ORIGINAL ARTICLES

Efficacy and Toxicity of Cyclosporine in Chro& Progressive Multiple Sclerosis: A Randomized, Double-blinded, Placebo-controlled Clinical Trial The Multiple Sclerosis Study Group”

Patients with clinically definite multiple sclerosis, mild to moderately severe neurological disability (entry score on the Expanded Disability Status Scale (EDSS) between 3.0 and 7.0),and a progressive course defined by an increase in the EDSS of between 1and 3 grades in the year prior to entry were randomized to receive either cyclosporine (n = 273) or placebo (n = 274) in a 2-year, double-blinded, multicenter trial. Treatment groups at entry proved balanced for age, gender, duration of illness, and neurological disability. Cyclosporine dosage was adjusted for toxicity and a median trough whole-blood level was maintained between 310 and 430 nglml. The mean increase in EDSS score was 0.39 2 1.07 grades for cyclosporine-treated patients and 0.65 1.08 grades for placebo-treated patients from entry until the time of early withdrawal or completion of the study ( p = 0.002). Of three primary efficacy criteria, cyclosporine delayed the time to becoming wheelchair bound ( p = 0.038; relative risk, 0.765), but statistically significant effects were not observed for “time to sustained progression” or on a composite score of “activities of daily living.” Active treatment did have a favorable effect on several secondary measures of disease outcome. A large and differential withdrawal rate (44% for cyclosporine-treatedpatients, 32% for placebo-treated patients) complicated the analysis but did not appear to explain the observed effect of cyclosporine in delaying disease progression. Multivariate analysis did not show institutional effects but did demonstrate substantial effects of baseline neurological disability on outcome. Nephrotoxicity and hypertension were common troublesome toxicities and accounted for most of the excess loss of patients in the cyclosporine arm of the study. Thus, chronic cyclosporine therapy was associated with a statistically significant but clinically modest delay of progression of disability in a group of patients with multiple sclerosis selected for moderately severe and progressive disease. Close supervision by physicians familiar with cyclosporine is mandatory to minimize known adverse effects, particularly nephrotoxicity, when considering the use of this immunosuppressant.

*

The Multiple Sclerosis Study Group. Efficacy and toxicity of cyclosporine in chronic progressive multiple sclerosis: a randomized, double-blinded, placebo-controlled clinical trial. Ann Neurol 1990;27:591-605

T h e Multiple Sclerosis Study Group comprises the following. Participating institutions, principal investigators (italics), and investigative teams: Albert Einstein College of Medicine-Lube C. Scheinberg, MD, Charles R. Smith, MD, Barbara S. Giesser, MD, Ute Traugott, MD, Mindy Aisen, MD, Nicholas La Rocca, PhD, Kate Robbins, CCOT; Duke University-Barrie J. Hurwitz, MD, Steven J. Greenberg, MD; Louis M. Fredane, MD, Rebecca Herbstreith, RN, Jean G. Hurwin, BS Pharm, James R. Burke, MD; The National Institutes of Health-Henry F. McFarland, MD, Andrew Goodman, MD, Dale E. McFarlin, MD, Helen Krebs, RN, Heidi Maloni, RN, Joe Debronso, PhD Pharm; The Rocky Mountain MS Center-Gary M. Franklin, MD, MPH, Jack S. Burks, MD, Lorene Nelson, MS, Carolyn Wangaard, RN, CANP; The University of Arizona-William Sibley, MD, Jose Laguna, MD, Joan Laguna; The University of Chicago-Burg G. W. Arnason, MD, Raymond P. Roos, MD, Anthony T. Reder, MD, Mark A. Agius, MD, Roberta A. Martia, RN; T h e University of Maryland-Kenneth P. Johnson, MD; Hillel Panitch, MD, Carol Lee Koski, MD, Paul Fishman, MD, Sue Haley, RN; T h e University of Texas Health Science Center at Dallas-Jonathan Walker, MD, Richard S. A. Tindall, MD, J. Theodore Phillips, MD, PhD, Ovella Taylor, Julia Rollins, RN; The S. University of Texas Health Science Center at Houston-]erg Wohsky, MD, E. Simon Sears, MD, Avindra Nath, MD, Catherine A. Weisbrodt, RN; The University of Utah-Jack H. Petajan, MD, PhD, Patrick F. Bray, MD, John W. Rose, MD, David J. Thurman, MD, William A. Galster, MS; The Veterans Adminis-

tration Wadsworth and The University of California Los Angeles Medical Centers-Wallace W. Tourtellotte, MD, PhD, Robert W. Baumhefner, MD, George W. Ellison, MD, Lawrence W. Myers, MD, Karl Syndulko, PhD, Lavonna Newton, RN, Iris Rosario, RN, Bridgett Fahy-Chandon, BA, Diane C. Guntrip; Washington University-John L. Trotter, MD, David B. Clifford, MD, Larry Smith, MD, Jane McInnis, RN. External Advisoty Committee: Chairman, Stanley van den Noort, MD, The University of California Irvine; Jack Antel, MD, The Montreal Neurological Institute; Joe R. Brown, MD, The University of California San Diego; Paul Keown, MD, The University of British Columbia; Leslie Weiner, MD, The University of Southern California; Ruth B. Loewenson, PhD, The University of Minnesota; E. David Mellits, ScD, The Johns Hopkins University; Stephen C. Reingold, PhD, The National Multiple Sclerosis Society. Sandoz Phamceuticah Research Institute Coordinators: George Belendiuk, MD, PhD, Lisa Beth Ferstenberg, MD, William Mietlowski, PhD, Michael Remavich, Barbara J. Kasper, RN, Suzanne Solch, RN, Debra J. Resta, RN, Kathleen Sandford, RN. Received Sep 13, 1787, and in revised form Dec 4. Accepted for publication Dec 9, 1787. Address correspondence to Dr Wolinsky, The University of Texas Health Science Center at Houston, PO Box 20708, Houston, TX 77225.

Multiple sclerosis is the most common, nontraumatic cause of neurological disability among young adults and afflicts an estimated 250,000 individuals in this country 111. While the cause and pathogenesis of the disease remain uncertain, several lines of evidence point toward an immunopathological mechanism in the generation of the demyelinated lesions that characterize this disorder 121. These include the immunoarchitecture of active plaques [3-61, intrathecal synthesis of immunoglobulins of restricted heterogeneity 17, 81, association of disease frequency with major histocompatibility antigen complex phenotypes 191, systemic immune abnormalities consistent with immune activation 110-121, and loss of suppressor cell function or number 113, 141. Further, certain forms of immunomodulatory therapy have been reported to transiently benefit patients at selected stages of the disease 115171, while others appear to induce clinical attacks 1181. Cyclosporine A is a cyclic undecapeptide that exhibits potent immunosuppressive activity by a relatively selective inhibitory effect on the helper population of T lymphocytes. This is probably mediated through inhibition of the transcription of messenger RNA (mRNA) for interleukin 2 (IL-2) and other lymphokines 119-22). Cyclosporine has been successfully used in transplantation for the prevention of graft rejection 1231. In principle, treatment with cyclosporine should be most effective in preventing the development of antigen-specific effector immune responses when therapy is initiated before the introduction of a foreign antigen. Nonetheless, preliminary reports on the use of cyclosporine in controlling putative autoimmune disease in humans 124-261 and the drug’s ability to modify the course of experimental allergic encephalomyelitis 127, 281, a T-cell-mediated animal model for multiple sclerosis, suggested that cyclosporine might be of value in the management of multiple sclerosis 129-31). This report details the results of a double-blinded, placebo-controlled, 2-year, multicenter, parallel group study of the efficacy and safety of cyclosporine in the treatment of chronic progressive multiple sclerosis.

Methods Study Organization and Design This study was initially conceived as a multi-institutional, double-blinded, placebo-controlled trial with the preliminary trial design formulated by Sandoz Research Institute coordinators and formally presented to representatives of participating centers in January 1984. Following additional input from the investigators and the Food and Drug Administration (FDA), a final study design was agreed on and approved by the institutional review boards of each of the 12 participating centers. An independent external advisory committee was developed, and the study begun in January 1985. For determination of sample size, it was assumed to be 592

Annals of Neurology Vol 27

N o 6 June 1990

important to detect a 0.5-point difference in the mean change from baseline of the Expanded Disability Status Scale (EDSS) score 132) for the cyclosporine-treated group relative to the placebo-treated group at the time of scheduled efficacy analyses. Selected published 133, 341 and unpublished natural history data (L. W. Myers, M. R. Mickey, and G. W. Ellison; D. W. Paty and J. Petkau, personal communications, 1984) on change from baseline EDSS scores over time for patients with multiple sclerosis were used for this purpose, although these data were admittedly less than ideal. Efficacy and safety analyses were planned when 50% of the randomized patients had been followed for 12 months and when all randomized patients were followed for 12 and 24 months. An analysis of safety alone was planned when 50% of the randomized patients were followed for 6 months. Estimates of sample size utilized a Tukey-type repeated significance test procedure 1351 with a power of 0.8 at an overall significance level of 0.05. The projection was that 108 patients per treatment group were required to detect a 0.5 difference in the mean changes of the EDSS score over time at a variance of less than 1.5 for a two-way analysis of covariance model, using center and treatment as classification factors and a one-tailed test. To satisfy an FDA requirement of two pivotal studies in chronic progressive multiple sclerosis for a new drug application for cyclosporine, the 12 centers were arbitrarily assigned to one of two studies, with each center to enter a minimum of 30 patients. Patient accession was closed on December 31, 1985, with the last entered patient completing the blinded phase of treatment at the end of 1987. Premature stopping rules for unacceptable toxicity or drug efficacy were never reached for the overall study. Study coordination, data collection, drug usage monitoring, and initial statistical evaluations were organized and effected by Sandoz Pharmaceuticals. The investigative team at each center was composed of at least an unblinded investigator who had no direct patient contact and whose sole role was to monitor toxicity in individual patients and adjust study medication levels; a blinded study coordinator who arranged all follow-up visits, entered data, and administered instrumented performance examinations; and two blinded neurologists who performed the clinical evaluations. An external advisory committee was organized and consisted of individuals, independent of Sandoz and the participating centers, who had expertise in clinical trials in multiple sclerosis, in study design and evaluation, andor in the nephrotoxicity of cyclosporine. They were selected with advice from The National Multiple Sclerosis Society and charged with the responsibility of reviewing the results of the interim analyses and determining the ethics of continued conduct of the trial. Their decisions were communicated to the investigators without any details concerning results of the analyses. Efficacy was assessed by changes in scores on the EDSS and the incapacity status scales (ISS), and functional system scores of the Multiple Sclerosis Minimal Record of Disability (MSMRD) 136, 37); standardized neurological examination; quantitative examination of neurological function (QENF) 138, 39); ambulation index (AI) 116); physical examination; and clinical evaluation. Safety was assessed by physical examination, laboratory evaluation, electrocardiographic results, vital signs, and adverse reaction profile.

Patient Characteristics Entry criteria included the diagnosis of clinically definite multiple sclerosis for a minimum of 1 year [40), an entry rating on the EDSS of grade 3.0 to 7.0 (inclusive), and an age of 18 to 5 5 years. Patients were required to have shown chronic and progressive clinical deterioration of at least 1 grade but not more than 3 grades on the EDSS in the 12 months prior to entry, with some clinical decline within the previous 6 months, but no acute relapse in the 3 months immediately prior to randomization. They were not to have received any immunosuppressive drug such as azathioprine, corticosteroids, or interferon in the 3 months prior to randomization; to have discontinued unproven treatments for multiple sclerosis such as hyperbaric oxygen, gangliosides, or snake venom at least 1 month prior to study entry; and to never have been treated with cyclophosphamide or radiation. All patients were required to be properly motivated and able to cooperate for the study, and to be otherwise physically healthy, without uncontrolled hypertension (systolic pressure >I70 mm Hg or diastolic pressure >I10 mm Hg), malignancy, recent myocardial infarction, chronic pulmonary disease, active infection, hepatic or renal dysfunction, or other neurological disorders. Medications known to interfere with the metabolism of cyclosporine were prohibited. Patients were specifically excluded from participation for known sensitivity or adverse reactions to immunosuppressive agents, severe dementia, paraplegia or gait ataxia sufficient to prevent walking, or severe upper extremity ataxia precluding independent feeding or dressing.

Drag Administration Cyclosporine at a concentration of 100 mglml was provided in an olive oil-Labrafil (Gattefosse Etablissements, France) base, with matching placebo consisting only of the pharmacologically inert ingredients. Study medication was initiated at a dosage of 6 mglkg of the suspension diluted at least 1: 10 in milk or orange juice and taken each morning with breakfast. Whole-blood cyclosporine trough levels were determined on all patients by radioimmunoassay based on a polyclonal antibody (Sandoz Pharmaceuticals) as performed in a reference laboratory (MetPath, Teterboro, NJ). Dosage was adjusted by an unblinded observer at each institution to initially attempt to maintain a trough level between 400 and 600 nglml and later reduced to 300 to 500 ngjml, to avoid a serum creatinine level of more than 1.5 times the baseline level and to not exceed a maximum dose of cyclosporine of 10 mglkgl day. Study medication bottles were prepared, coded, packaged, and labeled by Sandoz Pharmaceuticals and preassigned to patient numbers using a computerized random allocation system and a block size of 4.Prospective patients were prescreened by the investigative teams at individual centers and entered as groups of 4. This ensured the availability of both placebo- and cyclosporine-treated patients for whom the unblinded investigator, who had no contact with study patients, could make changes in the study medication dosage in such a fashion as to preserve the double-blinded study design. Patients were formally entered into the study following informed consent and the medication was begun on the morning following entry. Changes in concomitant medications which might beneficially affect neurological function (i.e., baclofen, dantro-

lene) were proscribed as was the institution of new or enhanced programs of physical, occupational, or speech therapy. Treatment with other immunosuppressants including corticotropin or corticosteroids was not allowed.

Clinical Assessment At entry, all patients underwent extensive baseline evaluation including an entry checklist; background information; a demographic and risk factor questionnaire; hematological and blood chemical studies; urinalysis and urine culture; cerebrospinal fluid examination; electrocardiography; standard physical examination with vital signs recorded; and a standardized neurological examination which included a definition of abnormalities to be recorded, MSMRD, QENF, and AI. Participating neurologists were trained in the application of the EDSS during a joint session which included repeated rating of videotaped presentations of patients with multiple sclerosis who had varying levels of disability. All clinical coordinators underwent detailed training in the administration of the instrumented performance tests of the QENF. Patients at seven centers also underwent magnetic resonance imaging (MRI). All of the above safety and efficacy measurements were to be conducted at completion of the study or if the patient was withdrawn from the study prematurely. Blood and urine laboratory studies were repeated at 2 weeks after study entry and then at monthly intervals. All measures of neurological function were repeated at 3-month intervals by the blinded study coordinator and blinded evaluating neurologist. Adverse drug reactions and concomitant medications were enumerated by the blinded study coordinator on each return visit. Exacerbations were recorded at the 3-month interval visits. All serious potential adverse drug reactions or deteriorations in neurological status were immediately reported to one of the blinded investigators.

Withdrawals All randomized patients who failed to complete the 24month study were to undergo a complete exit examination as detailed. Patients were prematurely withdrawn from the study on their request or for protocol violation, serious intercurrent illness precluding continued treatment with study medication, poor study compliance, inability to tolerate study medication, significant adverse reactions, or unrelenting continued progression of disease. Efforts were made to follow all patients who were prematurely withdrawn from the study by continued monthly clinic visits for monitoring laboratory data and vital signs and at quarterly visits for clinical assessment for up to 6 months. Postwithdrawal evaluations included detailed standardized recorded examinations. In addition, after completion of the study, all patients who were withdrawn early were contacted using a standardized questionnaire. This was done as an attempt to retrieve milestone information (wheelchair-bound state, dependence in activities of daily living attained in the 24-month period during which patients would have been in the study) to complete an intention-to-treat analysis (see also below). All patients who received the study medication for 24 months were given the option to begin cyclosporine under open label conditions, assuming that they showed acceptable renal function as determined by the unblinded investigator. MS Study Group: Cyclosporine in CPMS

593

Statistical Analysis The EDSS was the principal efficacy variable on which the study sample size was estimated. The a priori choice of the EDSS was based on its wide use as a measure of disability in patients with multiple sclerosis, its reduction of scores from eight diverse functional systems into a single grade, and its presumed ordinal description of disability. It was also assumed that an end point analysis would adequately handle “dropout” patients who were prematurely withdrawn from the study. Based on an early analysis of baseline data and the actual observed frequency and timing of withdrawals, it was thought that the above assumptions might prove unreliable and thus require the use of more refined analytical strategies. Spearman rank correlation analysis of the baseline EDSS with 69 other efficacy variables showed a high correlation only with the A1 and the ambulation item on the ISS, and displayed a near linear correlation with the A1 over the EDSS range of 5.5 to 7.0 (inclusive) C41). These observations suggested that the EDSS might not behave in an ordinal manner and resulted in the use of a “collapsed EDSS (cEDSS) in which patients with an EDSS score of 3.0 to 5.5 (inclusive) were arbitrarily assigned an EDSS grade of 5.5 for some statistical analyses (a more detailed discussion will be published separately). A high withdrawal rate was encountered as the study progressed. By study end this totaled 37.3% of all patients. Both the extent and variable occurrence of withdrawals over the course of the study and relative to fixed efficacy evaluation visits every third month raised concern for the adequacy of the original statistical approach and the evaluable sample size. These considerations were raised prior to viewing and independent of any knowledge of an efficacy analysis. Thus, primary efficacy criteria were reexamined and redefined in survival analytical terms. This technique was adopted since the primary efficacy objective to determine if cyclosporine delays progression of disease in patients with chronic progressive multiple sclerosis could readily be posed in terms of time to significant clinical milestones of deterioration. Such an approach appeared superior in dealing with the high withdrawal rate. As a consequence of the above, three additional primary outcome criteria were selected for efficacy analysis: (1) time to sustained progression defined as the occurrence of either a 0.5-grade increase on the cEDSS or a l-point increase on the A1 relative to baseline, which was sustained for two consecutive efficacy assessments or evident at the exit visit; (2) first time to becoming wheelchair bound defined by either an EDSS grade of 7.0 or an A1 score of 7; and (3) a composite factor analysis of the baseline dressing, grooming, and feeding items on the ISS extracted as an “activities of daily living” (ADL) rating, which provided the end point of time to a total ADL rating of 6. Four secondary efficacy criteria that were chosen included: (1)an end point analysis of the change from baseline cEDSS for both patients completing 12 and 24 months of the study (“completers”); (2) overall assessments of the blinded neurologists; (3) patient’s global assessment of change from baseline to 12 and 24 months for completers; and (4) comparisons of the duration of patient follow-up in the two study arms. Primary efficacy criteria were analyzed using KaplanMeier survival analytical methods [42}. The proportioned

594 Annals of Neurology Vol 27 No 6 June 1990

hazards model of Cox 143) was used to test for the presence of treatment by center interactions and the model was adjusted for the prognostic factors of QENF of the lower extremities for the time to sustained progression and the time to becoming wheelchair bound, and of QENF for the upper extremities for the time to progression to an ADL score of 6. Secondary measures of efficacy were analyzed by a two-way analysis of covariance applied to the cEDSS to assess treatment by center interactions; the absence of such interaction allowed the data to be pooled and a one-way analysis of covariance model used with the lower extremity QENF composite as a covariant. Nonparametric techniques including rank analysis of covariance {44] were applied to assess consistency and robustness of the parametric results. Similar approaches were applied to the other secondary measures of efficacy.

Results EfFcacY

A total of 554 patients were entered into the study. Seven patients were excluded from further consideration; 4 had been randomized to receive cyclosporine and 3, to receive placebo. Of these, 5 never received treatment and declined continued participation after enrollment in the study and 1 was lost to follow-up after the baseline evaluation. The remaining, placebo-randomized patient was unable to tolerate the medication and discontinued the study before the week 2 visit; no followup data were obtained. Thus, of the remaining 547 patients, 2 73 were randomized to receive cyclosporine and 274, to receive placebo as outlined. The randomization was successful as judged by the similar proportion of patients allocated to each study arm by age, gender, race, height, weight, abnormalities recorded on general physical examination, duration of disease from first symptoms or diagnosis to entry, neurological disability, and distribution of neurological abnormalities and their severity by functional system scores of the EDSS components of the standardized neurologic examination or the multiple individual measures of the QENF (representative data are shown in Table 1). The distribution of patients by disability as measured by EDSS and A1 into the two study arms is shown in Table 2. For one-third of all patients the entry EDSS was collapsed to 5.5 for those analyses based on the cEDSS, and such patients were required to undergo proportionally greater accumulation of disability to be scored as having progressed clinically as defined already. PATIENT CHARACTERISTICS AT ENTRY.

E N D POINT OUTCOMES AND WITHDRAWAL EFFECTS.

Only 56% of those patients randomized to receive cyclosporine completed 24 months of continuous therapy, whereas 68% of those randomized to placebo successfully completed the trial ( p = 0.003). The mean durations of treatment for the cyclosporine- and

Table 1. Selected Patient Characteristics at Entry

Age (yr) Female :male ratio White patients Height (cm) Weight (kg) Married Divorced Employed outside home Homemaker Years of education Months from diagnosis Months from first symptom Expanded Disability status Scale Ambulation Index

Cyclosporine (n = 273)

Placebo (n = 274)

40.5 ? 7.7" 1.35: 1 92 % 169.3 ? 9.8 69.5 ? 16.3 74% 12% 36%

40.6 ? 8.2a 1.32: 1 93% 169.5 11.2 69.0 k 15.9 71% 10% 35%

23% 14.3 ? 2.7 72.0 ? 61.5 123.0 ? 82.8

23% 14.3 74.7 129.5

*

5.4

k

1.2

* 2.5 * 68.7 * 88.3 5.4 * 1.2

3.8

5

1.8

3.8

* 1.7

"Mean 2 I standard deviation.

placebo-treated patients were similar (18.4 & 7.6 and 19.3 k 7.6 months, respectively; p = 0.18). Yet, the proportion of patients withdrawn from the trial over any time interval was fewer in the placebo-treated arm ( p = 0.009; relative risk, 1.436). However, the completers in both arms of the study had comparable baseline EDSS distributions ( p 1 0.6). At the time of exit from the study (whether by completion of 24 months as planned or at the time of withdrawal from the study for any reason) the cyclosporine-treated patients displayed a mean deterioration of neurological function, as measured by the EDSS, of 0.39 -+ 1.07 compared to a deterioration of EDSS of 0.65 ? 1.08 by their placebo-treated counterparts ( p = 0.002).' The mean deterioration in EDSS score for patients withdrawn from the study during any 6-month interval favored treatment with cyclosporine over placebo, and more placebo-treated patients were prematurely with'Nonparametric one-way rank analysis of covariance of entry and exit EDSS scores for all evaluable patients were also statistically significant ( p < 0.001).

Table 2. Distribution of Extended Disability Status Scale (EDSS)and Ambukation index (Ai) Scores at Randomization

EDSS

Cyclosporine"

Placebo"

A1

Cyclosporine"

Placebo"

140 mm Hg) andor diastolic (> 90 mm Hg) blood pressure were recorded at least once in 50 and 5996, respectively, of all cyclosporinetreated patients; most patients with sustained hypertension were treated with a variety of antihypertensive agents. Noncompleters had mean elevations in systolic and diastolic blood pressure that were similar to the blood pressures of the completers and of all patients entered into the trial (data not shown). Only 8 patients

\

3003

'\

g F 250 :1 I 0 2 :I c200 1 ; 9 : I 0 I 150

C

.-'0

(21

3

100

I - I

: I

I - 1 :I 5 0 " '

I

'

'

I

"

'

Fig 6. Median trough whole-blood cyclosporine lwek as determined by radioinmunoassaywith the Sanhz polyclonal antibody for actively treatedpatients aver the course of the study. The median level at 24 months of treatment was 306 nglml, with a broad range of observed values fmm undetectable to 1,604 nglml.

-

a

OTHER LABORATORYABNORMALITIES. Mean serum

magnesium levels fell precipitously following the initiation of cyclosporine therapy (Fig 8), with 20% of patients showing magnesium levels below the normal range within 1 month of randomization. Hypomagnesemia can be associated with a variety of neurological symptoms 1471 and might be anticipated to increase spasticity in patients with multiple sclerosis. However, no early adverse clinical effects were reported to suggest that this occurred during the trial, and mean magnesium levels among patients who were withdrawn were similar to those of completers in the active treatment arm. No systematic attempts by the unblinded investigators were made to correct hypomagnesemia. Both mean hemoglobin and hematocrit values dropped during the course of the trial; changes were very minor for placebo-treated patients and reasonably

an abnormal creatinine value (see Fig 5 ) and 84% of all cyclosporine-treated patients showed an elevated serum creatinine level on at least one visit. If clinically significant nephrotoxicity to cyclosporine is defined as a serum creatinine level of 1.5 times baseline or greater, or 2 mgldl or greater over three consecutive visits despite attempts at reductions in cyclosporine dosage, then 56 (21%) of the 273 patients with multiple sclerosis who were randomized to cyclosporine treatment developed such toxicity. Up to 13% of cyclosporine-treated patients developed creatinine levels

Table 5 . Renal Dysfunction: Comparison of Noncompleters to Completers and to AIl Patients Exit Valuesa Months Followed

No.

BUN (mg/dl)

34 30 42 15

18.1 ? 19.4 f 20.4 f 18.9 ?

~~

Creatinine (mg/dl)

Cyclosporine (ndml) -

1.2 ? 1.3 ? 1.3 ? 1.1 f

0.4 0.4 0.5 0.3

310 5 378 146 -I- 130 236 5 229 158 4 122

0.3 0.4

355 2 241 297 2 257

Noncompleters 1-6 1-12 13-18 19-23

7.9 7.9 9.0 6.0

Completers 24

Total

-

152 273

23.1 21.4

"Mean f 1 standard deviation.

BUN = blood urea nitrogen.

600 Annals of Neurology Vol 27 No 6 June 1990

?

6.6

1.3

?

?

7.5

1.3

5

~~

~

,

_..

-- CY-

-- CYdow=+-m

-ob=ba

--p(Kebo

c ' \ W

w E

I

I

, , o L L - - d 0 1 3

6

-

fa!

I

I

1.9

70LI--LL-'----LU?

9 12 16 18 21 24

0 1 3

Month After Entry

6

9

12 15 18 21 24

Month After Entry

Fig 7. Change in mean sitting systolic (left panel) and diastolic (right panel) blood pressure for qclosporine-treated and placebotreated patients under active treatment over the course of the study. Representative actual mean systolic pressure values +- 1 standard deviation for all remaining patients at 3 and 24 months offollw-up were: in the cyclosporinegroup, 123.5 & 16.7 mm Hg (n = 262) and 125.7 16.1 mm Hg (n = 153); in the placebo group, 115.1 C 13.6 mm Hg (n = 260) and 114.4 & 14.7 mm Hg (n = 185), respectively. Mean diastolic pressure values for these patients at 3 and 24 months of follow-up were: in the cyclosporine group, 81.8 C 11.8 mm Hg and 83.2 c 10.5 mm Hg; in the placebo group, 75.4 C 8.9 mm Hg and 76.0 2 9.9 mm Hg, respectively.

cyclosporine placebo

*

attributed to the effects of repeated phlebotomy. This contrasted with cyclosporine-treated patients for whom a maximal mean fall in hemoglobin of 2.1 gm/dl from baseline occurred 1 year into the study in association with a maximal mean reduction of hematocrit of 7.2%. Changes in mean values for these hematological indices were similar for patients who were withdrawn and patients who completed the study (data not shown). Apart from an associated fall in the total red blood-cell count for cyclosporine-treated patients, no other statistically significant or clinically important differences were seen in the hematological profiles. No statistical differences at entry or at exit were noted between the treatment groups for serum sodium, calcium, chloride, inorganic phosphorus, glucose, total protein, glutamic-oxaloacetic transaminase (SGOT), gamma glutamyl transpeptidase (GGTP), or C-reactive protein levels. Statistically significant but clinically unimportant increases (potassium, direct bilirubin, alkaline phosphatase, lactic acid dehydrogenase) or decreases (bicarbonate, albumin, serum glutamicpyruvic transaminase {SGPT)) were observed at exit for the cyclosporine-treated group (data not shown). Increases in serum cholesterol and triglyceride levels were found in both treatment groups over time; the levels were higher in the cyclosporine-treated than in the placebo-treated group (placebo group at 24 months: increased mean cholesterol level, 8.4 mg/dl,

1.6'

' ' 0 1 3

1

I

6

9

I

I

I

I

I

12 15 18 21 24

Month After Entry Fig 8. Mean serum magnesium levelsfor actively treated patients in both study arms over 24 months. Actual mean magnesium values c 1 standard deviationfor all remaining patients at 1 and 24 months of follow-up were: in the qclosporine group, 1.65 0.14 mEqlL (n = 265) and 1.75 C 0.17 mEqlL (n = 153); in the placebo group, 1.89 0.13 mEqIL (n = 272) and 1.91 C 0.14 mEqlL (n = IS>), respectiveb.

*

*

and triglyceride level, 17.7 mg/dl; cyclosporine group at 24 months: increased mean cholesterol level, 26.7 mg/dl, and triglyceride level, 36.1 mg/dl). A mean increase of 1.3 mg/dl in uric acid levels in the cyclosporine-treated group was also evident at 24 months of the trial. The clinical impact of the latter changes are uncertain. Symptoms suggesting adverse effects were sought at each follow-up visit by the patient coordinator using a standardized interview approach, and abnormalities noted on the physical examination were systematically recorded by the patient coordinator or blinded neurologist. Any adverse effect recorded for a particular patient at a given visit was scored only once for purposes of determining the frequency of these events. The 10 most common adverse reactions recorded in each treatment arm and their frequency of occurrence are shown in Table 6. For the most part, these events did not require specific treatment or necessitate withdrawal of patients from the study (see also below). COMMONLY REPORTED ADVERSE EFFECTS.

MS Study Group: Cyclosporine in CPMS 601

Table 6. Most Commonly Reported Adverse Reactions C yclosporine

Placebo

~~

Hirsutism Headache Hypertension Gingival hyperplasia Nausea Paresthesia Edema Abdominal discomfort Fatigue

66.5% 38.7% 34.7% 32.7%

Headache Hirsutism Rash“ Nausea

20.7% 16.4% 15.6% 15.3%

28.4% 27.3% 25.8% 21.1%

12.7% 11.6% 10.5% 7.8%

Rash”

14.5%

Fatigue Dizziness Edema Gingival hyperplasia Extremity abnormalityb Ataxia

14.7%

7.8% 8.4%

“Includes other reported skin changes in addition to rash. bPrimarily recorded as distal sensory symptoms.

Table 7. Distribution of Causes for the 207 Patients (38%) Who Discontinued the Study Cyclosporine ~~~

Placebo

~~

Treatment failure (study drug ineffective) Protocol violator Unrelated illness Other Subtotal

47 (17%)

57 (22%)”

27 (10%) 12 (4%) 3 (1%) 87 (33%)

17 (7%)

Adverse reactions

31 (11%) 120 (44%)

Total ~~

6 (2%) 2 (1%) 86 (31%) 1 (lo0 mm Hg for diastolic blood pressure in about 20% of all cyclosporine-treated patients). Only 8 patients were withdrawn on the basis of uncontrolled hypertension despite therapy and/or a reduction in cyclosporine dosage. Other side effects of cyclosporine that were severe enough to cause patient withdrawal included gastric distress ( 5 patients); headaches (2 patients); and diarrhea, leg discomfort, and gingival hyperplasia (1 patient each). However, in some instances multiple unacceptable side effects may have occurred with only one cause recorded. In an unknown number of instances, the occurrence of otherwise acceptable side effects may have contributed to patient withdrawal in the face of a lack of perceived clinical improvement. Thus, while the adverse effect profile of cyclosporine in carefully monitored and managed patients in this and other studies E30, 31) of multiple sclerosis is not atypical, cyclosporine cannot be viewed as a drug without substantial potential risk. In conclusion, this study documented that continuous oral therapy with cyclosporine is associated with a statistically significant but clinically modest delay in the accumulation of functional neurological disability in patients with chronic progressive multiple sclerosis who were selected for actively evolving disease. Whether the nominal delay in clinical progression shown would be sustained beyond the 2-year duration of the study is unknown. The burdensome and potentially serious toxicity of cyclosporine must temper one’s consideration of its use in this disease. The clinical study was supported by Sandoz Pharmaceuticals with some additional support from The National Multiple Sclerosis Society. The assistance of Mrs Shirley Roe in the preparation of this manuscript is appreciated. ~

~~

References 1. Baum HM, Rothschild BB. The incidence and prevalence of reported multiple sclerosis. Ann Neurol 1981;10:420-428 2. Waksman BH. Autoimmunity in demyelinating diseases. Ann N Y Acad Sci 1988;540:13-24 3. Traugott U, Reinherz EL, Raine CS. Multiple sclerosis. Distribution of T cells, T cell subsets and Ia-positive macrophages in lesions of different ages. J Neuroimmunol 1983;4:201-221

604 Annals of Neurology Vol 27 No 6 June 1990

4. Hauser SL, Bhan AK, Gilles F, et al. Immunohistochemical analysis of the cellular infiltrate in multiple sclerosis lesions. Ann Neurol 1986;19:578-587 5. McCallum K, Esiri MM, Tourtellotte WW, Booss J. T cell subsets in multiple sclerosis. Gradients at plaque borders and differences in nonplaque regions. Brain 1987;110:1297-1308 6.Traugott U, Lebon P. Interferon-gamma and Ia antigen are present on astrocytes in active chronic multiple sclerosis lesions. J Neurol Sci 1988;84:257-264 7. Tourtellotte WW, Staugaitis SM, Walsh MJ, et al. The basis of intra-blood-brain-barrier IgG synthesis. Ann Neurol 1985; 17:21-27 8. Walsh MJ, Tourtellotte WW. Temporal invariance and clonal uniformity of brain and cerebrospinal IgG, IgA, and IgM in multiple sclerosis. J Exp Med 1986;163:41-53 9. Odum N, Hyldig-Nielsen JJ, Morling N, et al. HLA-DP antigens are involved in the susceptibility to multiple sclerosis. Tissue Antigens 1988;31:235-237 10 Rose LM, Ginsberg AH, Rothstein TL, et al. Selective loss of a subset of T helper cells in active multiple sclerosis. Proc Natl Acad Sci USA 1985;82:7389-7393 11 Morimoto C, H d e r DA, Weiner HL, et al. Selective loss of the suppressor-inducer T-cell subset in progressive multiple sclerosis. Analysis with anti-2H4 monoclonal antibody. N Engl J Med 1987:3 16:6?-72 I L . Kerman RH, Wolinsky JS, Nath A, Sears ES Jr. Serial immune evaluation of cyclosporine- and placebo-treated multiple sclerosis patients. J Neuroimmunol 1988;18:325-331 13. Antel JP, Bania MB, Reder A, Cashman N. Activated suppressor cell dysfunction in progressive multiple sclerosis.J Immunol 1986;137:137- 141 14. Hafler DA, Weiner HL. T cells in multiple sclerosis and inflammatory central nervous system diseases. Immunol Rev 1987; 100:307-332 15. Rose AS, Kuzma JW, Kurtzke JF, et al. Cooperative study in the evaluation of therapy in multiple sclerosis. ACTH vs placebo-final report. Neurology 1970;20:1-59 16. Hauser SL, Dawson DM, Lehrich JR, et al. Intensive immunosuppression in progressive multiple sclerosis. N Engl J Med 1983;308: 173- 180 17. Milligan NM, Newcombe R, Compston DA. A double-blind controlled trial of high dose methylprednisolone in patients with multiple sclerosis. J Neurol Neurosurg Psychiatry 1987;50: 511-516 18. Panitch HS, Hirsch RL, Schindler J, Johnson KP. Treatment of multiple sclerosis with gamma interferon: exacerbations associated with activation of the immune system. Neurology 1987; 37:109?-1102 19. Hess AD, Colombani PM, Esa AH. Cyclosporinc and the immune response: basic aspects. CRC Crit Rev Immunol 1986; 6:123-149 20. Colombani PM, Hess AD. T-lymphocyte inhibition by cyclosporine. Potential mechanisms. Biochem Pharmacol 1987;36: 3789-3793 21. Drugge RJ, Handschumacher RE. Cyclosporine: mechanism of action. Transplant Proc 1988;2O(suppl 2):301-309 22. Hess AD, Esa AH, Colombani PM. Mechanisms of action of cyclosporine: effect on cells of the immune system and on subcellular events in T cell activation. Transplant Proc 1988; ~ O ( S U P P 2):29-40 ~ 23. Cohen DJ, Loertscher R, Rubin MF, et al. Cyclosporine: a new immunosuppressive agent for organ transplantation. Ann Intern Med 1984;101:66?-682 24. Nussenblatt RB, Palestine AG, Rook AH, et al. Treatment of intraocular inflammatory disease with cyclosporin A. Lancet 1983;2:235-238 1 1

25. Stiller CR, Dupre J, Gent M, et al. Effects of cyclosporine immunosuppression in insulin-dependent diabetes mellitus of recent origin. Science 1984;223:1362-1367 26. Schindler R. Ciclosporin in autoimmune diseases. Berlin: Springer-Verlag, 1985 27. Bolton C, Bore1 JF, Cuzner ML, et al. Immunosuppression by cyclosporine A of allergic encephalomyelitis.J Neurol Sci 1982; 56~147-153 28. Hinrichs DJ, Wegmann KW, Peters BA. The influence of cyclosporine A on the development of actively induced and passively transferred experimental allergic encephalomyelitis. Cell Immunol 1983;77:202-209 29. Johnson KP, Belendiuk G. Use of cyclosporine in neurological autoimmune disease. Arch Neurol 1985;42:1043-1044 30. Kappos L, Panold U, Dommasch D, et al. Cyclosporine versus azathioprine in the long-term treatment of multiple sclerosis: results of the German multicenter study. Ann Neurol 1988; 23156-63 31. Rudge P, Koetsier JC, Mertin J, et al. Randomized double blind controlled trial of cyclosporin in multiple sclerosis. J Neurol Neurosurg Psychiatry 1989;52:559-565 32. Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983; 3311444-1452 33. Millar JHD, Zilkha KJ, Langman MJ, et al. Double-blind trial of linoleate supplementation of the diet in multiple sclerosis. Br Med J 1973;1:765-768 34. Bates D, Fawcett PRW, Shaw DA, Weightman D. Polyunsaturated fatty acids in the treatment of acute remitting multiple sclerosis. Br Med J 1978;2:1390-1391 35. Tukey JW. Some thoughts on clinical trials, especially problems of multiplicity. Science 1977;198:679-684 36. LaRocca MG, Scheinberg LC, Slater RJ, et al. Field testing of a minimal record of disability in multiple sclerosis; the United States and Canada Acta Neurol Scand 1984;7O(suppl 101): 126-138 37. Slater RJ. Criteria and uses of the Minimal Record of Disability in multiple sclerosis. Acta Neurol Scand 1984;7O(suppl 101): 16-20 38. Potvin AR, Tounellotte WW. Quantitative examination of neurologic function. Boca Raton, FL: CRC Press, 1985 39. Tourtellotte WW, Syndulko K. Quantifying the neurologic examination. In: Munsat T, ed. Quantification of neurologic deficit. Stoneham, MA: Butterworths, 1989

Kunzke JF, et al. Cooperative study in 40. Rose AS, Kuzma JW, the evaluation of therapy in multiple sclerosis; ACTH vs placebo in acute exacerbations. Preliminary report. Neurology 1968;18(suppl):1- 10 41. Multiple Sclerosis Study Group. Measuring disease status in a clinical trial of cyclosporine in chronic progressive multiple sclerosis. Should a unified rating scale be used? Ann Neurol 1988;24:144A (Abstract) 42. Lee ET. Statistical methods for survival data analysis. Belmont, CA: Lifetime Learning, 1980 43. Cox DR. Analysis of survival data. New York: Chapman & Hall, 1984 44. Conover WJ, Iman RL.. Analysis of covariance using the rank transformation. Biometrics 1982;38:715-724 45. Bhaskar R, Reitman D, Sacks HS, et al. Loss of patients in clinical trials that measure long-term survival following myocardial infarction. Controlled Clin Trials 1986;7:134-148 46. Amato MP, Fratiglioni L, Groppi C, et al. Interrater reliability in assessing functional systems and disability on the Kurtzke scale in multiple sclerosis. Arch Neurol 1988;45:746-748 47. Thompson CB, June CH, Sullivan KM, Thomas ED. Association between cyclosporin neurotoxicity and hypomagnesaemia. Lancet 1984;2:1116-1120 48. Weinschenker BG, Ebers GC. The natural history of multiple sclerosis. Can J Neurol Sci 1987;14:255-261 49. Palestine AG, Nussenblatt RB, Chan CC. Side effects of systemic cyclosporine in patients not undergoing transplantation. Am J Med 1984;77:652-656 50. vonGraffenried B, Harrison WB. Renal function in patients with autoimmune diseases treated with cyclosporine. Transplant Proc 1985;17(suppl 1):215-23 1 51. Penn I. Cancers after cyclosporine therapy. Transplant Proc 1988;2O(~~ppl 1):276-279 52. Weidle PJ, Vlasses PH. Systemic hypertension associated with cyclosporine: a review. Drug Intell Clin Pharm 1988;22:44345 1 53. Rubinstein LV, Gail MH, Santner TJ. Planning the duration of a comparative clinical trial with loss to follow-up and a period of continued observation. J Chronic Dis 1981;34:469-479 54. Paty DW. Magnetic resonance imaging in the assessment of disease activity in multiple sclerosis. Can J Neurol Sci 1988; 151266-272 55. Keown PA. Optimizing cyclosporine therapy: dose, levels, and monitoring. Transplant Proc 1988;2O(suppl 2):382-389

MS Study Group: Cyclosporine in CPMS 605