PLA2R Antibody Levels and Clinical Outcome in Patients with Membranous Nephropathy and Non-Nephrotic Range Proteinuria under Treatment with Inhibitors of the Renin-Angiotensin System

Patients with primary membranous nephropathy (MN) who experience spontaneous remission of proteinuria generally have an excellent outcome without need of immunosuppressive therapy. It is, however, unclear whether non-nephrotic proteinuria at the time of diagnosis is also associated with good prognosis since a reasonable number of these patients develop nephrotic syndrome despite blockade of the renin-angiotensin system. No clinical or laboratory parameters are available, which allow the assessment of risk for development of nephrotic proteinuria. Phospholipase A2 Receptor antibodies (PLA2R-Ab) play a prominent role in the pathogenesis of primary MN and are associated with persistence of nephrotic proteinuria. In this study we analysed whether PLA2R-Ab levels might predict development of nephrotic syndrome and the clinical outcome in 33 patients with biopsy-proven primary MN and non-nephrotic proteinuria under treatment with blockers of the renin-angiotensin system. PLA2R-Ab levels, proteinuria and serum creatinine were measured every three months. Nephrotic-range proteinuria developed in 18 (55%) patients. At study start (1.2±1.5 months after renal biopsy and time of diagnosis), 16 (48%) patients were positive for PLA2R-Ab. A multivariate analysis showed that PLA2R-Ab levels were associated with an increased risk for development of nephrotic proteinuria (HR = 3.66; 95%CI: 1.39–9.64; p = 0.009). Immunosuppressive therapy was initiated more frequently in PLA2R-Ab positive patients (13 of 16 patients, 81%) compared to PLA2R-Ab negative patients (2 of 17 patients, 12%). PLA2R-Ab levels are associated with higher risk for development of nephrotic-range proteinuria in this cohort of non-nephrotic patients at the time of diagnosis and should be closely monitored in the clinical management.


Introduction
Membranous nephropathy (MN) is the most common cause of nephrotic syndrome in adults. The clinical outcome is variable and ranges from spontaneous remission of proteinuria to end stage renal failure. Statistical models have delineated elevated serum creatinine, male gender, hypertension, older age and high proteinuria as predictors of poor renal outcome [1]. Even though high level of proteinuria at the time of diagnosis represents a risk for loss of renal function, long term follow-up studies have shown that spontaneous remission of nephrotic range proteinuria is a frequent event in patients with MN. Patients with spontaneous remission of proteinuria have an excellent long term renal prognosis and do not need immunosuppressive therapy [2]. Similarly, patients with non-nephrotic proteinuria at onset of the disease, which persists during the follow-up, also have a good long-term prognosis [3]. However, non-nephrotic proteinuria at the time of diagnosis does not always indicate a good prognosis since nephrotic syndrome may develop during the course of the disease [3]. The course of the disease can only be detected by follow-up measurements of proteinuria in these patients. PLA R-Ab are 2 present in many patients with primary MN and high antibody levels are associated with a longer persistence of nephrotic range proteinuria during treatment [4,5]. We therefore prospectively analysed whether PLA R-Ab levels at the time of diagnosis of 2 primary MN are associated with the long-term clinical outcome of patients with non-nephrotic range proteinuria already treated with inhibitors of the renin-angiotensin system.

Patients and study design
Inclusion criteria for participating in this prospective multicenter open clinical study were histologic diagnosis of primary MN, a serum test for PLA R-Ab within six months of renal biopsy, 2 proteinuria of ,3.5 g/24 h and no immunosuppressive therapy prior to inclusion in the study. Secondary MN led to the exclusion of patients from the study. Patients were screened by the treating physicians. This included serologic tests for lupus erythematodes, hepatitis, a detailed medical history and a screening for malignancies depending on the age and risk factors of the patient. During the study follow-up the treating physicians decided on the therapeutic strategy without any recommendations. However, the most common factors for starting immunosuppressive treatments were severe clinical symptoms such as nephrotic syndrome and edema not responding to supportive therapy. A further parameter for start of immunosuppression was the decline of renal function. Since the role of PLA R-Ab levels in predicting the development 2 of disease or treatment success was unclear at the beginning of the study, all treatment decisions were based on the clinical experience of the treating physicians and not the PLA R-Ab levels. 2 PLA R-Ab levels, 24-hour protein excretion and serum 2 creatinine were measured every three months. The study was approved by the local ethics committee of the chamber of physicians in Hamburg and conducted in accordance with the ethical principles stated by the Declaration of Helsinki. Written informed consent was obtained from all participating patients.
Measurement of proteinuria, serum albumin and PLA R-Ab levels 2 Serum levels of PLA R-Ab were measured by an ELISA which 2 was published and validated earlier [6]. According to the manufacturer, the ELISA results were considered positive at a level .20 U/ml. Proteinuria is given as total 24-hour excretion and serum creatinine in mg/dl. In patients who developed nephrotic-range proteinuria during the study follow-up, remission of proteinuria was defined as proteinuria of ,3.5 g/24 h and at least 50% reduction from the highest level of proteinuria prior to remission, complete remission of proteinuria was defined as proteinuria ,0.5 g/24 h. A significant increase of serum creatinine was defined as an increase by $25% compared to the time of study inclusion and a serum creatinine $1.3 mg/dl.

Statistical analysis
Data are given as mean values 6 SD. Statistical significance was defined as p,0.05 (a = 0.05). Survival analysis was done on the endpoint defined as development of nephrotic range proteinuria (proteinuria .3.5 g/24 h) by computing Kaplan-Meier curves for PLA R-Ab positive and negative patients. The Kaplan-Meier 2 curves were compared by log rank tests for significant differences. A multivariate Cox regression analysis was performed on the endpoint development of nephrotic range proteinuria, employing clinical variables measured at the beginning of the study (age, sex, total IgG PLA R-Ab levels, proteinuria) as explanatory variables. 2 The explanatory variable PLA R-Ab levels were ln-transformed 2 prior to this analysis. The strength and significance of effects of the explanatory variables were indicated by hazard ratios and their associated confidence intervals and p values in a forest plot. All statistical analyses were done using SPSS version 21.

Clinical Baseline Characteristics
Thirty-three patients with biopsy proven primary MN and proteinuria ,3.5 g/24 h were included in this study ( Table 1). The time between renal biopsy and study inclusion (first measurement of PLA R-Ab) was 1.2 2 61.5 months. Patients were prospectively followed for 25.368.8 months after recruitment. At the time of study start 30 (91%) patients were treated with an ACE-inhibitor or angiotensin ll-receptor-blocker, 20 (61%) patients received diuretics, 15 (45%) patients received statins and five (15%) patients received anticoagulants, although they had non-nephrotic proteinuria. Two patients were put on treatment with anticoagulants after thromboembolic events before inclusion in the study (one renal vein thrombosis and one lung artery embolism). A third patient had proteinuria of 1.8 g/24 h, but a serum albumin of 19 g/l, which resulted in the treatment with anticoagulants. The additional two patients were treated with anticoagulants because of unrelated conditions. At the time of inclusion in the study 16 (48%) patients had detectable PLA R-Ab in the serum. Proteinuria at inclusion in the study was higher in PLA R-Ab positive 2 2 patients compared to PLA R-Ab negative patients but this difference did not reach statistical significance. Serum creatinine and age were not different between 2 PLA R-Ab positive and PLA R-Ab negative patients. Significantly more PLA R-Ab positive patients received immunosuppressive treatment during the follow-up time. During the follow-up of the study all 33 patients received an ACE-inhibitor or an angiotensin-receptor-blocker. Fifteen (45%) patients received an immunosuppressive treatment during the study follow-up. At study start, 17 (52%) of the 33 patients were negative for PLA R-Ab in the serum, 16 (48%) patients were 2 positive for PLA R-Ab ( Figure 1). 2 Data on the PLA 2 R staining of biopsy specimens are available from eight of the 17 PLA R-AB negative patients. In three of these 2 biopsy specimens an enhanced PLA 2 R staining could be detected, five patients showed no enhancement for PLA 2 R staining.

Development of nephrotic-range proteinuria during follow-up
Within three months after inclusion in the study, proteinuria increased from 2.060.9 g/24 h to 3.463.3 g/24 h in the whole study cohort. This increase was due to the 138% increase in proteinuria in PLA R-Ab positive patients in the first three 2 months of the study (Figure 2). At the same time, in PLA R-Ab 2 negative patients proteinuria remained stable at 1.861.0 g/24 h. At every single time point during the study proteinuria was higher All three patients with enhanced PLA 2 R staining in the biopsy and negative PLA R-Ab remained negative for PLA R-Ab during 2 2 the study follow-up. Two of them did not develop nephrotic range proteinuria and had a complete remission of proteinuria at the end of the study follow-up. The third patient developed nephrotic range proteinuria, but had a spontaneous remission of proteinuria at the end of the study follow-up.
Analysis of PLA R-Ab levels as risk factor for 2 development of nephrotic proteinuria We performed a univariate and a multivariate Cox regression analysis to identify risk factors associated with the development of nephrotic range proteinuria in our patient cohort (Figure 3). The hazard ratios and their confidence intervals in the multivariate analysis indicate that on average, and adjusting for the effects of all other variables, the risk for developing nephrotic-range proteinuria increases by 9% with every year increase in age and by 266% per unit increase in the natural logarithm of the total IgG PLA R-Ab pressants: six patients received alkylating agents plus steroids, three of them were later switched to rituximab; one patient received alkylating agents plus steroids and was later switched to a calcineurin inhibitor plus steroids; two patients received calcineurin inhibitors and were later switched to rituximab; two patients received calcineurin inhibitors in combination with   a serum creatinine of 5.01 mg/dl and 5.32 mg/dl respectively (both 24 months after inclusion in the study). In these patients serum creatinine levels at study start were 1.7 mg/dl and 1.64 mg/dl respectively, showing a major decline in renal function during the follow-up. When considering the PLA R-Ab levels at the end of the study 2 more PLA R-Ab positive patients had no remission of proteinuria 2 and had a significant increase of the serum creatinine during follow-up (six out of eight and five out of eight, respectively) compared to patients who were negative for PLA R-Ab (one out 2 of 25 for both parameters) (Fishers exact test: p,0.005 for both parameters).

Discussion
Proteinuria is the hallmark of membranous nephropathy and it is used to assess the need for immunosuppressive treatment [1]. The outcome of patients with primary MN ranges from spontaneous remission to end-stage renal disease. Immunosuppressive treatment is beneficial concerning remission of proteinuria and maintenance of renal function, but associated with potential severe toxicity [7][8][9][10]. Since patients with high nephrotic range proteinuria have a worse clinical outcome [2], most clinical studies in patients with primary MN focussed on these patients. Only a few studies are available concerning patients with non-nephrotic proteinuria. These studies, however, show that patients with nonnephrotic range proteinuria do not always have an excellent prognosis. In some cases proteinuria increases and can be associated with bad clinical outcome [3,11]. There are no clinical parameters, which predict whether non-nephrotic patients will develop nephrotic proteinuria. Clinical management of these patients depends on a watch and wait approach. Even if nephrotic proteinuria develops, it is difficult to decide whether immunosuppressive treatment should be started or not since some patients may still undergo spontaneous remission. The finding that PLA R-Ab are present in most patients with primary MN and 2 associated with disease activity was a major step forward in the clinical management of these patients [4,5]. In this study, we prospectively analysed the development of proteinuria and clinical outcome in patients with MN and non-nephrotic proteinuria under treatment with inhibitors of the renin-angiotensin system, and assessed which potential role PLA R-Ab might play for the 2 clinical management of these patients.
Although all patients were non-nephrotic at study start, more than half of the patients developed nephrotic-range proteinuria during follow-up. This change resulted in the decision to treat most of these patients with immunosuppressants. Since the study was not blinded, it cannot be excluded that in some cases treatment decisions were influenced by the PLA R-Ab levels, this 2 however, was not part of the study protocol.
Some patients with MN, who were not nephrotic at the time of diagnosis, might experience a progression of their clinical disease activity and perhaps an unfavourable outcome. Nephrotic range proteinuria developed more frequently and faster in PLA R-Ab 2 positive patients compared to PLA R-Ab negative patients. In addition to PLA R-Ab, age was also significantly associated 2 with development of nephrotic proteinuria in the multivariate Cox regression analysis. Male sex on the other hand showed an increased hazard ratio of 4.77 for the development of nephrotic proteinuria, but this was not statistically significant. One variable not analysed in this study was sodium excretion and sodium intake restriction, which might result in changing values of proteinuria over time particularly in patients on ACE inhibitors or angiotensin II receptor blockers. A considerable part of the patients included in this study had low levels of proteinuria but high PLA R-Ab levels. This 2 observation confirms previous data showing that PLA R-Ab levels 2 are not directly correlated with proteinuria at a defined time [5,12]. At the same time, this finding is very important, since serial PLA R-Ab measurements allow to differentiate between two 2 groups of patients: i) patients who were PLA R-Ab positive and 2 nephrotic, but are now undergoing remission and probably do not need immunosuppressive therapy from ii) patients who are PLA R-Ab positive but the proteinuria is still sub-nephrotic and 2 might eventually increase in the future. These patients would need special attention.
The clinical outcome of the three patients with an enhanced PLA 2 R staining in renal biopsy specimens and no detection of serum PLA R-Ab suggests that these patients either had 2 PLA R-Ab before inclusion in the study, but were in immuno-2 logical and clinical remission, or that these patients have an early phase of the disease where all antibodies (present in low levels) are bound in the kidney. The third option would be that a number of patients may have primary MN with circulating PLA R-Ab but 2 their levels are too low to be detected in the serum and can only be diagnosed by a renal biopsy [12].
Several studies have already shown a close association of changes in serum PLA R-Ab levels and proteinuria over time 2 [5,13]. The same observation was made in the present study. Patients who remained positive for PLA R-Ab at the end of the 2 study, had fewer remissions of proteinuria and showed more often an increase in serum creatinine compared with patients who were negative for PLA R-Ab. These data suggest that patients who 2 become PLA R-Ab negative, either spontaneously or under 2 immunosuppressive treatment, have a better outcome than patients remaining positive for PLA R-Ab. 2 A major open question which remains in the management of patients with primary MN is whether and how PLA R-Ab levels 2 could guide immunosuppressive treatment. Prospective, controlled studies will be necessary to show an eventual tight relationship of the PLA R-Ab decrease and remission of proteinuria. Should such 2 studies confirm that patients with non-nephrotic proteinuria and high PLA R-Ab are in an active immunological status of the 2 disease, it would suggest that these patients might benefit from early immunosuppressive treatment upon development of nephrotic proteinuria, without waiting and leaving them for over 6 months on high proteinuria.
In conclusion, our data show that the PLA R-Ab status in 2 patients with primary MN and non-nephrotic proteinuria at the time of diagnosis might be an important marker for the course of the disease. ''1'' shows a statistically significant difference (p,0.05) between PLA R-Ab positive patients and PLA R-Ab negative patients. 2 2 (TIF)