The Use of Serological Tests in the Care of Patients with Lupus Nephritis.

Abstract

Introduction Lupus nephritis is a frequent and serious manifestation of SLE, occurring in 50% of patients. Early recognition and treatment of lupus nephritis are essential to control glomerular and tubulointerstitial inflammation that would otherwise progress to chronic damage and kidney failure. There are now several options for treating lupus nephritis, including two newly approved drugs. However, managing lupus nephritis after the initial episode, which is generally informed by a kidney biopsy, is difficult. This difficulty arises because proteinuria and eGFR, the currently available clinical biomarkers of kidney disease activity, cannot accurately distinguish between ongoing inflammatory injury and chronic kidney damage. Serologic markers, such as complement system fragments and autoantibodies, that reflect the overall activity of the immune system in an individual patient help inform lupus nephritis management, but these too are plagued by limitations. Patient 1 A 29-year-old woman with SLE complicated by class 4 lupus nephritis maintained in remission on mycophenolate mofetil (MMF) 750 mg twice daily and hydroxychloroquine presents for routine follow-up. She complains of worsening fatigue. Her BP is 125/70 mm Hg. Complement component C3 and C4 fell from 95 to 70 mg/dl (reference range: 87–200) and from 20 to 14 mg/dl (reference range: 18–52), respectively. Testing for previously positive antinuclear antibody (ANA) and antidouble-stranded DNA (anti-DsDNA) antibody shows that they are negative. Her eGFR is 120 ml/min per 1.73 m2. Her urine sediment shows five eumorphic red blood cells per high-power field and several acanthocytes. Her urine protein-creatinine increased from 0.4 g/g at her last visit 4 months ago to 2 g/g. Infectious workup is negative. Discussion of Patient 1 The classic and alternative complement pathways play an important and complex role in the pathogenesis of lupus nephritis. An acute decrease in circulating C3 and C4 reflects complement consumption and is thought to be associated with active nephritis. However, in general, hypocomplementemia can be multifactorial in SLE/lupus nephritis due to variations in synthesis and individual genetics. In an unadjusted analysis when compared with mean baseline complement levels, C3 and C4 decreased significantly at lupus nephritis flare, but the sensitivity and specificity of these changes (75% and 71%, respectively, and 48% and 71%, respectively) were modest. In the same study and by multivariable regression analysis, only a lower level of C3 was independently associated with lupus nephritis flare (1). Comparing the National Institutes of Health activity index of lupus nephritis with complement levels, decreased C3 was more pronounced in patients with high histologic activity (activity index greater than ten) than those with low to moderate lupus nephritis activity (2). Relying on C3 and C4 levels as biomarkers of flare in lupus nephritis remains controversial as neither test can confirm active nephritis, especially in lupus nephritis class 5 where complement levels can remain normal. Interestingly, our patient has serologic conversion to negative ANA and anti-DsDNA. Although ANA is a hallmark of SLE, it is nonspecific and may be absent at times in up to 30% of patients (3). On the other hand, anti-DsDNA is a specific biomarker for SLE and is a component of the SLE disease activity index. Despite that, only about 50% of patients with SLE ever test positive for anti-DsDNA. It is challenging to distinguish between absence of autoantibody due to natural disease fluctuations, the effect of therapy, and variability attributed to testing methods. Nonetheless, the negative autoantibody testing in this patient, in particular for anti-DsDNA, does not rule out active nephritis but may affect the choice of therapies. The drops in complement levels and in hematuria and proteinuria are suggestive of active lupus nephritis flare in this patient with known remote class 4 lupus nephritis. Whether this patient has active proliferative lupus nephritis, class 5 lupus nephritis, or both is not clear as class transition may occur. Regardless, this will not affect our decision to intensify immunosuppression in this case. For this reason and given the high suspicion for lupus nephritis flare, we will defer kidney biopsy and proceed with adjusting therapy, unless enrollment in an lupus nephritis drug trial is feasible. Available treatment options include (1) cyclophosphamide, (2) a higher dose of mycophenolate, and (3) combined drugs, such as belimumab plus either cyclophosphamide or mycophenolate or voclosporin plus mycophenolate. Both belimumab and voclosporin were recently Food and Drug Administration approved for active lupus nephritis. Cyclophosphamide should be avoided if possible to preserve fertility. Given that this is not the first active lupus nephritis episode in this patient, the addition of belimumab might be of benefit. The Belimumab in Lupus Nephritis (BLISS-LN) trial showed that the addition of belimumab to background therapy not only improved response but also decreased the risk for future flares (4). However, subgroup analysis of the phase 2 belimumab trials and post hoc analysis of the phase 3 belimumab trials showed that patients with positive ANA and anti-DsDNA had better response to belimumab compared with the serologically negative patients (5). Mechanistically, this makes sense, as Blys levels are higher in serologically active patients. Hence, adding belimumab in this particular case might not be the best approach. In addition to steroids, we can either increase the dose of MMF or add volcosporin. The latter regimen may be more beneficial given the need of less steroid exposure, the rapid drop in proteinuria, and superior chances of remission. Patient 2 A 33-year-old woman with biopsy-proven lupus nephritis (3 + 5) is maintained in clinical remission on MMF for 36 months. Recent laboratory tests showed no anti-DsDNA antibodies, a C3 level that had returned to normal, and persistently low C4 at 12 mg/dl (reference range: 18–52). Her kidney function is normal, and urine protein-creatinine is 0.35 g/g, with no hematuria. She wants to discuss discontinuing immunosuppression. Discussion of Patient 2 Serial monitoring of serum C3 and C4 is used in assessing disease activity and as a guide to clinical management of SLE/lupus nephritis; however, the predictive value of histologic activity/quiescence is modest at best (6,7). Concordance values between normalization of C3 and C4 at the time of repeat biopsy and histologic status are only about 0.56 and 0.66, respectively (7). Although chronically low C4 is uncommon, it is a risk factor for SLE. Early classic complement components are important for clearance of cellular debris. Low C4 impairs this clearance and permits exposure of antigen-presenting cells to self-antigens and production of autoantibodies with development of SLE/lupus nephritis or SLE-like diseases. About half of the patients with SLE and complete C4 deficiency develop lupus nephritis at some point. However, isolated low C4 is not necessarily associated with lupus nephritis activity, especially if it is persistent (8), and can be the result of altered C4 gene copy number (9). This patient appears to be in clinical remission for about 3 years, but given the discordance between serologies and histology, we cannot confirm that she is in histologic remission. The dilemma is between continuing unnecessary immunosuppression and predisposing her to a lupus nephritis flare that may occur in 25%–30% of patients after withdrawal of therapy (10) and that is attributed, for the most part, to residual histologic activity (6). This risk of flare decreases to 9%–13% in patients with confirmed histologic activity index of zero at the time of withdrawal of therapy (6,11). After discussing the data with the patient, she opts to taper off mycophenolate and remain under close surveillance for future lupus nephritis flares. Patient 3 A 39-year-old woman who completed induction therapy with low-dose (Euro-lupus) cyclophosphamide for active lupus nephritis class 4 (activity index 12/24) presents at 16 months for follow-up while on MMF. Her serum creatinine is 1.2 mg/dl. Before lupus nephritis diagnosis, serum creatinine was 0.9 mg/dl; at diagnosis, serum creatinine was 1.3 mg/dl; and 12 months after treatment was started, serum creatinine was 0.9 mg/dl. Her urine protein-creatinine is 0.9 g/g compared with 4 g/g at lupus nephritis flare. Her urine sediment currently has three acanthocytes per high-power field. Anti-DsDNA remained positive, C3 increased from 70 to 86 (reference range: 87–200), and C4 increased from 16 to 19 mg/dl (reference range: 18–52). Discussion of Patient 3 In summary, after 16 months of therapy, kidney function has not improved and has fluctuated considerably. Proteinuria improved but did not fall below 0.5 g/d. Although this proteinuria cutoff is considered a clinical response criteria in lupus nephritis, it does not correlate with histology. Conversely, ongoing proteinuria does not distinguish ongoing histologic activity from chronic kidney damage. Protocol kidney biopsies after 6 months of induction therapy showed that 62% of patients with histologic remission had 24-hour proteinuria >0.5 g/d. As mentioned in the previous two patients, circulating anti-DsDNA and complement levels are not strong diagnostic markers of lupus nephritis activity. This patient’s complement levels almost normalized, but this does not help in deciding whether to intensify therapy for ongoing active lupus nephritis or intensify conservative therapy to control proteinuria attributed to chronic kidney damage. During the visit, her BP was relatively elevated at 120/75 mm Hg from her usual levels of about 105/65 mm Hg. This and the fluctuation in serum creatinine raise the concern for vascular lesions in addition to lupus nephritis–related inflammatory damage. Thrombotic microangiopathy is not uncommon in patients with SLE. Antiphospholipid antibodies and lupus anticoagulants were ordered and were found to be positive. The patient underwent another kidney biopsy. This biopsy showed improved but persistently active lupus nephritis (activity index 5/24 and chronicity index 3/12) along with an acute thrombotic microangiopathy lesion. Immunosuppression was intensified to improve the chance of achieving remission and minimize accrual of chronic damage. Additionally, anticoagulation with warfarin was started to treat antiphospholipid nephropathy. The current clinical serologies used to manage SLE (anti-DsDNA, C3, and C4) do not strongly correlate with lupus nephritis histology, so we continue to rely on the kidney biopsy as the gold standard diagnostic and management tool in lupus nephritis. Identifying and validating novel biomarkers that will help in the stratification, prognostication, and management of lupus nephritis are essential. Disclosures I. Ayoub reports receiving honoraria from ACR and serving as a scientific advisor or member of the Journal of Clinical Nephrology Editorial Board and the Lupus Foundation of America Advisory Board. B.H. Rovin reports consultancy agreements with Alexion, AstraZeneca, Aurinia Pharmaceuticals, Biogen, Bristol Myers Squibb (BMS), Calliditas Therapeutics, Chemocentryx, EMD-Serono/Merck, Genentech, Human Genome Sciences (GSK), Idorsia, Janssen, MedImmune, Morphosys, Novartis, Omeros, Retrophin, RILITE Foundation, Roche, and Vistera; receiving honoraria from Alexion, AstraZeneca, Aurinia Pharmaceuticals, Biogen, BMS, Calliditas Therapeutics, Chemocentryx, EMD-Serono/Merck, Genentech, Human Genome Sciences (GSK), Idorsia, Janssen, Morphosys, Novartis, Omeros, Retrophin, RILITE Foundation, Roche, and Vistera; serving as a scientific advisor or member of the American Society of Nephrology (ASN) Kidney Week, CureGN, Kidney Disease Improving Global Outcomes, Kidney International, Kidney International Reports, the Lupus Foundation of America, Nephrology Dialysis and Transplantation, and UpToDate; and work with ASN (mostly educational courses), work with the National Kidney Foundation and the International Society of Nephrology, and work with Lupus Foundation of America. Funding This work was supported by National Institutes of Health grant RO1 AR071947 (Predictive Biomarkers of Disease Activity and Organ Damage in Patients with Lupus).