Pharmacological Therapy of Lupus Nephritis

Downloaded from www.jama.com at Johns Hopkins University, on November 23, 2005

GRAND ROUNDS CLINICIAN’S CORNER AT THE JOHNS HOPKINS BAYVIEW MEDICAL CENTER

Derek M. Fine, MD Kidney involvement is common in systemic lupus erythematosus, occurring CASE PRESENTATION in up to 60% of affected adults during the course of their disease. Diffuse Mrs P, a 31-year-old woman, developed proliferative lupus nephritis (World Health Organization class IV), the most a blood pressure of 170/104 mm Hg at ominous variant, has traditionally been treated with cyclophosphamide and 38 weeks of pregnancy. Her obstetri- glucocorticoids. With cyclophosphamide, women of childbearing potential cian performed a laboratory evaluation must weigh the risks of sustained amenorrhea, infertility, increased suscep- ϩ that revealed proteinuria (2 ), throm- tibility to infection, bone marrow suppression, hemorrhagic cystitis, and ma- bocytopenia (platelet count, 121ϫ103/µL lignancy against the benefits of better disease control compared with glu- [normal range: 150-350ϫ103/µL]), a serum creatinine level of 0.8 mg/dL (70.7 cocorticoids alone. Because of the host of adverse effects associated with µmol/L) (normal range: 0.4-1.1 mg/dL cyclophosphamide, alternative approaches to the treatment of lupus nephri- [35.3-97.2 µmol/L]), uric acid level of 9.0 tis are desirable. A 31-year-old woman developed class IV lupus nephritis mg/dL (normal range: 2.4-5.7 mg/dL), in the postpartum period. Seeking to preserve fertility and avoid other known and aspartate and alanine aminotrans- toxicities of cyclophosphamide, she chose to undergo therapy with myco- ferase levels of 170 U/L and 190 U/L, phenolate mofetil. In the treatment of severe lupus nephritis, mycopheno- respectively (normal range: 0-35 U/L). late mofetil has emerged as an alternative to cyclophosphamide, offering a Mrs P was diagnosed with the HELLP major advance in the therapy of lupus nephritis. (hemolysis, elevated liver enzymes, low platelets) syndrome1 and underwent an JAMA. 2005;293:3053-3060 www.jama.com emergent cesarean delivery of a healthy male newborn. Despite this interven- quent 2 months. Her transient conges- At the time of Mrs P’s nephrology tion, Mrs P continued to have thrombo- tive heart failure was attributed to post- evaluation, her medications included cytopenia and hypertension. In addi- partum cardiomyopathy. Five months candesartan (32 mg/d) and metopro- tion, she developed temperatures up to after delivery, she presented with fe- lol succinate (150 mg/d). The family 102° F. She was treated empirically with ver, fatigue, arthralgias, malar rash, a medical history was noncontributory. antibiotics and received heparin briefly papular eruption on her forearms, and She had previously worked as a neo- for a presumptive diagnosis of pelvic alopecia. Her platelet count was still low natal nurse and did not smoke or drink. ϫ 3 thrombophlebitis. Fevers continued and (69 10 /µL) and she remained ane- Her child was now 6 months old; she she developed weakness and fatigue. mic (hematocrit, 26% [normal range: and her husband were eager to have Four months after delivery, she de- 38%-47%]). These findings prompted more children. veloped progressive dyspnea on exer- a serological workup that revealed high On physical examination the pa- tion and orthopnea. Echocardiogra- titer positive assay for antinuclear an- tient was afebrile with a blood pres- phy revealed an ejection fraction of tibodies (ANAs) and antibodies to sure of 132/80 mm Hg, a resting pulse 30%. Therapy was initiated with an an- double-stranded DNA (dsDNA). Her se- Author Affiliation: Division of Nephrology, Depart- giotensin-converting enzyme inhibi- rum complement levels were low, with ment of Medicine, Johns Hopkins University School tor and a loop diuretic, with resolu- the C3 and C4 complements measur- of Medicine, Baltimore, Md. ing 43 mg/dL (normal range: 79-152 Corresponding Author: Derek M. Fine, MD, Division tion of her symptoms and normalization of Nephrology, Department of Medicine, Johns Hop- of her cardiac function over the subse- mg/dL) and 5 mg/dL (normal range: kins University School of Medicine, 1830 E Monu- 12-42 mg/dL), respectively. Her eryth- ment St, Suite 416, Baltimore, MD 21205 (dfine1 @jhmi.edu). See also Patient Page. rocyte sedimentation rate was el- Grand Rounds at The Johns Hopkins Bayview Medi- evated at 107 mm/h (normal range: 4-25 cal Center Section Editors: John H. Stone, MD, MPH, CME available online at mm/h). The diagnosis of systemic lu- Charles Weiner, MD, Stephen D. Sisson, MD, The Johns www.jama.com Hopkins Hospital, Baltimore, Md; David S. Cooper, MD, pus erythematosus (SLE) was made. Contributing Editor, JAMA.

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS of 80/min, and a respiratory rate of plications. Most of the 22 glomeruli in mofetil in lupus nephritis, Mrs P chose 12/min. She had thinning hair and a ma- the sample showed diffuse mesangial ex- to undergo therapy with mycopheno- lar rash, but her skin was otherwise pansion and proliferative changes, with late mofetil. clear. Her cardiac examination re- mesangial and endocapillary prolifera- vealed a regular rate and rhythm, nor- tion. There were segmental necrotizing DISCUSSION mal S1 and S2, and no murmurs. Her lesions in 4 of the glomeruli and fibro- Mrs P’s presentation is instructive be- lungs were clear to auscultation. There cellular crescents in 2, indicating signifi- cause it highlights the fact that SLE can was no organomegaly, and she had no cant activity (representative glomeruli develop during or after pregnancy and clubbing, cyanosis, edema, or arthritis shown in FIGURE 1). In addition, im- that it can mimic preeclampsia and the in her extremities. The neurological ex- munofluorescence revealed a “full HELLP syndrome. Her case also dem- amination was normal. Additional labo- house” of immunoreactants, with onstrates how to approach difficult ratory results were notable for a se- granular staining for IgG, IgA, C3, C1q, therapeutic issues by involving the pa- rum creatinine level of 1.1 mg/dL (97.2 and ␬ and ␭ light chains. The biopsy tient in the decision-making process. In µmol/L) with a urine protein to creati- was consistent with diffuse prolifera- Mrs P’s case, the particular concern re- nine ratio of 2.5 mg protein/mg creati- tive (World Health Organization lated to the potential for both short- and nine (proteinuria approximately 2.5 [WHO] class IV) glomerulonephritis. long-term adverse effects of cyclophos- g/d). Microscopic examination of her A meeting including the patient, her phamide, particularly those related to urine showed 10 to 15 red blood cells/ husband, the rheumatologist, and the fertility. The choice was between a stan- high power field, but no red cell casts. nephrologists was convened to dis- dard medication associated with sub- Further serological workup revealed an- cuss treatment options. Although cy- stantial risks of toxicity and a newer tibodies to Ro and Sm. clophosphamide was considered the agent with a limited track record. Her A kidney biopsy was performed to de- standard of care in this setting, the pa- case illustrates emerging data that termine the histological nature and se- tient and her husband were con- strongly support mycophenolate mofetil verity of her renal process. However, cerned about the potential effects of that as an alternative to cyclophosphamide given that her platelet count remained medication on fertility, the risk of ma- in the treatment of lupus nephritis. low, biopsy was delayed until after a lignancy, and the possible infectious The diagnosis of SLE was suspected methylprednisolone pulse, 1 g/d for 3 consequences of severe immunosup- in this patient when she developed a days, followed by 60 mg/d of predni- pression. In light of these concerns, malar rash, alopecia, arthralgias, and sone. The methylprednisolone led to a treatment with mycophenolate mofetil worsening fatigue, accompanied by platelet count increase to greater than was considered. Presented with a bal- thrombocytopenia and hemolytic ane- 100ϫ103/µL within 5 days, and the kid- anced discussion of the data now avail- mia. The diagnosis was confirmed by ney biopsy was undertaken without com- able on the use of mycophenolate positive assays for ANAs, antibodies to

Figure 1. Light Microscopic Findings of Representative Glomeruli

A B Fibrocellular Crescent

A, Hyperlobulated glomerulus with global involvement with endocapillary and mesangial hypercellularity with matrix expansion (white arrowheads) and wireloop lesions (black arrowheads). B, Glomerulus with global endocapillary proliferation, leukocyte influx, mesangial expansion, and crescent formation (hematoxylin-eosin stain; original magnification x 400).

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS double-stranded DNA (dsDNA), and Table 1. Classification and Treatment of the Different Forms of Lupus Nephritis the findings of her renal biopsy. Her WHO Immunosuppression presentation was confusing because Class* Description Recommendations† both preeclampsia and the HELLP syn- I Minimal mesangial lupus nephritis No specific therapy drome can mimic SLE closely, and her II Mesangial proliferative lupus nephritis No specific therapy stage of pregnancy corresponded pre- III Focal (proliferative) lupus nephritis Mild: none or glucocorticoids cisely to the time at which these enti- Severe: see treatment for class IV below ties occur. The persistence of her clini- IV Diffuse (proliferative) lupus nephritis Induction (6 mo): cyclophosphamide or cal and laboratory abnormalities for mycophenolate mofetil months during the postpartum period Maintenance: mycophenolate mofetil or azathioprine were consistent with SLE as the sole V Membranous lupus nephritis Glucocorticoid with or without cyclosporine cause of her presentation during preg- VI Advanced sclerosing lupus nephritis No specific therapy nancy. In the absence of postpartum Abbreviation: WHO, World Health Organization. complications, for example, the plate- *Subclasses omitted (see Weening et al6 for full classification). 7 let counts in the HELLP syndrome nor- †Recommendations derived from Rose et al. mally rebound toward normal within 1 week of delivery.1 The confusion in of Nephrology and the Renal Pathol- score is associated with a worse prog- the postpartum period was com- ogy Society.6 The general structure in- nosis.15,17,18 The absence of chronic pounded further by her development cludes 6 principal pathological pat- changes is particularly informative as of cardiomyopathy. Although post- terns (classes I-VI) (TABLE 1). these patients have an excellent prog- partum cardiomyopathy is certainly a The focal and diffuse proliferative nosis.19 Therefore, when chronicity is possible explanation, her cardiac forms of lupus nephritis (classes III and limited, therapeutic interventions are dysfunction may also have been IV, respectively) are distinguished from likely to have maximum benefit. related to her SLE.2,3 Moreover, SLE- each other only by the percentage of In Mrs P’s case, the presence of an related cardiomyopathy has also been glomeruli involved. These renal le- elevation of her creatinine level with reported in the postpartum period.4 sions generally present with micro- hematuria and moderate levels of pro- Systemic lupus erythematosus cardi- scopic hematuria with or without red teinuria suggested most likely the omyopathy, like postpartum cardio- blood cell casts, varying degrees of pro- presence of a class III or IV prolifera- myopathy, may be associated with a teinuria, and progressive renal failure. tive lesion, although without biopsy waxing and waning course and spon- In contrast, the membranous lesion these could not be distinguished. Par- taneous resolution. (class V) typically presents with ne- ticularly important was the need to phrotic-range proteinuria. Several stud- determine the severity of the activity, Kidney Involvement in SLE ies, however, have illustrated the un- as the clinical presentation associated The kidney is a major target organ of reliability of diagnoses rendered on the with mild histopathologic activity can SLE. Up to 60% of patients with SLE basis of clinical features alone.8-10 Be- be very similar to one seen with more will develop renal manifestations at cause the optimal treatment varies with active histopathology earlier in its some point in their course, with 25% the type of glomerular disease, kidney development. Indeed, in this case, to 50% presenting with kidney involve- biopsy should be performed to make a despite relatively mild clinical fea- ment early.5 The clinical presentation definitive diagnosis.11,12 In addition to tures, significant disease activity was of kidney involvement is highly vari- histopathologic confirmation of the present. able, ranging from mild asymptomatic WHO class, biopsy provides impor- The differential diagnosis of kidney proteinuria to rapidly progressive glo- tant prognostic information by permit- involvement in SLE extends beyond the merulonephritis. Features generally ting the pathologist to assess both ac- WHO classification to include renal include varying degrees of glomerular tivity and chronicity.13,14 thrombotic microangiopathy, usually involvement with proteinuria— Histological activity and chronicity related to presence of antiphospho- nephrotic in 45% to 65% of cases5—as scores have been developed to allow for lipid antibodies, which may be pres- well as hematuria with red cell casts improved prediction of the progres- ent in 15% to 90% of SLE patients.20 and/or acute renal failure. sion of lupus nephritis.15 Of the fea- Other nonlupus causes of kidney dis- The histopathologic manifestations tures indicating activity, the presence of ease that may affect a person of simi- of lupus nephritis are classified into sev- cellular crescents13,15,16 and fibrinoid ne- lar age or sex (eg, IgA nephropathy, thin eral categories designated by the WHO crosis15 appear to have the highest pre- basement membrane disease) must classification. These criteria have un- dictive value for poorer outcome. When also be considered. The differentia- dergone several revisions, the most re- using the chronicity index, which in- tion of these disorders is once again en- cent of which evolved under the aus- cludes features such as glomerular scle- hanced by performance of a kidney pices of both the International Society rosis and interstitial fibrosis, a higher biopsy.

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS

Current Treatment Options tion with glucocorticoid therapy. also be recognized, however, that ear- in Lupus Nephritis Following the protocol popularized by lier trials of cyclophosphamide used The optimal treatment regimen7 in lu- the NIH investigators, intravenous cy- higher doses for longer durations than pus nephritis varies according to WHO clophosphamide is used as induction the current standard regimens. Among class5,14 (Table 1). Patients with the therapy, administered monthly at a dose patients treated with cyclophospha- mildest forms of lupus nephritis (WHO of0.5to1g/m2 of body surface area for mide in one study, 26% developed infec- class I or II) generally do well without 6 months. During the maintenance tion compared with 7% in the glucocor- specific intervention. In the absence phase, cyclophosphamide is adminis- ticoid group.27 However, there was no of appropriate immunosuppressive tered at the same dose as induction difference in infection rates on long- therapy, however, the proliferative therapy for 4 to 6 additional cycles. In- term follow-up, although trends con- forms (class III and IV) of lupus ne- travenous methylprednisolone treat- sistently show higher infection rates with phritis typically progress to chronic re- ment (1 g/d for 3 days) is frequently ad- cyclophosphamide treatment. This is nal failure.21 The benefits of early treat- ministered at initiation of therapy, illustrated by the higher incidence of her- ment are well documented.22 This has followed by tapering oral doses start- pes zoster infection in 15% of patients led to a propensity to treat all patients ing at 0.5 to 1.0 mg/kg/d. Based on stud- (vs 4% with glucocorticoids) and cer- with proliferative lesions regardless of ies that have evaluated the efficacy of vical dysplasia in 11% of patients (vs 0% severity. In patients such as Mrs P, who intermittent glucocorticoid pulses, with glucocorticoids). Other notable have the most severe forms of lupus ne- many patients also receive monthly complications during treatment include phritis, aggressive immunosuppres- methylprednisolone pulses on the the development of avascular necrosis sive therapy is warranted. same day they receive cyclophospha- in 11% of patients (vs 22% with gluco- mide.27,29 corticoids alone). However, the only Cyclophosphamide in the adverse effect to achieve statistical sig- Treatment of Lupus Nephritis Cyclophosphamide Toxicity nificance between the cyclophospha- Early treatment regimens for class IV Mrs P’s concerns about the potential mide and glucocorticoid only groups was lupus nephritis involved predomi- toxicities of cyclophosphamide and her amenorrhea (52% vs 10%; PϽ.001).27 nantly the use of high-dose glucocor- desire to avoid the drug were well ticoids. Remissions on low doses were founded. Immediate toxicity of pulse Cyclophosphamide difficult to maintain, and most pa- cyclophosphamide includes nausea, and Ovarian Failure tients required high doses of glucocor- vomiting, hair loss, and fatigue. Major The devastating complication of go- ticoids for long periods of time to toxicities include cytopenias, serious in- nadal failure is well described with the achieve control of the disease. Due to fections, hemorrhagic cystitis, malig- NIH regimen, with rates of ovarian fail- the significant toxicity and poor long- nancy, and, of great importance to this ure ranging from 26% to 52%.27,32,33 The term outcome, the search for more patient, gonadal failure.30 Serious in- risk of toxicity increases with both dose effective and glucocorticoid-sparing fections are common in SLE, and death and duration of cyclophosphamide regimens began. In early trials, cyclo- from infection correlates with the therapy. Moreover, amenorrhea is more phosphamide in combination with glu- recent use of glucocorticoids and likely to occur in older women. cocorticoids demonstrated improved re- cyclophosphamide.30,31 Boumpas et al33 found that 12% of those nal survival over glucocorticoid therapy Although their renal outcomes were treated with 7 monthly cyclophospha- alone23 and achieved lower rates of re- significantly improved and they had mide doses developed amenorrhea, currence.24 Intravenous cyclophospha- reduced glucocorticoid exposure, compared with 39% of those who re- mide became preferred over the oral patients treated with cyclophospha- ceived more than 14 doses (P=.07). agent due to perceived lower levels of mide under NIH protocol had more Rates also increased with older age, with toxicity. Subsequent studies showed adverse effects in both short- and long- only 12% of patients 25 years or that longer duration of therapy during term follow-up when compared with younger developing sustained amen- the maintenance phase improved re- glucocorticoids alone.27,29 Mortality was orrhea, compared with 27% of pa- mission rates.25 7.4%27 with initial cyclophosphamide tients 26 to 30 years old and 62% of pa- Based on investigations conducted at administration (mean follow-up, 5 years) tients older than 30 years (P=.04). the National Institutes of Health (NIH) and 19% at a median 11 years of follow- Although young women may not de- over the past 20 years,23,25-27 intrave- up,29 compared with 0% and 4%, respec- velop ovarian failure at rates as high in nous cyclophosphamide became the tively, in the glucocorticoid treatment short-term follow-up, they are likely to standard therapy for class IV lupus ne- arm. These high mortality rates with the experience menopause earlier. In an ex- phritis.28 Had Mrs P undergone therapy use of cyclophosphamide, insuffi- tended follow-up of up to 11 years of with this agent, she would have re- ciently acknowledged in some reviews, an earlier NIH cohort,27 60% of those ceived cyclophosphamide in an intra- must be considered when contemplat- treated with cyclophosphamide devel- venous form as a bolus in combina- ing the use of cyclophosphamide. It must oped amenorrhea.

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS

Preservation of ovarian function by with cyclophosphamide.39 Although al- Table 2. Selected Adverse Reactions inducing gonadal quiescence with go- ternatives to cyclophosphamide have Reported in More Than 10% of Patients nadotropin-releasing hormone antago- been assessed, until the introduction of Treated With Mycophenolate Mofetil* nists has been recommended.34,35 Al- mycophenolate mofetil, none had Frequency Frequency though no large studies have been shown the potential to rival cyclophos- Adverse in Treatment in Placebo Reaction Groups, % Groups, % performed, several small studies sug- phamide in efficacy while surpassing it Abdominal pain 12-28 11 gest that these agents have the poten- in safety. Diarrhea 16-36 14 tial to prevent ovarian failure.34,36-38 The Vomiting 13-14 . . . Use of Mycophenolate Mofetil in Infection† 18-21 14 largest study in SLE patients involved Urinary tract 37-46 . . . 36 women treated with cyclophospha- the Treatment of Lupus Nephritis infection Leukopenia 12-35 4 mide. Chronic amenorrhea occurred in Mycophenolate mofetil is metabo- Anemia 25 . . . 11% of those treated with concomi- lized to the active immunosuppres- *Ellipses indicates comparative placebo data not reported. tant leuprolide and 39% of those with- sant mycophenolic acid. Through the †Most common in transplant patients receiving combination immunosuppression (includes cytomegalovirus, her- out it (P=.06). The most serious ad- inhibition of the enzyme inosine mono- pes zoster, and mucocutaneous fungal infections).44-47 verse effect of leuprolide is the potential phosphate dehydrogenase by its me- for bone mineral density loss caused by tabolite, mycophenolate mofetil blocks relative estrogen deficiency.35 Under de novo synthesis of purines, a path- domized trials began. Two trials that as- ideal circumstances, leuprolide is ad- way essential for the synthesis of DNA sessed the role of mycophenolate ministered 3 to 4 weeks before the first in lymphocytes.40 Through this mecha- mofetil in remission induction and one dose of cyclophosphamide, but in the nism, it inhibits B and T cell prolifera- that evaluated the use of mycopheno- presence of acute, life-threatening dis- tion, antibody formation, and genera- late mofetil as a remission mainte- ease, this is usually not an option. In tion of cytotoxic T cells. In addition, nance agent are discussed below. All 3 such instances, leuprolide may be given mycophenolate mofetil inhibits the ex- of these investigations were random- closer to the first dose.35 Other op- pression of adhesion molecules on en- ized, but not blinded. tions include oocyte cryopreservation dothelial cells41 and down-regulates me- or preservation of fertilized eggs. These sangial cell proliferation.41 Mycophenolate Mofetil alternatives also suffer from the time Due to its history as an immunosup- and Induction Therapy constraints necessary for the induc- pressant in solid-organ transplanta- for Lupus Nephritis tion of hyperovulation. tion, mycophenolate mofetil gener- The first randomized trial of induction ated interest as a possible therapy for therapy with mycophenolate mofetil Need for Non–Cyclophosphamide- lupus nephritis. The use of mycophe- was conducted in Hong Kong.48 Forty- Based Regimens nolate mofetil in the treatment of hu- two patients with diffuse proliferative Although many patients with prolif- man glomerular disease was pio- lupus nephritis (WHO class IV) were erative lupus nephritis achieve remis- neered by Briggs and colleagues,42 who assigned to receive either mycopheno- sions with the current cyclophospha- treated 2 patients with proliferative lu- late mofetil (1 g twice a day) or oral mide regimens, there remain significant pus nephritis effectively with myco- cyclophosphamide (2.5 mg/kg/d). numbers of treatment failure and re- phenolate mofetil. Dooley and col- Patients in both treatment arms received nal disease relapses. In view of the fail- leagues43 subsequently reported a series prednisolone (starting dose 0.8 mg/kg/d ure of cyclophosphamide-based regi- of 13 patients with lupus nephritis, 12 tapered to 10 mg/d by 6 months). For mens to induce lasting remissions in of whom had class IV disease, who did maintenance therapy during the sec- many patients and the substantial tox- not respond to cyclophosphamide ond 6 months of therapy, the patients icity associated with this medication, therapy. In that series, only 1 patient treated with mycophenolate mofetil had the development of alternative ap- experienced an elevation in serum cre- a 50% reduction of their dose and the proaches to treatment is essential. In a atinine level over the course of the patients treated with cyclophospha- recent study39 in which women were study; 2 patients had increasing pro- mide were switched to azathioprine (2.5 asked about hypothetical treatment teinuria. Adverse effects reported in- mg/kg/d). Both groups continued tak- preferences, 98% stated that they would cluded pancreatitis (n=1), herpes sim- ing prednisolone at low to moderate choose azathioprine over cyclophos- plex stomatitis associated with severe doses. Complete remission, defined as phamide if the 2 medications were con- leukopenia (n=1), pneumonia with- a stable serum creatinine level (Ͻ15% sidered equally effective. Even given a out leukopenia (n=1), asymptomatic above baseline), normal urine sedi- theoretical probability of renal sur- leukopenia (n=2), and nausea/ ment, and less than 0.3 g/d of protein- vival of 100% at 5 years with cyclo- diarrhea (n=2). Based on the appar- uria, was achieved in 81% of those phosphamide, a substantial minority of ent success in this and other reports in treated with mycophenolate mofetil and 31% would still have preferred azathio- addition to a potentially more limited 76% of those treated with oral cyclo- prine, given the risk of ovarian failure adverse effect profile (TABLE 2),44-47 ran- phosphamide (PϾ.99)(TABLE 3). The

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS

Table 3. Summary of Major Mycophenolate Mofetil Trial Outcomes No. of Subjects Source and Follow-up Treatment Regimen Primary Outcome Toxicity Chan et al,48 42 Induction mycophenolate Mycophenolate mofetil: Mycophenolate mofetil: 2000 21 Mycophenolate mofetil mofetil vs oral Complete remission, 81%* Amenorrhea, 0% 21 Cyclophosphamide cyclophosphamide Partial remission, 14%† Severe infection, 19% 1 y follow-up (see text) Relapse, 15% Leukopenia, 0% Cyclophosphamide: Death, 0% Complete remission, 76%* Cyclophosphamide: Partial remission, 14%† Amenorrhea, 23% Relapse, 11% Severe infection, 33% Leukopenia, 10% Death, 10% Appel et al,49 140 Induction mycophenolate Mycophenolate mofetil: Mycophenolate mofetil: 2003 71 Mycophenolate mofetil mofetil vs intravenous Complete remission, Amenorrhea, no data yet 69 Cyclophosphamide cyclophosphamide 21%‡§ reported 24-wk outcome Partial remission, 29%|| Severe infection, 6% Cyclophosphamide: Cyclophosphamide: Complete remission, 6%‡§ Amenorrhea, no data yet Partial remission, 25%|| reported Severe infection, 13% Contreras et al,50 59 Maintenance mycophenolate Mycophenolate mofetil: Mycophenolate mofetil: 2004 20 Mycophenolate mofetil mofetil vs azathioprine Chronic renal failure, 95%¶ Amenorrhea, 6%§ 19 Azathioprine vs intravenous Renal relapse, 15%#§ Major infection, 2%§ 20 Cyclophosphamide cyclophosphamide Cyclophosphamide: Leukopenia, 2% 1- to 5-y follow-up (after intravenous Chronic renal failure, 74%¶ Death, 5% (cumulative rates) cyclophosphamide Renal relapse, 40%#§ Cyclophosphamide: induction in both) Amenorrhea, 32%§ Major infection, 25%§ Leukopenia, 10% Death, 20% *Complete remission: less than 0.3 g/d protein, normal serum albumin, normal urine sediment, creatinine no more than 15% above baseline. †Partial remission: 0.3-2.9 g/d protein, albumin level greater than 3.0 g/dL. ‡Complete remission: urine protein less than 0.5 g/d and normal sediment, serum creatinine level normal. §PϽ.05. ||Partial remission: more than 50% improvement. ¶Doubling of the nadir serum creatinine level or the development of end-stage renal disease. #Doubling of urine protein:creatinine ratio or increase in serum creatinine level of at least 50%.

frequency of severe adverse effects was Figure 2. Mrs P’s Urine Protein Changes and Glucocorticoid Tapering Over the First greater in the cyclophosphamide group, 9 Months of Treatment With Mycophenolate Mofetil including death (10%) and amenor- 9 rhea (23%), compared with none of either in the mycophenolate mofetil 8 group. It is important to recognize that 7 in this trial, oral cyclophosphamide was 6 used, contrary to the intravenous route of administration according to the NIH 5 regimen. Nevertheless, the equal effi- 4 cacy of the less toxic drug, mycophe- Proteinuria 3 nolate mofetil, has been a major advance in defining its role in the treatment of 2 severe lupus nephritis. 1 49

(Random Urine: mg Protein/mg Creatinine) (Random Urine: mg Protein/mg A more recent study presented in 0 abstract form compared remission in- 3/1/2004 6/1/2004 9/1/2004 12/1/2004 duction with mycophenolate mofetil to Date of Assessment conventional intravenous cyclophos- Intravenous Methylprednisolone 1 g/d for 3 d phamide (Table 3). In this multi-

Prednisone center trial based in the United States, 140 patients were randomized to re- 60 55 40 30 25 17.5 15 12.5 10 7.5 5 2.5 mg/d ceive treatment with intravenous cy-

Mycophenolate Mofetil 1g Twice Daily clophosphamide or mycophenolate mofetil, with a target dose of mycophe-

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS nolate mofetil (1500 mg twice daily) mofetil and azathioprine groups. Al- jor issues surrounding the risk of fu- higher than that used in the Hong Kong though these results support the use of ture pregnancy—the risk of an SLE study.48 The rationale for the higher my- mycophenolate mofetil (and azathio- flare, preeclampsia, premature deliv- cophenolate mofetil dose was evi- prine) in maintenance therapy, there are ery, or poor fetal outcome55-58— dence in the transplantation literature several major limitations.54 First, the continue to be discussed with Mrs P on that African Americans, who com- definition of remission—reduction in a regular basis. prised 56% of the trial cohort in the US the urine protein:creatinine ratio to less study, require higher doses of myco- than 3 if nephrotic at enrollment, and CONCLUSION phenolate mofetil than whites.51 Al- a ratio 50% of baseline if subne- Mycophenolate mofetil has substan- though mycophenolate mofetil was phrotic—was less stringent in compari- tial appeal as a new approach to the superior to cyclophosphamide in re- son to other trials. Second, many pa- treatment of lupus nephritis. Addi- mission induction, the preliminary data tients did not reach a satisfactory tional data related to the use of myco- showed much lower rates of remis- remission by the end of the induction phenolate mofetil, including long- sion with induction than in the Hong phase, which made it likely that they term remission and relapse rates, ideal Kong trial. Only 21% of those receiv- would fare poorly with the same drug length of treatment, optimal glucocor- ing mycophenolate mofetil and 6% of (cyclophosphamide) continued in the ticoid tapering, and long-term toxici- those receiving cyclophosphamide maintenance phase. Overall, the poor ties, are the subject of ongoing studies. achieved complete remissions in the US response to induction seen in this study For now, as Mrs P’s case demon- study (P=.005). Some of the disparity can in large part be accounted for by strates, mycophenolate mofetil is a rea- between the 2 studies can be explained the large proportion of Hispanic and sonable option for many patients with by racial differences between the co- black patients (5% were white).52 lupus nephritis who seek a safer, effec- horts (black race predicts poorer out- tive alternative to cyclophosphamide. come in class IV lupus nephritis52,53). CASE RESOLUTION Financial Disclosures: None reported. There may also have been differences in Mrs P began receiving mycophenolate Acknowledgment: I thank my patient for sharing her the histological severity and clinical se- mofetil, 1000 mg twice daily in addi- story and Allan Gelber, MD, MPH, PhD, John Stone, MD, MPH, and Michael Choi, MD, for reading and verity of disease in the 2 trials. Other po- tion to the 60 mg daily of prednisone. improving the manuscript. tential explanations include the route of Within weeks, her serum creatinine therapy (intravenous vs oral) and the level had improved from 1.1 mg/dL REFERENCES doses of glucocorticoids used. The defi- (97.2 µmol/L) to 0.8 mg/dL (70.7 1. Martin JN Jr, Blake PG, Perry KG Jr, McCaul JF, Hess nitions of remission were similar and µmol/L) and her thrombocytopenia had LW, Martin RW. The natural history of HELLP syndrome: patterns of disease progression and regression. probably cannot explain the differ- resolved. Her hematocrit increased to Am J Obstet Gynecol. 1991;164:1500-1509. ences seen. Because detailed data from 35% and her complement levels nor- 2. Borenstein DG, Fye WB, Arnett FC, Stevens MB. the latter trial are not available, these ex- malized. After an initially sharp in- The myocarditis of systemic lupus erythematosus: as- sociation with myositis. Ann Intern Med. 1978;89:619- planations remain speculative. crease in proteinuria, within 3 months, 624. her proteinuria had decreased to less 3. Moder KG, Miller TD, Tazelaar HD. Cardiac in- Mycophenolate Mofetil volvement in systemic lupus erythematosus. Mayo Clin than 1 g/mg creatinine (FIGURE 2). Proc. 1999;74:275-284. and Maintenance Therapy Prednisone was tapered and discontin- 4. Raskin RJ, Haddock JB, Lawless OJ. Systemic lupus erythematosus with myocarditis complicating Another major trial recently evaluated ued 5 months after the initiation of my- pregnancy. South Med J. 1983;76:258-260. the use of mycophenolate mofetil in re- cophenolate mofetil. No longer taking 5. Cameron JS. Lupus nephritis. J Am Soc Nephrol. mission maintenance.50 After induc- prednisone for over 4 months, Mrs P 1999;10:413-424. 6. Weening JJ, D’Agati VD, Schwartz MM, et al. The 2 tion with 4 to 7 months of 0.5 to 1 g/m has maintained improved renal func- classification of glomerulonephritis in systemic lupus of intravenous cyclophosphamide, pa- tion (serum creatinine level, 0.8 mg/dL erythematosus revisited. Kidney Int. 2004;65:521-530. 7. Rose BD, Schur PH, Falk RJ, Appel GB. Treatment tients were randomized to 1 of 3 treat- [70.7 µmol/L]) and proteinuria (0.4 of lupus nephritis. UpToDate Web site. Available at: ment groups (Table 3): quarterly in- g/mg creatinine), a stable hematocrit http://patients.uptodate.com/topic.asp?file=glomrdis /16594&title=Lupusϩnephritis. Accessibility verified travenous cyclophosphamide, daily oral and platelet count, and normocomple- May 17, 2005. mycophenolate mofetil, or daily oral mentemia. Mild fatigue and a minimal 8. Huong DL, Papo T, Beaufils H, et al. Renal involvement in systemic lupus erythematosus: a study of 180 azathioprine. Although the cumula- malar rash remain her only current patients from a single center. Medicine (Baltimore). tive rate of renal survival did not differ symptoms. We plan to complete 18 1999;78:148-166. significantly between the groups, those months of maintenance therapy, based 9. Nossent JC, Henzen-Logmans SC, Vroom TM, Huy- sen V, Berden JH, Swaak AJ. Relation between sero- in the cyclophosphamide group had an on the experience when using cyclo- logical data at the time of biopsy and renal histology increased relapse rate compared with phosphamide. Because mycopheno- in lupus nephritis. Rheumatol Int. 1991;11:77-82. 10. Gladman DD, Urowitz MB, Cole E, Ritchie S, Chang mycophenolate mofetil, increased mor- late mofetil is contraindicated in preg- CH, Churg J. Kidney biopsy in SLE, I: a clinical- tality compared with azathioprine, and nancy, future pregnancy could be morphologic evaluation. QJMed. 1989;73:1125- 1133. increased drug-related morbidity com- considered after the maintenance course 11. Kashgarian M. Lupus nephritis: lessons from the pared with both the mycophenolate if she remains in remission. Other ma- path lab. Kidney Int. 1994;45:928-938.

PHARMACOLOGICAL THERAPY OF LUPUS NEPHRITIS

12. Grande JP, Balow JE. Renal biopsy in lupus 28. Mok CC, Wong RW, Lai KN. Treatment of se- cal observations. J Am Soc Nephrol. 1999;10:833-839. nephritis. Lupus. 1998;7:611-617. vere proliferative lupus nephritis: the current state. Ann 44. Halloran P, Mathew T, Tomlanovich S, Groth C, 13. Moroni G, Pasquali S, Quaglini S, et al. Clinical Rheum Dis. 2003;62:799-804. Hooftman L, Barker C; The International Mycophe- and prognostic value of serial renal biopsies in lupus 29. Illei GG, Austin HA, Crane M, et al. Combination nolate Mofetil Renal Transplant Study Groups. My- nephritis. Am J Kidney Dis. 1999;34:530-539. therapy with pulse cyclophosphamide plus pulse meth- cophenolate mofetil in renal allograft recipients: a 14. Contreras G, Roth D, Pardo V, Striker LG, Schultz ylprednisolone improves long-term renal outcome pooled efficacy analysis of three randomized, double- DR. Lupus nephritis: a clinical review for practicing without adding toxicity in patients with lupus nephritis. blind, clinical studies in prevention of rejection. nephrologists. Clin Nephrol. 2002;57:95-107. Ann Intern Med. 2001;135:248-257. Transplantation. 1997;63:39-47. 15. Austin HA III, Muenz LR, Joyce KM, et al. Prog- 30. Petri M. Cyclophosphamide: new approaches for 45. Mycophenolate mofetil (Cellcept) product nostic factors in lupus nephritis: contribution of renal systemic lupus erythematosus. Lupus. 2004;13:366-371. information. Nutley, NJ: Roche Laboratories; 1995. histologic data. Am J Med. 1983;75:382-391. 31. Hellmann DB, Petri M, Whiting-O’Keefe Q. Fa- 46. European Mycophenolate Mofetil Cooperative 16. Austin HA III, Boumpas DT, Vaughan EM, Ba- tal infections in systemic lupus erythematosus: the role Study Group. Placebo-controlled study of mycophe- low JE. High-risk features of lupus nephritis: impor- of opportunistic organisms. Medicine (Baltimore). nolate mofetil combined with cyclosporin and corti- tance of race and clinical and histological factors in 166 1987;66:341-348. costeroids for prevention of acute rejection. Lancet. patients. Nephrol Dial Transplant. 1995;10:1620-1628. 32. Mok CC, Lau CS, Wong RW. Risk factors for ovar- 1995;345:1321-1325. 17. Austin HA III, Boumpas DT, Vaughan EM, Ba- ian failure in patients with systemic lupus erythema- 47. Sollinger HW; U.S. Renal Transplant Mycophe- low JE. Predicting renal outcomes in severe lupus ne- tosus receiving cyclophosphamide therapy. Arthritis nolate Mofetil Study Group. Mycophenolate mofetil phritis: contributions of clinical and histologic data. Kid- Rheum. 1998;41:831-837. for the prevention of acute rejection in primary ca- ney Int. 1994;45:544-550. 33. Boumpas DT, Austin HA III, Vaughan EM, Yar- daveric renal allograft recipients. Transplantation. 1995; 18. Nossent HC, Henzen-Logmans SC, Vroom TM, boro CH, Klippel JH, Balow JE. Risk for sustained amen- 60:225-232. Berden JH, Swaak TJ. Contribution of renal biopsy data orrhea in patients with systemic lupus erythematosus 48. Chan TM, Li FK, Tang CS, et al; Hong Kong- in predicting outcome in lupus nephritis: analysis of receiving intermittent pulse cyclophosphamide therapy. Guangzhou Nephrology Study Group. Efficacy of 116 patients. Arthritis Rheum. 1990;33:970-977. Ann Intern Med. 1993;119:366-369. mycophenolate mofetil in patients with diffuse pro- 19. Mittal B, Rennke H, Singh AK. The role of kidney 34. Blumenfeld Z, Shapiro D, Shteinberg M, Avivi I, liferative lupus nephritis. N Engl J Med. 2000;343:1156- biopsy in the management of lupus nephritis. Curr Opin Nahir M. Preservation of fertility and ovarian func- 1162. Nephrol Hypertens. 2005;14:1-8. tion and minimizing gonadotoxicity in young women 49. Appel GB, Ginzler EM, Radhakrishnan J, et al. Mul- 20. Nzerue CM, Hewan-Lowe K, Pierangeli S, Harris with systemic lupus erythematosus treated by ticenter controlled trial of mycophenolate mofetil as EN. “Black swan in the kidney”: renal involvement in chemotherapy. Lupus. 2000;9:401-405. induction therapy for severe lupus nephritis [abstract]. the antiphospholipid antibody syndrome. Kidney Int. 35. Slater CA, Liang MH, McCune JW, Christman GM, J Am Soc Nephrol. 2003;14:38A. 2002;62:733-744. Laufer MR. Preserving ovarian function in patients re- 50. Contreras G, Pardo V, Leclercq B, et al. Sequen- 21. Appel GB, Cohen DJ, Pirani CL, Meltzer JI, Estes ceiving cyclophosphamide. Lupus. 1999;8:3-10. tial therapies for proliferative lupus nephritis. N Engl D. Long-term follow-up of patients with lupus ne- 36. Blumenfeld Z. Ovarian rescue/protection from che- J Med. 2004;350:971-980. phritis: a study based on the classification of the World motherapeutic agents [abstract]. J Soc Gynecol Investig. 51. Neylan JF; U.S. Renal Transplant Mycopheno- Health Organization. Am J Med. 1987;83:877-885. 2001;8(suppl):S60-S64. late Mofetil Study Group. Immunosuppressive therapy 22. Esdaile JM, Joseph L, MacKenzie T, Kashgarian 37. Dooley MA, Patterson CC, Hogan SL, et al. Pres- in high-risk transplant patients: dose-dependent ef- M, Hayslett JP. The benefit of early treatment with ervation of ovarian function using depot leuprolide ac- ficacy of mycophenolate mofetil in African-American immunosuppressive agents in lupus nephritis. etate during cyclophosphamide therapy for severe renal allograft recipients. Transplantation. 1997;64: J Rheumatol. 1994;21:2046-2051. lupus nephritis [abstract]. Arthritis Rheum. 2000; 1277-1282. 23. Austin HA III, Klippel JH, Balow JE, et al. Therapy 43(suppl):2858. 52. Barr RG, Seliger S, Appel GB, et al. Prognosis in of lupus nephritis: controlled trial of prednisone and 38. McCune J, Somers EC, Ognenovski V, Christ- proliferative lupus nephritis: the role of socio- cytotoxic drugs. N Engl J Med. 1986;314:614-619. man G. Use of leuprolide acetate for ovarian protec- economic status and race/ethnicity. Nephrol Dial 24. Donadio JV Jr, Holley KE, Ferguson RH, Ilstrup DM. tion during cyclophosphamide therapy of women with Transplant. 2003;18:2039-2046. Treatment of diffuse proliferative lupus nephritis with severe SLE [abstract]. Arthritis Rheum. 2001;42(suppl) 53. Dooley MA, Hogan S, Jennette C, Falk R; Glo- prednisone and combined prednisone and :2006. merular Disease Collaborative Network. Cyclophos- cyclophosphamide. N Engl J Med. 1978;299:1151-1155. 39. Fraenkel L, Bogardus S, Concato J. Patient pref- phamide therapy for lupus nephritis: poor renal sur- 25. Boumpas DT, Austin HA III, Vaughn EM, et al. erences for treatment of lupus nephritis. Arthritis vival in black Americans. Kidney Int. 1997;51: Controlled trial of pulse methylprednisolone versus two Rheum. 2002;47:421-428. 1188-1195. regimens of pulse cyclophosphamide in severe lupus 40. Allison AC, Eugui EM. Mycophenolate mofetil and 54. Gelber AC, Christopher-Stine L, Fine DM. Se- nephritis. Lancet. 1992;340:741-745. its mechanisms of action. Immunopharmacology. 2000; quential therapies for proliferative lupus nephritis. 26. Steinberg AD, Steinberg SC. Long-term preser- 47:85-118. N Engl J Med. 2004;350:2518-2520. vation of renal function in patients with lupus nephri- 41. Blaheta RA, Leckel K, Wittig B, et al. Mycophe- 55. Meng C, Lockshin M. Pregnancy in lupus. Curr tis receiving treatment that includes cyclophospha- nolate mofetil impairs transendothelial migration of al- Opin Rheumatol. 1999;11:348-351. mide versus those treated with prednisone only. logeneic CD4 and CD8 T-cells. Transplant Proc. 1999; 56. Moroni G, Quaglini S, Banfi G, et al. Pregnancy Arthritis Rheum. 1991;34:945-950. 31:1250-1252. in lupus nephritis. Am J Kidney Dis. 2002;40:713-720. 27. Gourley MF, Austin HA III, Scott D, et al. Meth- 42. Briggs WA, Choi MJ, Scheel PJ Jr. Successful my- 57. Petri M. Prospective study of systemic lupus ery- ylprednisolone and cyclophosphamide, alone or in cophenolate mofetil treatment of glomerular disease. thematosus pregnancies. Lupus. 2004;13:688-689. combination, in patients with lupus nephritis: a ran- Am J Kidney Dis. 1998;31:213-217. 58. Rahman FZ, Rahman J, Al-Suleiman SA, Rah- domized, controlled trial. Ann Intern Med. 1996;125: 43. Dooley MA, Cosio FG, Nachman PH, et al. My- man MS. Pregnancy outcome in lupus nephropathy. 549-557. cophenolate mofetil therapy in lupus nephritis: clini- Obstet Gynecol Surv. 2004;59:754-755.