Chapter 50 Immunosuppressive Drug Therapy

Immunosuppressive agents are widely used for serious manifestations of systemic lupus erythematosus (SLE) to minimize irreversible injury and reduce toxicity from corticosteroids. In the past decade, efforts have focused on minimizing the use of cyclophosphamide (CyX) for even the most severe manifestations, particularly nephritis, by (1) using sequential therapy with CyX for induction of remission, followed by maintenance therapy with mycophenolate mofetil (MMF) or azathioprine (AZA); (2) shortening the period of induction with CyX; and (3) substituting MMF for CyX for remission induction in nephritis. The goal of substituting new biologic agents for conventional immunosuppressives for lupus nephritis (LN) has not yet been realized, although rituximab has been successfully substituted for immunosuppression in some patients with cytopenias. MMF and methotrexate (MTX) have been increasingly used for nonrenal lupus in place of AZA.

Most studies of immunosuppressive agents in lupus have been performed on nephritis. The availability of histologic examination and relatively accurate tests of renal function allow for a more accurate estimation of the response to therapy than trials in nonrenal lupus. The duration of nephritis trials (historically up to 20 years) has been significantly shortened by using primary endpoints such as complete remission after 24 weeks of induction treatment, rather than long-term preservation of renal function after many years. The duration of most current clinical trials is therefore much less than the anticipated survival of most patients with lupus.

This chapter focuses on controlled trials of widely used immunosuppressive agents, emphasizing nephritis trials, and reviews the use of the alkylating agents AZA, cyclosporine (CS), tacrolimus (TACRO), MTX, leflunomide, and MMF, and their roles in induction, as well as sequential therapies after treatment with CyX.

Alkylating Agents

Of the more than 12 alkylating agents that are currently in use, CyX, chlorambucil, and mechlorethamine (nitrogen mustard) have been most widely used to treat patients with SLE. The earliest use of alkylating agents, reported by Osborne and associates ¹ in 1947, was the topical application of nitrogen mustard in cutaneous lupus, followed in 1949 with the description by Chasis² of rapid and dramatic responses to nitrogen mustard in LN—patients with nephrotic syndrome were sometimes observed to begin diuresing within 1 day of treatment. Mechlorethamine has since been largely abandoned because of toxicity, although it is arguable that those patients with the worst of symptoms might yet benefit even now from such aggressive therapy during initiation of long-term treatment with a better-tolerated compound such as MMF.

Cyclophosphamide

CyX, despite significant toxicity, particularly gonadal failure, remains a mainstay of treatment of many patients with severe SLE. Its clinical effects, both therapeutic and toxic, vary, depending on the dose, route of administration, duration of administration, and cumulative dose.

CyX is a mechlorethamine derivative that is inactive as administered. It is metabolized by mitochondrial cytochrome P-450 enzymes in the liver to a variety of active metabolites, an increasing number of which have been shown to have both therapeutic and toxic actions. It has been proposed that various genetic polymorphisms of the P-450 enzymes are associated with the toxicity of CyX as well as the clinical response to the drug in patients with LN.

Active metabolites of CyX include 4-hydroxycyclophosphamide, aldophosphamide, phosphoramide mustard, and acrolein, all of which have differing rates of synthesis, half-lives, immunologic effects, and toxicities.³ Serum levels of these metabolites are not routinely measured; hence, dose adjustment in patients with renal or hepatic failure is largely empiric. Doses should be reduced approximately 30% in patients with a creatinine clearance of less than 30 mL/min. Some investigators have proposed stepwise reduction as renal function declines.⁴ Furthermore, CyX is incompletely cleared by dialysis; therefore the dose should be lowered for dialysis patients as well. The effect of hepatic insufficiency on CyX toxicity is incompletely understood, in part because the liver is responsible for both the production of active metabolites and their degradation. CyX is metabolized not only in the liver but also in lymphocytes and transitional epithelial cells in the bladder, which may result in local toxicity or immunosuppression or both. CyX may have toxic and/or therapeutic effects in cells that are not actively dividing, as well as in dividing cells.

CyX is well absorbed orally, and the oral and intravenous doses are equivalent. Large boluses of CyX can be administered orally, achieving comparable serum levels versus intravenous administration. Approximately 20% is excreted by the kidney, and 80% is processed by the liver.

The immunologic effects of CyX have been described. Direct effects of CyX on DNA result in cell death. These effects may occur at any stage during the cell cycle. Direct immunomodulatory effects may also occur and may be responsible for the relatively rapid onset of therapeutic efficacy of CyX (i.e., within 2 to 4 days) that is observed in some patients at a time when attrition of immunocompetent cells would not be expected. Putative mechanisms of action include alteration of macrophage function, increased production of prostaglandin E₂, alteration of gene transcription, and direct functional effects on lymphocytes. Intravenous CyX (IVC) induces suppression of T-cell activation; however, modulation of T-cell function has not been convincingly shown to play an important role in the treatment of lupus.

CyX produces dose-related lymphopenia. IVC reduces the population of cluster of differentiation 4 (CD4⁺) and cluster of differentiation 8 (CD8⁺) lymphocytes and B cells, with a more significant reduction of CD4⁺ lymphocytes and B cells during monthly therapy.⁵^–⁷ After the cessation of monthly therapy, B-cell populations rapidly return to baseline, but CD4⁺ populations remain relatively suppressed during less intensive IVC therapy, resulting in prolonged reduction of the CD4⁺/CD8⁺ ratio.⁶

Persistent reduction of the number of cluster of differentiation 19 (CD19⁺) lymphocytes 6 months after the completion of therapy has been reported,⁸ and specific reduction of B-cell function has been described.⁹ Reduction of autoantibody production has been demonstrated in patients with SLE who are treated with both oral CyX and IVC and in patients with rheumatoid arthritis (RA) who are treated with oral CyX. Despite the reduction of pathogenic autoantibody production, reduction of overall levels of immunoglobulin (Ig) G, IgA, and IgM, and IgG subclasses has not been observed in the authors’ patient population. This suggests that specific suppression of autoantibody production is a function of CyX when used in therapeutic doses and may underlie its beneficial action in patients with SLE.

Low doses of CyX in both animals and humans can heighten immune responses. This has been noted in both antibody-mediated and cell-mediated immunity, and it has been theorized that low doses of CyX could enhance antitumor immunity in humans. Low doses of CyX accelerate the production of diabetes in the nonobese, diabetic mouse. The mechanism of action of CyX in these situations is unclear, but it may represent functional alterations, as well as a depletion of lymphocyte subsets. These observations suggest that tapering the dose of CyX may produce unexpected effects, although no clinical data support the hypothesis that during tapering of immunosuppressive drugs, particularly CyX, immunosuppression is supplanted by immunostimulation.

Daily oral CyX, which has been used for induction in some recent nephritis trials, ^10,¹¹ is usually initiated at 1 to 2 mg/kg/day. The use of a standard maximum dosage of 2 mg/kg/day, with dose reduction in the presence of leukopenia (white blood cell [WBC] count of less than 3500 cells/mm³) or neutropenia (WBC count of less than 1000 cells/mm³), is a common practice. Gradually increasing the dose of CyX with the goal of producing mild leukopenia is another treatment strategy. Although these approaches have not been directly compared in a single trial, it is likely that the avoidance of leukopenia, coupled with prophylaxis against Pneumocystis carinii, may significantly reduce morbidity and mortality from infection during daily CyX therapy. Monitoring for toxicity includes weekly complete blood counts (CBCs) initially advancing to monthly when stable, urinalyses to detect hemorrhagic cystitis, and annual urine cytologic studies.

Monthly bolus IVC usually begins with a dose of 500 to 750 mg/m² body surface area administered over 1 hour in normal saline. For each subsequent monthly treatment, the dose may be increased 10% to 25% with a goal of achieving a nadir of the WBC count between 2000 and 3000 cells/mm³. Dose reduction should occur if the nadir of the CBC is a WBC count of less than 2000 cells/mm³ or a granulocyte count of less than 1000/mm³. Many physicians limit the maximum CyX dose to 1 g/m², with a downward dose adjustment in renal failure. The current evidence-based period of induction using monthly IVC is 6 months. Administering IVC (500 mg) for six doses every 2 weeks, followed by AZA, has also been shown to be effective. Additionally, sodium 2–mercaptoethane sulfonate (MESNA) totaling 80% of the IVC dose (as calculated for intravenous MESNA dosing) is routinely administered in divided doses over 12 hours. The oral dose administered in tablet form is double the intravenous dose; therefore the tablet size of 400 mg is appropriate for patients receiving approximately 1 g IVC. Despite the lack of compelling evidence that this practice is effective in patients with SLE, the use of MESNA has been associated with a very low incidence of IVC-related bladder complications in patients with lupus. Patients unable to empty the bladder completely, such as those with neurogenic bladders, may require catheter drainage or irrigation during treatment. In the authors’ institution, two patients with decreased urine output who received IVC without bladder irrigation developed severe hemorrhagic cystitis after treatment. Antiemetic medications, such as granisetron or ondansetron, are also routinely administered; the initial administration of 5 to 20 mg of dexamethasone, 25 to 50 mg of diphenhydramine, and/or 1 mg of lorazepam may also be helpful.

Hemorrhagic Cystitis and Carcinomas of the Bladder

Acrolein, which is directly toxic to the bladder, can cause hemorrhagic cystitis, a premalignant lesion identified in 50% of patients receiving CyX who eventually develop transitional cell carcinoma of the urinary tract; and bladder fibrosis, which has been reported in 5% to 34% of patients receiving daily oral CyX. Hemorrhagic cystitis, which may include either microscopic or gross hematuria (and may be life threatening), mandates permanent discontinuation of CyX, and lifetime annual urologic follow-up. The risk of bladder carcinoma associated with daily CyX therapy is dose dependent and is significantly increased after a total dose of 30 g. In a study of patients with RA, 9 of the 119 patients treated with CyX developed bladder carcinomas after 20 years; of these, 7 received more than 80 g of CyX.¹²^–¹⁵

In comparison with oral administration, monthly IVC for patients with SLE is rarely complicated by bladder injury except in patients who did not receive intravenous hydration, have urinary tract obstruction, or do not maintain adequate urine output during the 24 hours after treatment. However, IVC cannot be safely administered after bladder complications of daily CyX.

Other Malignancies

Development of malignancies after CyX administration is well described in patients with rheumatic diseases, particularly RA and granulomatosis with polyangiitis (Wegener granulomatosis). Non–urinary tract neoplastic complications of CyX include skin cancers and hematologic malignancies, as well as cervical atypia, which can be observed even in patients who have received cumulative doses of CyX of less than 10 g. In those patients who have received 80 to 120 g cumulative CyX doses, myelodysplastic syndromes are observed, characterized by monozomy-5 or monozomy-7 or both. Long-term follow-up studies by Baltus and associates ¹³ and Baker and colleagues¹⁴ of patients with RA and treated with oral CyX have established approximately 10% additional incidence of malignancy, compared with age-matched controls after a total dose of 30 g. Doses of less than 10 g are almost certainly safer; doses of 100 g or more are even more likely to produce malignancies. Radis and others¹⁵ reported a 20-year follow-up of the original study by Baker and colleagues¹⁴ and showed continued occurrence of CyX-induced malignancies; after 20 years, only 40% of the original patient population remained free of cancer.

IVC therapy of patients with lupus has not yet been associated with a statistically significant increase in solid tumors, probably because of the lower cumulative doses and the use of intravenous hydration to protect the urinary tract, although a significant increase in cervical intraepithelial neoplasia exists.¹⁶

Hematologic Toxicity

The acute effects of CyX on the bone marrow are usually benign; stem cells are resistant to CyX. After pulse therapy, the nadir of the lymphocyte count occurs on approximately day 7 to 10 and that of the granulocyte count on approximately day 10 to 14. Usually, a prompt recovery from granulocytopenia occurs after 21 to 28 days. In some patients, the recovery period may be prolonged, necessitating longer dose intervals. Prior use of alkylating agents may be associated with delayed recovery. Immunologically mediated cytopenias often improve after treatment with appropriate doses of IVC, whereas they are more likely to worsen after AZA administration.

Gastrointestinal Toxicity

IVC can be associated with short-term nausea and gastrointestinal (GI) dysmotility during treatment. Occasionally, significant hepatic toxicity may occur with the doses used for autoimmune diseases. With oral or IVC treatment, both anorexia and nausea may occur, particularly with high doses.

Pulmonary Toxicity

Pulmonary toxicity is an infrequent complication during therapy with CyX. Acute interstitial pneumonitis is the most frequently encountered pulmonary involvement of CyX therapy. Pulmonary injury as a result of CyX therapy should be suspected in patients treated with CyX during the previous 6 months before presentation who have bilateral reticular or nodular diffuse opacities on chest x-ray examination or peripheral ground-glass opacities in the upper lung fields on a computed tomographic (CT) scan of the chest. Additionally, a late-onset pneumonitis associated with fibrosis may insidiously develop after months to years of CyX therapy, even with relatively low doses. These late conditions are minimally responsive to corticosteroids, are irreversible, and usually result in terminal respiratory failure or lung transplantation.¹⁷

Gonadal Toxicity and Teratogenicity

Gonadal failure is an important side effect of CyX in both men and women. CyX is toxic to the granulosa cell and reduces serum estradiol levels and progesterone production, inhibits the maturation of oocytes, and reduces the number of ovarian follicles, ultimately resulting in ovarian failure. Studies in patients with breast cancer receiving CyX show that in women in their 40s, 30s, or 20s, the respective cumulative doses of CyX required to produce ovarian failure were 5, 9, or 20 g, respectively. Amenorrhea or premature ovarian failure is less likely to occur in patients who receive short-term (approximately 6 months) monthly IVC. Some women who develop amenorrhea from CyX subsequently resume menses and are able to bear children.

In addition to minimizing exposure, proposed approaches to fertility preservation include the following: (1) preservation of oocytes, embryos, or ovarian tissue, and (2) the use of depot gonadotropin releasing hormone analogs (GnRH-a) to suppress the metabolism of the ovaries during cytotoxic therapy with CyX. The authors of this text and others have reported favorable results of open trials of depot GnRH-a administration for ovarian protection during monthly IVC therapy. In the authors’ study,¹⁸ premature ovarian failure occurred in only 1 of 20 (5%) patients treated with depot GnRH-a, versus 6 of 20 (30%) controls matched for age (mean = 23 years) and cumulative CyX dose (mean = 12 g) (P = <0.05, McNemar test). A metaanalysis has suggested that depot GnRH-a for ovarian protection in women receiving CyX is both safe and effective, and randomized trials are ongoing.¹⁹

Azoospermia frequently occurs in men after treatment with CyX, and therefore sperm banking should be considered before therapy. In addition, testosterone supplementation has been reported but not proven to offer protection of testicular function in men during CyX therapy.

CyX and its metabolites cross the placenta and appear in breast milk. CyX is a potent teratogen that can cause severe birth defects after administration of as little as 200 mg during early pregnancy. Reported abnormalities included absence of the thumbs, absence of the great toes or all toes, palatal abnormalities, and a single coronary artery.²⁰ Because fertility is preserved in most patients with lupus, highly effective contraceptive techniques, such as intrauterine devices (IUDs), oral contraceptives, or injected progestins in appropriately selected patients, should be strongly considered. The use of CyX in life-threatening lupus during late pregnancy is controversial but may be appropriate in special circumstances because fetal loss is extremely likely when severe maternal flares are uncontrolled. Major CyX-induced toxicities are believed to occur during the first half of pregnancy.

Infections

The risk of bacterial infections and herpes zoster is increased with CyX therapy, as is the risk of Pneumocystis carinii pneumonia (PCP). This risk is further increased when 20 mg/day or more of prednisone or bolus corticosteroids are administered concomitantly. The authors of this text use prophylaxis against PCP, administering either trimethoprim-sulfamethoxazole three times weekly (in patients with lupus already known to tolerate this drug), dapsone (100 mg/day) in patients without glucose-6-phosphate dehydrogenase deficiency, or atovaquone (1500 mg/day). Preexisting or treatment-related hypogammaglobulinemia should be considered in patients who develop infections. Granulocyte colony–stimulating factor (GCSF) can potentially help decrease the morbidity and mortality associated with accidental severe drug-induced leukopenia during CyX therapy. Although the possibility of inducing lupus flares is a concern, the authors of this text do not use this compound except when the possibility of infection is present. Finally, the syndrome of inappropriate antidiuretic hormone (SIADH) can also occur after administering IVC.

Clinical Trials Administering Cyclophosphamide for Lupus Nephritis

In this discussion, trials of monthly bolus CyX are emphasized, but many modified regimens have been proposed, such as weekly, biweekly, or a once-every-3-weeks bolus CyX given intravenously, and boluses of CyX given orally. These regimens have been reported to be safe and effective in small series. Results of controlled trials are summarized in eTables 50-1 and 50-2. eTable 50-3 summarizes the results of the controlled trials of sequential therapy using CyX for induction of remission.

eTABLE 50-1 Controlled Trials of Cyclophosphamide and/or Azathioprine in the Treatment of Lupus Nephritis

STUDY	PATIENTS (n)	RESULTS
Fries and associates, 1973²³	14	P; then CyX alone
Garancis and Piering, 1973¹³⁴	22	P plus CyX; then P and AZA
Donadio and associates 1972, 1974^60,⁶¹	26	More recurrences with P; P versus P plus CyX results in survival with patients on dialysis
Ginzler and associates, 1976⁶⁴	14	P plus AZA versus P plus CyX
Balow and associates, 1984²¹	111	P plus IVC; then P and AZA and CyX; then P and AZA; then P alone
Boumpas and associates, 1992²²	65	IVC for 30 months; then IVCX for 6 months; then MP alone
Sesso and associates, 1994¹³⁵	29	IVC or MP (both were unsuccessful)
Gourley and associates, 1996¹³⁶	80	IVC; then MP; trend for IVC plus MP and then IVC

AZA, Azathioprine; CyX, cyclophosphamide; IVC, intravenous cyclophosphamide; MP, bolus methylprednisolone; P, prednisone.

eTABLE 50-2 Controlled Trials Including Bolus Methylprednisolone, Cyclosporin, or Intravenous Immunoglobulin in Systemic Lupus Erythematosus

AUTHOR	THERAPEUTIC ARMS	RESULTS
Boumpas and associates, 1992²²	IVC short-term IVC long-term Bolus MP	Long-term IVC; then short-term bolus MP < either IVC
Sesso and associates, 1994¹³⁵	IVC Bolus MP	Equivalent outcome; 38% renal failure in 15 months
Gourley and associates, 1996¹³⁶	IVC Bolus MP IVC and bolus MP	IVC; then bolus MP Trend for IVC and bolus MP; then IVC
Fu and associates, 1998⁸⁰	CS without P POC and P	Similar renal outcome; 38% renal failure in 15 months
Boletis and associates, 1999¹³⁷	IVC IVIG	Equivalent short-term results

CS, Cyclosporine; IVC, intravenous cyclophosphamide; IVIG, intravenous immunoglobulin; MP, methylprednisolone; P, prednisone; POC, oral cyclophosphamide.

eTABLE 50-3 Controlled Trials Using Sequential Therapy with Azathioprine or Mycophenolate Mofetil After Induction with Cyclophosphamide in the Treatment of Lupus Nephritis

Study	Patients (n)	Results
Chan and colleagues, 2005¹⁰	42	MMF; then AZA and CyX
Houssiau and colleagues, 2002³⁵	90	Low-dose IVC and AZA, versus high-dose IVC and AZA
Contreras and colleagues, 2004²⁶	59	MMF and AZA; less toxic than quarterly IVC and MP
Yee and colleagues, 2004¹¹	32	IVC plus bolus MP, then followed by AZA; less toxic than POC and bolus MP, then followed by AZA

AZA, Azathioprine; CyX, cyclophosphamide; IVC, intravenous cyclophosphamide; MP, methylprednisolone; POC, oral cyclophosphamide.

In a seminal 20-year clinical trial at the National Institutes of Health (NIH), patients with proliferative nephritis received either prednisone alone or prednisone plus one of the following: AZA (2 mg/kg/day), AZA (1 mg/kg/day) plus CyX (1 mg/kg/day), CyX (2 mg/kg/day), or bolus IVC for approximately 2 to 4 years.²¹ Several key findings were revealed: (a) Differences in progression to renal failure were not apparent until more than 5 years had elapsed. After 10 years, however, significant differences in renal survival became apparent, favoring any regimen that included CyX over the administration of prednisone alone. (b) Patients treated with either prednisone alone or with oral CyX had higher death rates than those groups given IVC or AZA plus CyX, which was likely because of the toxicity of daily CyX and the ineffectiveness of prednisone. (c) Oral AZA (1 mg/kg/day) plus CyX (1 mg/kg/day) was equivalent to IVC in terms of preventing renal failure or survival. (d) In serial biopsies, progression of chronic change initially occurred in all patients. Patients who were treated with immunosuppressive agents appeared to stabilize after an initial period of scarring; patients who were treated with prednisone had progressive scarring.

Intravenous Bolus Cyclophosphamide for the Treatment of Lupus Nephritis

Monthly bolus IVC was first described as a treatment for lupus in the 1980s ⁶ and for many years was the standard of care for treating severe lupus. Extensive studies have highlighted the issues discussed in the following text.

Relationship of Efficacy and Toxicity to the Effective Dose of Intravenous Cyclophosphamide

In a retrospective study at the NIH,³ 62 patients with proliferative LN treated with CyX were genotyped for common variant alleles of the P-450 enzyme. Homozygosity or heterozygosity for a particular variant allele (CYP2C19*2) predicted not only lower rates of ovarian toxicity, but also a worse clinical response, including an increased risk of end-stage renal disease and the doubling of serum creatinine, suggesting that efficacy and toxicity were both related to the effective dose given. This study provides the best evidence that higher effective doses of monthly IVC increase both therapeutic response and toxicity.

Advantage of Maintenance Immunosuppression Therapy after Intravenous Cyclophosphamide Induction Therapy

Boumpas and associates ²² confirmed early observations that patients treated with monthly IVC for only 6 months had a high rate of subsequent flares. Seven monthly pulses of IVC, followed by an every-3-month IVC maintenance regimen for 2 years, resulted in significantly fewer flares and fewer doublings of serum creatinine, compared with seven monthly pulses of CyX without maintenance pulses (Figure 50-1). The longer IVC regimens, which became generally accepted as the standard of care for LN, resulted in more toxicity, particularly ovarian failure. It has since been suggested that patients who achieve a complete remission after 6 months may have a lower risk of flare. Nonetheless, prolonged immunosuppression (currently using sequential therapy if IVC is initially used) remains appropriate for the majority of patients.

FIGURE 50-1 Treatment of severe lupus nephritis. A, Probability of not doubling serum creatinine levels in 65 patients with severe active lupus nephritis randomly assigned to receive intravenous methylprednisolone (MP) (1.0 g/m² monthly for 6 months); short-course intravenous cyclophosphamide (CY-S) (0.5 to 1.0 g/m² monthly for 6 months); or long-course intravenous cyclophosphamide (CY-L) (0.5 to 1.0 g/m² monthly for 6 months), followed by quarterly infusions for 24 months (Gehan test comparing CY-L with MP, P = .037). B, Probability of no exacerbation of lupus activity on completion of monthly pulses in groups randomly assigned to receive CY-L and CY-S (Gehan test, P = .006). Numbers of patients that remain at risk at various times are shown in brackets along the abscissa.

(From Boumpas DT, Austin HA, Vaughn EM, et al: Controlled trial of pulse methylprednisolone versus two regimens of pulse cyclophosphamide in severe lupus nephritis. Lancet 340:741–745, 1992. Used with permission.)

Concomitant Daily Corticosteroids

All published IVC trials have administered daily oral corticosteroids during induction usually beginning with prednisone (1 mg/kg/day) or equivalent . Administering CyX without corticosteroids for LN (e.g., because of a patient’s refusal to take prednisone) is not evidence based and, in the opinion of the authors of this text, unnecessarily exposes patients to a toxic drug using an unproven regimen. One small trial, by Fries and others,²³ addressed this issue in 1973 and compared oral CyX alone with prednisone alone for a mean of 9 weeks in 14 patients with lupus and 10 with nephritis. CyX without prednisone failed to control either minor or major manifestations, despite the development of leukopenia and significant additional toxicity. Patients who were changed to prednisone from CyX fared better. These results suggest that CyX and prednisone may act synergistically, and CyX without prednisone may be less effective.

Combining Bolus Methylprednisolone with Intravenous Cyclophosphamide

The initial treatment of LN with bolus corticosteroids (e.g., methylprednisolone [MP] [1000 mg/day] for 1 to 3 days) is widely used (including by the authors of this text), especially in patients with severe disease such as crescentic nephritis or acute renal failure. Serial administration of both agents in combination is supported by an NIH study ²⁴ that compared monthly bolus IVC, monthly bolus MP, and the combination of monthly IVC plus bolus MP for LN. During follow-up for a median of 11 years, an intention-to-treat survival analysis revealed the likelihood of treatment failure to be significantly lower in the groups who received CyX (P = 0.04) and combination therapy (P = 0.002) than in the group who received MP alone. Furthermore, the proportion of patients who had doubling of serum creatinine levels was significantly lower in the combination group than in the CyX group. No additional adverse events occurred in the group treated with the combination therapy versus CyX alone, except that patients who received MP pulses had more osteonecrosis.

Racial Differences in Response to Intravenous Cyclophosphamide

Several studies have suggested poorer responses to IVC in African-American versus Caucasian patients. For example, Dooley and associates ²⁵ described poorer renal survival in African Americans during the initial period of monthly IVC administration, with several patients rapidly progressing to renal failure, and further disparity appearing during long-term follow-up studies with renal survival after 5 years at 94.5% for Caucasians and 57% for African Americans. These observations are further supported by the results of trials of induction with MMF versus IVC described in the text that follows.

Sequential Therapy for Lupus Nephritis

In a 2004 study by Contreras and colleagues,²⁶ 59 patients with LN and the World Health Organization (WHO) class III, IV, or Vb nephritis were treated with monthly IVC for seven doses. Patients were then randomized to maintenance dosing for 1 to 3 years after the initial therapy consisting of quarterly IVC, MMF, or AZA. During maintenance therapy, four deaths occurred in the CyX group and one death was reported in the MMF group; chronic renal failure occurred in three patients in the CyX group, one patient in the MMF group, and one patient in the AZA group. The 72-month, event-free survival, which is defined as no death or progression to hemodialysis, was higher in groups treated with MMF (P = <0.05) and AZA (P = <0.01) versus CyX. The relapse-free survival was also higher in the MMF versus the CyX groups (P = <0.02) (Figure 50-2).

FIGURE 50-2 Mycophenolate mofetil for the treatment of lupus nephritis.

(From Contreras G, Pardo V, Leclercq B, et al: Sequential therapies for proliferative lupus nephritis. N Engl J Med 350:971–980, 2004.)

More recently, patients participating in the Aspreva Lupus Management Study (ALMS) were entered into a maintenance phase study that compared maintenance treatment with either MMF or AZA after patients had completed a randomized trial of induction therapy with IVC versus MMF for LN (see the description provided later in this chapter). MMF was superior to IVC in maintaining remissions (Figure 50-3). The data are consistent with the differences observed in the Contreras trial, which showed a statistically significant advantage of MMF but not AZA over IVC administered every 3 months for maintenance. Interestingly, patients who received MMF for maintenance tended to fare better if they had been randomized to receive IVC rather than MMF as induction therapy in the preceding part of the trial.²⁷

FIGURE 50-3 Kaplan-Meier curves for time-to-treatment failure and time-to-renal flare.

(From Wofsy D, Appel GB, Dooley MA, et al: Aspreva Lupus Management Study maintenance results. Lupus 19:S27, 2010.)

The intention-to-treat population was made up of 227 patients, of whom 116 were given MMF and 111 were given AZA. Figure 50-3 shows the time to treatment failure (see part A) and the time to renal flare from reference (see part B).²⁷

Houssiau and associates,²⁸ on the other hand, did not identify a difference between MMF and AZA as maintenance therapy for LN. Several additional trials have administered daily oral CyX, followed by AZA. For example, Chan and others²⁹ studied 42 patients with somewhat active diffuse proliferative glomerulonephritis who were randomized to daily CyX for 6 months, followed by AZA versus high-dose MMF for 12 months and then by low-dose MMF for 6 months. At long-term follow-up in the MMF group, 81% experienced a complete remission and 14% experienced a partial remission. In the group randomized to CyX and AZA, 76% experienced a complete remission and 14% experienced a partial remission. Oral CyX appeared to be more toxic than MMF.

As previously noted, the EURO-Lupus study ²⁸ compared high-dose versus low-dose CyX in patients with lupus and proliferative nephritis; they were then switched to maintenance therapy with AZA.

Another group that used sequential therapy, the European League Against Rheumatism (EULAR), conducted a randomized controlled trial of pulse CyX and MP versus continuous CyX and prednisolone, followed by AZA and prednisolone in LN ¹¹; this study suggested that no significant differences were observed between these two regimens. However, enrollment was difficult, and only 32 patients were treated. The authors encountered cytopenias in the group who received oral CyX (2 mg/kg) and concluded, “…the initial dose of 2 mg/kg oral CyX was felt by the investigators to be too toxic to persist with. The intermittent intravenous pulse regimen appears to be better tolerated than oral continuous treatment, with less severe adverse effects.”¹¹

Induction Therapy: Comparisons of Intravenous Cyclophosphamide with Other Agents

Intravenous Cyclophosphamide versus Mycophenolate Mofetil

Considerable excitement has been generated by randomized trials that in aggregate have suggested that MMF is either equivalent or superior to IVC as induction therapy for mild to moderately severe LN. Considerable variation in comparative responses to these agents has been demonstrated, both in different racial and ethnic groups and in regions of the world, emphasizing the continuing need to individualize therapy. The following randomized trials have been selected from a larger number to illustrate key points:

Chan and colleagues ²⁹ compared induction with MMF (2 g/day) with long-term MMF maintenance versus daily oral CyX for 6 months, followed by AZA. Patients were from China and had overall moderately active disease. In the MMF group, 81% experienced a complete remission, and 14% experienced partial remission. In the group randomized to CyX followed by AZA, 76% experienced a complete remission and 14% experienced a partial remission. At long-term follow-up, comparable preservation of renal function and reduction of proteinuria were observed.

Ginzler and associates,³⁰ in a trial that included a high proportion of African-American patients, compared MMF (3 g/day) versus monthly IVC. The higher MMF dose was chosen because of a concern that MMF at 2 g/day was less effective for African Americans than for Caucasians in allogeneic renal transplantation. Patients were required to have creatinine clearances greater than 30 mL/min and serum creatinine levels less than 3.0 mg/dL; overall, they had moderate to very active disease. Patients who did not respond to one regimen were allowed to cross over to the other. At 6 months, the primary endpoint, complete remission, was achieved at a higher rate with MMF than with IVC; however, after 6 months the mean serum creatinine levels and urinary protein excretion were identical when all patients in both groups were considered. A trend revealed that African-American patients responded better to MMF.

Recently, the ALMS trial ³¹ randomized 370 patients with LN to IVC versus MMF. In contrast to the previously mentioned Ginzler trial,³⁰ the overall outcomes in terms of both achievement of remission and serious adverse events and mortality were no different in the two groups. However, as detailed in the following text, a significantly higher response rate to MMF than to was achieved in patients of African-American and mestizo descent and individuals of Hispanic origin.³²

The concern that MMF might not be as effective as IVC in preventing the progression of irreversible renal injury in serial biopsies is addressed in two trials. Ong ³³ compared renal biopsies before and after 6 months of MMF (2 g/day) versus IVC and found comparable reduction of NIH activity scores with somewhat greater increases in chronicity scores in the IVC group than in the MMF group.

Hu ³⁴ compared serial renal biopsies in 25 patients treated with MMF versus IVC and showed comparable reductions in activity indices and slight increases in chronicity indices in both groups.

The possibility that brief administration of CyX might be effective in inducing remission was suggested by the Euro-Lupus Nephritis Trial.³⁵ In this study, 90 patients with proliferative glomerulonephritis were randomized to either high-dose CyX (six monthly and two quarterly pulses, increased according to their WBC nadir) or low-dose CyX (six doses of 500 mg CyX every 2 weeks). Maintenance therapy was with AZA. Renal remission was achieved in 71% of patients receiving low-dose CyX versus 54% of patients who were given high-dose CyX; renal flares were observed in 27% of patients in the low-dose group and 29% of those in the high-dose group.

Pulse Cyclophosphamide versus Intravenous Immunoglobulin

Boletis and others ³⁶ compared IVC with 10 immunoglobulin infusions and found equivalent results over an 18-month period. Proteinuria actually increased slightly in the IVC group.

Oral Cyclophosphamide

Long-term daily CyX has been abandoned for LN, but daily oral CyX continues to be used for induction of remission in some centers.

Daily Oral CyX for Induction

As previously noted, Chan and colleagues ³⁷ randomized 42 patients to either daily CyX for 6 months, followed by AZA for 6 months, or high-dose MMF for 6 months, followed by low-dose MMF. Complete remissions occurred in 81% of those in the MMF group versus 76% of those in the oral CyX group, suggesting that daily CyX is effective for remission induction although it unfortunately results in three times higher cumulative CyX exposure than IVC exposure for a comparable period. In a second study, long-term outcomes in a cohort of patients with lupus and diffuse proliferative nephritis were studied; this cohort received sequential therapy with oral CyX and prednisolone for induction, followed by AZA for maintenance therapy. Of the 66 patients included in the study, 82.4% achieved complete remission, of whom 39.1% experienced relapse during the follow-up period of 91.7 months, ±36.7 months. No end-stage renal failure or death occurred among the patients, although three patients (4.4%) had doubling of baseline creatinine.

Guidelines for Treating Lupus Nephritis

A committee sponsored by the American College of Rheumatology (ACR) has formulated guidelines for the treatment of LN, which should have already been published when this text is available. Either CyX or MMF is considered acceptable first-line therapy. The use of initial bolus MP is encouraged for patients initiating therapy with either CyX or MMF. In the authors’ experience, bolus MP will more likely be used in clinical practice in conjunction with IVC rather than with MMF, although the rationale for aggressively initiating corticosteroid treatment is arguably stronger in patients given MMF, since the dose of MMF is gradually increased over 1 to 2 weeks.

Intravenous Cyclophosphamide in Nonrenal Lupus

In general, severe nonrenal manifestations of lupus that result from immune complex disease respond more rapidly and completely to IVC plus corticosteroids than to corticosteroids alone. The required treatment duration and size of the individual IVC boluses vary with different disease manifestations. For example, transverse myelitis may respond to a shorter course of treatment than severe nephritis, and immune-mediated thrombocytopenia may respond to lower than usual doses. During treatment of severe lupus with IVC, improvement of minor manifestations, including constitutional symptoms, fevers, arthralgias, rash, pleurisy, and serologic abnormalities, as well as reduced prednisone requirements, usually occur within 3 months.

Analysis of the recent ALMS trial suggested that nonrenal disease manifestations, in general, responded well to either IVC or MMF and that no significant differences were reported in the responses to one or the other.³⁸

Neuropsychiatric Lupus

No clear guidelines exist regarding therapy for neuropsychiatric lupus (NP-SLE) with various modalities, including corticosteroids, CyX, and/or anticoagulation. Distinction among various primary pathogenic mechanisms, such as immune complex–mediated vasculitis, antibody-mediated cerebral injury, microangiopathy, and thrombosis, and secondary causes, such as atherosclerosis or infection, is notoriously difficult and is further complicated by the multifactorial etiologic origin of many events. In many cases, skilled physicians must make a “seat-of-the-pants” decision regarding the use of immunosuppression, anticoagulation, or both, based on clinical, serologic, or magnetic resonance imaging evidence, unless there is biopsy evidence of tissue inflammation or cerebrospinal fluid pleocytosis. In many series, treatment decisions have been made (apparently appropriately) on the basis of clinical judgment rather than on specific inclusion criteria.

Active, steroid-refractory cerebral lupus that is adjudged to be secondary to immunologically mediated injury has responded well to IVC with or without bolus MP in most cases. Anticoagulation has been simultaneously used when distinguishing thrombotic from inflammatory disease has been impossible or to rule out the possibility that vascular inflammation is contributing to the development of thrombosis. Neither the presence of antiphospholipid antibodies nor the involvement of one or more large vessels rules out the use of immunosuppression as opposed to (or in addition to) anticoagulation. Boumpas and associates ³⁹ treated nine patients with monthly doses of IVC, three of whom had transverse myelitis and five of whom had focal neurologic findings, seizures, or both. The duration of symptoms ranged from 3 to 45 days. All nine patients had findings suggesting an inflammatory process, including anti-DNA antibodies, and five had cerebrospinal fluid pleocytosis. Five of these patients concomitantly had antiphospholipid antibodies. All patients recovered either partially or completely. These observations suggest that in selected patients who have antiphospholipid antibodies that may not be the major cause of their events, IVC administration is associated with clinical improvement.

Other series of IVC in NP-SLE report favorable results. Neuwelt and others ⁴⁰ retrospectively reviewed 31 patients with NP-SLE who were treated with IVC and in whom a variety of prior therapies had failed, including corticosteroids, warfarin, chlorambucil, and AZA. Indications included organic brain syndrome in 55% of patients, strokes in 35%, peripheral mononeuropathies in 32%, seizures in 29%, and transverse myelitis in 16%. Patients with anticardiolipin antibodies were treated with warfarin. Treatment regimens varied from low-to-high doses of IVC, and plasmapheresis was added in some patients when they appeared not to improve after IVC. Overall, 61% of patients were reported to improve, of whom 26% were not initially improved after 9 months of therapy and appeared to respond to the addition of plasmapheresis. The failure rate for patients with organic brain syndrome was 83%, compared with 37% for other indications. Malaviya and associates⁴¹ treated 14 patients with a variety of focal and diffuse neurologic deficits. All patients except the two with seizures stabilized or improved.

Numerous studies have demonstrated improvement of transverse myelitis with IVC, with or without bolus corticosteroids. Because of the catastrophic nature of transverse myelitis and the importance of prompt therapeutic intervention, it may be appropriate to have a very low threshold for prompt institution of IVC when this syndrome appears suddenly, with or without concomitant high-dose (1 g/m²) MP. In the authors’ institution, prompt use of bolus CyX for transverse myelitis has been associated with the preservation of the ability to ambulate in most patients, although many have continued to have neurogenic bladders.

Ten patients with bilateral corticosteroid-refractory optic neuritis and severe visual compromise were treated with bolus IVC for 6 months.¹³³ Of the 20 patients, the eyes recovered completely in 10 patients and partially in 6, but the eyes of 4 patients did not recover.

Baca ⁴² treated seven children with NP-SLE (including seizures, focal neurologic deficits, transverse myelitis, and organic brain syndromes) with monthly bolus CyX combined with three initial boluses of MP (30 mg/kg). Three patients had anticardiolipin antibodies but did not undergo anticoagulation. Six patients recovered completely, and one had a minor residual deficit.

Neuromyelitis optica (NMO), which is characterized by antibodies to aquaporin-4, transverse myelitis and optic neuritis, occurs with increased frequency in patients with SLE. In the authors’ lupus cohort, transverse myelitis occurred in 23 of 856 patients, and NMO and NMO-spectrum disorders occurred in one third of these patients (unpublished data). Although the neurologic literature encourages the use of AZA or rituximab or both in patients without SLE, the authors of this text have chosen to use IVC plus bolus corticosteroids in their patients, with success comparable to their other patients with transverse myelitis. The increased risk of recurrence in NMO encourages prolonged immunosuppression.

Only gold members can continue reading. Log In or Register to continue