There is no cure for SLE, and complete sustained remissions are rare. Therefore, the physician should plan to induce remissions of acute flares and then maintain improvements with strategies that suppress symptoms to an acceptable level and prevent organ damage. Usually patients will endure some adverse effects of medications. Therapeutic choices depend on (1) whether disease manifestations are life-threatening or likely to cause organ damage, justifying aggressive therapies; (2) whether manifestations are potentially reversible; and (3) the best approaches to preventing complications of disease and its treatments. Therapies, doses, and adverse effects are listed in Table 378-5.
CONSERVATIVE THERAPIES FOR MANAGEMENT OF NON-LIFE-THREATENING DISEASE
Among patients with fatigue, pain, and autoantibodies indicative of SLE, but without major organ involvement, management can be directed to suppression of symptoms. Analgesics and antimalarials are mainstays. NSAIDs are useful analgesics/anti-inflammatories, particularly for arthritis/arthralgias. However, two major issues indicate caution in using NSAIDs. First, SLE patients compared with the general population are at increased risk for NSAID-induced aseptic meningitis, elevated serum transaminases, hypertension, and renal dysfunction. Second, all NSAIDs, particularly those that inhibit cyclooxygenase-2 specifically, may increase risk for myocardial infarction. Acetaminophen to control pain may be a good strategy, but NSAIDs are more effective in some patients. The relative hazards of NSAIDs compared with low-dose glucocorticoid therapy have not been established. Antimalarials (hydroxychloroquine, chloroquine, and quinacrine) often reduce dermatitis, arthritis, and fatigue. A randomized, placebo-controlled, prospective trial has shown that withdrawal of hydroxychloroquine results in increased numbers of disease flares; hydroxychloroquine also reduces accrual of tissue damage, including renal damage, over time. Because of potential retinal toxicity, patients receiving antimalarials should undergo ophthalmologic examinations annually. A placebo-controlled prospective trial suggests that administration of dehydroepiandrosterone may reduce disease activity. If quality of life is inadequate despite these conservative measures, treatment with low doses of systemic glucocorticoids may be necessary. The clinician may also consider treatment with belimumab (anti-BLyS) in these patients, although published clinical trials enrolled patients who had failed to respond to conservative therapies. Lupus dermatitis should be managed with topical sunscreens, antimalarials, topical glucocorticoids, and/or tacrolimus, and if severe or unresponsive, systemic glucocorticoids with or without mycophenolate mofetil.
LIFE-THREATENING SLE: PROLIFERATIVE FORMS OF LUPUS NEPHRITIS
Guidelines for management of lupus nephritis have been published recently by the American College of Rheumatology and the European League Against Rheumatism (encompassed and referenced in Fig. 378-2 and Table 378-5). The mainstay of treatment for any inflammatory life-threatening or organ-threatening manifestations of SLE is systemic glucocorticoids (0.5–1 mg/kg per day PO or 500–1000 mg of methylprednisolone sodium succinate IV daily for 3 days followed by 0.5–1 mg/kg of daily prednisone or equivalent). Evidence that glucocorticoid therapy is life-saving comes from retrospective studies from the predialysis era; survival was significantly better in people with DPGN treated with high-dose daily glucocorticoids (40–60 mg of prednisone daily for 4–6 months) versus lower doses. Currently, high doses are recommended for much shorter periods; recent trials of interventions for severe SLE use 4–6 weeks of 0.5–1 mg/kg per day of prednisone or equivalent. Thereafter, doses are tapered as rapidly as the clinical situation permits, usually to a maintenance dose ranging from 5 to 10 mg of prednisone or equivalent per day. Most patients with an episode of severe SLE require many years of maintenance therapy with low-dose glucocorticoids, which can be increased to prevent or treat disease flares. Frequent attempts to gradually reduce the glucocorticoid requirement are recommended because virtually everyone develops important adverse effects (Table 378-5). High-quality clinical studies regarding initiating therapy for severe, active SLE with IV pulses of high-dose glucocorticoids are not available. Most recent clinical trials in lupus nephritis have initiated therapy with high-dose IV glucocorticoid pulses (500–1000 mg daily for 3–5 days). This approach must be tempered by safety considerations, such as the presence of conditions adversely affected by glucocorticoids (e.g., infection, hyperglycemia, hypertension, osteoporosis).
Cytotoxic/immunosuppressive agents added to glucocorticoids are recommended to treat serious SLE. Almost all prospective controlled trials in SLE involving cytotoxic agents have been conducted in combination with glucocorticoids in patients with lupus nephritis. Therefore, the following recommendations apply to treatment of nephritis. Either cyclophosphamide (an alkylating agent) or mycophenolate mofetil (a relatively lymphocyte-specific inhibitor of inosine monophosphatase and therefore of purine synthesis) is an acceptable choice for induction of improvement in severely ill patients; azathioprine (a purine analogue and cycle-specific antimetabolite) may be effective but is slower to influence response and associated with more flares. In patients whose renal biopsies show ISN grade III or IV disease, early treatment with combinations of glucocorticoids and cyclophosphamide reduces progression to ESRD and death. Shorter-term studies with glucocorticoids plus mycophenolate mofetil (prospective randomized trials of 6 months, follow-up studies of 36 months) show that this regimen is similar to cyclophosphamide in achieving improvement. Comparisons are complicated by effects of race, since higher proportions of African Americans (and other non-Asian, nonwhite races) respond to mycophenolate than to cyclophosphamide, whereas similar proportions of whites and Asians respond to each drug. Regarding toxicity, diarrhea is more common with mycophenolate mofetil; amenorrhea, leukopenia, and nausea are more common with cyclophosphamide. Importantly, rates of severe infections and death are similar in meta-analyses. Two different regimens of IV cyclophosphamide are available. For white patients with northern European backgrounds, low doses of cyclophosphamide (500 mg every 2 weeks for six total doses, followed by azathioprine or mycophenolate maintenance) are as effective as standard high doses, with less toxicity. Ten-year follow-up has shown no differences between the high-dose and low-dose groups (death or ESRD in 9–20% of patients in each group). The majority of the European patients were white; it is not clear whether the data apply to U.S. populations. High-dose cyclophosphamide (500–1000 mg/m2 body surface area given monthly IV for 6 months, followed by azathioprine or mycophenolate maintenance) is an acceptable approach for patients with severe nephritis (e.g., multiple cellular crescents and/or fibrinoid necrosis on renal biopsy, or rapidly progressive glomerulonephritis). Cyclophosphamide and mycophenolate responses begin 3–16 weeks after treatment is initiated, whereas glucocorticoid responses may begin within 24 h.
For maintenance therapy, mycophenolate and azathioprine probably are similar in efficacy and toxicity; both are safer than cyclophosphamide. In a recently published multicenter study, mycophenolate was superior to azathioprine in maintaining renal function and survival in patients who responded to induction therapy with either cyclophosphamide or mycophenolate. The incidence of ovarian failure, a common effect of high-dose cyclophosphamide therapy (but probably not of low-dose therapy), can be reduced by treatment with a gonadotropin-releasing hormone agonist (e.g., leuprolide 3.75 mg intramuscularly) prior to each monthly cyclophosphamide dose. Patients with high serum creatinine levels (e.g., ≥265 μmol/L [≥3.0 mg/dL]) many months in duration and high chronicity scores on renal biopsy are not likely to respond to any of these therapies. In general, it may be better to induce improvement in an African-American or Hispanic patient with proliferative glomerulonephritis with mycophenolate mofetil (2–3 g daily) rather than cyclophosphamide, with the option to switch if no evidence of response is detectable after 3–6 months of treatment. For whites and Asians, induction with either mycophenolate mofetil or cyclophosphamide is acceptable. Cyclophosphamide may be discontinued when it is clear that a patient is improving. The number of SLE flares is reduced by maintenance therapy with mycophenolate mofetil (1.5–2 g daily) or azathioprine (1–2.5 mg/kg per day). Both cyclophosphamide and mycophenolate mofetil are potentially teratogenic; patients should be off either medication for at least 3 months before attempting to conceive. Azathioprine can be used if necessary to control active SLE in patients who are pregnant. If azathioprine is used either for induction or maintenance therapy, patients may be prescreened for homozygous deficiency of the TMPT enzyme (which is required to metabolize the 6-mercaptopurine product of azathioprine) because they are at higher risk for bone marrow suppression.
Good improvement occurs in ~80% of lupus nephritis patients receiving either cyclophosphamide or mycophenolate at 1–2 years of follow-up. However, in some studies, at least 50% of these individuals have flares of nephritis over the next 5 years, and re-treatment is required; such individuals are more likely to progress to ESRD. Long-term outcome of lupus nephritis to most interventions is better in whites than in African Americans. Methotrexate (a folinic acid antagonist) may have a role in the treatment of arthritis and dermatitis but probably not in nephritis or other life-threatening disease. Small controlled trials (in Asia) of leflunomide, a relatively lymphocyte-specific pyrimidine antagonist licensed for use in rheumatoid arthritis, have suggested it can suppress disease activity in some SLE patients. Cyclosporine and tacrolimus, which inhibit production of IL-2 and T lymphocyte functions, have not been studied in prospective controlled trials in SLE in the United States; several studies in Asia have shown they are effective in lupus nephritis. Because they have potential nephrotoxicity but little bone marrow toxicity, the author uses them for periods of a few months in patients with steroid-resistant cytopenias of SLE or in steroid-resistant patients who have developed bone marrow suppression from standard cytotoxic agents.
Use of biologics directed against B cells for active SLE is under intense study. Use of anti-CD20 (rituximab), particularly in patients with SLE who are resistant to the more standard combination therapies discussed above, is controversial. Several open trials have shown efficacy in a majority of such patients, both for nephritis and for extrarenal lupus. However, recent prospective placebo-controlled randomized trials, one in renal and one in nonrenal SLE, did not show a difference between anti-CD20 and placebo when added to standard combination therapies. In contrast, recent trials of standard therapy plus belimumab (anti-BLyS, which binds soluble BLyS/BAFF, which is required for maturation of naïve and transitional B cells to plasma cells and memory B cells) showed improvement in 51% of SLE patients compared to 36% of those on placebo; these differences were statistically significant. The U.S. Food and Drug Administration (FDA) has approved belimumab for treatment of seropositive patients with SLE who have failed standard treatments. The belimumab trial did not include patients with active nephritis or CNS disease. Post hoc analyses have shown that the SLE patient most likely to respond to belimumab has fairly robust clinical activity (a Systemic Lupus Erythematosus Disease Activity Index [SLEDAI] score of ≥10), positive anti-DNA, and low serum complement. SLEDAI is a widely used measure of SLE disease activity; scores >3 reflect clinically active disease. At this time, it is useful to add belimumab to the therapeutic armamentarium in SLE, and it is clear that some patients benefit. However, its role in management of lupus nephritis is not yet known.
SPECIAL CONDITIONS IN SLE THAT MAY REQUIRE ADDITIONAL OR DIFFERENT THERAPIES
Crescentic Lupus Nephritis
The presence of cellular or fibrotic crescents in glomeruli with proliferative glomerulonephritis indicates a worse prognosis than in patients without this feature. There are no large prospective multinational controlled trials showing efficacy of cyclophosphamide, mycophenolate, cyclosporine, or tacrolimus in such cases. Most authorities currently recommend that high-dose cyclophosphamide is the induction therapy of choice, in addition to high-dose glucocorticoids. One prospective trial from China showed superiority of mycophenolate to cyclophosphamide.
Membranous Lupus Nephritis
Most SLE patients with membranous (INS-V) nephritis also have proliferative changes and should be treated for proliferative disease. However, some have pure membranous changes. Treatment for this group is less well defined. Some authorities do not recommend immunosuppression unless proteinuria is in the nephrotic range (although treatment with angiotensin-converting enzyme inhibitors or angiotensin II receptor blockers is recommended). In those patients, recent prospective controlled trials suggest that alternate-day glucocorticoids plus cyclophosphamide or mycophenolate mofetil or cyclosporine are all effective in the majority of patients in reducing proteinuria. It is more controversial whether they preserve renal function over the long term.
Fertility rates for men and women with SLE are probably normal. However, rate of fetal loss is increased (approximately two- to threefold) in women with SLE. Fetal demise is higher in mothers with high disease activity, antiphospholipid antibodies, and/or active nephritis. Suppression of disease activity can be achieved by administration of systemic glucocorticoids. A placental enzyme, 11-β-dehydrogenase 2, deactivates glucocorticoids; it is more effective in deactivating prednisone and prednisolone than the fluorinated glucocorticoids dexamethasone and betamethasone. Glucocorticoids are listed by the FDA as pregnancy category A (no evidence of teratogenicity in human studies); cyclosporine, tacrolimus, and rituximab are listed as category C (may be teratogenic in animals but no good evidence in humans); azathioprine, hydroxychloroquine, mycophenolate mofetil, and cyclophosphamide are category D (there is evidence of teratogenicity in humans, but benefits might outweigh risks in certain situations); and methotrexate is category X (risks outweigh benefits). Therefore, active SLE in pregnant women should be controlled with hydroxychloroquine and, if necessary, prednisone/prednisolone at the lowest effective doses for the shortest time required. Azathioprine may be added if these treatments do not suppress disease activity. Adverse effects of prenatal glucocorticoid exposure (primarily betamethasone) on offspring may include low birth weight, developmental abnormalities in the CNS, and predilection toward adult metabolic syndrome. It is likely that each of these glucocorticoids and immunosuppressive medications gets into breast milk, at least in low levels; patients should consider not breastfeeding if they need therapy for SLE. In SLE patients with antiphospholipid antibodies (on at least two occasions) and prior fetal losses, treatment with heparin (usually low-molecular-weight) plus low-dose aspirin has been shown in prospective controlled trials to increase significantly the proportion of live births; however, a recent prospective trial showed no differences in fetal outcomes in women taking aspirin compared to those taking aspirin plus low-molecular-weight heparin. An additional potential problem for the fetus is the presence of antibodies to Ro, sometimes associated with neonatal lupus consisting of rash and congenital heart block with or without cardiomyopathy. The cardiac manifestations can be life-threatening; therefore the presence of anti-Ro requires vigilant monitoring of fetal heart rates with prompt intervention (delivery if possible) if distress occurs. Recent evidence shows that hydroxychloroquine treatment of an anti-Ro-positive mother whose infant develops congenital heart block significantly reduces the chance that subsequent fetuses will develop heart block. There is some evidence that dexamethasone treatment of a mother in whom first- or second-degree heart block is detected in utero may sometimes prevent progression of heart block. Women with SLE usually tolerate pregnancy without disease flares. However, a small proportion develops severe flares requiring aggressive glucocorticoid therapy or early delivery. Poor maternal outcomes are highest in women with active nephritis or irreversible organ damage in kidneys, brain, or heart.
Lupus and Antiphospholipid Syndrome (APS)
Patients with SLE who have venous or arterial clotting and/or repeated fetal losses and at least two positive tests for antiphospholipid antibodies have APS and should be managed with long-term anticoagulation (Chap. 379). A target international normalized ratio (INR) of 2.0–2.5 is recommended for patients with one episode of venous clotting; an INR of 3.0–3.5 is recommended for patients with recurring clots or arterial clotting, particularly in the CNS. Recommendations are based on both retrospective and prospective studies of posttreatment clotting events and adverse effects from anticoagulation.
Microvascular Thrombotic Crisis (Thrombotic Thrombocytopenic Purpura, Hemolytic-Uremic Syndrome)
This syndrome of hemolysis, thrombocytopenia, and microvascular thrombosis in kidneys, brain, and other tissues carries a high mortality rate and occurs most commonly in young individuals with lupus nephritis. The most useful laboratory tests are identification of schistocytes on peripheral blood smears, elevated serum levels of lactate dehydrogenase, and antibodies to ADAMS13. Plasma exchange or extensive plasmapheresis is usually life-saving; most authorities recommend concomitant glucocorticoid therapy; there is no evidence that cytotoxic drugs are effective.
Patients with any form of lupus dermatitis should minimize exposure to ultraviolet light, using appropriate clothing and sunscreens with a sun protection factor of at least 30. Topical glucocorticoids and antimalarials (such as hydroxychloroquine) are effective in reducing lesion severity in most patients and are relatively safe. Systemic treatment with retinoic acid is a useful strategy in patients with inadequate improvement on topical glucocorticoids and antimalarials; adverse effects are potentially severe (particularly fetal abnormalities), and there are stringent reporting requirements for its use in the United States. Extensive, pruritic, bullous, or ulcerating dermatitides usually improve promptly after institution of systemic glucocorticoids; tapering may be accompanied by flare of lesions, thus necessitating use of a second medication such as hydroxychloroquine, retinoids, or cytotoxic medications such as methotrexate, azathioprine, or mycophenolate mofetil. In therapy-resistant lupus dermatitis there are reports of success with topical tacrolimus (caution must be exerted because of the possible increased risk for malignancies) or with systemic dapsone or thalidomide (the extreme danger of fetal deformities from thalidomide requires permission from and supervision by the supplier).
Prevention of complications of SLE and its therapy include providing appropriate vaccinations (the administration of influenza and pneumococcal vaccines has been studied in patients with SLE; flare rates are similar to those receiving placebo) and suppressing recurrent urinary tract infections. Vaccination with attenuated live viruses is generally discouraged in patients who are immunosuppressed. Strategies to prevent osteoporosis should be initiated in most patients likely to require long-term glucocorticoid therapy and/or with other predisposing factors. Postmenopausal women can be protected from steroid-induced osteoporosis with either bisphosphonates or denosumab. Safety of long-term use of these strategies in premenopausal women is not well established. Control of hypertension and appropriate prevention strategies for atherosclerosis, including monitoring and treatment of dyslipidemias, management of hyperglycemia, and management of obesity, are recommended.
Studies of highly targeted experimental therapies for SLE are in progress. They include targeting (1) activated B lymphocytes with anti-CD22 or TACI-Ig, (2) inhibition of IFN-α, (3) inhibition of B/T cell second signal coactivation with CTLA-Ig, (4) inhibition of innate immune activation via TLR7 or TLR7 and 9, (5) induction of regulatory T cells with peptides from immunoglobulins or autoantigens; (6) suppression of T cells, B cells, and monocyte/macrophages with laquinimod; and (7) inhibition of lymphocyte activation by blockade of Jak/Stat. A few studies have used vigorous untargeted immunosuppression with high-dose cyclophosphamide plus anti-T cell strategies, with rescue by transplantation of autologous hematopoietic stem cells for the treatment of severe and refractory SLE. One U.S. report showed an estimated mortality rate over 5 years of 15% and sustained remission in 50%. It is hoped that in the next edition of this text, we will be able to recommend more effective and less toxic approaches to treatment of SLE based on some of these strategies.