Proc. Natl. Acad. Sci. USA
`Vol. 93, pp. 1716-1720, February 1996
`Medical Sciences
`
`The treatment of chronic progressive multiple sclerosis
`with cladribine
`(immunosuppression/double-blind study/crossover/2-chlorodeoxyadenosine/deoxypurine)
`E. BEUTLER*t, J. C. SIPEt, J. S. ROMINEt, J. A. KoZIOL*, R. McMILLAN*, AND J. ZYROFF§
`*Department of Molecular and Experimental Medicine, The Scripps Research Institute, and *Division of Neurology and §Department of Radiology, Scripps Clinic
`and Research Foundation, La Jolla, CA 92037
`Contributed by E. Beutler, November 6, 1995
`
`ABSTRACT
`A 2-year, placebo-controlled, double-blind,
`crossover study was started in 1992 to evaluate cladribine, an
`immunosuppressive drug, in the treatment of chronic pro-
`gressive multiple sclerosis. In the first year patients were given
`cladribine 0.10 mg/kg per day for 7 days as four monthly
`courses for a total of 2.8 mg/kg or placebo. During the second
`year patients treated with placebo during the first year were
`given i.v. infusions of 0.10 mg, 0.05 mg, and 0.05 mg of
`cladribine per kg of body weight per day for 7 consecutive days
`in three successive monthly courses, for a total dose of 1.4
`mg/kg. Patients who had been treated previously with
`cladribine were crossed over to placebo. Analysis of the results
`revealed a favorable influence on the neurological perfor-
`mance scores, both in the Kurtze extended disability status
`and the Scripps neurological rating scale, and on MRI findings
`in patients treated with cladribine. In the first year the most
`striking finding was that while clinical deterioration contin-
`ued in the placebo-treated patients, the condition of patients
`who received cladribine stabilized or even improved slightly.
`Toxicity and therapeutic response were dose-related.
`Although the underlying cause of multiple sclerosis (MS)
`remains a mystery, considerable evidence exists that damage
`to the central nervous system is mediated by immunopatho-
`logic mechanisms (1, 2). For this reason, immunosuppression
`is a rational approach to treatment of this disorder.
`2-Chlorodeoxyadenosine (2-CdA; cladribine; Leustatin) is
`an adenosine deaminase-resistant purine nucleoside, designed
`by Carson et al. (3-5) to simulate the immunodeficiency state
`of hereditary adenosine deaminase deficiency by causing the
`accumulation of deoxynucleotides in lymphocytes. This simple
`compound has been widely used for the treatment of lymphoid
`malignancies (6) and has a very favorable toxicity profile
`relative to other lymnphocytolytic drugs.
`Since there is no satisfactory treatment for chronic progres-
`sive MS, the use of cladribine was considered because of its
`relatively low toxicity and the long-lasting lymphopenia that it
`produces. Open-label feasibility studies were begun in 1990
`with a small number of patients. The results in terms of both
`apparent benefit on neurological performance and lack of
`toxicity were favorable and encouraged us to proceed with a
`larger 2-yr placebo-controlled crossover study to further ex-
`plore issues of safety and therapeutic effect. The first-year
`results of this study showed a positive effect (7). We now report
`observations from the entire 2 yr of this double-blind study and
`an additional 6-mo unblinded follow-up.
`
`METHODS
`Patient Selection. The study subjects were 51 patients with
`clinically definite or laboratory-supported definite chronic
`The publication costs of this article were defrayed in part by page charge
`payment. This article must therefore be hereby marked "advertisement" in
`accordance with 18 U.S.C. §1734 solely to indicate this fact.
`
`1716
`
`progressive MS (8) for >2 yr. The patients had been followed
`at Scripps Clinic by the neurology group for periods of from
`6 mo to 20 yr. The study plan and risks and potential benefits
`were explained to each patient in detail, and all patients gave
`informed consent to participate in the investigations, which
`were performed under investigator-initiated INDs no. 29,111
`and no. 93,777 from the Food and Drug Administration.
`Patient characteristics are summarized in Table 1.
`Study Design. Drug dosage. A phase III double-blind cross-
`over study was started in January of 1992. In the first year of
`this study patients on the cladribine arm were given four
`monthly 7-day courses of 0.10 mg of cladribine per kg per day
`(0.7 mg/kg per course) for a total of four courses (total dose
`= 2.8 mg of cladribine per kg), except as noted below. During
`the second year blinding was maintained, and the patients who
`had received placebo were given active drug, but at one-half
`the total dose that had been administered during the first year.
`The four infusions given to these patients were divided so that
`the first consisted of 0.7 mg ofcladribine per kg, the second and
`third each consisted of 0.35 mg of cladribine per kg of body
`weight, and the fourth consisted of saline placebo (total dose
`= 1.4 mg of cladribine per kg). The patients who had originally
`received cladribine were given four monthly saline placebo
`infusions from the beginning of the second year.
`Cladribine or placebo was administered monthly on an
`outpatient basis by 7-day i.v. infusions through a central venous
`catheter using a portable infusion pump. Blood counts and
`chemistry panels were done before each infusion, and the
`counts, reviewed by an unblinded hematologist/internist, had
`direct contact with only four of the patients when a medical
`problem required internal medicine consultation.
`Throughout the study, patients, neurologists, nurses, and the
`neuroradiologist were unaware of the treatment assigned to
`each patient. Cladribine causes no symptoms on infusion and
`cannot be distinguished from placebo by patients or profes-
`sional staff. Drug was withheld on occasions when blood
`counts did not meet the safety standards that had been
`established; this occurred on four occasions in patients receiv-
`ing active drug and in two patients receiving placebo. Corti-
`costeroid therapy was permitted when the examining neurol-
`ogist considered them necessary for treatment of the patient.
`Two patients received corticosteroids during the study; one
`patient required one course during placebo administration,
`and another patient required two courses during cladribine
`therapy.
`Sample size and patient compliance. On the basis of the
`results of the open label study, we had estimated that a sample
`size of 22 pairs of patients would be sufficient to detect a 15%
`improvement in the Scripps neurological rating scale (SNRS)
`Abbreviations: MS, multiple sclerosis; SNRS, Scripps neurological
`rating scale; EDSS, Kurtze extended disability status.
`tTo whom reprint requests should be addressed at: Department of
`Molecular and Experimental Medicine, The Scripps Research Insti-
`tute, 10666 North Torrey Pines Road, La Jolla, CA 92037.
`
`Downloaded from https://www.pnas.org by 73.210.63.126 on October 4, 2022 from IP address 73.210.63.126.
`
`Petitioner TWi Pharms., Inc.
`EX1027, Page 1 of 5
`
`

`

`Medical Sciences: Beutler et al.
`
`Table 1.
`
`Demographic characteristics of patients at study entry
`Initial
`Initial
`cladribine
`placebo
`18/9
`16/8
`27/0
`22/2
`43.4 (28-54)
`42.5 (21-54)
`
`Characteristic
`Sex (F/M)
`Race (White/other)
`Mean age, yr (range)
`Mean duration of clinical
`symptoms of MS, yr (range)
`
`10.5 (2-31)
`
`12.7 (2-24)
`
`while on cladribine, compared with no improvement while
`receiving placebo, with a statistical power of 0.90, using a
`one-sided test at an a level of 0.05. Initially 24 pairs of patients
`were matched by age, sex, and disease severity. Each matched
`pair was randomized by the statistician (J.A.K.) using random
`number tables so that one patient was assigned to the group
`initially receiving cladribine and the other to the group re-
`ceiving placebo. One additional individual, for whom no
`suitable match was identified, was started on the cladribine
`arm; this individual left the study after 8 mo on the protocol.
`Two patients, both initially assigned to cladribine, were lost to
`follow-up at 2 and 3 mo on the study, respectively. Because the
`loss of these two patients on the initial cladribine arm was not
`attributed to treatment, we recruited two additional patients,
`appropriately matched by the "blinded" neurologists (J.C.S.
`and J.S.R.) and assigned by the statistician to cladribine, as
`replacements. One patient receiving placebo withdrew from
`the study after 4 mo on the protocol for reasons related to
`treatment (i.e., lack of stabilization of disease); this patient was
`not replaced. The analyses reported are based on the experi-
`ence of the 24 matched pairs of patients and exclude the three
`patients initially on cladribine who did not complete a full year
`of the study, as described above. Four additional patients were
`lost from the study in the second year for various reasons
`unrelated to the study. Accordingly, 24-mo or longer follow-up
`of 21 of the 24 patients who had received cladribine in the first
`year and 22 of the 24 patients who had received placebo in the
`first year was possible.
`Deviations from the original protocol. There were a few
`deviations from the stated drug dosages. In the original
`protocol a total of six monthly doses had been planned, but this
`was modified to four courses after the study had been initiated
`because of a greater than expected decline in the platelet count
`in some patients. One patient received five courses before the
`decision to reduce dosage had been made. One patient re-
`ceived only two courses, and two patients received only three
`courses because of thrombocytopenia. At the beginning of the
`second phase of the study, five patients who were to have
`received placebo were given a single 0.7 mg/kg of infusion
`cladribine by error. Separate analysis showed that the response
`of these patients was not greater than those of the other
`patients, and the data of these patients have been retained.
`Observations and Data Analysis. Neurologic evaluation.
`Patients were evaluated monthly by means of the Kurtzke
`extended disability status scale (EDSS) (9) and with the SNRS
`(10). MRI brain scans with contrast enhancement were per-
`formed in these patients before treatment and at 6-mo intervals.
`To assess inter-rater variability 20 patients were indepen-
`dently assessed by each examiner (J.S.R., J.C.S.) on the same
`day. Interrater agreement (11) on each scoring instrument was
`quite high: the weighted K coefficient of agreement was 0.976
`for the EDSS scores and 0.828 for the SNRS scores. Interrater
`.1.0 EDSS points,
`agreement, defined as a difference of
`reached 100% for all sets of examinations. This result com-
`pared favorably with results reported in other clinical trials of
`investigative therapeutic agents in MS (12). In comparison,
`interrater agreement on the SNRS was 85%, with agreement
`defined as a difference of .10 SNRS points. In a separate
`evaluation 18 patients were assessed by the same examiner
`
`1717
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`twice on the same day, the period between examinations
`ranging from 135 min to 240 min. Intra-rater agreement on the
`EDSS was perfect with both examiners, and weighed K coef-
`ficients of agreement between the two SNRS scores were 0.978
`(J.S.R.) and 0.998 (J.C.S.).
`Hematologic evaluation. Monthly blood counts were ob-
`tained during drug administration. Less frequent counts were
`obtained thereafter. Lymphocyte subset evaluation was done
`monthly during the course of drug administration and less
`frequently thereafter.
`MRI. MRI was performed on a 1.5-T General Electric Signa
`scanner. T2 and proton density-weighted images were obtained
`by using a conventional spin-echo sequence with repetition
`times of 2500 msec and echo-delay times of 30 and 90 msec.
`Axial scans of 3-mm thickness and 0 interslice gap were
`obtained -10 min after gadopentetate dimeglumine (Magne-
`vist, Berlex Laboratories) injection to assure optimal time for
`transmigration of the contrast agent across the blood-brain
`barrier.
`Statistical Methods. The SNRS was designated as the primary
`outcome parameter. Summary statistics are reported as mean
`and range or SEM. Analyses of the neurological scores and
`MRI findings were undertaken with parametric methods
`appropriate for two-period repeated-measurements crossover
`designs (13) as well as a nonparametric repeated-measures
`ANOVA technique (14). Last available observations were
`carried forward for patients who had completed at least 18 mo
`of the study. Similar analyses were done in which these data
`remained missing and in which they were modeled under the
`representation that they were missing at random; these anal-
`yses yielded similar results to those reported here. Kaplan-
`Meier curves and log-rank statistics were also used to compare
`neurological rating-scale outcomes between the two treatment
`groups during the first year of the study. Two-sided P values are
`reported throughout.
`
`RESULTS
`Fig. 1
`Neurologic Findings. Clinical performance scores.
`depicts the EDSS and SNRS scores of the patients. The
`average EDSS and SNRS scores of patients receiving cladrib-
`ine improved modestly during the first year of the study,
`whereas the scores of patients with placebo continued to
`deteriorate. The improvement in SNRS scores appeared to
`18 mo and be well maintained for the 24 mo of
`peak at
`follow-up in the patients treated with 2.8 mg of cladribine per
`kg, even though they received no active drug after the first 4
`mo of the study. After 24 mo, in unblinded observations, fairly
`rapid deterioration was documented. While the scores of the
`patients who received placebo in the first year of the study
`deteriorated during that year, the lower dose of cladribine they
`received (1.4 mg/kg) also seemed to be effective in stabilizing
`their disease, albeit for a shorter time period, with peak
`8 mo after treatment initiation. Inspection
`improvement at
`of the curves thus suggests that the stabilization of disease
`produced by the lower dose of cladribine may be of shorter
`duration than that seen with the larger dose and that a rebound
`worsening of disease may occur between 24 and 30 mo after
`initiation of therapy with the higher dose. ANOVA based on
`the two-period crossover design with absolute changes in
`EDSS and SNRS as end points revealed no significant car-
`ryover effects between subjects or period effects within sub-
`jects, but highly significant treatment effects: the F-statistics
`for assessing treatment effects with subjects were F1,44 = 10.19,
`P = .0026 for EDSS and F, 44 = 23.46, P < 0.0001 for SNRS.
`Kaplan-Meier time-to-failure plots show that whether fail-
`ure is defined as a gain of 0.5, 1, or 1.5 points on the EDSS scale
`or loss of 5, 10, or 15 points on the SNRS scale, patients
`receiving cladribine fared better than those who received
`placebo in the first study year. Plots showing time-to-gain of 1.0
`
`Downloaded from https://www.pnas.org by 73.210.63.126 on October 4, 2022 from IP address 73.210.63.126.
`
`Petitioner TWi Pharms., Inc.
`EX1027, Page 2 of 5
`
`

`

`1718
`
`Medical Sciences: Beutler et al.
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`
`0 570170
`
`1111
`
`1
`
`6.0-
`
`6.5 (
`
`0 Initial Cladnbine
`a Initial Placebo
`
`%
`
`"
`
`65
`
`60
`
`P6
`
`0 3
`
`6
`
`9
`
`12
`
`0 3
`
`6
`
`9
`
`12
`
`24 27 30
`
`15 18 21 24 27 30
`15
`18 21
`MONTH
`MONTH
`FIG. 1.
`The EDSS (A) and SNRS (B) of 48 patients treated with cladribine in a double-blind crossover study. The solid symbols indicate months
`during which drug was administered. The group of patients denoted by circles was given 0.7 mg of cladribine per kg of body weight in a 7-day infusion
`during 4 mo. The group that received placebo first (aJ) was given 0.7 mg of cladribine per kg of body weight as a first infusion (first solid square)
`and two subsequent 7-day infusions of 0.35 mg per kg of body weight (second and third solid square). The fourth infusion in these patients was
`placebo, as denoted by the open square. Bars represent 1 SEM. Dashed lines represent changes occurring after the study had been unblinded.
`EDSS or loss of 10 SNRS points are shown in Fig. 2. Surpris-
`they had lesions that disappeared. However, the other 12
`ingly, time-to-improvement plots (data not shown) also showed a
`patients were scored as having lesions: 10 because enhancing
`statistically significant advantage for patients receiving cladribine.
`lesions persisted throughout the 12-mo period, and 2 because
`Log-rank statistics were used to compare times-to-failure
`they developed new enhancing lesions. In contrast, at the end
`between the two treatment groups of the first year of the study,
`of the first year, 22 of the patients who had received cladribine
`where failure was defined as a gain of 1 or 1.5 points on the
`were classified as having no lesions: 11 patients had had no
`EDSS or a loss of 10 or 15 points on the SNRS relative to
`enhancing volumes at baseline, and 11 had enhancing lesions
`baseline value. Each of these statistics was highly significant
`that disappeared. Only two individuals in this group had
`(AEDSS = 1.0:L = 6.313, P = 0.012; AEDSS = 1.5:L = 5.254,
`lesions: one had continuing enhancing lesions, and one devel-
`P = 0.024; ASNRS = -10:L = 8.299, P = 0.004; ASNRS =
`oped new enhancing lesions where there had been none (P <
`- 15:L = 6.800, P = 0.009), indicating that patients receiving
`0.001, McNemar's test).
`cladribine fared better than those who received placebo.
`Further analysis was made of patients after the crossover of
`MRI findings. Neither nonparametric ANOVA nor para-
`treatments at 12 mo. Twenty-two of the patients in the group
`metric ANOVA based on the two-period crossover design
`that received placebo first and who were then treated with 1.4
`revealed a significant treatment effect on demyelinated vol-
`mg ofcladribine per kg ofbodyweight were evaluated at 24 mo.
`umes on MRI (data not shown). In contrast, a highly signifi-
`Eleven had no lesions at both 12 and 24 mo, one had lesions
`cant difference was seen in the enhancing volumes, a mea-
`at both 12 and 24 mo, and 10 had lesions at 12 mo but none
`surement of current disease activity (15). For analysis, enhanc-
`at 24 mo (P < 0.001, McNemar test). From this crossover
`ing volume findings were dichotomized as being either present
`analysis, there is clear evidence that the reduced dose of
`or absent; the results of this analysis are summarized in Table
`cladribine received by the initial placebo group during the
`2. At the end of 1 yr, 12 patients in the group given placebo
`second year on the study significantly reduced the occurrence
`were scored as having no enhancing lesions: 9 because they had
`of enhancing lesions on MRI scans. Among the 20 patients in
`no enhancing lesions throughout the period, and 3 because
`the group that received cladribine first who were evaluated at
`
`6
`8
`TIME (months)
`TIME (months)
`FIG. 2.
`Kaplan-Meier time-to-failure plots of patients in the double-blind crossover study. (Left) Time-to-increase of EDSS score by 1.0 point.
`(Right) Time-to-failure as defined by a decrease of SNRS score by 10 points. The difference in time-to-failure estimated by logarithmic-rank statistics
`was highly significant (EDSS, P = 0.012; SNRS, P = 0.004).
`
`0
`
`2
`
`4
`
`10
`
`12
`
`Downloaded from https://www.pnas.org by 73.210.63.126 on October 4, 2022 from IP address 73.210.63.126.
`
`Petitioner TWi Pharms., Inc.
`EX1027, Page 3 of 5
`
`

`

`Medical Sciences: Beutler et al.
`
`Table 2.
`
`Presence of enhancing lesions by MRI over the study
`Outcome
`
`Initial placebo
`Lesions
`Lesions
`present
`absent
`13
`11
`11
`13
`12
`12
`1
`22
`1
`21
`
`Initial cladribine
`Lesions
`Lesions
`present
`absent
`12
`12
`6
`18
`2
`22
`0
`23
`0
`20
`
`Time
`Baseline
`6 mo
`12 mo
`18 mo
`24 mo
`
`24 mo, 19 had no lesions both at 12 and at 24 mo, and one had
`lesions at 12 but none at 24 mo. Thus, the treatment effect
`found in the group that received cladribine in the first year
`seemed to carry over for an additional 12 mo.
`Toxicity and adverse events. The average platelet counts
`remained normal throughout the study, but the platelet counts
`of seven patients receiving the larger cladribine dose in the first
`year of the study fell to <100,000/ul, and one of these, a
`patient who had taken carbamazepine (Tegretol) and who was
`ingesting large amounts of phenytoin (Dilantin), developed
`severe thrombocytopenia with platelet counts <10,00QOO,l.
`The effect of cladribine on the blood counts of these patients
`has been reported in detail elsewhere (16). In the second year
`of the study, when patients received only one-half of the dose
`of cladribine administered in the first year, platelet counts of
`<100,0001,ul were encountered in only one patient, the lowest
`count observed being 83,000/,ul. Changes in lymphocyte sub-
`sets are depicted in Fig. 3. There was marked depletion of CD4
`cells with both the high- and low-dose schedule.
`
`1719
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`Six patients developed herpes zoster, one of them only after
`retreatment with cladribine under another protocol. All of
`these infections were segmental, mild, and responded rapidly
`to oral acyclovir. One patient developed fatal, fulminating,
`newly acquired hepatitis B immediately after her second dose
`of cladribine. As discussed elsewhere (7), it seems unlikely that
`the administration of cladribine played a role in this patient's
`illness. One patient developed a severe Salmonella infection
`with near-perforation of the bowel while receiving placebo
`during the first study year. She responded well to antibiotic
`therapy and continued the study.
`
`DISCUSSION
`Although numerous immunosuppressive and immunomodu-
`latory drugs have been given to patients with MS, conventional
`immunosuppression has not demonstrated sufficient promise
`to be considered as a routine treatment for the chronic
`progressive form of MS (17-19). Why then, might another
`immunosuppressive agent produce a greater effect? The sup-
`pression of CD4 cells and sparing of CD8 cells that was
`observed with cladribine administration is much greater than
`that observed with other immunosuppressive agents. Treat-
`ment of MS patients for a year with chlorambucil (18) and
`cyclophosphamide (20) each produced a relatively transient
`2-fold decline in the CD4/CD8 ratio. In contrast, treatment
`with cladribine for only a few months produced an -4-fold
`decrease in this ratio, and the effect was sustained for many
`months after a 4-mo course of the drug.
`Because of its selective and prolonged effect on T cells,
`cladribine appeared to be a reasonable candidate drug for MS
`treatment. Our studies clearly show that cladribine retards the
`progression of neurologic impairment of patients with chronic
`
`o Inrtia Ckxk*3*i
`o Inial Pkcebo
`
`3
`
`6
`
`9
`
`12
`
`15
`
`18
`
`,
`
`,
`
`21
`
`I
`
`24
`
`200
`
`150
`
`100
`
`50
`
`0
`
`0
`
`I c
`
`k
`040
`
`1800
`
`1500
`
`o lnmloicsodik1n
`o InlcdPkcebo
`
`-
`
`I
`
`12003
`
`6
`
`9
`
`12
`
`15 182124
`
`6100 --
`300
`
`06
`o~~~~~~of
`
`120
`
`240
`
`160-
`
`:
`
`80
`
`o nlkid C1okKktln
`o Ir dPkxebo
`
`18
`
`21
`
`24
`
`0
`
`3
`
`6
`
`9
`
`12
`
`15
`
`18
`
`21
`
`24
`
`< ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
`l
`
`Month
`
`Month
`
`500 .wl
`
`400
`
`I
`
`bceb
`
`1.
`
`200
`
`co00
`0(.
`
`00
`
`3
`
`6
`
`9
`
`1 51 12
`
`o InIta Clad mt
`o Inrik PkWcebo
`
`aw
`
`0
`
`3
`
`6
`
`9
`
`12
`
`15
`
`Month
`
`12001
`
`I.
`
`0 0
`
`0
`
`3
`3
`
`6
`
`9
`
`12
`
`15
`
`18
`
`21
`
`24
`
`Month
`
`0
`
`0
`
`3
`
`6
`
`9
`
`12
`
`15
`
`18
`
`21
`
`24
`
`0
`
`3
`
`6
`
`9
`
`12
`
`15
`
`18
`
`21
`
`24
`
`Month
`
`Month
`
`Changes in lymphocyte subsets in patients. Symbols and drug dosages are as in Fig. 1. The decline of T lymphocytes (CD3) was more
`FIG. 3.
`marked than that of B lymphocytes. A rapid and profound fall in the inducer/helper (CD4) T lymphocytes was accompanied by a more modest
`and less sustained decline in the cytotoxic/suppressor (CD8) T lymphocytes. Accordingly, there was an #4-fold decline in the CD4/CD8 ratio.
`The number of activated T and B lymphocytes (CD25) fell -5-fold and remained at subnormal levels for the 14 mo after cessation of drug
`administration. The slight dip in some of the lymphocytes in the group that received cladribine first is due to the erroneous administration of a
`single dose of drug to a few of the patients in that group on month 13.
`
`Downloaded from https://www.pnas.org by 73.210.63.126 on October 4, 2022 from IP address 73.210.63.126.
`
`Petitioner TWi Pharms., Inc.
`EX1027, Page 4 of 5
`
`

`

`1720
`
`Medical Sciences: Beutler et al.
`
`progressive MS. The effect is of sufficient magnitude that
`highly statistically significant results could be obtained by the
`study of only 24 pairs of patients in a double-blind study. The
`smaller effects produced by other drugs have required that
`much larger groups of patients be investigated. At a total dose
`of 2.8 mg/kg given over a period of 4 mo, the stabilization of
`disease appears quite durable.
`As for other cytotoxic agents such as methotrexate and
`cyclophosphamide, marrow suppression is a known side-effect
`of cladribine (6, 21). In the present study occasional patients
`developed significant thrombocytopenia at higher drug dos-
`ages; one patient developed severe but reversible marrow
`suppression. The lower dose of cladribine used in the second
`arm of the present study appears to provide a larger margin of
`safety. From this point of view, we were encouraged by the
`therapeutic effect of 1.4 mg/kg of drug given over a 3-mo
`period, a dose that appeared to be very well tolerated. The
`therapeutic effect of the lower dose appeared to be less durable
`than that observed at higher doses, this dosage-response
`relationship lending further weight to the validity of our
`results.
`The usefulness of MRI as a measure of clinical activity and
`potentially as an objective means for assessing response to
`therapy has recently been documented (19). In our studies we
`observed marked improvement in the appearance and disap-
`pearance of lesions visualized on MRI after gadolinium en-
`hancement. These highly statistically significant changes per-
`sisted for the full 2 yr of observation in the patients who had
`received 4 mo of therapy in the first arm of the study and were
`seen in patients receiving the low cladribine dose as well as in
`those receiving the high dose.
`Treatment with immunosuppressive drug is not without risk.
`The occurrence of herpes zoster in six of the patients is surely
`a manifestation of their immunosuppressed state, but no other
`opportunistic infections attributed to cladribine therapy were
`encountered, in spite of the fact that the CD4 cell count
`remained low for long time periods. Many immunosuppressive
`drugs are also myelosuppressive, and cladribine does exhibit a
`dose-dependent effect on hematopoietic stem cells. In the
`current study thrombocytopenia was noted in some patients
`receiving higher doses of the drug, and severe, transient
`marrow suppression occurred in one patient who was also
`ingesting other potentially myelosuppressive drugs (16). Al-
`though the short-term risks of cladribine seem acceptable for
`patients with severely progressive disease, little can be written
`about the long-term risk. We began using this drug, primarily
`in patients with malignant disorders, in 1981. Most patients
`treated before 1987 had end-stage malignant disease, and only
`a few are alive today. In the past 8 yr, however, many patients
`with hairy cell leukemia and with relatively early-stage lym-
`phomas have been treated with the drug. No long-term toxic
`results have been documented, but since the drug is incorpo-
`rated into DNA (3), the possibility of malignancies occurring
`long after administration of this purine analogue cannot be
`dismissed.
`Cladribine was given by continuous i.v. infusion on an
`outpatient basis to all of the patients reported here. This route
`of infusion was used because the safety and efficacy of the drug
`given by other routes had not been established at the time this
`study was initiated. We now recognize that the drug may be
`
`Proc. Natl. Acad. Sci. USA 93 (1996)
`given by the much more convenient subcutaneous route with-
`out causing local irritation and with pharmacokinetic (22) and
`therapeutic (23) results that appear entirely equivalent to the
`i.v. route. We have designed a study of the effect in both
`chronic progressive and relapsing/remitting MS of subcuta-
`neously administered cladribine, 0.07 mg/kg per day given as
`six monthly courses each consisting of five daily injections. It
`is notable, that this method of administration has recently been
`used in MS patients by Grieb et al. (24) who report preliminary
`data suggesting beneficial results in the treatment of relapsing/
`remitting disease with cladribine.
`This is manuscript 9176-MEM from The Scripps Research Institute.
`This work was supported by National Institutes of Health Grants
`NS30218 and RR00833, Food and Drug Administration Grant FD-
`R-000280, and the Stein Endowment Fund.
`Ford, H. C. (1987) Med. Hypotheses 24, 201-216.
`1.
`Weiner, H. L. & Hafler, D. A. (1988) Ann. Neurol. 23, 211-222.
`2.
`Carson, D. A., Wasson, D. B., Taetle, R. & Yu, A. (1983) Blood
`3.
`62, 737-743.
`Carson, D. A., Wasson, D. B. & Beutler, E. (1984) Proc. Natl.
`Acad. Sci. USA 81, 2232-2236.
`Carson, D. A., Wasson, D. B., Kaye, J., Ullman, B., Martin,
`D. W., Jr., Robins, R. K. & Montgomery, J. A. (1980) Proc. Natl.
`Acad. Sci. USA 77, 6865-6869.
`Beutler, E. (1994) Semin. Hematol. 31, 40-45.
`Sipe, J. C., Romine, J. S., Koziol, J. A., McMillan, R., Zyroff, J.
`& Beutler, E. (1994) Lancet 344, 9-13.
`Poser, C. M., Paty, D. W., Scheinberg, L., McDonald, W. I.,
`Davis, F. A., Ebers, G. C., Johnson, K. P., Sibley, W. A., Silber-
`berg, D. H. & Tourtellotte, W. W. (1983) Ann. Neurol.
`13,
`227-231.
`Kurtzke, J. F. (1983) Neurology 33, 1444-1452.
`Sipe, J. C., Knobler, R. L., Braheny, S. L., Rice, G. P., Panitch,
`H. S. & Oldstone, M. B. (1984) Neurology 34, 1368-1372.
`Cohen, J. (1968) Psychol. Bull. 70, 213-220.
`Goodkin, D. E., Cookfair, D., Wende, K., Bourdette, D., Pul-
`licino, P., Scherokman, B. & Whitham, R. (1992) Neurology 42,
`859-863.
`Jones, B. & Kenward, M. G. (1989) Design and Analysis of
`Cross-Over Trials (Chapman & Hall, London).
`Koziol, J. A. & Maxwell, D. A. (1983) Stat. Med. 1, 83-89.
`Miller, D. H., Barkhof, F., Berry, I., Kappos, L., Scotti, G. &
`Thompson, A. J. (1991) J. Neurol. Neurosurg. Psychiatr. 54, 683-
`688.
`Beutler, E., Koziol, J., McMillan, R., Sipe, J. C., Romine, J. S. &
`Carrera, C. J. (1994) Acta Haematol. (Basel) 91, 10-15.
`Goodkin, D. E., Bailly, R. C., Teetzen, M. L., Hertsgaard, D. &
`Beatty, W. W. (1991) Neurology 41, 20-25.
`Chiappelli, F., Myers, L. W., Ellison, G. W., Liao, D. & Fahey,
`J. L. (1991) Int. J. Immunopharmacol. 5, 455-461.
`Khoury, S. J., Guttmann, C. R., Orav, E. J., Hohol, M. J., Ahn,
`S. S., Hsu, L., Kikinis, R., Mackin, G. A., Jolesz, F. A. & Weiner,
`H. L. (1994) Neurology 11, 2120-2124.
`Moody, D. J., Fahey, J. L., Grable, E., Ellison, G. W. & Myers,
`L. W. (1987) J. Neuroimmunol. 14, 175-182.
`Beutler, E. (1992) Lancet 340, 952-956.
`Liliemark, J., Albertioni, F., Hassan, M. & Juliusson, G. (1992)
`J. Clin. Oncol. 10, 1514-1518.
`Juliusson, G., Liliemark, J., Hippe, E., Blichfeldt, P., Wallman,
`K., Hagberg, H., Stolt, C.-M., Bergheim, J., Winquist, I., Hede-
`nus, M., Carlsson, M., Velin, L., Hellquist, L., Kristofferson, R.,
`Ly, B. E. & Vaart, J. (1992) Blood 80, Suppl. 1, 359 (abstr.).
`Grieb, P., Ryba, M., Stelmasiak, Z., Nowicki, J., Solski, J. &
`Jakubowska, B. (1994) Lancet 344, 538.
`
`20.
`
`21.
`22.
`
`23.
`
`24.
`
`4.
`
`5.
`
`6.
`7.
`
`8.
`
`9.
`10.
`
`11.
`12.
`
`13.
`
`14.
`15.
`
`16.
`
`17.
`
`18.
`
`19.
`
`Downloaded from https://www.pnas.org by 73.210.63.126 on October 4, 2022 from IP address 73.210.63.126.
`
`Petitioner TWi Pharms., Inc.
`EX1027, Page 5 of 5
`
`