Big MAC attack: Outcomes in membranous nephropathy & membrane attack complex on biopsy

Written by: Brian Rifkin, MD, Ana Cecilia Farfan Ruiz, MD MSc & Miquel Blasco Pelicanom MD, PhD

Inforgraphics by: Brian Rifkin, MD & Corina Teodosiu, MD

AcademicCME (www.academiccme.com) is accrediting this educational activity for CE and CME for clinician learners. Please go to https://academiccme.com/kicr_blogposts/ to claim credit for participation.

Membranous nephropathy (MN) is responsible for nearly 30% of nephrotic syndrome presentations and remains one of the most common causes of nephrotic syndrome among non-diabetic adults, with an estimated incidence of 8-10 cases per 1 million in Western countries. Most patients present with the triad of nephrotic range proteinuria, edema and low serum albumin (with or without kidney failure). High blood pressure, elevated cholesterol, and the potential risk for thrombosis are also often present in patients with membranous nephropathy.

70-85% of MN cases are classified as primary membranous glomerulonephritis— previously known as idiopathic. Secondary MN accounts for roughly 30% of patients with MN and typically occurs in the setting of autoimmune diseases, infections, malignancies, or with certain medication exposures. (FIGURE 1)

Figure 1.

Much of what we know of the pathogenesis of MN comes from the classic Heymann Nephritis (HN) model from 1959, where the glycoprotein 330 (megalin) was identified as the antigen involved in the autoimmune process observed in rat models. Megalin provoked podocyte injury by forming immune complexes that activated complement.

In humans, autoantibodies against the M-type phospholipase A2 receptor on podocytes were eventually discovered, and were present in 70-80% of patients with MN without a concomitant secondary disease. The additional discovery of other podocyte antigens causing MN, and their respective antibodies, have been shown to be associated with malignant tumors: thrombospondin 1 domain- containing 7 A (THSD7A) discovered in 2014 and NELL-1 (neural epidermal growth factor- like 1). The list of MN associated antibodies continues to grow in recent years with the identification of EXT1/EXT2 in 2019, Semaphorin 3B and Protocadherin 7 (PCDH7) in 2020, and serine protease HTRA1 in 2021.

Subepithelial immune deposits occur in situ when circulating antibodies bind to intrinsic, fixed antigens forming dense “spike” antigen-antibody complexes visible on electron microscopy. (FIGURE 2)

Figure 2.

The immune complexes serve as activators that trigger the formation of C5b-C9 complexes (FIGURE 3) or membrane attack complexes (MAC), that insert on the glomerular epithelial cells membrane resulting in damage. This, in turn, stimulates release of proteases and oxidants by the mesangial and epithelial cells, damaging the capillary walls and causing them to become "leaky". On light microscopy, there is glomerular basement thickening as evidenced by epimembranous spikes on silver stain. IgG1 and IgG3 deposits are seen with immunofluorescence earlier in the course of the disease. Later IgG4 deposits can be demonstrated along the capillary loop, in addition to the deposition of C3.

Figure 3.

In terms of prognosis, it is generally understood that about a third of untreated patients have spontaneous remission, another third progress to require dialysis and the last third continue to have proteinuria, without worsening kidney function. Features that have classically been taught to be indicative of progressive disease on kidney biopsy include the presence of segmental sclerosis including the extent of interstitial fibrosis and tubular atrophy. In addition, in order to define the initial approach to treatment in these patients, there are clinical and laboratory risk factors for disease progression that have been determined. These include PLA2R levels, the amount of proteinuria, baseline kidney function and life-threatening nephrotic syndrome. (FIGURE 4) Given that spontaneous remission occurs in 1/3 of patients with 32% occurring during the first 14 months after presentation (including spontaneous remission in patients with proteinuria of up to 12 g/day), we need better tools to decide which patients to treat early and aggressively with immunosuppressive therapy.

Figure 4. KDIGO 2021 Guidelines Chapter 3, Membranous Nephropathy

PLA2R antibodies levels have been found to correlate to the amount of proteinuria and the likelihood of remission. However, a single level of PLA2R antibodies matched to a single quantity of proteinuria is of limited value, as this may only reflect a phase of the disease. As explained in the review byLerner et Al, primary membranous nephropathy can be divided into 5 phases (FIGURE 5) that include PLA2R antibodies (deposits in the kidney and ultimately appearing in the serum) and proteinuria (as a marker of podocyte injury that takes time to resolve even after suppression of autoimmune activity) as immunological and clinical parameters. Complement activation, which is a key component in this process, could potentially help with stratification of treatment and prognosis.

Figure 5. Five phases of membranous nephropathy

In the study by Teysseyre et al, they theorize that components of the complement system seen on biopsy could be used as markers of disease in MN. As a reminder, C3 is involved in the membranous lesion. The intensity of its deposition is associated with the severity of the level of eGFR at the time of the biopsy in the glomeruli of patients with MN, and also seemed to be correlated with the level of proteinuria. The formation of MAC is central to the pathogenesis of glomerular damage and back in 1992 Brenchley et al, described an association between high urinary C5b-9 levels and unstable clinical courses in MN. The use of urinary MAC excretion as a prognostic marker has also been studied. However, it is not specific for MN and can be present in diabetic nephropathy and FSGS with nephrotic range proteinuria. Also complement markers of activation (like C3dg) can be derived from the renal tubules and not from the glomeruli, as in tubulointerstitial nephritis, reminding us that C3 synthesis occurs in proximal tubular cells, hence the expression of MAC.

The authors retrospectively reviewed a database from the Department of Pathology of Montpellier University Hospital, France. To be included, patients had to have undergone a kidney biopsy confirming MN from December 2004 to December 2015. Exclusion criteria included: age < 18 years, kidney transplant, patients lost to follow-up, insufficient biopsy (< 2 glomeruli), MN associated with another nephropathy, and patients with an uncertain diagnosis. Historical samples were stained for PLA2R, anti-human C4d, and anti-human C9 neoepitope (which is highly specific to C5b-9 fixation in the membrane). Two renal pathologists, who were blinded to the clinical data, graded the intensity of staining from negative to strong (0-3+). There was no discordance between the two pathologists regarding positivity of the staining in the given set of samples.

The primary endpoint was renal survival from the date of the biopsy until the date of renal failure, as defined by an estimated GFR <30 ml/min/1.73m2 calculated by the CKD-EPI formula. Secondary endpoints were clinical remission of the nephrotic syndrome (partial or complete) as defined by the 2012 KDIGO guidelines at 6 months, 12 months and the end of follow-up.

64 patients’ biopsies were included for study, including 45 with primary MN and 19 with secondary MN (FIGURE 6) with a median follow-up of 94.5 months. Patients were primarily male (59.4%) with a median age of 54 years. 65.6% had nephrotic syndrome at the time of biopsy. All but one patient received renin-angiotensin-aldosterone system (RAAS) inhibitors. Forty-five (70.3%) received at least one immunosuppressive therapy during follow-up (corticosteroids, mycophenolate mofetil, alkylating agents and/or rituximab).

Figure 6.

C5b-9 glomerular staining was positive in 29 patients (45.3%). The deposits were granular and diffuse and located in the subepithelial space (FIGURE 7). C5b-9 staining was weak (1+) in 17 patients, moderate (2+) in 9 patients and strong (3+) in 3 patients. There was no C5b-9 staining positivity in the control kidney biopsy. C5b-9 positive patients had higher proteinuria and lower eGFRs compared with C5b-9 negative patients. There were no other significant differences in patient characteristics or administered treatments. More importantly, there were no differences in classical histologic grading characteristics including vascular lesions, interstitial fibrosis and tubular atrophy between groups. This showed that classical markers of severe disease on kidney biopsy really did not help in differentiating outcomes.

Figure 7.

C5b-9 positive patients did have significantly decreased remission rates (as defined by level of proteinuria) at 6 months, 12 months and at the end of follow-up (FIGURE 8). In addition, analysis showed a faster time to remission in patients without C5b-9 glomerular deposition, a difference of about 9 months on average. Of the 45 patients receiving immunosuppression, 26 achieved remission, including 38.5% with C5b-9 deposits versus 61.5% without deposits. Overall, glomerular deposition of C5b-9 was associated with a lower remission rate even after corrections for eGFR and proteinuria at baseline. C5b-9 deposits were also strongly associated with renal failure. 51.7% (15/29) of the patients with deposits versus 11.4% (4/35) progressed to an eGFR < 30 ml/min/1.73 m2 (p= 0.0004). Baseline eGFR and glomerular deposition of C5b-9 were most strongly associated with renal failure.

Figure 8.

The complement system is being increasingly recognized as having a significant role in patients with glomerulonephritis. In this study, C5b-9 positive patients had more severe nephrotic syndrome at diagnosis, more treatment failures and lower renal survival rates than patients without glomerular deposition of C5b-9. Furthermore, in this study of MN nearly 95% of patients had staining with C3 and C4d, suggesting that complement activation is almost universal, even though only 45.3% of MN patients demonstrated C5b-9 deposits on biopsy. “Complete” activation of the complement cascade and MAC may carry a more devastating prognosis. Complement is regulated at both the systemic and glomerular levels. Podocytes express complement regulatory proteins that may incompletely activate the complement cascade. Similarly, glomerular lesions may be secondary to complement activation in addition to antibody mediated mechanisms.

As more therapeutics are developed to interrupt complement activation, it is possible that C5b-9 positivity identified on biopsies may help guide which subset of MN patients could benefit from such interventions. Eculizumab, a C5 binding agent that blocks the terminal formation of MAC, previously did not show benefits in patients with MN, however the doses regimens used in the trial were lower than the usual for complement mediated disease (C3 glomerulopathy or atypical HUS). It is alsopossible that treating only the subpopulation of patients with C5b-9 positive biopsies might give more significant results, and warrants further study. In addition, we currently have two new C5 blockers (ravulizumab, cemdisiran), as well as new molecules with action at different complement levels. Among them, there are three phase II clinical trials for MN (pegcetacoplan, BCX9930, Narsoplimab). (FIGURE9) As previously described inKI reportscommunity blog, iptacopan, a new potent oral and highly selective factor B inhibitor of the alternative complement pathway is being evaluated for C3GN but could eventually have broader use in other glomerular diseases. The authors note several limitations of the current study including: retrospective nature, single center, number of patients lost to follow-up, historical treatments (low rituximab use), inclusion of secondary MN (mostly SLE), serum PLA2R not evaluated (no serum bank), and use of a single kidney biopsy as a control. The authors intend to replicate this study with a more modern cohort. If this study can be reproduced, the addition of C5b-9 glomerular staining on MN biopsies could be a revolutionary prognostic, and hopefully, therapeutic indicator.

Figure 9.

AcademicCME (www.academiccme.com) is accrediting this educational activity for CE and CME for clinician learners. Please go to https://academiccme.com/kicr_blogposts/ to claim credit for participation.