A Biomarker for Calcineurin Inhibitor Responsiveness in Primary Membranous Nephropathy

Viewpoint on:

Yu X, Cai J, Jiao X, Zhang S, Liu H, Ding X: Response Predictors to Calcineurin Inhibitors in Patients with Primary Membranous Nephropathy.Am J Nephrol 2018;47:266–274

The ability to reliably and accurately predict the response to treatment of glomerular disease with specific regimens would be a great advance. Factors such as renal function, scarring on renal biopsy analysis, urine protein composition, and genetic background (for focal segmental glomerulosclerosis [FSGS]) are already used for this purpose, but an easily accessible serum biomarker with high predictive capability would be a practice altering development.

Yu and co-workers have attempted to fill this gap by conducting a novel pilot study in calcineurin-inhibitors (CNI) treated patients with presumed primary membranous nephropathy (PMN) followed for at least 12 months for a remission (complete or partial) or no remissions. About 90% of the patients were anti-phospholipase A2 receptor (anti-PLA2R) antibody positive by ELISA. Since all patients with PMN studied were treated with a CNI it is not possible to determine how many of the observed remissions were “spontaneous” or whether the predication based on one or more biomarkers also applies to other treatment regimens (such as rituximab or cyclophosphamide) or whether the findings are specific only to CNI treatment. These are serious limitations of the study.

Using nano-HPLC-MS technology they were able to identify several differentially expressed serum biomarkers predictive of remission or no remission following CNI therapy. The leading candidate was serum amyloid A1 protein (SAA-1) – a well-known acute phase reactant synthesised in the liver in response to interleukin-6. Decreased levels of SAA-1 were found in non-responsive patients compared to those with a complete remission. The area-under-the-urve of a receiver operating characteristic analysis (AUROC) was 0.91 for SAA-1, 0.93 for anti-PLA2R, and 0.96 for the combination of SAA-1 and anti-PLA2R, indicating an additive effect when two biomarkers were combined, but also showing that SAA1 and anti-PLA2R are comparable for prediction of remission in CNI treated subjects with PMN (see Figure 1). However, it must be recognized that the initial response to CNI in PMN is very high, so long as the treatment is continued, but relapses very commonly develop when the treatment is discontinued and as stated above, the study design cannot discern the outcomes in terms of treatment induced or a spontaneous remission. It is possible that CNI exhibited their effects on remission locally by altering podocyte biology rather that promoting an immunological remission of PMN. The high relapse rate after stopping CNI in PMN attests to this possibility. Interestingly SAA-1 serum levels correlate inversely with anti-PLA2R antibody levels in PMN.

Figure 1
Figure 1: ELISA validation of serum SAA1 and its diagnostic value. a SAA1 levels (ELISA) in PMN no remission (NR), PMN CR, healthy controls, and MCD patients. * p < 0.05 versus control group; # p < 0.05 versus MCD group; Δ p < 0.05 versus CR group. b ROC curve of SAA1, PLA2R antibody, and their combination for the detection of CNI NR PMN patients. c ROC analysis of SAA1, PLA2R antibody, and their combination for the detection of CNI no remission PMN patients. The AUC of SAA1 and PLA2R antibody are 0.901 and 0.934 respectively. With the combination of SAA1 and PLA2R antibody, the AUC value from ROC curve analysis is 0.956.

Despite its weaknesses, this study is of interest as it provides a glimpse of how a personalized approach to treatment of PMN may evolve in the future. Perhaps a panel of highly validated serum / urine biomarkers will be utilized to predict responses to specific regimens of therapy, this allowing for treatment to be tailored to individual patients with PMN. Anti-PLA2R antibody levels are already widely accepted as one such biomarker [1], but assessing anti-PLA2R antibody levels do not help much (yet) in deciding which one of several treatment regimens (CNI, Rituximab, Cyclophosphamide, other) are best suited to the individual patient, except that low anti-PLA2R titers predict a spontaneous remission and no immediate need for any specific treatment at all [1]. The study of Yu et al. does not contribute much to this aspect of individualized treatment of PMN. Also, CNIs are not the first choice for treatment of PMN, except perhaps in China, so the generalization of the findings of this novel and preliminary study to the universe of PMN is limited. Nevertheless, studies of this type are likely to increase in frequency over time and may eventually lead to a change in practice.

Richard Glassock


1. De Vriese AS, Glassock RJ, Nath KA, Sethi S, Fervenza FC. A Proposal for a Serology-Based Approach to Membranous Nephropathy. J Am Soc Nephrol. 2017 ;28:421-430. J Am Soc Nephrol. 2017 ;28:421-430.

Quoted Karger Article

Response Predictors to Calcineurin Inhibitors in Patients with Primary Membranous Nephropathy