Management Of Systemic Lupus ... Health Article

Two groups of cytotoxic drugs have been used in large numbers of patients with severe SLE—purine antagonists and alkylating agents. There is increasing use of MTX and of the T cell-specific immunosuppressive agent mycophenolate mofetil, and new cytotoxic drugs are being studied. The first cytotoxic agent used in SLE was nitrogen mustard. ³⁷ The two drugs used most frequently are azathioprine (Imuran) and cyclophosphamide (Cytoxan); of the two, cyclophosphamide is more effective and more toxic. Cyclophosphamide suppresses both humoral and cellular immune responses more effectively than azathioprine, and in contrast to azathioprine is not cell cycle-dependent. Mycophenolate mofetil is coming into wide usage. Doses of these drugs, their side effects, and guidelines to combination therapies are reviewed in Table 76-6.

With regard to azathioprine in doses of 2 to 3 mg/kg/day, short-term studies (1 to 2 years in duration) have failed to show a better outcome in patients with severe SLE treated with glucocorticoids plus azathioprine compared with those treated with glucocorticoids alone. ³⁸ In studies that follow patients for 5 to 15 years, however, individuals who receive combination therapy have fewer chronic changes on renal biopsy, better renal function, fewer severe disease flares, and lower glucocorticoid requirements. ^{39, 40, 41} Significant adverse side effects of chronic azathioprine therapy include increased rates of infection with opportunistic organisms (especially herpes zoster), ovarian failure, bone marrow suppression (especially leukopenia), hepatic damage, and increased susceptibility to malignancies. ^{41, 42, 43} For patients with severe SLE, some experts use daily or pulse glucocorticoid regimens and a cytotoxic drug (azathioprine in slowly progressive disease; cyclophosphamide or mycophenolate mofetil in rapidly progressive disease).

In general, the physician should allow 6 to 12 months for azathioprine to be effective. After disease is well controlled and glucocorticoid doses are tapered to the lowest possible maintenance levels, doses of azathioprine should be slowly tapered and discontinued if disease is well controlled for a year or more.

The benefits of cyclophosphamide therapy have been somewhat controversial due to conflicting reports regarding its efficacy in severe SLE and its numerous undesirable side effects. Some authorities have suggested that it is effective over the short term in only a small proportion of patients with severe lupus nephritis ^{44, 45} and that it does not change long-term outcome. Furthermore, many patients prefer azathioprine because it less toxic, even when informed that cyclophosphamide may be more efficacious. ⁴⁶ Other data ^{31, 33, 34, 39, 40, 41} suggest that cyclophosphamide is effective in most patients with severe lupus nephritis over both the short and the long term; many experts believe that it should be included in initial therapeutic regimens in most SLE patients with severe nephritis (or other life-threatening, rapidly progressive organ involvement). Most experts agree that a high rate of relapse of nephritis (in patients with diffuse proliferative nephritis) occurs during the 5 to 10 years after initiation of combined glucocorticoid and cytotoxic regimens. In one recent long-term study, ⁴⁷ the cumulative risk of renal flare in patients treated with IV cyclophosphamide was 45 percent at 10 years, and 27 percent of patients progressed to renal failure. This is similar to long-term outcomes reported previously. ^{48, 49} I have reviewed this work and formed the following opinions. When cyclophosphamide is given intravenously once a month for 6 months and then discontinued, 50 to 80 percent of patients can be expected to improve. That improvement is lost in more than half during the subsequent 6 to 24 months. ^{31, 33, 47, 48, 49} In contrast, if IV cyclophosphamide is given monthly for 6 months, then at longer intervals for an additional 12 to 24 months, numbers of disease flares and preservation of renal function are better than in groups treated with glucocorticoids alone. ³¹ After cyclophosphamide is discontinued, about 25 percent of patients experience flares of SLE within 5 years, and about 50 percent within 10 years. ^{47, 49}

Adverse reactions to cyclophosphamide are more common, however, with longer duration of therapy. For example, irreversible ovarian failure occurs in a much higher proportion of women on a 30-month regimen of IV cyclophosphamide (39 percent) than on a 6-month regimen (12 percent). In fact, 100 percent of women over the age of 30 years had ovarian failure in one study. ⁵⁰ Some data ⁵¹ suggest that serious infections are more frequent in patients receiving both cyclophosphamide and steroids (49 percent) compared to steroids alone (29 percent), particularly if the nadir of total leukocyte count after cyclophosphamide doses is below 3000 cells/dl. In my experience, introducing granulocyte colony-stimulating factor (G-CSF) treatments at the time of this nadir, particularly if an infection is present, will prevent infection or enhance antibiotic therapies without activating disease. Others have reported that administration of G-CSF to patients with SLE is relatively safe: There is a small chance of aggravating a disease flare. ⁵² In many patients, lupus nephritis is a process that progresses over a period of many years. Acute changes can be reversed by aggressive immunosuppression with glucocorticoids or glucocorticoids plus cytotoxics. Scarring may progress inexorably in some individuals, however, and that process probably occurs more rapidly in patients treated with steroids alone, but it does occur in those with combination therapy as well. It may be useful and more acceptable to patients to induce improvement with glucocorticoid plus cyclophosphamide than to maintain the improvement with a safer regimen, such as low-dose daily or alternate-day steroids plus daily azathioprine, mycophenolate mofetil, or IV cyclophosphamide at long intervals (every 2 or 3 months). Data supporting this approach are appearing in the literature, and a prospective controlled trial comparing continuation of therapy (after administration of monthly IV cyclophosphamide) with azathioprine versus quarterly IV cyclophosphamide is in progress. ^{53, 54, 55} Recent studies have challenged the current community standard for treatment of severe lupus nephritis with cyclophosphamide among rheumatologists in the United States. Neither route of administration nor dose are known to be ideal. The six monthly IV cyclosphosphamide doses usually used (0.5 to 1 g/m ² body surface area, beginning at the lower end and escalating until leukopenia occurs), followed by two quarterly pulses, was compared to lower doses in a recent prospective randomized study conducted in Europe. ⁵³ The comparison group received a fixed dose of 500 mg of cyclophosphamide intravenously once every 2 weeks for six doses. Both groups, after conclusion of cyclophosphamide, were maintained on azathioprine. At the beginning of the trial all patients in both groups received IV pulses of methylprednisolone, followed by 0.5 mg/kg of prednisolone daily. After a median follow-up of 41 months, the two groups had similar rates of renal remission (71 percent in the low-dose group and 54 percent in the high-dose group), and renal flares occurred in 27 percent of the low-dose and 29 percent of the high-dose groups. Severe infections were twice as frequent in the high-dose group, but differences did not reach statistical significance. These data suggested that we can administer lower doses of cyclophosphamide than are standard, for shorter periods of time, with lower glucocorticoid doses. One cautionary note should be added: The vast majority of patients were Caucasian; data cannot be compared directly to most trials in SLE patients conducted in the United States.

IV cyclophosphamide is effective in some patients with serious extrarenal manifestations of SLE, including diffuse CNS disease, thrombocytopenia, and interstitial pulmonary inflammation. ^{56, 57} If in the initial suppression of severe lupus, intermittent IV cyclophosphamide and daily glucocorticoids are not effective, alternative strategies (added to daily glucocorticoids) include the following: 1) monthly IV pulses of methylprednisolone (Solu-Medrol, discussed earlier); 2) daily doses of oral cyclophosphamide; 3) daily oral doses of both cyclophosphamide and azathioprine; 4) mycophenolate mofetil; 5) cyclosporine; and 6) the experimental therapies plasmapheresis or immunoablative doses of cyclophosphamide, with or without stem cell infusion.

Several regimens combine daily oral cytotoxic drugs and glucocorticoids. Combination therapies studied in patients with lupus nephritis include glucocorticoids plus 1) daily oral cyclophosphamide, 2 to 3 mg/kg/day; 2) daily oral cyclophosphamide, 1.5 to 2.5 mg/kg/day, plus daily oral azathioprine, 1.5 to 2.5 mg/kg/day ^{39, 40, 41, 48} ; and 3) daily oral mycophenolate mofetil. ^{55, 55a, 55b} The oral regimens have the advantages of convenience and daily immunosuppression of severe disease. Oral administration of cyclophosphamide, however, carries a substantial risk of urinary bladder toxicity (hemorrhagic cystitis, sclerosing chronic cystitis, and carcinoma of the bladder). That risk is much lower with IV administration, especially if cyclophosphamide doses are accompanied by administration of mesna, which inactivates the oxazaphosphorine metabolite of cyclophosphamide that irritates the bladder. ⁵⁸ (I give 1 mg of IV mesna for every 1 mg of IV cyclophosphamide, half of the mesna just before and half 2 hours after the cyclophosphamide infusion.) IV pulse cyclophosphamide is effective in most patients with severe disease. In my experience, daily oral administration may be more effective than intermittent high-dose pulses (and more toxic) in some patients, and combination therapy of oral daily glucocorticoids plus azathioprine plus cyclophosphamide is effective in some patients who fail standard glucocorticoids-plus-pulse-cyclophosphamide regimens. There are no prospective controlled studies in human SLE that support these views. Addition of mycophenolate can also be effective, and there are increasing published data supporting this approach. ^{55, 55a, 55b}

Several questions about therapy with IV or oral cyclophosphamide remain unanswered. The minimal effective dose is unknown and probably varies from patient to patient. Some authorities recommend that the dose be increased until the leukocyte nadir 10 to 14 days later is below 3500 cells/mm; there is no evidence that such a nadir is required for clinical response. Another problem is when and how to stop this therapy, as discussed earlier. My approach is to give monthly doses of the highest cyclophosphamide dose tolerated until response is achieved, then either continue IV cyclophosphamide at longer intervals (every 2 months instead of every 1 month, then, if response is sustained, every 3 months), or to add azathioprine or mycophenolate mofetil, continuing the treatments for a total of 2 years. The methods of using cytotoxic drugs are reviewed in Table 76-6.

Most patients treated with cyclophosphamide experience adverse side effects. ^{39, 42, 43, 44, 50, 51, 53, 54, 55} The most important adverse side effect is disabling or life-threatening infections. The most common infections are herpes zoster and staphylococcal, gram-negative bacterial, Candida, and Pneumocystis carinii infections. These infections must be carefully sought if appropriate symptoms are present, because, (with the exception of herpes zoster) they are often fatal. Other toxicities include bone marrow suppression (especially leukopenia) and increased malignancies. Infertility is common in both men and women (banking of sperm should be considered before cyclophosphamide regimens are instituted in men) and may be reversible if detected early in young patients so the drug can be stopped. Loss of ovarian function may also contribute to osteopenia and to accelerated atherosclerotic disease in women receiving chronic glucocorticoid therapy. Hair loss is common, as are gastrointestinal side effects, including nausea, diarrhea, and dyspepsia. Some patients experience malaise and fatigue, which improve after cytotoxic drugs are discontinued (see Table 76-6 for a review of adverse effects).

MTX is reported to be effective in SLE arthritis, myositis, vasculitis, rash, serositis, and nephritis in some patients. ^{59, 60, 61} Most studies of MTX therapy in SLE have evaluated mainly cutaneous and joint involvement. The most recent prospective, randomized, placebo-controlled trial ⁶⁰ showed lower pain scores, lower numbers of cutaneous lesions, higher complement levels, lower systemic lupus erythematosus disease activity index (SLEDAI) scores of disease activity and lower prednisone doses in the patients receiving MTX compared to the placebo control group. Two studies evaluating response of lupus nephritis reached different conclusions regarding efficacy. Currently, MTX has a place in the treatment of patients with relatively mild disease, particularly cutaneous and articular, who have not achieved adequate responses to NSAIDs, antimalarials, and low-dose glucocorticoids.

Mycophenolate mofetil, widely used to prevent rejection of allografts, has been successful in treatment of murine lupus. ⁶² Small open trials and two recent prospective controlled trial in patients have suggested that many individuals respond. ^{55, 55a, 55b, 62, 63, 64} This drug is a selective, noncompetitive reversible inhibitor of inosine monophosphate dehydrogenase, which blocks de novo guanosine nucleotide incorporation into DNA. T and B lymphocytes depend on de novo synthesis of purines and lack salvage pathways present in other cell types. Therefore, mycophenolate is relatively specific for lymphocytes; therapy should spare ovarian and testicular function (not proven). Mycophenolate suppresses antibody formation and glycosylation of adhesion molecules. Adverse side effects are similar to those of azathioprine, with a higher incidence of nausea and diarrhea and a rare association with gastrointestinal ulceration and perforation. Infection (including herpes zoster), leukopenia, lymphoma, and nonmelanoma carcinoma are more common than with placebo ⁶⁴ (similar to azathioprine).

At a dose of 2 to 3 g/day, mycophenolate is superior to azathioprine (1 to 2 mg/kg/day) in preventing graft rejection. A prospective, randomized trial ⁵⁵ comparing daily oral cyclophoshamide given for 6 months (followed by daily azathioprine) to daily oral mycophenolate mofetil given for 12 months in patients with diffuse proliferative glomerulonephritis showed complete or partial remission in 90 percent of the cyclophophamide group and 95 percent of the mycophenolate group over a 12-month period. The rate of infections was higher in the cyclophosphamide group, as were amenorrhea, alopecia, leukopenia, and death. Rates of relapse did not differ in the first 12 months (11 to 15% in the two groups). However, at 24 months, the flare rate was higher in the mycophenolate group. ⁶³ All patients in the study were Chinese, and the severity of nephritis was probably less than that of patients included in the National Institute of Health (NIH) studies of cyclophosphamide. Ethnic and disease-severity differences might account for the high response rate in this group. Much data supports the fact that outcomes of nephritis are worse and responses to therapy less in patients with lupus nephritis who are African American. ^{66, 67} Thus, interpretation of the results of clinical trials must consider the ethnic composition of patients studied as well as different criteria for inclusion in the study, different therapeutic regimens, and different definitions of the outcomes measured. A recent prospective trial in the United States ^{55a, 55b} compared treatment of severe lupus nephritis with IV cyclophosphamide versus mycophenolate mofetil 2 to 3 g a day. At 6 months, complete or partial remissions occurred in 49% of mycophenolate and 26% of cyclophosphamide—a statistically significant difference favoring mycophenolate. Side effects were less in the mycophenolate group. The durability of the responses and the acquisition of more responses beyond 6 months of therapy are not known. To administer mycophenolate mofetil, I begin 250 mg twice daily and gradually increase the dose over a few weeks to a total of 2 to 3 g daily. In my experience, response occurs more rapidly to mycophenolate than to azathioprine. In addition, I concur with the opinion of other experts that some patients refractory to high-dose glucocorticoids and cyclophosphamide respond to mycophenolate. ⁶⁴

Several other cytotoxic drugs have been used in uncontrolled studies of patients with SLE, including nitrogen mustard, chlorambucil, cyclosporine, and leflunomide. I have used nitrogen mustard or chlorambucil (with some success) in patients who are unresponsive to azathioprine and in whom bladder toxicity from cyclophosphamide has developed. Cyclosporine can be used as a second agent added to glucocorticoids; reports of the utility of cyclosporine in SLE give mixed results. One open randomized study compared cyclosporine (5 mg/kg/day) to prednisolone plus cyclophosphamide (2 mg/kg/day) for 1 year in children with lupus DPGN: Both groups showed a reduction of urine protein and maintenance of renal function, with better growth in the neoral group. ⁶⁸ An open trial (unblinded) assigned patients with moderately severe active SLE to pulse methylprednisolone plus daily glucocorticoids plus cyclosporine (less than 5 mg/kg/day) or to steroids alone; the cyclosporine group had a significantly lower 12-month mean cumulative dose of prednisone and few disease flares. ⁶⁹ In contrast, a retrospective review suggested that cyclosporin is useful primarily in management of lupus thrombocytopenia and that side effects or failure to control disease activity led to withdrawal of the therapy in most patients. ⁷⁰ In uncontrolled trials of cyclosporine in adults with lupus nephritis, ⁷¹ some of whom have failed steroid-plus-cytotoxic therapies, at least half show diminished disease activity scores, reduced proteinuria, and stable or improved renal function. Deterioration in renal function is a frequent side effect of cyclosporine; other side effects include hypertension, hypertrichosis, gingival hyperplasia, paresthesias, tremor, and gastrointestinal symptoms. Neverthe-less, these drugs are acceptable alternatives in patients doing poorly on standard, proven regimens who have normal or close-to-normal renal function and good control of blood pressure.

Leflunomide, a new agent released for treatment of RA, which is immunosuppressive by impairing lymphocyte function, has also been used in patients with SLE. One retrospective analysis of a small number of patients treated with 100-mg loading doses daily for 3 days, then maintenance daily doses of 20 mg, reported improvement in 10 of 14 patients. ⁷² The major adverse effects of leflunomide include nausea, diarrhea, infections, allergic reactions, and hepatotoxicity. It is not yet clear that this agent has an important role in the management of SLE.

Finally, in patients with life-threatening disease who are refractory or intolerant to all these "standard" strategies, maximal "immunoablative" cytotoxic or immunosuppressive therapy can be considered with or without infusion of autologous stem cells. ^{73, 74, 75, 76} These procedures are discussed in this chapter in a later section on experimental therapies.