Pharmacotherapeutic Options

Introduction

This section covers the individual pharmacotherapeutic options for SLE, including their appropriate uses in SLE and the efficacy and safety data supporting their use:

Hydroxychloroquine

Hydroxychloroquine (HCQ) is a 4-aminoquinoline compound with antimalarial and immunomodulatory properties. It was synthesized in the 1946 by hydroxylation of chloroquine, an earlier antimalarial and received FDA approval for the treatment of malaria and SLE symptoms in 1955. For many decades, HCQ has been the cornerstone therapy for nearly all patients with SLE.

Indications and Dosage

Hydroxychloroquine is indicated for the treatment of SLE in adults. It is available as 200 mg tablets. The maximum recommended dosage is 5 mg per kilogram of body weight per day. It should be used with care in patients with renal or hepatic impairment, with possible dose modification. Hydroxychloroquine is administered orally, with food or milk. It is contraindicated in individuals with hypersensitivity to 4-aminoquinoline compounds.

Mechanism of Action

The immunomodulatory mechanism of HCQ is incompletely understood. It is known to accumulate in lysosomes, and that it inhibits TLRs and the cGAS-STING pathway, a key mediator of inflammation; this in turn affects cytokine release, cell signaling, antigen presentation, T-cell polarization, and apoptosis, among other processes. The potential effects of HCQ action in immune cells, including antigen-presenting cells, T cells and B cells, are shown in Figure 8-1.

Efficacy

Most of the evidence for the efficacy of HCQ in SLE comes from observational data. The use of HCQ is associated with a reduction in tissue damage, thrombotic event rate, glomerulonephritis and disease activity, as well as improvements in overall survival. In one analysis of observational data from 608 patients in the US-based multiethnic LUMINA cohort, mortality was lower in the HCQ group (17 patients, or 5%) than in the non-HCQ group (44 patients, or 17%) (P <0.0001). The effect of discontinuing HCQ on SLE outcomes was tested in a small randomized clinical trial conducted by the Canadian Hydroxychloroquine Study Group. A total of 47 patients were randomized to continue their prior HCQ regimen (n = 25) or to receive a placebo (n = 22) for 24 weeks. Clinical flare-ups of SLE were significantly more common in the placebo group (16/22 patients; 73%) than in the continued HCQ group (9/25 patients; 36%). A 2022 case-crossover study of SLE patients with ≥1 flare found an adjusted odds ratio (AOR) of 1.98 for any flare occurrence associated with HCQ doses of ≤5 mg/kg/day compared to >5 mg/kg/day. The AOR was higher – 6.04 – for moderate or severe lupus flares. Another study of 1460 patients with SLE found that the risk of first SLE flare was higher with HCQ dose reduction or HCQ discontinuation compared to HCQ maintenance at the initial dose, with respective adjusted hazard ratios of 1.20 and 1.56.

Safety

Hydroxychloroquine is generally safe and well tolerated. The most common adverse events (AEs) include GI disturbance (including nausea, emesis and diarrhea; ~7-37%), skin hyperpigmentation (~10-25%), myopathy (~1.3-12.6%) and retinopathy (~0.29-7.5% after 5 years, with the risk increasing with duration of use). Gastrointestinal and cutaneous AEs typically improve with time, and rarely require cessation. Clinically significant myopathy is also uncommon. The most serious AEs are ophthalmologic, including keratopathy and retinopathy, the latter typically manifesting as “bull’s eye” maculopathy. The mechanism of retinal toxicity of HCQ is not known, but may due to changes in photoreceptor metabolism. In patients with risk factors (Table 8-1), yearly ophthalmologic monitoring is indicated; in those without, age-appropriate monitoring is recommended for the first 5 years, followed by yearly check-ups thereafter.

Other adverse events observed in patients taking HCQ include conduction abnormalities (including QR prolongation and torsade des pointes), cardiomyopathy, neuropsychiatric events, hypoglycemia, serious cutaneous eruptions, oxidative hemolysis, cytopenias and liver enzyme abnormalities. These AEs are rare and should be addressed on a case-by-case basis.

Guideline Recommendations

The 2019 EULAR guidelines for the management of SLE recommend HCQ (at a maximum dose of 5 mg/kg/real body weight) for all patients with SLE, regardless of disease severity, unless contraindicated. To minimize the risk of retinal AEs, in patients without risk factors for retinal toxicity (such as pre-existing retinal disease), the guidelines recommend ophthalmological screening by visual fields examination and/or spectral domain-optical coherence tomography be done at baseline, after 5 years and annually after that.

Glucocorticoids

Glucocorticoids (GCs) are sterol-derived compounds with potent anti-inflammatory and immunosuppressive properties that have been used to treat rheumatic diseases, including SLE, since their discovery in 1948. Like all sterols, GCs possess three six-carbon rings and one five-carbon ring. Glucocorticoids include the endogenous hormone cortisol (hydrocortisone) and its synthetic derivatives prednisolone, methylprednisolone, triamcinolone, dexamethasone and betamethasone. Although the term GCs is often used synonymously with corticosteroids, corticosteroids are a broader category that includes all steroids produced in the adrenal cortex, i.e. both GCs (which regulate carbohydrate metabolism) and mineralocorticoids (which regulate electrolyte balance).

Indications and Dosage

Prednisolone and its prodrug prednisone are the most commonly prescribed GCs in Europe and the United States, respectively. Prednisone is indicated during an exacerbation or as maintenance therapy in selected cases of SLE. Prednisone and other systemic GCs are typically administered orally, in a single daily dose and are available as tablets whose dosage strengths vary (for prednisone, tablets are available in 5 mg, 10 mg and 20 mg forms). The actual dosage and administration method is decided on an individual basis, depending on the type and severity of the SLE manifestation. For prednisone, a low or maintenance dose is 0.1-0.25 mg/kg/day; a moderate dose is 0.5 mg/kg/day; a high dose is 1-3 mg/kg/day; and a massive dose is 15-30 mg/kg/day. Other GCs are often administered in “prednisone equivalent” doses, as presented in Table 8-2. Localized SLE skin manifestation may be treated with topical preparations, including hydrocortisone (0.125%-1.0%; short-acting), triamcinolone (0.025%-0.5%; intermediate-acting) and betamethasone (0.01%-0.1%; long-acting). For cases of life-threatening complications of SLE, a pulse therapy of 3-5 days of 1g/day intravenous methylprednisolone can be used. Because of the long-term adverse events associated with GC use, GC should be tapered to the lowest possible maintenance dose, making every effort to discontinue them.

Mechanism of Action

The mechanism of action of GCs is complex and incompletely understood. Glucocorticoids act through “genomic” and “non-genomic” mechanisms (Figure 8-2.). Genomic mechanisms involve direct control of transcription via GC receptors, which are expressed in nearly all cells and can bind GC response elements (GREs) to promote transcription or negative GC response elements (nGREs) to repress it. At high GC doses, the (n)GREs may become saturated, enabling the GC to act through non-genomic mechanisms, including through cell-surface receptors and second messengers.

Glucocorticoid-mediated transcriptional regulation results in anti-inflammatory effects by increasing neutrophil count and decreasing neutrophil trafficking, decreasing macrophage, monocyte and lymphocyte counts and trafficking and impairing immune cell function (phagocytosis, antigen presentation, cytokine production); GC-associated immunosuppression is effected through reduced antibody production.

Efficacy

The efficacy of GC use in SLE has been established clinically in the preceding seven decades of their widespread use. Case reports in the 1960s and 1970s established that GCs can dramatically improve renal function in patients with lupus nephritis. Glucocorticoids are efficacious in the treatment of a variety of lupus manifestations, including cutaneous lesions, arthralgia, acute lupus pneumonitis, myocarditis, lupus nephritis, gastro-intestinal manifestations and neuropsychiatric lupus.

Safety

Although GCs are efficacious for most active manifestations of SLE, their long-term use is associated with serious adverse events such as metabolic changes (including diabetes), cardiovascular events, bone density loss (osteoporosis) and increased infection risk. These adverse events are dose-dependent; a EULAR task force concluded that doses of ≤5 mg and >10 mg prednisone equivalent per day represent low and high risk, respectively (the risk with intermediate doses depends on patient-specific factors). However, there is evidence to suggest that sustained doses of ≥7.5 mg per day may cause irreversible organ damage. Therefore, the use of GCs should be restricted to short-term (2-4 weeks) treatment of flare-ups; the GC should then be tapered to a minimal maintenance dose and eventually discontinued. In contrast to chronic GC use, high-dose pulses of methylprednisolone generally have no serious adverse events.

Guideline Recommendations

The 2019 EULAR guidelines for the management of SLE state that GCs can be used at the dose and administration route appropriate for the specific type and severity of organ involvement. To minimize the GC dose, the guidelines recommend using intravenous methylprednisolone pulses initially (usually 250-1000 mg per day, for 1-3 days), as this allows for a lower oral GC dose and quicker tapering; the guidelines also state that rapid initiation of immunomodulatory therapy may achieve the same result. Finally, when GC maintenance treatment is required, the guidelines recommend using a prednisone equivalent dose of <7.5 mg/day.

Azathioprine

Azathioprine (AZA) is a purine analogue derived from 6-mercaptopurine (6-MP). First synthesized in 1957 as part of a program to develop drugs for the inhibition of adenine and guanine biosynthesis, AZA, together with 6-MP, was initially investigated as a potential chemotherapeutic drug. After the discovery of its anti-inflammatory properties in 1958, AZA was tested for use in transplantation medicine and for the treatment of rheumatic disease, including SLE. It remains part of the wider SLE drug armamentarium today, though it is generally less efficacious than cyclophosphamide and mycophenolate mofetil (MMF).

Indications and Dosage

Azathioprine is only FDA approved for the treatment of rheumatoid arthritis and for prevention of organ rejection in renal transplant/transplantation. However, it is used off label for both induction and maintenance therapy for SLE, including in lupus nephritis (LN) – although its efficacy for the treatment of LN is lower than that of MMF. Azathioprine is available as tablets and is administered orally, typically with food. The dosage varies depending on the SLE manifestation and severity, but a starting dose of 1 mg/kg is typically used, followed by 2-3 mg/kg/day in 1-3 doses. In LN, a maximum daily AZA dose of 2 mg/kg is recommended; a longitudinal cohort study of AZA use reported a median maintenance dose of 1.5 mg/kg/day. Because the bioactive metabolites AZA are excreted renally, AZA dosage is typically adjusted for kidney activity, with a dose reduction of 25% and 50% in patients with creatinine clearance of 10-30 mL/min and <10 mL/min, respectively. In patients on hemodialysis, AZA is administered post-dialysis.

Mechanism of Action

Azathioprine is a prodrug that is converted in several steps into active metabolites as depicted in Figure 8-3. After the non-enzymatic cleavage of AZA into 6-MP, the latter is converted via a series of sequential enzymatic reactions into 6-thioinosine monophosphate, 6-thioxanthosine monophosphate and finally 6-thioguanosine monophosphate, a cytotoxic purine analogue which can be incorporated into DNA and RNA molecules, resulting in the inhibition of cell growth and division. Thus, AZA exerts its main anti-inflammatory and immunosuppressive effects through blocking lymphocyte proliferation. Two other mechanisms are believed to contribute to AZA efficacy in SLE. First, AZA can be cleaved non-enzymatically into imidazole derivatives, which may have a weak immunomodulatory effect on lymphocyte function and migration. Second, the intermediate metabolite of enzymatic 6-MP conversion, 6-thioinosine monophosphate, can also be enzymatically converted to 6-methylthioinosine 5ʹ-monophosphate, which may reduce intracellular nucleotide availability.

Efficacy

The efficacy of AZA in SLE has primarily been established clinically since its introduction in the late 1950s. In the context of maintenance therapy for LN, the MAINTAIN trial reported a similar renal flare rate for AZA and MMF (25% vs 19%), and the rates was comparable in a long-term follow-up study. However, the patient population in MAINTAIN was predominantly White. The ALMS trial, which had a more racially heterogeneous patient population (41% White, 10% Black, 34% Asian, 15% other), found that MMF was superior to AZA with respect to time to treatment failure (hazard ratio [HR], 0.44; P = 0.003), time to renal flare (HR, 0.50; P = 0.03) and time to rescue therapy (HR, 0.39; P = 0.02). Thus, while AZA can be used as induction in mild proliferative (class III or IV) LN in patients with contraindications to the standard of care therapy of MMF or cyclophosphamide, it is not the preferred agent; it can be used in maintenance therapy, although MMF is preferred. Azathioprine is effective and widely used for moderately severe extra-renal manifestations of SLE, including thrombocytopenia, serositis and neuropsychiatric SLE.

Safety

Azathioprine is generally safe and well tolerated at a daily dose of 2-2.5 mg/kg. The principal adverse events observed in patients taking AZA are gastrointestinal and hematologic. The most common gastrointestinal symptoms are nausea and vomiting, although more rarely diarrhea, fever, malaise and myalgias may occur. Hematologic abnormalities, including leukopenia and thrombocytopenia may occur even with low doses of AZA. Liver enzyme abnormalities may occur, but serious hepatotoxicity is rare. Finally, AZA is associated with a slightly increased risk of lymphoma and its prescribing information therefore carries a black box warning for malignancy. Azathioprine is safe to use during pregnancy, making it particularly useful in SLE, whose patient population consists largely of women of child-bearing age.

Guideline Recommendations

The 2019 EULAR guidelines for the management of SLE recommend considering AZA for immunosuppression in patients who do not respond to hydroxychloroquine (HCQ) or HCQ in combination with glucocorticoids. They also recommend the use of AZA for maintenance therapy in patients with hematological manifestations of SLE or with LN.

Methotrexate

Methotrexate (MTX) is an antimetabolite initially synthesized in the early 1950s from the related molecule aminopterin, a 4-amino derivative of folic acid. Methotrexate was initially developed for use in hematologic and solid malignancy; however, during the 1960s and 1970s, it was tested in several rheumatic diseases and since the 1980s has been the standard therapy for rheumatoid arthritis. Its broad anti-inflammatory and immunosuppressive properties make it a useful part of the SLE drug armamentarium.

Indications and Dosage

Methotrexate is FDA approved only for the treatment of certain hematological cancers, rheumatoid arthritis, polyarticular juvenile idiopathic arthritis and psoriasis; therefore, its use in SLE is off-label. In the context of SLE, MTX is used off-label in patients who require chronic, steroid-sparing immunosuppression to control symptoms such as arthralgia and arthritis, cutaneous manifestations, serositis and others. Methotrexate can be administered orally (tablets) or parenterally (subcutaneous injection). The typical dose in SLE is 20-25 mg once weekly, with concomitant folic acid supplementation to reduce adverse event risk.

Mechanism of Action

Methotrexate is an inhibitor of dihydrofolic acid reductase, an enzyme that converts dihydrofolates to tetrahydrofolates, which are key intermediaries in nucleotide synthesis (Figure 8-4a). This results in impaired DNA synthesis and negatively affects actively proliferating cells (including cancer cells and bone marrow cells), reducing the neutrophil and lymphocyte count. Methotrexate also promotes the release of adenosine, which has potent anti-inflammatory effects (Figure 8-4b). It also inhibits the synthesis of polyamines (Figure 8-4.c), although this mechanism is more important in rheumatoid arthritis, as the role of polyamines in SLE is unclear. Finally, via several, incompletely understood mechanisms, MTX inhibits the activity of the pro-inflammatory transcription factor NF-κB; one such mechanism is activation of JNK kinase, which inhibits NF-κB (Figure 8-4d).

Efficacy

The efficacy of MTX in SLE has been demonstrated in numerous case series and a handful of randomized controlled clinical trials (RCTs), mostly in the 1990s. Two double-blind RCTs are particularly notable. A study of 41 patients with SLE, published in 1999, found that MTX treatment led to a significantly lower incidence of articular complaints and cutaneous lesions, as well as lower mean SLEDAI scores and a greater steroid-sparing effect, compared to the placebo (P <0.001 for all outcomes). Another trial which randomized 86 patients to either MTX or placebo found a significantly lower prednisone dose (P <0.010) and a significant improvement in the revised Systemic Lupus Activity Measure (SLAM-R) score (P <0.039) with MTX treatment. A systematic review of 3 RCTs (including the two just mentioned) and 6 observational studies showed a significant reduction in the SLE Disease Activity Index (SLEDAI) score among patients who received MTX compared to patients in control groups and a significant reduction in the average glucocorticoid dose.

Safety

Methotrexate is generally well tolerated when taken at doses typical for rheumatic disease; long-term studies have revealed few serious adverse events. Common adverse reactions include ulcerative stomatitis, leukopenia, nausea, abdominal distress, liver enzyme abnormalities and mild alopecia. Other clinically relevant adverse events include infection, malaise, fatigue, chills, fever and dizziness.

Guideline Recommendations

The 2019 EULAR guidelines for the management of SLE recommend MTX as an option for patients who do not achieve their treatment goal with hydroxychloroquine (HCQ) or HCQ in combination with glucocorticoids (GCs), or for patients unable to titrate GCs to doses compatible with long-term use. Methotrexate is also recommended as an option for the treatment of cutaneous manifestations and arthritis in non-responsive patients or those who require high-dose GCs.

Cyclophosphamide

Cyclophosphamide (CYC) is an oxazaphosphorine with a long history of use in systemic lupus erythematosus (SLE) as a cytotoxic agent. The first reports of CYC efficacy in SLE, specifically lupus nephritis (LN), date back to the 1950s, followed by controlled trials in the 1980s and 1990s. With its proven immunosuppressive efficacy, CYC remains a useful option for the treatment of SLE. However, its gonadotoxic and teratogenic effects require a careful patient-provider discussion on the topic of risks and benefits, especially in a population that is predominantly composed of childbearing-age women. The risk of infertility and the need for contraception should be considered and weighed against adverse SLE outcomes, including death.

Indications and Dosage

Cyclophosphamide is FDA approved only for the treatment of certain malignant diseases. However, it is widely used off-label in SLE, including as induction therapy for proliferative (class III or IV) and membranous (class V) LN and as monthly pulses in patients with severe non-renal manifestations, including thrombocytopenia, anemia, neuropsychiatric lupus, acute pneumonitis, alveolar hemorrhage, abdominal vasculitis and extensive cutaneous involvement. Although oral formulations exist, CYC is predominantly administered intravenously in SLE. Two commonly used administration protocols, EuroLupus and the NIH protocol, are shown in Table 8-5, as are dosage adjustment recommendations.

Mechanism of Action

Cyclophosphamide is metabolized in the liver to 4-hydroxycyclophosphamide and aldophosphamide, with the latter metabolite undergoing further conversion in the target cells to phosphoramide mustard and acrolein. Although 4-hydroxycyclophosphamide, acrolein and phosphoramide mustard are all active metabolites, the primary therapeutic effect of CYC is mediated by phosphoramide mustard, which alkylates guanines at the N7 positions, leading to DNA crosslinking (Figure 8-5). This in turn leads to inhibition of cell division and to apoptosis. Rapidly proliferating cells and tissues, including B and T lymphocytes, are disproportionately affected, which leads to immunosuppression and to several of the adverse effects.

Efficacy

The efficacy of CYC in SLE was originally demonstrated in several landmark trials in patients with LN. After a number of trials in the 1970s which established the superiority of oral CYC to prednisolone, in 1986 intravenous pulse CYC in combination with low-dose prednisone was shown to be superior (P = 0.027) to high-dose prednisone for the preservation of renal function in a trial of 107 patients with LN. In another landmark trial of 65 patients with severe LN, published in 1992, intravenous pulse CYC was shown not only to be superior to pulse methylprednisolone for the preservation of renal function, significantly (P <0.04) reducing the probability of doubling serum creatinine. Furthermore, intravenous CYC was shown to have comparable efficacy with fewer adverse events (AEs) compared to oral CYC. Cyclophosphamide has also demonstrated efficacy in clinical trials in patients with severe neuropsychiatric, hematologic and pulmonary manifestations.

Safety

Like other cytotoxic agents, CYC is associated with a number of AEs, including common events like nausea, vomiting, stomatitis, and alopecia and serious AEs like bone marrow, gonadal, and bladder toxicity. To mitigate bladder toxicity, intravenous or oral MESNA (at 20% of the CYC dose) can be administered 0, 2, 4 and 6 hours after CYC treatment. Gonadal toxicity may be offset with subcutaneous leuprolide 3.75 mg two weeks prior to each CYC dose in females and intramuscular testosterone 100 mg every two weeks in males. One study found no difference in the anti-Müllerian hormone levels between patients with SLE who received CYC by the EuroLupus protocol and those who never received cytotoxic therapy, suggesting that EuroLupus may not be gonadotoxic. Rarer AEs include cardiac, hepatic, and pulmonary toxicity, and malignancy. Cyclophosphamide also has high teratogenic potential; while not contraindicated in pregnancy, it should be avoided unless the benefits of treatment outweigh the risk to the fetus.

Guideline Recommendations

Because of its gonadotoxic effects, the 2019 EULAR guidelines are generally conservative in their recommendations for CYC use. It is recommended as an option for severe organ-threatening or life-threatening cases and as rescue therapy for patients who do not respond to other immunosuppressive drugs. For LN, low-dose CYC is recommended as induction treatment. Finally, CYC is also recommended for refractory cases with hematologic manifestations.

Mycophenolic Acid

Mycophenolic acid (MPA) is a benzofuran-derived immunosuppressive drug. Isolated as a natural product from the mold Penicillium stoloniferum, it was initially described in 1913 and found to contain both antibiotic and anti-inflammatory properties. It came into wider clinical use in the 1970s to prevent organ rejection following transplantation. An ester of MPA, mycophenolate mofetil (MMF), was developed to improve its bioavailability and tolerance, and came into clinical use in the 1990s. Enteric coated mycophenolate sodium (eMPA) is also available. Mycophenolate is widely used today both in transplantation medicine and in rheumatic disease, including SLE.

Indications and Dosage

Mycophenolate mofetil and eMPA are currently FDA approved only for prophylaxis of organ rejection; their use in SLE is therefore off-label. Mycophenolate is used as induction therapy in moderately severe proliferative lupus nephritis (LN) and pure membranous LN with nephrotic-range proteinuria, as maintenance therapy in moderately severe to severe proliferative LN and pure membranous LN and for management of extra-renal disease in patients who do not respond to azathioprine (AZA). Mycophenolate (MMF or eMPA) is typically administered orally (in the form of tablets, capsules, or oral suspension), although MMF can also be given intravenously. For non-delayed release tablets, the typical daily dosage is 1-3 g, split in two divided doses. In patients with creatinine clearance below 25 mL/min, the daily dose should not exceed 1 g. For the delayed-release tablet formulation, the recommended dosage is 720 mg twice daily (1440 mg daily total dose), but the dosage may range in practice from 360 mg to 1080 mg twice daily (720 mg to 2160 mg daily total dose).

Mechanism of Action

Once inside the target cells, MMF is converted into MPA, which inhibits inosine monophosphate dehydrogenase, a key enzyme in the de novo guanine nucleotide synthesis pathway (Figure 8-6.). This reduces DNA synthesis and inhibits cell division. Because lymphocytes cannot utilize the salvage pathway of guanine synthesis and thus rely solely on the de novo pathway, they are disproportionately affected by MMF/MPA. The immunosuppressive effects of MMF/MPA thus arise from a reduction in lymphocyte proliferation, although antibody production and antigen proliferation are also negatively affected.

Efficacy

Mycophenolate has established efficacy in LN and other manifestations of SLE, with the exception of neuropsychiatric lupus. After initial reports of efficacy in patients with LN in the late 1990s and early 2000s, in studies in the later 2000s daily oral MMF showed at least equal and sometimes superior efficacy to monthly intravenous cyclophosphamide (CYC) in the management of LN. Among the 227 patients with LN in the Aspreva Lupus Management Study (ALMS), MMF maintenance treatment resulted in longer time to treatment failure (hazard ratio [HR], 0.44; P = 0.003), time to renal flare (HR, 0.50; P = 0.03) and time to rescue therapy (HR, 0.39; P = 0.02), compared to the AZA. Other data from the ALMS trial also suggest that MMF is superior as induction therapy to intravenous CYC in Black (53.9% vs 40.0% responders; P = 0.39) and Hispanic (60.9% vs 38.8%; P = 0.011) patients with equivalent efficacy in White (54.2% vs 56.0%; P = 0.83) and Asian (63.9 vs 53.2%; P = 0.24) patients. In 2017, eMPA was shown to be superior to AZA in SLE with respect to the proportion of patients achieving clinical remission (i.e., clinical SLEDAI = 0: 32.5% vs 19.2%; P <0.001), time to remission (HR, 1.43; P = 0.017), frequency of flares (BILAG A/B flares: 50% vs 71.7%; P = 0.001; BILAG B flares: 8.3% vs 21.67%; P = 0.004) and time to first flare (BILAG A/B flare: HR, 1.81 P = 0.0004; BILAG B flare: HR, 2.84; P = 0.003).

Safety

The most common adverse events of mycophenolate are gastrointestinal, including nausea, emesis and diarrhea; these can be mitigated with the enteric-coated eMPA formulation of mycophenolate. Leukopenia, and rarely thrombocytopenia, may also occur, as can opportunistic bacterial, fungal, and viral infections. Mycophenolate has high teratogenic potential and the 2020 ACR reproductive health guidelines recommend discontinuing it within 3 months prior to attempting conception.

Guideline Recommendations

The 2019 EULAR guidelines recommend mycophenolate as an option for immunosuppression in patients who do not respond to hydroxychloroquine (with or without glucocorticoids [GCs]) or in patients unable to titrate their GCs to a dose compatible with long-term use. For cutaneous manifestations, it is recommended in cases refractory to first-line therapy (topical agents, antimalarials, systemic GC) and those who require high-dose GCs. For hematological manifestations, it is recommended as an option for response maintenance. In LN, it is recommended as an option for both induction therapy and maintenance therapy. Finally, in patients with severe nephrotic syndrome or incomplete renal response who do not have uncontrolled hypertension, high chronicity index at kidney biopsy and/or reduced GFR, the guidelines state that mycophenolate can be combined with a low-dose calcineurin inhibitor.

Calcineurin Inhibitors

Calcineurin inhibitors (CNIs) are a diverse class of compounds that inhibit calcineurin, an intracellular phosphatase which stimulates T-cell activation. Originally developed for organ rejection prophylaxis, CNIs have also been investigated for rheumatic diseases. Three have been tested in SLE: cyclosporine A (CsA), tacrolimus (TAC) and voclosporin (VC). Cyclosporin A, a cyclical polypeptide containing 11 amino acids, was first isolated in the 1970s from the ascomycete fungus Tolypocladium inflatum and began to be used for the treatment of lupus nephritis (LN) in the late 1980s. Tacrolimus, a macrolide, was isolated by Japanese researchers in 1982 from another fungus, Streptomyces tsukubaensis. Found to be up to 100-fold more potent at immunosuppression in vitro than CsA, TAC testing in SLE began two decades after CsA. Voclosporin, a semisynthetic structural derivative of CsA, is the newest CNI, with improved potency, less toxicity and more stable pharmacokinetics compared to its parent molecule. First synthesized in the 1990s, it received FDA approval in 2021 for use in LN therapy.

Indications and Dosage

Cyclosporine A is currently FDA approved only for the prophylaxis of transplant rejection, rheumatoid arthritis and psoriasis It can be used off-label as induction or maintenance therapy in non-severe proliferative LN and membranous LN, and in non-renal manifestations of SLE, including cutaneous and hematological. Cyclosporine A is typically taken orally, with or between meals, on a consistent schedule. Depending on clinical need, CsA is administered at a dose of 1.5-6 mg/kg per day, normally split into two doses; a relatively new microemulsion formulation allows for dosing at the lower end of the spectrum. Cyclosporine A should be avoided in patients with creatinine clearance <60 mL/min, severe uncontrolled hypertension, and advanced tubule-interstitial disease and tubular atrophy on renal biopsy.

Like CsA, oral TAC is used off-label in lupus, as it is currently FDA approved only for the prophylaxis of transplant rejection. It is useful as induction therapy in proliferative LN and induction or maintenance therapy in membranous LN, as well as in non-renal SLE, including cutaneous (either in the oral formulation or as a topical preparation) and musculoskeletal manifestations. The typical dose range is 1-4 mg daily, taken in two doses every day at a consistent time.

Voclosporin is FDA approved, in combination with a background immunosuppressive therapy regimen, for the treatment of adult patients with active LN. The recommended dosage of VC is two 23.7 mg oral doses per day, taken at regular times. The dosage should be modified based on renal function, assessed by estimated glomerular filtration rate (eGFR), as shown in Table 8-7. In the first month, eGFR should be assessed every two weeks and every four weeks after the first month. Voclosporin should be avoided in patients with a baseline creatinine clearance <45 mL/min/1.73m unless the benefits outweigh the risks.

Mechanism of Action

As the name suggests, CNIs inhibit the activity of calcineurin, a calcium/calmodulin-dependent phosphatase. However, CNIs cannot bind calcineurin directly; once inside the target cells, they bind to specific receptors: cyclophilin (for CsA and VC) or FKBP12 (for TAC). This complex can then bind to calcineurin and block its activity (Figure 8-7.1). Calcineurin is required for the nuclear import of transcription factors involved in IL-2 transcription. The inhibition of calcineurin also decreases the transcription of several other inflammatory cytokines, including TNF-α, IL-1β, IL-6, IL-10 and IFNγ, and decreases T-cell proliferation. Calcineurin inhibitors also exert an inhibitory effect on antibody production, B-cell activation, and class switching; these mechanisms all result in immunosuppression. Calcineurin inhibitors also have anti-proteinuric properties mediated by the effects of CNIs on podocyte stabilization and afferent arteriole vasoconstriction; these mechanisms make CNIs an attractive option for the treatment of LN, particularly class V LN.

Efficacy

The efficacy of CsA in LN has been tested in several small randomized controlled trials, primarily in LN. In a trial of 75 patients, CsA demonstrated efficacy equivalent to azathioprine as maintenance therapy for diffuse (WHO LN class IV, Vb, or Vc) LN. Another trial in 42 patients with membranous LN showed that CsA, in combination with prednisone, resulted in significantly greater probability of remission over 12 months of follow-up than prednisone alone. In a trial of 40 patients with proliferative LN, CsA demonstrated comparable efficacy to cyclophosphamide (CYC) as a combination induction/maintenance regimen; the efficacy of the two regimens remained comparable during long-term (>7 years) follow-up. In patients with severe SLE (including non-renal manifestations) randomized to either open-label CsA or azathioprine, CsA demonstrated corticosteroid-sparing efficacy similar to that of azathioprine. In the modern era, CsA has largely been replaced by agents with better efficacy and safety, including mycophenolate mofetil (MMF).

The efficacy of TAC in LN was also assessed in a small number of controlled trials, following promising results from an open-label pilot study. In a double-blind, randomized trial in 63 patients with LN, TAC significantly decreased LN disease activity compared to the placebo. In two small open-label studies in patients with LN, TAC demonstrated comparable efficacy to CYC. A meta-analysis of 9 small, open-label, controlled studies in patients with LN found that TAC is superior to CYC but not MMF for the induction of complete renal remission. Finally, a few small studies have demonstrated the efficacy of topical TAC preparations for the management of cutaneous symptoms of SLE.

The efficacy of VC for the management of LN has been assessed in AURORA 1, one of the largest trials to date in patients with LN. AURORA 1 was a multicenter, international, double-blind, phase 3 trial that randomized 357 patients (1:1) to either VC 23.7 mg twice daily (n=179) or placebo (n=178). All patients received a background MMF regimen (1 g twice daily) and glucocorticoid treatment: an intravenous methylprednisolone pulse on day 1 and 2, followed by rapidly-tapered low-dose oral prednisone. To be eligible for inclusion, patients must have been diagnosed with SLE and active LN according to the ACR criteria, and with a kidney biopsy confirming the LN as class III, IV, or V (alone or in combination with class III or IV). The primary endpoint was complete renal response at week 52, defined as a composite of urine protein/creatinine ratio [UPCR] ≤0.5 mg/mg, eGFR ≥60 mL/min/1.73 m² or no eGFR decrease from baseline of ≥20%, no use of rescue medication, and no >10 mg/day prednisone equivalent for ≥3 consecutive days or for ≥7 days during weeks 44-52. As shown in Figure 8-8A, a significantly higher proportion of patients in the VC group (41%) achieved complete renal response compared to those in the placebo group (23%; odds ratio 2.65; 95% confidence interval 1.64-4.27; P <0.0001). Compared to the placebo, VC treatment also resulted in a significantly higher proportion of patients who achieved the secondary endpoints of complete renal response at week 24 (20% vs 32%; P = 0.002; Figure 8-8A), partial renal response (≥50% UPCR reduction from baseline) at week 24 (50% vs 70%; P <0.001; Figure 8-8B) and at week 52 (52% vs 70%; P <0.001; Figure 8-8B), and significantly shortened the time to UPCR ≤0.5 mg/mg (372 days vs 169 days; P <0.001) and time to 50% reduction in UPCR (63 vs 29 days; P <0.001). The efficacy of VC for the treatment of non-renal manifestations of SLE has not yet been tested.

Safety

As a class, CNIs are associated with a number of AEs, including hyperglycemia, hyperlipidemia, hyperuricemia, hypertension, dose-dependent nephrotoxicity (including tubular toxicity), hypomagnesemia, gingival hyperplasia, hypertrichosis, gastrointestinal complaints and neurotoxicity. Calcineurin inhibitors are also associated with an increased risk of malignancy and serious infection, which is indicated in the black box warnings in the labels for CsA, TAC, and VC. However, data on the risk of malignancy in SLE patients is lacking. Compared to CsA, TAC is associated with a lower frequency of gingival hyperplasia, hypertrichosis, nephrotoxicity, hypertension and hypercholesterolemia. Tacrolimus also has a good pregnancy safety profile, making it an attractive option for immunosuppression. In AURORA 1, VC demonstrated a favorable safety profile, with an AE profile similar to that of the placebo and consistent with the background MMF regimen and the study population; these promising results need to be confirmed in long-term data.

Guideline Recommendations

The 2019 EULAR guidelines contain the following recommendations for the use of CNIs in SLE:

• Cyclosporine can be used for maintenance of response in patients with hematologic manifestations.
• Topical CNIs are recommended as an option for the first-line treatment of cutaneous manifestations.
• A low-dose CNI in combination with mycophenolate is recommended for cases of severe nephrotic syndrome or incomplete renal response, if none of uncontrolled hypertension, high chronicity index at kidney biopsy and reduced GFR is present.

The publication of the 2019 EULAR-ACR guidelines predates the publication of the AURORA 1 trial results and the FDA approval of VC; therefore, no recommendations for the use of VC are provided.

Rituximab

Rituximab (RTX) is a chimeric mouse/human monoclonal IgG antibody against the B-cell surface antigen CD20. Designed to specifically target malignant B cells, RTX was first synthesized in the 1980s using the then-new hybridoma cell lines capable of producing chimeric antibodies combining the mouse variable region with a human Fc region. Successfully tested in the 1990s in randomized controlled clinical trials (RCTs) in patients with non-Hodgkin lymphoma (NHL), RTX received initial FDA approval for this indication in 1997. Following this, it was quickly repurposed for use in patients with SLE. Although RTX failed in its main efficacy RCTs for SLE, it has demonstrated efficacy in “real world” settings and continues to be used for SLE treatment, particularly for cases refractory to classic immunosuppressants such as azathioprine (AZA) and mycophenolate mofetil (MMF).

Indications and Dosage

Rituximab is FDA-approved for certain hematological cancers (including NHL) and inflammatory diseases (including rheumatoid arthritis [RA]). Although not FDA approved for SLE, RTX is often used off-label, particularly in refractory cases. Rituximab is administered by intravenous injection, with two commonly used dosage regimens: 375 mg/m2/week for 4 weeks (the lymphoma treatment regimen) and two doses of 1000 mg separated by 14 days (the regimen used in RCTs of RTX in SLE). Other than hypersensitivity, there are no contraindications for the use of RTX and no formal studies have investigated the use of RTX in patients with renal or hepatic impairment.

Mechanism of Action

Rituximab binds and inactivates CD20, an antigen expressed on the surface of early immature B cells and mature B cells. Little is understood about the physiological role of CD20, other than that it is a transmembrane protein with a role in calcium intake. It appears to be involved in B-cell development by regulating interactions with the cellular microenvironment. Crucially, it is not known to be expressed on any other cells, making it an attractive target for B cell-specific inhibition. Once bound to CD20, RTX efficiently induces B-cell lysis either via complement-mediated or antibody mediated cellular toxicity (Figure 8-9). The effects of RTX-induced B-cell depletion are long-lasting, with data from patients with RA suggesting that B-cell levels take 6-9 months to increase to pre-treatment levels. Early B-cell reconstitution is indicative that a patient will have a poor response to rituximab therapy.

Efficacy

After the initial successes of RTX in NHL and RA, its efficacy in SLE was tested in a number of small, uncontrolled, open-label trials, including a dose-escalating trial, performed between 2002 and 2009. Encouraging results from these early reports suggested a possible benefit in both SLE and lupus nephritis (LN) and led to the design of two RCTs to formally test the efficacy of RTX in non-renal SLE (the EXPLORER trial) and LN (the LUNAR trial).

The EXPLORER trial was a double-blind, placebo-controlled, multicenter trial that randomized (2:1) a total of 257 patients to either 1000 mg RTX at day 1, 15, 168 and 182 (n = 169), or a matching placebo (n = 88). Eligible patients must have had active non-renal SLE, defined as active disease at screening, ≥1 organ system with a British Isles Lupus Assessment (BILAG) A score (severe disease activity) or ≥2 organ systems with a BILAG B score (moderate disease activity). Patients with severe central nervous system involvement or organ-threatening disease were excluded. All patients were on an immunosuppressant drug – AZA, MMF, or methotrexate (MTX) – and continued their regimen throughout the trial. All patients also received daily prednisone. The primary endpoint was the achievement and maintenance of a major clinical response, a partial clinical response, or no clinical response at week 52, assessed by BILAG score improvement criteria. Despite achieving B-cell depletion in more than 90% of the patients and a median anti-dsDNA reduction of 76% (P = 0.006 compared to the placebo), RTX failed to improve any of the primary (Figure 8-10A) or secondary efficacy endpoints (including, in addition to individual BILAG-assessed response components, the time to the first moderate or severe disease flare, LupusQoL-assessed quality of life, and steroid-sparing major clinical response). Rituximab did result in a higher primary endpoint achievement rate (P = 0.0408) in the African American/Hispanic patient subgroup (Figure 8-10B).

The LUNAR trial was a double-blind, placebo-controlled trial that randomized (1:1) 144 patients to either 1000 mg RTX (n = 72) at day 1, 15, 168 and 182 (the same dosage and administration protocol as in the EXPLORER trial), or a matching placebo (n = 72). Patients with class III or IV (with or without class V) LN – with a supporting renal biopsy – and proteinuria were eligible for enrolment. Patients with ≥50% glomerular sclerosis, interstitial fibrosis, or an estimated glomerular filtration rate (eGFR) below 25 mL/minute/1.73 m² were ineligible. In addition to RTX or placebo, all patients also received a daily dose of MMF, a pulse of 1000 mg methylprednisolone on day 1 and again within 3 days, another pulse of 100 mg methylprednisolone before RTX/placebo administration on days 15, 168 and 182 and daily oral prednisone. The primary efficacy endpoint was renal response – complete, partial, or no response – at week 52, assessed by serum creatinine levels, urinary sediment, and urine protein:creatinine ratio. Secondary endpoints included several other measures of renal response over time as well change in physical function score of the 36-Item Short-Form Health Survey questionnaire. While RTX treatment resulted in peripheral B-cell depletion in all but one patient, the primary (Figure 8-11) and secondary endpoint results were not different compared to the placebo. Although the proportion of responders was numerically higher with RTX (70%) than with placebo (45%) among African American patients, the difference was not statistically significant even in this patient population (P = 0.20).

Explanations for the failure of the EXPLORER and LUNAR trials generally focus on the study design, including the liberal use of glucocorticoids (EXPLORER) and MMF (LUNAR) in both RTX and placebo groups, and the powering of the trials to detect a composite (complete and partial) response, which may have underpowered them for the detection of a partial response.

Notwithstanding this failure in RCTs, many “real-world” reports from retrospective cohort studies support the efficacy of RTX in both SLE in general and renal and hematologic manifestations in particular. Data from the German GRAID registry showed that among 85 patients with SLE who received rituximab, 46.8% achieved a physician-estimated complete response (SELENA-SLEDAI score of ≤2 witha SELENA-SLEDAI flare index of 0), 34.2% a partial response (SELENA-SLEDAI score of ≥4 with no new orworsening of symptoms), and 19.0% no response. A study of 63 Japanese patients with SLE who received rituximab reported that 60% achieved a major clinical response (BILAG C scores or higher in all domains), 25% achieved a partial clinical response (at most one domain with a BILAG B score), and 15% achieved no clinical response. Among 164 European patients with LN who were receiving rituximab, one study reported a complete response (normal serum creatinine with inactive urinary sediment and 24-hour urinary albumin <0.5 g) rate of 30% and a partial response (>50% improvement in all baseline abnormal renal parameters and no deterioration in any parameter) rate of 37% at 1 year of treatment (with 33% non-responders). In a study of 71 French patients with hematologic manifestations – immune thrombocytopenia (ITP) and autoimmune hemolytic anemia (AIHA) – rituximab treatment resulted in a complete response (ITP: platelet count >100 ∙ 10⁹/L; AIHA: normal hemoglobin level in the absence of recent transfusion or ongoing hemolysis) rate of 60.5% and a partial response (ITP: platelet count >30 ∙ 10⁹/L with at least a doubling of the pretreatment value; AIHA: hemoglobin level >100 g/L with at ≥20 g/L increase from the pretreatment value) rate of 25.5%, with a failure rate of 15%.

Safety

Infusion-related reactions and infections are the most common adverse events (AEs) associated with RTX. The prescribing information for RTX contains a black box warning about potentially serious or fatal infusion-related reactions, severe or fatal mucocutaneous reactions, hepatitis B reactivation and progressive multifocal leukoencephalopathy, a severe viral brain infection that causes demyelination. Patients on anti-CD20 therapy, including rituximab, are also at much higher risk of an adverse outcome with COVID-19 infection.

In the EXPLORER trial, the incidence of AEs and serious AE (SAEs) was similar in the RTX (37.9%) and placebo groups (36.4%). Infusion-related SAEs were more common with placebo (17%) than RTX (9.5%), as were treatment-emergent infection-related SAEs (placebo: 17%; RTX: 9.5%). The only common SAE that was more frequent in the RTX group was neutropenia (3.6% vs 0 cases with placebo). In the LUNAR trial, SAEs were more common in the placebo group (40.8%) than in the RTX group (32.9%). Infection-related SAEs were well balanced (19.7% vs 19.2% for placebo and RTX, respectively). Infusion-related SAEs were rare, with 2 cases (2.8%) in the placebo group and 1 case (1.4%) in the RTX group.

Guideline Recommendations

The 2019 EULAR guidelines for the management of SLE state that RTX may be considered for patients with organ-threatening disease that is refractory to standard immunosuppressive agents. The guidelines also suggest RTX as an option for refractory cases of hematological manifestations.

Belimumab

Belimumab (BEL) is a fully human monoclonal IgG antibody that binds B-lymphocyte stimulator protein (BLyS; also known as B-cell activating factor [BAFF]). After the discovery and description of BlyS in the late 1990s, BEL was designed to block its interaction with receptors on B cells. Initial testing in the 2000s led to its approval by the FDA for the treatment of non-renal, non-central nervous system (CNS) systemic lupus erythematosus (SLE) in 2011 – the first new drug approved for SLE in more than 50 years. Research continued in 2010s and 2020s, leading to the approval of BEL for the treatment of childhood-onset SLE in 2019 and for lupus nephritis (LN) in 2020. Belimumab is an important second-line option for patients with SLE who do not respond to standard of care therapy.

Indications and Dosage

Belimumab is approved for the treatment of adult and pediatric patients ≥5 years of age who have active SLE or LN and who are receiving standard therapy. Belimumab has not been tested in patients with severe active CNS lupus, and it is therefore not recommended for the treatment of this manifestation. Belimumab is administered by injection, either as an intravenous (IV) infusion over a period of 1 hour (patients ≥5 years of age) or subcutaneously (SC; patients ≥18 years of age). The recommended IV dosage is 10 mg/kg every 2 weeks for the initial 3 doses and every 4 weeks thereafter. The recommended SC dosage depends on the manifestation: for non-renal SLE, 200 mg once weekly; for LN, 400 mg (two 200-mg injections) once weekly for the first 4 doses, followed by 200 mg once weekly thereafter. Belimumab has been tested in patients with mild, moderate and severe renal impairment; no dosage adjustment is recommended in patients with renal impairment. While no formal trials were performed to assess the effect of BEL on hepatic impairment, no dosage adjustment is recommended in patients with hepatic impairment.

Mechanism of Action

Belimumab binds BLyS, a B-cell survival factor, preventing its binding to its three receptors – BCMA, TACI and BR3 – on B cells. As many as 50% of patients with SLE have increased BLyS levels, and increased BLyS levels have been linked with greater disease activity, organ damage and increased risk of flares. By blocking BLyS activity, BEL inhibits B-cell (including autoreactive B-cell) survival and their differentiation into plasma cells – the main source of autoantibodies (Figure 8-12).

Efficacy

Following a phase I dose-escalation trial in healthy volunteers, BEL was tested in a dose-ranging phase II study with the goal of defining the dosage and the patient population for phase III trials. This randomized, double-blind, placebo-controlled trial assessed the safety and efficacy of three BEL IV doses – 1, 4 and 10 mg/kg – in patients with active SLE (excepting LN and CNS lupus) and a history of anti-nuclear antibodies (ANAs); importantly, eligible patients did not need to be ANA seropositive at screening. Belimumab was efficacious at reducing B-cell numbers and anti-dsDNA titers; furthermore, while it did not significantly affect disease activity in the patient population as a whole, it did significantly reduce disease activity in patients seropositive for ANAs or anti-dsDNA antibodies.

The results of these initial trials defined the patient population for the two phase III trials – BLISS-52 and BLISS-76 – which formed the basis for the initial FDA approval in 2011. Belimumab was also subsequently tested in a SC injection formulation (the BLISS-SC trial), in specific populations (East Asian [the BLISS-NEA trial], Black-identifying [the EMBRACE trial], pediatric [the PLUTO trial]), and in lupus nephritis (the BLISS-LN trial).

The BLISS-52 and BLISS-76 trials had a nearly identical design apart from the length (52 weeks in BLISS-52 and 76 weeks in BLISS-76). Both trials enrolled adult (≥18 years of age) patients with active SLE (ie, meeting the 1997 ACR criteria and with a Safety of Estrogens in Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index [SELENA-SLEDAI] score of ≥6) who were seropositive for ANAs or anti-dsDNA antibodies and were on a stable regimen of prednisone or non-steroidal anti-inflammatory (NSAID), antimalarial, or immunosuppressive agents. Patients with severe LN or CNS lupus were excluded. Patients (867 in BLISS-52, of which 865 received treatment; 826 in BLISS-76, of which 819 received treatment) were randomized (1:1:1) to BEL 1 mg/kg (n = 288 in BLISS-52; n = 271 in BLISS-76), BEL 10 mg/kg (n = 290 in BLISS-52; n = 273 in BLISS-76), or placebo (n = 287 in BLISS-52; n = 275 in BLISS-76), administered by IV infusion on days 0, 14, and 28, and every 28 days thereafter, in addition to the standard of care (SoC). The primary efficacy endpoint was the SRI-4 response at week 52 (a composite of a reduction of ≥4 points in the SELENA-SLEDAI score, no new British Isles Lupus Assessment Group [BILAG] A organ domain score, ≤1 new BILAG B organ domain score, and no worsening in the physician’s global assessment [PGA] score compared to baseline). In BLISS-52, both BEL groups met the primary endpoint (Figure 8-13), with 51% and 58% of patients in the BEL 1 mg/kg and 10 mg/kg groups, respectively, achieving the SRI response at week 52, compared to 44% in the placebo group (P = 0.0129 against BEL 1 mg/kg; P = 0.0006 against BEL 10 mg/kg). In BLISS-76, the SRI-4 response rate was significantly higher in the BEL 10 mg/kg group (43.2%; P <0.05), but not in the BEL 1 mg/kg group (40.6%; P >0.05), compared to the placebo group (33.5%). At 72 weeks (Figure 8-14), the SRI response rate was still numerically higher in the BEL 1 mg/kg (39.1%) and BEL 10 mg/kg (38.5%) groups, but not statistically different from that in the placebo group (32.4%). In BLISS-52, treatment with BEL resulted in significant improvements in steroid-sparing activity and health-related quality of life (assessed by changes from baseline in the 36-item Short-Form Health Survey physical component summary [SF-36 PCS]) at week 52. In BLISS-76, statistically fewer patients who received BEL 1 mg/kg and numerically fewer patients in the BEL 10 mg/kg group experienced one or more flares over 76 weeks and from week 24 to week 76.

The BLISS-SC was a double-blind, placebo-controlled, phase III trial which tested the efficacy of an SC formulation for BEL. The IV administration route is cumbersome for many patients and financially burdensome for the healthcare system, requiring the patients to go to a clinic/infusion center and qualified personnel to administer the infusion; by contrast, SC BEL could be self-administered at home. This formed the rationale for BLISS-SC. Eligible patients (≥18 years of age) must have had moderately to severely active (SELENA-SLEDAI score ≥8) SLE (per the 1997 ACR criteria) and were required to be seropositive for ANAs or anti-dsDNA antibodies; like in BLISS-52/76, patients with severe LN or CNS lupus were excluded. A total of 839 patients (836 of whom received treatment) were randomized (2:1) to receive weekly SC BEL 200 mg (n = 556) or a matching placebo (n = 280), in addition to SoC. The primary endpoint was the SRI-4 response at week 52, which was achieved by 61.4% of patients in the SC BEL group and 48.4% in the placebo group (P <0.0006; Figure 8-15). Belimumab also significantly (P <0.0004) improved the median time to (171 days vs 118 days) and risk of (hazard ratio 0.51) severe flare occurrence, and reduced glucocorticoid (GC) dosage by ≥25% to ≤7.5 mg/day in a greater proportion of patients (18.2% vs 11.9%; P = 0.0732).

The BLISS-NEA trial, a double-blind, placebo-controlled, phase III study, assessed the efficacy of BEL for the treatment of SLE among north-east Asian populations in China, Japan, and South Korea, since the number of patients of north-east Asian descent was low in prior phase III trials. Inclusion criteria included ≥18 years of age, active (SELENA-SLEDAI score ≥8) 1997 ACR criteria-defined SLE, seropositivity for ANAs, and a current, stable regimen of GC, antimalarial, NSAID, or other immunosuppressive therapy. Patients with severe LN, active nephritis, or CNS lupus were excluded. Of the 707 patients randomized (2:1), 471 were assigned to (and 470 received) IV BEL 10 mg/kg and 236 were assigned to (and 235 received) the placebo, along with SoC. The primary endpoint was the SRI-4 response at week 52. Belimumab proved effective in this population, with 53.8% and 40.1% of patients in the BEL and the placebo group, respectively, achieving SRI-4 (P = 0.0001; Figure 8-16). Severe flare risk was reduced by 50% with BEL compared to placebo (P = 0.0004), and the median cumulative prednisone equivalent dose was significantly lower in the BEL group than in the placebo group (4190.0 mg vs 4758.1 mg; P <0.0005).

The EMBRACE trial was a double-blind, placebo-controlled, phase III/IV study designed to test the efficacy of BEL in patients of Black African ancestry – a racial group whose representation in prior BEL trials was not reflective of its SLE prevalence. The trial, conducted in multiple centers in Brazil, Colombia, France, South Africa, the UK and the US, enrolled patients of self-declared Black race. Eligibility criteria included ≥18 years of age, 1997 ACR criteria-defined SLE with a SELENA-SLEDAI score ≥8, and seropositivity for ANAs. Patients with prior exposure to BEL, active nephritis, severe LN, or CNS lupus were ineligible. A total of 503 patients were randomized (2:1), of which 448 received treatment: 299 in the IV BEL 10 mg/kg group, and 149 in the placebo group; all patients also received SoC therapy. The primary endpoint was the SRI-4 response at week 52 with modified proteinuria scoring adapted from the SLEDAI-2K index (SRI-SLEDAI-2K). Although a numerically greater proportion of patients in the BEL group achieved the primary endpoint (48.7%) compared to those in the placebo group (41.6%), statistical significance was not reached (P <0.1068). This precluded declaring any other endpoints as significant. Belimumab appeared to have a greater effect on patients with high disease activity, high anti-dsDNA antibody titers, and low baseline C3/4 levels (Figure 8-17).

The efficacy of BEL for the treatment of pediatric SLE was tested in PLUTO, a double-blind, placebo-controlled, phase II trial. Inclusion criteria included age 5-17 years, 1997 ACR criteria-defined active (SELENA-SLEDAI score ≥8) SLE, and seropositivity for ANAs and/or anti-dsDNA antibodies. Exclusion criteria included active severe LN, active CNS lupus, systemic GC at a prednisone equivalent dose above 1.5 mg/kg/day, and prior B cell-targeted therapy (within 1 year) or BEL exposure (ever). Patients were split into three cohorts based on age and enrolment time: cohort 1 (age 12-17), cohort 2 (age 5-11), and cohort 3 (age 12-17). A total of 93 patients were randomized (5:1 in cohort 1 and 2; 1:1 in cohort 3) to receive either BEL 10 mg/kg (n = 53) or a placebo (n = 40). The primary endpoint was the SRI-4 response rate at week 52; however, because of the low prevalence of childhood-onset SLE, testing for statistical significance was not deemed to be feasible, and thus all results were considered descriptive. Belimumab treatment resulted in a numerically higher SRI-4 response rate (Figure 8-18) at week 52 (52.8%) compared to placebo (43.6%). Severe flare rate over 52 weeks was also 64% lower in the BEL group (17.0%) compared to the placebo group (35.0%), and the median time to first severe flare was longer with BEL (150 days) than with placebo (113 days). No differences in GC dose reduction were observed between the two groups.

Patients with severe active LN were excluded from the trials described above. As LN is one of the most serious manifestations of SLE, a specific trial was designed to assess the efficacy of BEL for LN – BLISS-LN. This multi-national, double-blind, placebo-controlled, phase III trial randomized (2:1) a total of 448 patients with active LN to receive IV BEL 10 mg/kg (n = 224) or a matching placebo (n = 224) in addition to SoC (mycophenolate mofetil [MMF] or cyclophosphamide-azathioprine [CYC-AZA]). Patients must have been ≥18 years of age, had 1997 ACR-criteria-defined SLE, been seropositive for ANAs and/or anti-dsDNA antibodies, had renal biopsy-proven (ie, showing active lesions) class III or IV LN (with or without coexisting class V LN) at screening or class V LN within 6 months before or during screening, and had a urine protein:creatinine ratio ≥1. Patients were excluded if they had dialysis within 1 year before trial start, an estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m², previous failure of both MMF and CYC induction therapy, CYC-based induction therapy within 3 months from trial start, or exposure to B cell-targeted therapy (including BEL) within 1 year before randomization. The primary efficacy endpoint was the primary efficacy renal response (PERR) at week 104, defined as a urine protein:creatinine ratio ≤0.7, an eGFR no worse than 20% below the pre-flare value or ≥60 mL/min/1.73 m², and no use of rescue therapy for treatment failure. Significantly more patients in the BEL group (43%) achieved the PERR response at week 104 compared to patients in the placebo group (32%; P = 0.03); Figure 8-19. Belimumab was also more efficacious than placebo with respect complete renal response (urine protein:creatinine ratio <0.5, eGFR no worse than 10% below the pre-flare value or ≥90 mL/min/1.73 m², and no use of rescue therapy), which was achieved by 30% and 20% of BEL group and placebo group patients, respectively (P = 0.02). Finally, BEL treatment significantly lowered the risk of renal-related event or death (hazard ratio 0.51; P = 0.001).

Safety

The overall clinical trial experience for BEL indicates that it is very well tolerated, with a favorable safety profile that does not appear to include increased infection risk or prolonged immunosuppression. The prescribing information for BEL contains warnings and precautions for serious infections (noting that the rate is not higher with BEL than with placebo), hypersensitivity reactions, depression and suicidality (rare cases reported in trials of BEL), and malignancy (increased risk seen with other immunosuppressive agents, unknown for BEL). In an analysis of pooled data from the initial dose-ranging phase II study and BLISS-52/76, the most common adverse events (AEs), occurring in ≥3% or more of patients and more commonly with BEL than with placebo, included: nausea, diarrhea, pyrexia, nasopharyngitis, bronchitis, insomnia, extremity pain, depression, migraine, pharyngitis, cystitis, leukopenia and viral gastroenteritis (Table 8-9). The most common serious AEs were serious infections, occurring in 6% of BEL-treated patients and 5.2% of placebo-receiving patients.

The safety profile of BEL in north-east Asian (BLISS-NEA) and Black-identified (EMBRACE) populations, as well as in patients with LN (BLISS-LN), was similar to that observed in BLISS-52 and BLISS-76, with no clinically meaningful differences. The safety results from PLUTO indicate that BEL is as well tolerated in children as in adults, and has a comparable AE profile. Data from the BLISS-SC trial showed an AE profile comparable to that from the IV BEL trials, with the exception of local injection site reactions, which were numerically more common in the SC BEL group (6.1%) than in the placebo group (2.5%).

Guideline Recommendations

The 2019 EULAR guidelines for the management of SLE state that add-on treatment with BEL may be considered with an inadequate response (frequent relapses and/or residual disease activity not permitting GC tapering) to a standard of care regimen (HCQ with GC, with or without other immunosuppressants).

Anifrolumab

Anifrolumab is a fully human monoclonal IgG antibody that binds and blocks the type I interferon (IFN) receptor (IFNAR). Rationally designed to interfere with type I IFN signaling which is implicated in SLE pathogenesis, anifrolumab was tested in the late 2010s and early 2020s in several RCTs. It received FDA approval for the treatment of SLE in 2021 – the second new drug (after belimumab) to be approved for SLE in more than 60 years. Although “real-world” efficacy data are still pending, anifrolumab is expected to become an important part of the armamentarium for the treatment of moderate to severe SLE, and particularly its cutaneous manifestations.

Indications and Dosage

Anifrolumab is FDA-approved for the treatment of moderate to severe SLE in adult patients who are receiving standard therapy. Since it has not been tested for the treatment of severe active lupus nephritis (LN) or nervous system lupus, it is not recommended for the treatment of these manifestations. Anifrolumab is administered by intravenous injection, with the recommended dosage of 300 mg every 4 weeks (the dosage regimen from the major RCTs of anifrolumab). No specific trials have been conducted to test the effect of anifrolumab on renal or hepatic impairment. No dosage adjustment is necessary in patients with renal or hepatic impairment.

Mechanism of Action

Anifrolumab binds subunit 1 of IFNAR (IFNAR1), a heterodimeric transmembrane receptor, blocking it from binding to type I IFNs (Figure 8-20). The binding of type I IFNs to IFNAR normally triggers downstream JAK/STAT signaling which induces expression of IFN-responsive genes – the “IFN signature” that characterizes many cases of SLE. Anifrolumab also promotes the internalization of IFNAR1, further disabling IFN signaling. Reduced IFN signaling suppresses downstream inflammatory processes, decreases plasma cell differentiation and normalizes T-cell subpopulations. Anifrolumab contains a triple mutation (L234F/L235E/P331S) in the IgG heavy chain that reduces binding to the FcγR receptor and its downstream function, including antibody-dependent cellular cytotoxicity and complement-dependent cytotoxicity.

Efficacy

After encouraging results from the initial phase I trial in healthy volunteers and several phase II trials in patients with SLE and lupus nephritis (LN), the efficacy of anifrolumab in non-renal/non-neuropsychiatric SLE was tested in three pivotal studies: the phase IIb trial MUSE and the phase III studies TULIP-1 and TULIP-2. The FDA approval of anifrolumab was based on results from these three studies.

The MUSE trial was a double-blind, placebo-controlled, multicenter study that randomized (1:1:1) a total of 305 patients with SLE (according to the 1997 ACR criteria) to receive intravenous anifrolumab 300 mg (n = 99), anifrolumab 1000 mg (n = 104), or placebo (n = 102), administered every 4 weeks for a total of 48 weeks. Eligible patients must have been seropositive for one or more of anti-nuclear autoantibodies (ANAs), anti-dsDNA antibodies, or anti-Sm antibodies, and must have had active SLE (measured by specific Systemic Lupus Erythematosus Disease Activity Index 2000 [SLEDAI-2K], British Isles Lupus Assessment Group [BILAG-2004], and physician’s global assessment [PGA] scores), excluding active severe LN or neuropsychiatric lupus. Patients also had to be on background therapy of at least one of oral prednisone, azathioprine (AZA), an antimalarial, mycophenolate mofetil/mycophenolic (MMF), or methotrexate (MTX). The primary efficacy endpoint was a composite of SLE Responder Index 4 (SRI-4) at week 24 and a sustained reduction in glucocorticoid (GC) use at weeks 12-24 (<10 mg/day and less or equal than the day 1 dose). Secondary efficacy endpoints included the SRI-4 response at week 52 with a sustained GC reduction at weeks 40-52 and reduction of GC dosage at week 52 to ≤7.5 mg/day in patients who were receiving ≥10 mg/day at baseline. Both doses of anifrolumab met the primary and secondary endpoints (Figure 8-21), with significantly more patients in the anifrolumab 300 mg (34.3%; P = 0.014) and the anifrolumab 1000 mg (28.8%; P = 0.063) achieving the SRI-4 with GC taper at week 24 compared to patients in the placebo group (17.6%).

The TULIP-1 trial was the first phase III trial of anifrolumab. This double-blind, placebo-controlled, multicenter study randomly assigned (1:2:2) a total of 457 patients with SLE (fulfilling the 1997 ACR criteria) to receive either anifrolumab 150 mg (n = 93), anifrolumab 300 mg (n = 180), or placebo (n = 184) intravenously every 4 weeks for 48 weeks. Inclusion criteria included active disease (assessed by specific SLEDAI-2K, BILAG-2004 and PGA scores), seropositivity for ANAs, anti-dsDNA, or anti-Sm antibodies, and stable treatment with one or more of prednisone (or equivalent GC), an antimalarial, AZA, mizoribine, MMF or mycophenolic acid, or MTX. Patients with severe LN or central nervous system lupus were excluded. The primary efficacy endpoint was the proportion of patients in the anifrolumab 300 mg and placebo groups who achieved an SRI-4 response at week 52. The SRI-4 response over time is shown in Figure 8-22A; at week 52, there was no significant difference in the proportion of patients in the anifrolumab 300 mg (36%) and the placebo (40%) group who achieved an SRI-4 response (P = 0.418). Since the primary endpoint was not significant, all secondary and other endpoints were, per the pre-specified analysis plan, counted as not significant. Numerically more patients in the anifrolumab 300 mg group (57%) achieved the CLASI response (≥50% improvement in the CLASI [cutaneous lupus erythematosus disease area and severity index] score from baseline), compared to patients in the placebo group (42%); Figure 8-22B. Patients in the anifrolumab 300 mg group also showed numerically higher rates of the BICLA (BILAG-based composite lupus assessment) response (37%) compared to patients who received the placebo (27%); Figure 8-22C.

The second phase III trial of anifrolumab was TULIP-2, likewise a double-blind, placebo-controlled, multicenter study. TULIP-2 enrolled 365 patients with 1997 ACR criteria-defined SLE. After randomization (1:1) to either anifrolumab 300 mg or placebo (unlike TULIP-1, TULIP-2 did not have an anifrolumab 150 mg group) administered intravenously every 4 weeks for 48 weeks, 3 patients dropped out so that 180 patients received the anifrolumab regimen and 182 patients received the placebo. Inclusion criteria were similar to those of TULIP-1, including moderately to severely active disease (measured by specific SLEDAI-2K, BILAG-2004, and PGA scores) and seropositivity for ANAs, anti-dsDNA antibodies, or anti-Sm antibodies. Patients also must have been on a stable regimen of one or more of: prednisone or equivalent, an antimalarial agent, AZA, mizoribine, MMF or mycophenolic acid, or MTX. Exclusion criteria included active severe LN and neuropsychiatric lupus. A secondary endpoint from TULIP-1, the BICLA response at week 52, was chosen as the primary endpoint in TULIP-1. The key secondary endpoints included the BICLA response at week 52 in patients with a high baseline IFN signature, sustained GC dose reduction to ≤7.5 mg/day in patients taking a baseline dose of ≥10 mg/day, ≥50% reduction from baseline in the CLASI score at week 12 among patients with CLASI ≥10 at baseline, ≥50% reduction in the swollen/tender joint count at week 52 in patients with ≥6 swollen or tender joints at baseline, and the annualized flare rate through week 52. Significantly more patients in the anifrolumab group (47.8%) achieved the primary endpoint compared to patients in the placebo group (31.5%; P = 0.001); see Figure 8-23A. A significant difference in favor of anifrolumab (48.0% vs 30.7%; P = 0.002) was observed in the BICLA response at week 52 among patients with high IFN gene expression as well. Compared to patients in the placebo group, a significantly greater proportion of patients in the anifrolumab group also achieved sustained GC dose reduction at weeks 40-52 (30.2% vs 51.5%; P = 0.01) and ≥50% CLASI score reduction from baseline at week 12 (25.0% vs 49.0%; P = 0.04). The proportion of patients achieving ≥50% reduction from baseline in the swollen/tender joint count at week 52 was comparable in the anifrolumab (42.2%) and the placebo (37.5%) group (P =0.55). Although the annualized flare rate was numerically lower in the anifrolumab group (0.43) than in the placebo group (0.64), the difference did not reach statistical significance (P = 0.08); see Figure 8-23B.

A pooled analysis of data from TULIP-1 and TULIP-2 found that patients (n = 360) in the anifrolumab 300 mg group experienced a lower annualized flare rate (0.51) compared to patients (n = 366) in the placebo group (0.67; nominal P =0.017). The median time to first flare was shorter in patients who received the placebo (119 days) than anifrolumab-treated patients (140 days; nominal P = 0.003), and a smaller proportion of patients in the anifrolumab group (33.6%) experienced at least one flare compared to those in the placebo group (42.9%; nominal P = 0.009).

Safety

Anifrolumab was generally well-tolerated by patients with SLE in MUSE, TULIP-1, and TULIP-2. The “Warnings and precautions” section of the prescribing information for anifrolumab lists serious infections, hypersensitivity reactions, and malignancies as possible serious adverse reactions, with malignancies listed on the basis of data from other immunosuppressive drugs rather than anifrolumab itself. Across the three trials, the most common adverse reactions (AEs) that occurred more often with anifrolumab 300 mg (n = 459) than with the placebo (n = 466) and in at least 2% of patients included upper respiratory tract infection (34% vs 23%), bronchitis (11% vs 5.2%), infusion-related reactions (9.4% vs 7.1%), herpes zoster (6.1% vs 1.3%), cough (5.0% vs 3.2%), and hypersensitivity (2.8% vs 0.6%). The rate of serious AEs (SAEs) was lower with anifrolumab (11.8%) than with placebo (16.7%). Vaccination for herpes zoster should be considered prior to starting anifrolumab.

Although “real-world” safety data is still lacking, an analysis of data from the long-term (3-year) extension of the TULIP trials revealed no new safety signals compared to the initial 1-year trial period. The SAE rate remained comparable in the anifrolumab group (22.6%; 683.5 patient-years of exposure) and the placebo group (25.0%; 250.3 patient-years of exposure).

Guideline Recommendations

The current (2019) EULAR guidelines pre-date the approval and marketing of anifrolumab and therefore do not contain recommendations for its use.