. Intensive immunosuppression in progressive multiple sclerosis. New Engl J Med 1983; 308: 173–180. 7 Likosky WH, Fireman B, Elmore R et al. Intense immunosuppression in chronic progressive multiple sclerosis: the Kaiser study. J Neurol Neurosurg Psychiatr 1991; 54: 1055–1060. 8 The Canadian Cooperative Multiple Sclerosis Study Group. The Canadian cooperative trial of cyclophosphamide and plasma exchange in progressive multiple sclerosis. Lancet 1991; 337: 441–446. 9 Weiner HL, Mackin GA, Orav EJ et al. Intermittent cyclophosphamide pulse therapy in progressive multiple sclerosis: final report of the Northeast Cooperative Multiple Sclerosis Treatment Group. Neurology 1993; 43: 910–918. 10 Bornstein MB, Miller A, Slagle S et al. A placebo-controlled, double-blind, randomized, two center, pilot trial of Cop 1 in chronic progressive multiple sclerosis. Neurology 1991; 41: 533–539. 11 Sipe JC, Romine JS, Koziol JA et al. Cladribine in treatment of chronic progressive multiple sclerosis. Lancet 1994; 344: 9–13. 12 van Bekkum DW. BMT in experimental autoimmune diseases. Bone Marrow Transplant 1993; 11: 183–187. 13 Marmont AM. Immune ablation followed by allogeneic or autologous bone marrow transplantation: a new treatment for severe autoimmune diseases? Stem Cells 1994; 12: 125–135. 14 Marmont AM, van Bekkum DW. Stem cell transplantation for

637


638

15 16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

severe autoimmune diseases: new proposals but still unanswered questions. Bone Marrow Transplant 1995; 16: 497– 498. Burt RK, Burns W, Hess A. Bone marrow transplantation for multiple sclerosis. Bone Marrow Transplant 1995; 16: 1–6. Tyndall A, Gratwohl A. Haemopoietic stem cell and progenitor cells in the treatment of severe autoimmune diseases. Ann Rheum Dis 1996; 55: 149–151. van Gelder M, van Bekkum DW. Treatment of relapsing experimental autoimmune encephalomyelitis in rats with allogeneic bone marrow transplantation from a resistant strain. Bone Marrow Transplant 1995; 16: 343–351. Slavin S, Karussis D, Weiss L et al. Immunohematopoietic reconstitution by allogeneic or autologous bone marrow grafts as a means for induction of specific unresponsiveness to donor-specific allografts and modified self in autoimmune disorders. Transplant Proc 1993; 25: 1274–1275. Karussis DM, Vourka-Karoussis U, Lehmann D et al. Immunomodulation of autoimmunity in MRL/1pr mice with syngeneic bone marrow transplantation (SBMT). Clin Exp Immunol 1995; 100: 111–117. Karussis DM, Vourka-Karoussis U, Lehmann D et al. Prevention and reversal of adoptively transferred, chronic relapsing experimental autoimmune encephalomyelitis with a single high dose cytoreductive treatment followed by syngeneic bone marrow transplantation. J Clin Invest 1993; 92: 765–772. van Gelder M, Kinwell-Bohre EPM, van Bekkum DW. Treatment of experimental allergic encephalomyelitis in rats with total body irradiation and syngeneic BMT. Bone Marrow Transplant 1993; 11: 233–241. van Gelder M, van Bekkum DW. Effective treatment of relapsing experimental autoimmune encephalomyelitis with pseudoautologous bone marrow transplantation. Bone Marrow Transplant 1996; 18: 1029–1034. Liu Yin JA, Jowitt SN. Resolution of immune-mediated diseases following allogeneic bone marrow transplantation for leukaemia. Bone Marrow Transplant 1992; 9: 31–33. Lowenthal RM, Cohen ML, Atkinson K, Biggs JC. Apparent cure of rheumatoid arthritis by bone marrow transplantation. J Rheumatol 1993; 20: 137–140. van Bekkum DW, Marmont AM, Tyndall A et al. Severe autoimmune diseases: a new target for bone marrow transplantation. Stem Cells 1996; 14: 460–472. Euler HH, Marmont AM, Bacigalupo A et al. Early recurrence or persistence of autoimmune diseases after unmanipulated autologous stem cell transplantation. Blood 1996; 88: 3621–3625. Tyndall A, Black C, Finke J et al. Treatment of systemic sclerosis with autologous haemopoietic stem cell transplantation. Lancet 1997; 349: 254. Snowden JA, Hannah EE, O’Donnell J et al. Resolution of longstanding lupus erythematosus (SLE) after autologous bone marrow transplantation for non-Hodgkin’s lymphoma: a case report. Bone Marrow Transplant 1996; 17 (Suppl. 1): 6 (Abstr. 70). Poser CM, Paty DW, Scheinberg L et al. New diagnostic criteria for multiple sclerosis: guidelines for research protocols. Ann Neurol 1983; 13: 227–231. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983; 33: 1444–1452. Sipe JC, Knobler L, Braheny SL et al. A neurologic rating scale (NRS) for use in multiple sclerosis. Neurology 1984; 34: 1368–1372. Rapp NS, Gilroy J, Lerner AM. Role of bacterial infection in exacerbation of multiple sclerosis. Am J Phys Med Rehabil 1995; 74: 415–418.

33 Patzold U, Pocklington PR. Course of multiple sclerosis: first results of a prospective study carried out on 102 MS patients from 1976–80. Acta Neurol Scand 1982; 65: 248–266. 34 Ebers GC. Treatment of multiple sclerosis. Lancet 1994; 343: 275–279. 35 Copelan EA, Deeg HJ. Conditioning for allogeneic marrow transplantation in patients with lymphohematopoietic malignancies without the use of total body irradiation. Blood 1992; 80: 1648–1658. 36 Burt RK, Hess A, Burns W et al. Syngeneic bone marrow transplantation eliminates autoreactive Vb8.2 T lymphocytes from the spinal cord of Lewis rats with experimental allergic encephalomyelitis. J Neurosci Res 1995; 41: 526–531. 37 van Besien K, Demuynck H, Lemaistre CF et al. High-dose melphalan allows durable engraftment of allogeneic bone marrow. Bone Marrow Transplant 1995; 15: 321–323. 38 Kaplan RS, Wiernik PH. Neurotoxicity of antitumor agents. In: Perry MC, Yarbro JW (eds). Toxicity of Chemotherapy. Grune & Stratton: Orlando, 1984, pp 365–431. 39 Burgue F, Maraninchi D, Olive D et al. Incidence and characterization of residual host lymphocytes after various conditioning regimens to bone marrow transplantation. Bone Marrow Transplant 1988; 3 (Suppl. 1). 199–200. 40 Powles RL, Milliken S, Helenglass G et al. The use of melphalan in conjunction with total body irradiation as treatment for acute leukaemia. Transplant Proc 1989; 21: 2955–2957. 41 Gassmann W, Uharek L, Erbersdobler A et al. High-dose cytostatic agents in allogeneic bone marrow transplantation: comparison of the engraftment promoting potential. Br J Haematol 1991; 78: 310–318. 42 Grau JM, Casademont J, Monforte R et al. Myasthenia gravis after allogeneic bone marrow transplantation: report of a new case and pathogenic considerations. Bone Marrow Transplant 1990; 5: 435–437. 43 Aldouri MA, Ruggier R, Epstein O, Prentice HG. Adoptive transfer of hyperthyroidism and autoimmune thyroiditis following allogeneic bone marrow transplantation for chronic myeloid leukaemia. Br J Haematol 1990; 74: 118–119. 44 Vialettes B, Maraninchi D, San Marco MP et al. Autoimmune polyendocrine failure – type I (insulin-dependent) diabetes mellitus and hypothyroidism – after allogeneic bone marrow transplantation in a patient with lymphoblastic leukaemia. Diabetologia 1993; 36: 541–546. 45 Moreau T, Thorpe J, Miller D et al. Preliminary evidence from magnetic resonance imaging for reduction in disease activity after lymphocyte depletion in multiple sclerosis. Lancet 1994; 344: 298–301. 46 Weinshenker BG, Bass B, Karlik S et al. An open trial of OKT3 in patients with multiple sclerosis. Neurology 1991; 41: 1047– 1052. 47 Lindsey JW, Hodgkinson S, Mehta R et al. Phase 1 clinical trial of chimeric monoclonal anti-CD4 antibody in multiple sclerosis. Neurology 1994; 44: 413–419. 48 Goodkin DE, Rudick RA, Van der Brug-Medendorp S et al. Low-dose (7.5 mg) oral methotrexate reduces the rate of progression in chronic progressive multiple sclerosis. Ann Neurol 1995; 37: 30–40. 49 Karussis DM, Meiner Z, Lehmann D et al. Treatment of secondary progressive multiple sclerosis with the immunomodulator linomide: a double blind, placebo-controlled pilot study with monthly magnetic resonance imaging evaluation. Neurology 1996; 47: 341–346. 50 Kolb HJ, Guenther W, Duell T et al for the IBMTR and EBMT/EULEP Study Group on Late Effects. Cancer after bone marrow transplantation. Bone Marrow Transplant 1992; 10 (Suppl. 1): 135–138.