Epidemiology of membranous glomerulonephritis: Clinical, laboratory, and management strategies

INTRODUCTION

Among adults, membranous nephropathy (MGN) remains the most common cause of nephrotic syndrome. Peak incidence occurs in the fourth and fifth decades of life, and overall incidence in adults is estimated at 1.2/100,000 per year.^[1] The natural history of untreated MGN may have a complete remission rate of 20–30% and a 60–80% 10-year kidney survival.^[2] Approximately 75% of MN cases are idiopathic membranous nephropathy (iMN), while secondary causes include autoimmune diseases, infection, drugs, and malignancy. The frequency and etiology of secondary causes vary in different geographic areas.^[2,3]

Primary membranous nephropathy (PMN) is a kidney-specific autoimmune disorder of the glomerulus, typically characterized by heavy proteinuria and a pathognomonic model of glomerular injury.^[4] The phospholipase A2 receptor (PLA2R) is the major target antigen in PMN. Recently, several types of immunoassay systems for anti-PLA2R antibodies have been developed.^[4-8] Proteinuria in PMN is preceded by an antibody response by weeks or months.^[9] PLA2R is an autoantigen present in glomerular podocytes. Combined with the anti-PLA2R antibodies produced by the body, it forms an in situ immune complex that activates the complement system, causing podocyte injury and resulting in the production of urine protein, which is the major pathogenic factor in the majority of PMN patients.^[9] However, several etiologies are involved, including viral infections (hepatitis B virus, hepatitis C virus, and human immunodeficiency virus), which may be associated with anti-PLA2R/THSD7A (Antiphospholipase A2 receptor/Thrombospondin type-1 domain-containing 7A), but these may represent patients who have coincidental PMN. Other possible etiologic factors or toxins, such as drugs, mercury, formaldehyde, and air contaminants.^[2-4]

As time progresses, a maximum amount of immunoglobulin G (IgG)4 and C5b-9 deposition is attained in some patients, which is adequate to trigger enough podocyte injury to raise urine protein excretion and end up with nephrotic syndrome.^[4,10] Clinical remission with proteinuria resolution lags immunologic remission with antibody disappearance by weeks/months.^[4,11] This offset between immunologic and clinical remissions reflects the prolonged time required to form enough deposits to cause proteinuria initially, and the time required to clear subepithelial deposits, repair podocyte and capillary wall damage, and restore glomerular permeability.^[4,11,12]

Although the specificity of the anti-PLA2R assay for PMN is essentially 100%, this finding has somewhat blurred the distinction between primary and secondary disease because some patients with secondary diseases such as hepatitis B and C, cancer , and sarcoidosis have been found to be anti-PLA2R-positive, suggesting the coincidental presence of PMN in some patients with an unrelated systemic disease rather than MN as a manifestation of, or secondary to, the systemic disease.^[4,13,14] However, in most centers, a diagnostic kidney biopsy remains the standard of care, even in anti-PLA2R/THSD7A-positive nephrotic patients.^[15]

Calcineurin inhibitors (CNIs), including cyclosporin (CSA) and tacrolimus (TAC), used either as monotherapy or in combination with low-dose steroids, which is thought to improve response and reduce nephrotoxicity, have also been shown to decrease proteinuria, reduce the rate of loss of kidney function, and decrease anti-PLA2R levels in PMN.^[4,16,17] Advantages of CNIs include the lower incidence of infection and malignancy compared with cytotoxic drugs and the efficacy of monotherapy if steroids are not used. Disadvantages include long-term nephrotoxicity with consequent need to monitor drug levels closely, increased incidence of hypertension and diabetes, especially with TAC, and some recent concerns about whether CNIs are effective at all in preventing end-stage kidney disease in the long term.^[4,16] CNIs have been shown to not only reduce anti-PLA2R levels but also to have a direct effect on stabilizing the podocyte actin cytoskeleton, which reduces protein filtration.^[18]

In Oman, CKD is rising progressively both in incidence and prevalence. There is 1% with severe kidney failure, 9% with moderate kidney failure, and 29% with mild kidney failure.^[19-21] There are 2500 patients on hemodialysis and over 200 on peritoneal dialysis. Additionally, 2,000 patients have undergone a kidney transplant. This study aims to investigate the characteristics and identify the main features of the MGN disease.

MATERIAL AND METHODS

The Royal Hospital is the only center where all patients from all hospitals and regions of the Ministry of Health (MOH) in Oman are referred for biopsy. It utilizes an internationally recognized electronic medical record system called AL SIFFA, which employs the International Classification of Diseases. All clinical, laboratory, and radiological data are collected prospectively. Hence, we carried out an analysis of prospectively collected data. This retrospective study was ethically approved by the Research and Ethics Review and Approval Committee of the Ministry of Health (MoH/CSR/18/9282).

A list of patients with a diagnosis of nephrotic syndrome and/or membranous glomerulonephritis was obtained from the medical records of the Royal Hospital from January 2007 to July 2018. The list was filtered, and patients with a diagnosis of nephrotic syndrome and a histological diagnosis of membranous glomerulonephritis (GN) were reviewed. Furthermore, the Medical Record Department provides an International Classification of Diseases, Ninth Revision, and clinical modification coding guidelines for MN, and these codes have been noted not to be adopted in practice or to provide inadequate distinction of glomerulonephritis.

Data of each patient were followed for four visits in the nephrology clinic at 0, 3 months, 1 year, and 2 years from the time of diagnosis.

The following variables were included for each patient: Age, sex, and weight, which was measured by kilogram, right and left kidney sizes, which were measured in centimeters, and APLA2, and in each visit to have creatinine, glomerular filtration rate (GFR), urea, urine protein creatinine ratio (UPCR), 24 h urine protein, calcium, phosphate, uric acid acid, lipid profile, glucose, hemoglobin, white cell count, and platelet count.

The variables were collected from the computerized health system medical records (Al-Shifa system) which collects data progressively that includes, but is not limited to, clinical, laboratory, and radiological data. These data are classified as normal, high, or low based on our laboratory references, as shown in Table 1.

RESULTS

During the study period, there were a total of 56 patients fitting the inclusion criteria of clinical nephrotic syndrome and histological diagnosis of membranous GN.

Table 2 shows the main characteristics of the patients. The mean (standard deviation [SD]) of age was 42.3^[12] years. There were 54.6% males and 44.5% females. Ninety-two (92%) of patients were Omani nationals and 7% were non-Omani citizens. The mean body mass index was 24.3 (7.1) kg/m². At the time of presentation, 66.04% of patients had their symptoms for >3 months of duration, while only 33.9% presented with symptoms for <3 months. Edema was present in 83.6% at the time of initial presentation. A total of 56.4% had high blood pressure. Type II diabetes mellitus was absence in 92.7%. The mean (SD) of the right kidney length was 10.5 cm, while the left kidney length was 10.6 cm, as per the radiologist’s report of the kidney ultrasound examination.

Table 3 shows the laboratory investigations at 0 time, 3 months, 1 year, and 2 years from the time of the presentation. The mean (SD) of serum creatinine level was 87.3 ± 48 umol/L at 0 time which gradually increased to 89.8 (50.4), 97.8 (61.8), and 100 (69.7) umol/L for which estimated GFR (eGFR) was dropping from 81.1 (17.4), then 79.2 (18.9), 78.1 (19.2), and 77.5 (18.9) mL/min/1.73 m at 3 months, 1 year, and 2 years of follow-up, respectively. Urea level was similarly going up 4.5 (3), 5.3 (2.8), 5.9 (2.7), and 6.1 (3.2) mmol/L from 0 time, 3 months, 1 year, and 2 years of follow-up-up from the time of the presentation, respectively.

The levels of serum albumin/protein were improving throughout follow-up from 21.8 (6.7)/51.7 (9), 26.9 (8.4)/55.1 (9.7), 32.2 (7.1)/63.5 (10.3), and 34.2 (6.4)/64.9 (9.8) g/L at 0 time, 3 months, 1 year, and 2 years from the time of the presentation, respectively. Similar improvements were observed for UPCR, with mean (SD) values of 895.1 (686.8), 574.4 (587.1), 227.7 (328.0), and 207.9 (337.1) at 0 time, 3 months, 1 year, and 2 years from the time of presentation, respectively. Similarly, the 24-hour urine collection improved, with mean (SD) values of 9.3 (7.8), 3.9 (3.6), 2.6 (3.9), and 0.5 (0.73) g of protein at 0 time, 3 months, 1 year, and 2 years from the time of presentation, respectively.

The mean (SD) hemoglobin level was 13.1 (2.2), 12.6 (2.1), 12.6 (2.3), and 12.9 (1.9) g/dL at 0 time, 3 months, 1 year, and 2 years from the time of presentation, respectively. The mean (SD) platelet counts were 317 (73.3), 313.2 (88.7), 293.5 (67.1), and 291.2 (71.8) at 0 time, 3 months, 1 year, and 2 years from the time of presentation, respectively. The mean (SD) of calcium and phosphate levels remained stable at 2.4 and 1.2 mmol/L throughout the follow-up period, respectively. Total cholesterol improved gradually during the visits with a mean (SD) of 7.6 (3) mmol/L at presentation time, 4.1 (1.9) mmol/L at 3 months, 5.5 (2.3) mmol/L at 1 year, and 5.0 (1.8) at 2 years.

Anti-PLA2R antibody was normal in 5.4% and high in 94.6%, with a mean (SD) of 46.5 (72).

Table 4 shows the medication doses in milligrams per day used during each visit. The mean (SD) doses of valsartan were 118.3 (40.9), 116.8 (40.7), 127.5 (39.5), and 126.1 (39.7) mg at 0 time, 3 months, 1 year, and 2 years from the time of presentation, respectively. Similarly, the lisinopril dose was 10.5 (6.1), 11.8 (8.6), 12.0 (6.1), and 12.8 (6.4) mg at 0 time, 3 months, 1 year, and 2 years from the time of the presentation, respectively. Prednisolone mean (SD) dose was 41.2 (21.1), 25.9 (16.1), 11.7 (11.4), and 8.4 (7.4) mg at 0 time, 3 months, 1 year, and 2 years from the time of the presentation, respectively. 22 patients were on valsartan, 25 patients were on lisinopril, and 8 patients were on both. 19 patients received TAC, 11 patients received cyclosporin, and 10 patients received rituximab.

The mean (SD) of TAC (mg/day) was 3.9 (1.7), 3.2 (1.3), 3.6 (1.5), and 3.3 (1.4) mg at 0 time, 3 months, 1 year, and 2 years from the time of the presentation, respectively. The mean (SD) of cyclosporine dose was 245.8 (130), 208.3 (62.4), 180.6 (98.7), and 180 (94.9) mg at 0 time, 3 months, 1 year, and 2 years from the time of the presentation, respectively. The mean (SD) dose of mycophenolate mofetil (MMF) was 1500 (500), 1625 (607.9), 1734.4 (442.2), and 1515.6 (520.1) mg at 0 times, 3 months, 1 year, and 2 years, respectively. The mean (SD) of rituximab dose was 2687.5 (955.6). Cyclophosphamide was not given.

DISCUSSION

To the best of our knowledge, this is the first paper to evaluate the MGN from Oman. It showed that most of the patients were young, of male gender, and mostly presented with nephrotic syndrome after 3 months of initial symptoms. Patients were managed from the time of initial presentation with ACEI (angiotensin-converting enzyme inhibitors) and ARB (angiotensin II receptor blockers), and prednisolone was administered at a suitable dosage throughout their follow-up period. Rituximab was used along with other immunosuppressants for the treatment of MGN. Hence, laboratory tests indicate stable results with preserved kidney function during the follow-up period.

In the present study, most of the patients’ age were between 25 and 64 years, which is the same age observed worldwide. In Saudi Arabia, Alfaadhel et al. reported that Saudi patients were aged between 18 and 65 years of age.^[22] Similarly, Xu et al. in the Chinese population^[23] and in Spain by Jatem Escalante et al.^[24] had reported a similar age range. In addition, a similar age range was reported among USA patients.^[4]

In our study, we showed that the disease is more common among male patients. Published data from Saudi Arabia^[22] showed the same, with an M: F ratio of 1.3:1, but in that, primary and secondary membranous glomerulonephritis were included. China^[23] and USA^[4] have reported more findings of MGN among male population. Additionally, there is a similar gender distribution among the chronic kidney disease population in Oman.^[19,20,25]

Most patients presented with nephrotic syndrome after three months, which can be explained by the gradual progress of subepithelial deposit accumulation and the damage to the podocyte at stage 1 of the disease, and that takes time to manifest clinically. Zou et al. in China report improvements in serum albumin levels and reductions in proteinuria in their patients over an 18-month period.^[26]

At presentation, most of the patients had normal creatinine, which was mildly elevated over time, after 2 years with a mild drop in eGFR. Proteinuria persisted over the same period but improved drastically with proper management, decreasing from almost 9 g/day to 0.5 g/day at the end of the follow-up period, despite patients continuing to have either persistent nephritic or proteinuria status. Similar observations have been made in other countries including China,^[27] Canada,^[28] and Saudi Arabia.^[29] This finding can be explained by the progression of the disease over time.^[4,23,30]

The serum albumin level has improved over time with good management strategies. There were not many studies focusing on this regard, but Zou et al.^[18] showed in their study the dramatic improvement in serum albumin levels with treatment over the time of disease progression.^[26]

Most patients tend to have high total cholesterol and other dyslipidemia abnormalities. This is well shown in our study. This, however, improved over time with various treatments for the main disease, both non-specific and specific. Gupta et al. reported a reduction in total cholesterol levels among their participants when they followed up their clinical and biochemical databases among a group of patients in two large tertiary hospitals in London.^[30]

Anti-PLA2R antibody testing was performed at our hospital, which is the only hospital designated by the MOH to handle kidney biopsies and their various management. The histopathology findings in the kidney biopsies revealed thickening of the basement membrane in all cases, as observed by light microscopy. Spikes, which are a result of subepithelial immunocomplex deposition, were reported in 35 patients. In immunofluorescence, granular capillary staining of IgG was observed in 14 patients, and APLA2R was stained in all kidney biopsies of the study population. In electron microscopy, foot process effacement with spikes was observed in 35 patients. Almost 95% of patients had positive anti-PLA2R antibodies, confirming the diagnosis of MGN, and this test was highly sensitive for diagnosing MGN without the need for a kidney biopsy. Clinical trials have shown that the anti-PLA2R antibody is sensitive to approximately 70% of patients with MGN.^[31]

The well-known immunosuppressant regimens include cyclophosphamide/steroids, calcineurin inhibitors (CNIs)/steroids, and rituximab. In our study, the latter two regimens were used. The CNIs were continued for the study period, and it is possible that immunological responses may take a long time for some patients, which could be due to the pathophysiology of the disease, which is not the same in all cases. Universally, the same regimens were used.^[32-38] Fervenza et al. reported that both regimens that we used in our study were effective in treating MGN.^[39] They used rituximab with two infusion, 1000 mg each, administered 14 days apart; repeated at 6 months in case of partial response or oral cyclosporine (starting at a dose of 3.5 mg/kg of body weight per day for 12 months)^[39] but in our study, we used rituximab dose as 375 mg/m² once weekly for a total of 4 doses and cyclosporine with an initial dose of 2.5 mg/kg/day in two divided doses.

This indicates that real-life clinical practice differs from that conducted in clinical trials. Indeed, the management based on clinical trials sounds almost perfect, but their implementation in clinical practice might prove difficult. Hence, the reporting and publication of such data are of utmost importance to further characterize our patients. A more difficult task would be to educate those subsets of patients who are achieving the optimal goals in their management strategies.

Rituximab, an anti-CD20 agent, is the preferred option in the management of primary membranous glomerulonephritis (PMGN) due to its ease of use, fewer side effects, and the absence of the need for monitoring. The effect of rituximab is similar to that of CNI, but the relapse rate is higher with CNI, as noted in the MENTOR Study. Despite rituximab being a good treatment option, it carries a risk of resistance and may be related to a high level of plasma cells that continuously produce IgG. For this reason, targeting plasma therapy (bortezomib and daratumumab) can be considered as a future treatment option.

Non-specific treatment (ACE-I and ARB) for MGN was prescribed in our patients, and some patients were on dual blockers. We used valsartan and lisinopril at different doses, with maximum doses of 160 mg/day and 20 mg/day, respectively. However, the use of non-specific treatment is not without a cost, especially in patients with chronic kidney disease, and reduces GFR as the risk of blocking the response of maintained an angiotensin II-induced increase in resistance at the efferent (post-glomerular) arteriole with an inhibitor of the renin-angiotensin system will sequentially relax the efferent arteriole, lower intraglomerular pressure, and reduce GFR and could be that the explanation of some patients who reached remission but had dropped in their eGFR.^[40]

Cattran et al. used combined dual blockers in a group of their patients who had spontaneous remission, 56 out of 196 (28%).^[41]

About 8% of our population was also diabetic, but their histological findings were going on with iMN. Both diabetic nephropathy (DM) and primary MGN present with nephrotic range proteinuria and hypoalbuminemia, but in the former, it indicates progression of the disease, and it gives histological characteristics of the secondary form of MGN.^[42] Moreover, the two diseases can coexist. Bhadauria et al. reported 16 patients with primary MGN out of 134 patients with DM.^[43]

The present study revealed certain clinical characteristics of MGN which are consistent with global trends. The prompt and early management of the MGN resulted in good clinical and laboratory changes over the study period. However, there is a need to increase the period of follow-up and participate in regional and international studies to have a more generalized statement and to improve the care of such patients.

All adult patients with idiopathic nephrotic syndrome should be screened initially for anti-PLA2R/THSD7A antibodies, as well as for the common causes of secondary MN, including hepatitis B and C, lupus, and sarcoid.

CONCLUSION

The characteristics of MGN align with global trends. However, since this is a single-center study, a national study may be necessary to gather more comprehensive information about the disease’s characteristics.