A randomized, double-blind study comparing the efficacy, safety and optimal dose of two formulations of cyclosporin, Neoral and Sandimmun, in patients with severe psoriasis

A randomized, double-blind study comparing the efficacy, safetyand optimal dose of two formulations of cyclosporin, Neoraland Sandimmun, in patients with severe psoriasis

J.KOO FOR THE OLP302 STUDY GROUPUniversity of California, San Francisco Medical Center, UCSF Psoriasis Treatment Center, San Francisco, CA 94143-1212, U.S.A.

Accepted for publication 6 February 1998

Summary This study compared the efficacy, safety and optimal dose of two formulations of cyclosporin,Sandimmun and Neoral , in patients with severe, chronic plaque-type psoriasis. Patients wererandomized on a 1 : 1 basis to 24 weeks of treatment with Neoral (n ¼ 152) or Sandimmun(n ¼ 157). The starting dose of each formulation was 2·5 mg/kg per day. Dose increases to maintainefficacy were allowed after 4 weeks. In patients who achieved remission, the dose was down-titratedat 4-week intervals from week 16. The maximum permitted dose for each formulation was 5·0 mg/kg per day. Neoral produced a more rapid response than Sandimmun: remission rates were higher forNeoral during the first 8 weeks of treatment. The number of dose reductions for safety was similar inboth treatment groups, but there were more dose increases to maintain efficacy in the Sandimmungroup (198) than the Neoral group (146). The number of patients with dose reductions after week16 was higher for Neoral (n ¼ 83) than for Sandimmun (n ¼ 73). The frequency and nature ofadverse events were similar for both treatment groups. The mean dose required to control the diseasewas < 10% lower with Neoral and fewer dose changes were needed. The increased bioavailabilityand reduced pharmacokinetic variability of cyclosporin provided by the Neoral formulation mayfacilitate short-course, intermittent therapy.

The efficacy of cyclosporin (CS), administered as com-mercially available Sandimmun soft gelatin capsules,was first described by Mueller and Herrmann in 1979 ina patient with psoriatic arthritis.1 Since then, theefficacy and safety of Sandimmun in psoriasis havebeen confirmed in a number of open, controlled andlong-term studies.2–5 In the majority of studies, thePsoriasis Area and Severity Index (PASI) score wasreduced by more than 70%, and in one large-scalemulticentre study, remission was achieved in 85% ofpatients.4 However, because the Sandimmun formula-tion of CS is an oil-in-water emulsion, the absorption ofCS is affected by food intake and fat content, by thedispersion of the agent in the gastrointestinal tract andby the secretion of bile. As a result, the use of Sandim-mun is limited by a degree of variability in CS bioavail-ability, both between patients and within an individualpatient over time.5

The experience gained with Sandimmun has facili-tated the development of a new oral formulation of CS,

Neoral , which is based on microemulsion technologyand is designed to reduce the pharmacokinetic varia-bility of CS and thus simplify its clinical use. The keycharacteristics that enable Neoral to achieve these aimsare the fast release of CS at the site of absorption,improved dissolution of CS in the gastrointestinaltract, and dispersion independent of the presence offood or the flow of bile.6 The increased bioavailabilityand reduced pharmacokinetic variability of the newformulation have been confirmed in psoriasis patients.7

Data suggest that the greatest changes in bioavailabilityappear to be in patients who were poor absorbers of CSadministered as the Sandimmun formulation.

In the light of these observations, a clinical trial wasdesigned to compare the efficacy of Neoral and Sandim-mun in patients with severe, chronic plaque-type psor-iasis; to compare the incidence and severity of adverseevents and laboratory abnormalities associated with thetwo formulations, particularly renal side-effects andblood pressure changes; and finally, to compare theoptimal dose of Neoral and Sandimmun, and to evaluatethe adequacy and validity of current dosing guidelinesin accordance with the dose-titration scheme.

British Journal of Dermatology 1998; 139: 88–95.

88 q 1998 British Association of Dermatologists

The Sandimmun formulation of cyclosporin has been withdrawn andreplaced by the Neoral formulation in the U.K. and many othercountries.

Subjects and methods

Study design

This prospective, randomized, double-blind, parallel-group study involved 43 centres in eight countries.After a 2-week screening period during which no treat-ment was permitted, patients with psoriasis were ran-domized on a 1 : 1 basis to 24 weeks of treatment withNeoral or Sandimmun. The starting dose of each for-mulation was 2·5 mg/kg per day. Dose increases tomaintain efficacy were allowed at intervals of 2 weeks,after week 4, and dose decreases for safety were allowedat any time (Table 1). In patients who achieved remis-sion, the dose was down-titrated at intervals of 4 weeksfrom week 16. The maximum permitted dose for eachformulation was 5·0 mg/kg per day. Patients were toreturn for evaluation at weeks 2, 4, 6, 8, 10, 12, 16, 20and 24.

Treatments

Sandimmun and Neoral were supplied in soft gelatincapsules of two strengths (25 and 100 mg) in Norwayand the U.S.A., and three strengths (25, 50 and100 mg) in other countries. Both formulations were tobe taken twice daily according to individual routinepractice, before or after meals and at 12 h intervals.During the study, concomitant antipsoriatic therapywas not permitted, with the exception of topical treat-ment with indifferent substances. Drugs known to

interfere with psoriasis were to be avoided, althoughpatients receiving beta-blockers at a steady dose couldenter the study. Drugs known to interfere with Sandim-mun pharmacokinetics or to potentiate the nephrotoxi-city of Sandimmun were not permitted. Non-steroidalanti-inflammatory drugs were permitted if the doseremained unchanged throughout the study.

Subjects

Subjects were eligible to enter the 2-week screeningperiod if they were aged from 18 to 70 years; hadgeneralized, chronic plaque-type psoriasis requiringsystemic treatment; had a minimum PASI score of 15;and if at least one regimen of previous treatment hadbeen ineffective or if another conventional systemicagent was considered to be inappropriate. Patientswere excluded if they were suffering from renal, liveror haematopoietic disease or from uncontrolled hyper-tension. Patients were randomized to enter the 24-weektreatment period at day þ 1 if their disease remainedstable between week – 2 and day 0.

Assessments

The primary efficacy variables were percentage changein PASI score from baseline to week 16, time frombaseline to remission and percentage of relapse-freepatients. Remission was defined as a decrease in thePASI score of $ 75% compared with the score at baseline,

TWO FORMULATIONS OF CYCLOSPORIN IN PSORIASIS 89

q 1998 British Association of Dermatologists, British Journal of Dermatology, 139, 88–95

Table 1. Protocol procedures for increasing and reducing study medication dose

Dose adjustment Procedure

Increase (lack of efficacy) Incremental increase (mg/kg per day): 2·5; 3·3; 4·2; 5·0Decrease (remission at 2·5 mg/kg per day) (from week 16 only) Incremental decrease (mg/kg per day): 1·9; 1·25Decrease (remission at >2·5 mg/kg per day) (from week 16 only) Incremental decrease (mg/kg per day): 5·0; 4·2; 3·3; 2·5; 1·9; 1·25Decrease (safety: renal dysfunction)Increase in serum creatinine compared with pretreatment value:

30–50% at two consecutive visits Reduce dose by 25–50%>50% but <100% Reduce dose by 50%$100% Discontinue study medication

Decrease (safety: liver enzyme elevation)Transaminase or alkaline phosphatase levels > three times upper normal Reduce dose by 25–50%limit or bilirubin level > two times upper normal limitDecrease (safety: hypertension)Increase in diastolic blood pressure $95 mmHg and up to Commence antihypertensive treatment with calcium110 mmHg at two consecutive visits antagonistDiastolic blood pressure still $95 mmHg at next visit Reduce study medication dose by 25–50%Decrease (safety: other adverse events)Hyperkalaemia Reduce dose by 25–50%Other clinical/biological abnormality causing clinical concern Consider dose reduction

while relapse was defined as a return in the PASI score to$ 50% compared with the score at pretreatment baselinein patients who had previously been in remission. Safetycriteria were incidence and severity of adverse events,physical examination and vital signs, laboratory tests(including levels of serum creatinine), biochemical andhaematological tests, and urine analysis.

Statistical methods

A sample size of 125 patients per group was determinedto provide 80% power to demonstrate equivalencebetween Neoral and Sandimmun with 95% confidence,assuming a mean reduction from baseline of 90% in thePASI score at week 16, with an SD of 20% and anequivalence interval of 6 7·5%.

For the safety analyses a single population wasdefined, comprising all randomized patients whoreceived at least one dose of study medication. Theefficacy analyses were performed on an intention-to-treat (ITT) population, which comprised all randomizedpatients who received at least one dose of study medica-tion and had at least one efficacy evaluation afteradministration of study medication.

Demographics and background characteristics werecompared between treatment groups using the Wil-coxon rank-sum test (for quantitative variables) andFisher’s exact test (for qualitative variables). The per-centage change from baseline in the PASI score at week16 was compared between treatment groups usinganalysis of covariance (ANCOVA) with baseline PASIscore as covariate, and treatment and (pooled) centreas factors. The times (since baseline) to remission (up toweek 16) and relapse were compared between treat-ment groups using the log-rank test. The time toremission was summarized using the Kaplan–Meierestimator, censoring patients who discontinued prema-turely without experiencing a remission.

Secondary efficacy variables (overall evaluation oftreatment, change from baseline in body surface areaaffected, and global evaluation) were compared usingthe Wilcoxon rank-sum test, ANCOVA and the van Elterentest, respectively. All significance tests (unless otherwisespecified) were conducted at the 5% level.

Results

Patient disposition

Three hundred and sixty-eight patients were screened,of whom 309 were randomized to treatment: 152 to the

Neoral group and 157 to the Sandimmun group. Twohundred and fifty patients completed the 24 weeks oftreatment, 125 in each treatment group. All rando-mized patients were eligible for safety analysis and allexcept one patient in the Sandimmun group wereeligible for the ITT population. Fifty-nine patients dis-continued the study (Neoral, n ¼ 27; Sandimmun,n ¼ 32): 22 for adverse events (Neoral, n ¼ 14; Sandim-mun, n ¼ 8); 15 for insufficient efficacy (Neoral, n ¼ 6;Sandimmun, n ¼ 9); five in each group were lost tofollow-up; four were withdrawn because of protocolviolations (Neoral, n ¼ 1; Sandimmun, n ¼ 3); threewithdrew their consent to participate (Neoral, n ¼ 1;Sandimmun, n ¼ 2); and five patients in the Sandim-mun group withdrew for other reasons.

Key patient characteristics at entry

There were no clinically significant differences inpatient or disease characteristics between treatmentgroups at entry that were considered to have anyinfluence on the outcome of the study. The mean PASIscore was < 25 and the mean affected body surface areawas < 40% for both treatment groups; these scores areindicative of patients with severe psoriasis. An exam-ination of previous antipsoriatic treatment indicatedthat in 35–39% of patients, treatment had failed orsymptoms had worsened after the use of systemictherapies, including psoralen plus ultraviolet A. Forthe remainder of the patients, another conventionalsystemic therapy had been considered inappropriate.

Efficacy analysis

For all of the efficacy parameters studied, an improve-ment was seen in both treatment groups over the courseof the study. With regard to the primary efficacy end-point (percentage reduction in PASI score at week 16),135 patients receiving Neoral and 137 patients receiv-ing Sandimmun completed 16 weeks of treatment andwere included in the analysis. Similar results were seenin both treatment groups, with a mean reduction of84·3% in the Neoral group and 81·8% in the Sandim-mun group (adjusted treatment difference: – 0·72%;95% confidence interval (CI) for the adjusted meandifference: – 4·61, 3·16) (Fig. 1). At the end-point, thePASI score had decreased by a mean of 79·6% in theNeoral group and 76·6% in the Sandimmun group.However, the mean percentage reductions in thePASI score at weeks 2, 4 and 6 were statisticallysignificantly larger (P ¼ 0·031, P<0·001, and

90 J.KOO FOR THE OLP 302 STUDY GROUP


P ¼ 0·008, respectively) in the Neoral group than in theSandimmun group, suggesting that while doses aresimilar, Neoral produces a more rapid and consistentresponse to treatment than Sandimmun.

Over the first 8 weeks of treatment, remission rateswere notably higher for Neoral than for Sandimmun(Fig. 2). Remission rates at week 8 were 51·1% for theNeoral group but only 38·2% for the Sandimmun group(12·9% difference; 95% CI: 1·7%, 24·1%). The differ-ence between treatment groups was reduced in terms ofproportion of patients in remission, between baselineand week 16 (Neoral 87·3%; Sandimmun 80·7%)—areflection of dose increases permitted to maintain effi-cacy. The difference between treatment groups in termsof time to remission between baseline and week 16achieved borderline statistical significance in favour ofNeoral (P ¼ 0·062), as assessed by the log-rank test.Under the conditions of the study, relapse was seen inonly 8% of the Neoral group and 4% of the Sandimmungroup. In the majority of these patients, the dose hadbeen reduced following remission.

Study medication dosage

The treatment groups began study medication at

similar mean daily doses (2·46 mg/kg per day forNeoral; 2·48 mg/kg per day for Sandimmun). Thepercentage of patients requiring dose increases ordecreases at each visit is illustrated in Fig. 3. The



Figure 1. Box and whisker plot of percentage reduction in Psoriasis Area and Severity Index (PASI) score from baseline by week (intention-to-treatpopulation). The top and bottom edges of the boxes are the upper and lower quartiles, and median values at each visit are joined by solid or dottedlines. Whiskers extend from the boxes to the most extreme data value up to a maximum of 1·5 times the interquartile range. Values beyond this areshown by a plot symbol. Boxes displaced to the left represent the Neoral group and those displaced to the right, the Sandimmun group. Patientsattending visit: NN, Neoral group; NS, Sandimmun (SIM) group.

Figure 2. Kaplan–Meier plot showing the cumulative percentage ofpatients in remission at each visit (intention-to-treat population).Number of patients at risk represents the number of patients failingto achieve remission at the previous visit and in whom treatment wasnot discontinued. SIM, Sandimmun.

number of dose increases required to maintain efficacywas lower in the Neoral group (146) than the Sand-immun group (198) and, as a result, the mean doseincrease from baseline at week 16 was 33% for Sand-immun but only 24% for Neoral. In addition, thenumber of patients with dose reductions after week 16was higher for Neoral (n ¼ 83) than for Sandimmun(n ¼ 73); most of these reductions were due to diseasestabilization. The overall effect of the dose adjustmentrequirements was that the mean Neoral dose remainedconsistently lower than the mean Sandimmun dosefrom week 6 onwards (P<0·05; Fig. 4). At week 16,before down-titration for stable remission was possible,the mean dose had increased by 0·58 mg/kg per day forNeoral and 0·81 mg/kg per day for Sandimmun. Byweek 24, the mean Neoral dose had been reduced to alevel similar to that at baseline, while the mean

Sandimmun dose was still higher than baseline by0·17 mg/kg per day.

Among patients who achieved remission (Neoral,n ¼ 127, Sandimmun, n ¼ 124), the majority firstattained remission at a dose of <3·5 mg/kg per day(Neoral, 91%; Sandimmun, 90%). Furthermore, 34% ofpatients who achieved remission in the Neoral group,but only 25% of those in the Sandimmun group, wereon a dose of <2·5 mg/kg per day when a remission firstoccurred.

Throughout the study the number of dose reductionsfor safety reasons was similar for both treatments. Atotal of 74 dose reductions was implemented for safetyreasons: 29 because of elevated serum creatinine, invol-ving 17 patients in the Neoral group and 11 in theSandimmun group; 10 because of hypertension (six inthe Neoral group and four in the Sandimmun group);five because of elevated liver enzymes (four in the Neoralgroup and one in the Sandimmun group); three becauseof hyperkalaemia (one in the Neoral group and two inthe Sandimmun group); and 27 because of otheradverse events (12 in the Neoral group and 15 in theSandimmun group).

Safety analysis

As dose adjustments were permitted, relatively fewpatients in either treatment group discontinued treat-ment prematurely because of laboratory abnormalitiesor adverse events (Neoral, n ¼ 14; Sandimmun, n ¼ 8).Adverse events were experienced by 75% of patientsreceiving Neoral and 68% of patients receiving Sandim-mun; 56% of patients in the Neoral group and 52% inthe Sandimmun group had adverse events that wereconsidered to be at least possibly related to studymedication. Most adverse events were of mild or mod-erate severity and reflected the known side-effect profileof CS. The most frequently occurring events were thoseassociated with the gastrointestinal system—mainlydiarrhoea and nausea (Neoral, 17%; Sandimmun,30%); central and peripheral nervous system com-plaints—mainly headache and paraesthesia (Neoral,26%; Sandimmun, 22%); and symptoms affecting thebody as a whole—mainly influenza-like (Neoral, 25%;Sandimmun, 24%).

No patients died during the study. Non-fatal seriousadverse events were reported in four patients receivingNeoral and seven patients receiving Sandimmun. Ofthese, events in two patients in the Neoral group andone in the Sandimmun group were considered to betreatment-related: one patient receiving Neoral had a



Figure 3. Percentage of patients with dose increases (A) and decreases(B) (safety population). Patients attending visit: NN, Neoral group; NS,Sandimmun (SIM) group.

serum creatinine level elevated >100% above baseline,and the other had a single episode of atrial fibrillation,while the remaining patient took an accidental overdoseof Sandimmun.

No clinically relevant differences between the treat-ment groups were reported for changes in vital signs orlaboratory parameters. The mean blood pressureincreased only slightly over time and to a similarextent in both treatment groups. In addition, the inci-dence and severity of new-onset hypertension weresimilar in both groups: intervention in the form ofinitiation of antihypertensive medication or an increasein the number of antihypertensive drugs was required in7% of patients receiving Neoral and 6% of patientsreceiving Sandimmun.

The mean serum creatinine levels were similarbetween the groups at baseline and remained stablethroughout the study. The percentage of patients with anotable increase in serum creatinine ($ 30% increasefrom baseline) was 20% for Neoral and 15% for Sand-immun. Of those patients who had a clinically notableserum creatinine level, a higher proportion of patientsin the Neoral group (27%) than in the Sandimmungroup (8%) first experienced a notable increase in serumcreatinine levels at a dose <2·5 mg/kg per day. Four

patients discontinued treatment because of elevatedserum creatinine values (two in the Neoral group andtwo in the Sandimmun group).

Discussion

This study was undertaken to compare the efficacy,safety and optimal dose of Neoral and Sandimmun inpatients with severe, generalized, chronic plaque-typepsoriasis for whom conventional therapy had beenineffective or was inappropriate. Remission rates werehigh in both Neoral and Sandimmun treatment groups,although time to remission was shorter with Neoralthan with Sandimmun (P ¼ 0·062). Based on a startingdose of 2·5 mg/kg per day, fewer dose increases wereneeded to maintain efficacy with Neoral compared withSandimmun. In addition, the number of dose reductionsthat were possible because of stable remission after week16 was higher with Neoral than with Sandimmun.Furthermore, throughout the study the number ofdose reductions for safety reasons was similar for bothtreatments. Consequently, the mean Neoral doseremained < 10% lower than the mean Sandimmundose from week 6 onwards. Tolerability was similar forthe two treatment groups, based on increases in serum



Figure 4. Box and whisker plot of mean daily study medication dosage by week (safety population). The top and bottom edges of the boxes are theupper and lower quartiles, and median values at each visit are joined by solid or dotted lines. Whiskers extend from the boxes to the most extremedata value up to a maximum of 1·5 times the interquartile range. Values beyond this are shown by a plot symbol. Boxes displaced to the left representthe Neoral group and those displaced to the right, the Sandimmun group. Patients attending visit: NN, Neoral group; NS, Sandimmun (SIM) group.

creatinine (the main safety parameter), blood pressure,adverse events and number of discontinuations. Thesimilar increase in serum creatinine concentration inthe two treatment groups supports the recommendationthat, as with Sandimmun, this marker should bemonitored throughout treatment with Neoral.8

Although serum creatinine concentration is establishedas a clinically valuable surrogate marker of renalfunction,9,10 the use of more accurate assessmentssuch as urinary creatinine clearance or radiometricmeasurement of glomerular filtration rate may warrantfurther investigation in controlled clinical trials withNeoral.

Overall, the findings of this study confirm theadequacy and validity of the current dosing guidelinesfor Neoral in psoriasis,8 which are based on theguidelines established for the use of Sandimmun.11

These guidelines indicate that, owing to the variabilityof the psoriatic condition and patients’ baseline his-tories, treatment must be individualized. For inductionof remission, the guidelines recommend an initial doseof 2·5 mg/kg per day orally, divided into two doses. Ifthere is no improvement after 1 month, the dailydose may be gradually increased but should notexceed 5 mg/kg per day. For patients in whom sufficientresponse of psoriatic lesions cannot be achieved within3 months on 5 mg/kg per day, or in whom the effectivedose is not compatible with the established safety guide-lines, treatment should be discontinued. Initial doses of5 mg/kg per day are justified in patients whose condi-tion requires rapid improvement. Once a satisfactoryresponse is achieved, Neoral may be discontinued andany subsequent relapse managed by the reintroductionof Neoral at the previous effective dose. In somepatients, continuous maintenance treatment may benecessary; doses should be titrated individually to thelowest effective level and should not exceed 5 mg/kg perday.

The risk of CS renal side-effects can be reduced ifthe established monitoring guidelines are adhered toclosely: serum creatinine should be monitored regularlyfor as long as CS is administered and the dose of CSdecreased if creatinine rises by $ 30% over the patient’spretreatment levels. Ideally, creatinine levels should bemaintained below 130% of the pretreatment levels. Arecent study has noted that older patients may be moresusceptible to CS renal side-effects;12 however, it mustbe emphasized that monitoring is recommended for allpatients, regardless of age.

This study has shown that Neoral demonstratessimilar efficacy and tolerability to Sandimmun in the

treatment of severe psoriasis, when used according toestablished guidelines. The predictable bioavailability ofthe new formulation enables more patients to respondat lower doses. Fewer changes in dose are required toobtain adequate control of the disease, so managementis simplified. The more consistent response seems likelyto facilitate short-term or intermittent treatment withCS, which is undoubtedly safer than long-term, contin-uous therapy.

Acknowledgments

The OLP 302 study group: B.Abrams, G.Albrecht,P.Altmeyer, F.A.Bahmer, S.Belaich, J.Berth-Jones,J.R.Bjerke, J.D.Bos, L.Braathen, G.Burg, D.Burrows,A.Claudy, L.A.Drake, L.Dubertret, R.Engst, M.I.Ettelt,E.Frenk, P.J.Frosch, J.Ganslandt, M.Goos, R.A.C.Gra-ham-Brown, J.J.Guilhou, U.F.Haustein, S.Helland,K.Hutchinson, H.Jacobi, E.G.Jung, S.Kang, P.Kind,J.Knop, N.J.Lowe, T.Luger, G.Mahrle, R.Marks, H.Mef-fert, J.Meyer, N.J.Mørk, U.Mrowietz, H.A.M.Neumann,E.A.Olsen, S.Rogers, T.Ruzicka, A-G.Schmidt, E.Schopf,J.L.Shupack, T.M.Starink, D.Tio, W.A.van Vloten,B.J.Vermeer and U.Wollina.

This study was sponsored by Novartis Pharma AG,CH-4002 Basel, Switzerland.

References1 Mueller W, Herrmann B. Cyclosporine A for psoriasis. N Engl J Med

1979; 301: 555.2 Ellis CN, Fradin MS, Messana JM et al. Cyclosporin for plaque-type

psoriasis: results of a multidose, double-blind trial. N Engl J Med1991; 324: 277–84.

3 van Joost TH, Bos JD, Heule F et al. Low dose cyclosporin A insevere psoriasis: a double-blind study. Br J Dermatol 1988; 118:183–90.

4 Ippolito F. Short- and long-term considerations concerning themanagement of plaque psoriasis with low-dose cyclosporin. Der-matology 1993; 187: 19–29.

5 Faulds D, Goa KL, Benfield P. Cyclosporin. A review of its pharma-codynamic and pharmacokinetic properties, and therapeutic usein immunoregulatory disorders. Drugs 1993; 45: 953–1040.

6 Vonderscher J, Meinzer A. Cyclosporin advantages and limitationsor rationale for the development of Sandimmun Neoral . In:International Congress on Immunointervention: the Role of Neoral

in Immune-based Diseases. Rome: Blackwell Healthcare Commu-nications, 1996; 21.

7 Elder CA, Moore M, Chang C-T et al. Efficacy and pharmacokineticsof two formulations of cyclosporine A in patients with psoriasis. JClin Pharmacol 1995; 35: 865–75.

8 Berth-Jones J, Voorhees JJ. Consensus conference on cyclosporin Amicroemulsion for psoriasis. Br J Dermatol 1996; 135: 775–7.

9 Feutren G, Mihatsch MJ. Risk factors for cyclosporin-inducednephropathy in patients with autoimmune diseases. International



Kidney Biopsy Registry of Cyclosporine in Autoimmune Diseases.N Engl J Med 1992; 326: 1654–60.

10 Powles AV, Cook T, Hulme B et al. Renal function and biopsyfindings after 5 years’ treatment with low-dose cyclosporin forpsoriasis. Br J Dermatol 1993; 128: 159–65.

11 Mihatsch MJ, Wolff K. Consensus conference on cyclosporin A forpsoriasis. Br J Dermatol 1992; 126: 621–3.

12 Lowe NJ, Weider JM, Rosenbach A et al. Long-term low-dosecyclosporine therapy for severe psoriasis: effects on renal functionand structure. J Am Acad Dermatol 1996; 35: 710–19.



Documents

A randomized, double-blind study comparing the efficacy, safety and optimal dose of two formulations of cyclosporin, Neoral and Sandimmun, in patients with severe psoriasis