Mycophenolate mofetil and tacrolimus versus tacrolimus alone for the treatment of idiopathic membranous glomerulonephritis: a randomised controlled trial

Membranous nephropathy (MN) is a relatively common cause of the nephrotic syndrome in adults affecting 5–10 per million population per year. Although spontaneous remissions are common, 40% of patients with untreated membranous nephropathy will develop end stage renal disease within 10–15 years [1].

Treatment of idiopathic membranous nephropathy can be challenging. The discovery of anti-phospholipase A2 receptor (anti-PLA2R) antibodies in 70% of patients with MN [2] supports the use of immunosuppression for the treatment of idiopathic MN. Treatment regimens including high dose steroids with chlorambucil or cyclophosphamide have proven to be effective in inducing remission [3, 4] however these regimens have significant adverse effects [5] and relapses are not infrequent.

When this trial was designed there were limited data on the use of rituximab in MN and only a small case series of 8 patients showed an effect of the agent on proteinuria [6]. Since then rituximab has gained popularity following data from clinical trials that support its use in MN [6‐9].The immunological effect of the agent was confirmed in an initial study that showed that anti-PLA2R antibodies disappeared in 68% (17 out of 25) of patients with MN treated with rituximab [10]. In addition, more recently both the GEMRITUX and the MENTOR trials showed that rituximab is effective in inducing remission of idiopathic MN [7, 8]. However there are significant numbers of patients who are resistant to rituximab and its use is still restricted in many centres due to cost. Therefore alternative treatment options are necessary and other agents used include calcineurin inhibitors and mycophenolate mofetil.

Calcineurin inhibitors (CNIs) such as cyclosporin and tacrolimus (TAC) are commonly used for the treatment of MN and are effective in inducing remission. Their antiproteinuric effect has been clearly documented [11‐13]; however relapses after discontinuation are common and there is a concern about nephrotoxicity after prolonged use. Although their main mode of action is immunomodulatory they also have a significant antiproteinuric effect which is attributed to their haemodynamic effect on glomerular perfusion as well as to their direct effect on podocytes [12, 14]. A study comparing tacrolimus to placebo showed that TAC alone is effective in achieving remission but patients often relapse after treatment withdrawal [15]. A recent trial comparing rituximab to cyclosporine showed a high relapse rate of the cyclosporine arm after treatment withdrawal [8].

Mycophenolate mofetil (MMF) has been used for the treatment of MN in regimens of variable dosing and duration, either alone or in combination with steroids [16‐18]. MMF monotherapy is not effective in inducing remission [17]; however it has been successfully used in combination with steroids [16, 19]. MMF has a long history of use in combination with CNIs in transplantation with a good safety profile and this combination has also been used in nephrotic syndrome secondary to other glomerulonephritides such as lupus nephritis [20, 21]. The presence of anti-PLA2R auto-antibodies in a significant number of patients with MN suggests that an immunosuppressive agent such as MMF, that suppresses B lymphocyte proliferation and antibody formation, may be particularly effective in inducing and maintaining immunological and clinical remission in MN.

We compared the efficacy of combination treatment with tacrolimus and mycophenolate mofetil versus tacrolimus alone for achieving sustained remission in patients with idiopathic membranous nephropathy.

Forty Patients were recruited to the trial between March 2009 and December 2014. 20 patients were randomly assigned to receive TAC alone and 20 patients to combination treatment with TAC and MMF (Fig. 1). The groups had similar baseline characteristics (Table 1). 39 out of 40 patients had a new diagnosis of MN. One patient in the TAC group had received previous treatment for membranous nephropathy with steroids and chlorambucil 10 years prior to entering the trial. This patient had since been off treatment and was recruited to the trial during a relapse.

11/20 patients in the TAC group and 13/20 in the TAC/MMF group were male. Baseline uPCR and renal function were equivalent in the two groups. In the TAC group median uPCR and serum creatinine at baseline were 704 mg/mmol (range 203–2159) and 0.8 mg/dl (range 0.5–1.4) respectively; in the TAC/MMF group median uPCR and serum creatinine were 756 mg/mmol (range 123–1784) and 0.8 mg/dl (range 0.5–1.20) respectively. All (20/20) patients in the TAC group and 17/20 patients in the TAC/MMF group were on an ACEI/ARB or a combination of both at the start of trial and for at least 3 months prior to enrolment; three patients in the TAC/MMF group were intolerant to ACEI/ARB due to hypotension or cough. The age, morphology and location of the electron dense deposits were recorded by electron microscopy. 12/20 biopsies in each group were histological stage II. PLA2R positivity on biopsy was detected in 10/20 patients in the TAC group and 12/20 in the TAC/MMF group. Interstitial fibrosis and tubular atrophy (IFTA) was significant (15%) in 2 cases in the TAC/MMF group but less than 10% in the rest of the biopsies at baseline. Demographic, laboratory and histological characteristics at baseline are shown on Table 1.

The median follow up in the TAC group was 6 years (range 6 months to 8.8 years) compared to 5.9 years (range 9 months to 8.8 years) in the TAC/MMF group, including patients that were lost to follow up early during the treatment phase.

Idiopathic MN is a common cause of the nephrotic syndrome in adults and untreated can lead to ESRD in up to 40% of patients within 10 years. The role of immunosuppression including biological agents in the management of MN is clearer now than when this trial started more than 10 years ago; however the optimal choice of agent remains controversial. In one small pilot study MMF was successfully used in a cohort of patients with MN resistant to steroids and where cytotoxics and cyclosporine failed [22]. In this study patients achieved significant reduction in proteinuria within 6 months of treatment with MMF monotherapy. However a more comprehensive study of 36 patients with idiopathic MN, treated with MMF or conservative treatment for 12 months, showed no benefit of MMF monotherapy in inducing remission compared to the control group receiving supportive care only [17]. Other randomised trials have shown a benefit from MMF when used in combination with steroids [16, 18, 19]; Chan et al. reported similar efficacy of MMF and steroids when compared to a modified Ponticelli regimen using chlorambucil [19], and Brantel et al. showed that a 12 month course of MMF and prednisolone was as effective as cyclophosphamide in decreasing proteinuria when compared to a historic group treated with the Ponticelli regimen [16]. Side effects from steroids are a concern however, and given the apparent effectiveness of MMF in MN when used in combination with other immunosuppressants, it was hypothesised in this trial that MMF may reduce the relapse rate of nephrotic syndrome when used in combination with TAC for the treatment of MN.

However this study does not show a benefit of MMF when added to TAC in reducing relapse rate. The trend of the TAC/MMF group to achieve remission sooner than the TAC alone group did not reach statistical significance but it perhaps suggests that in some patients the addition of MMF to TAC may have some effect in achieving effective remission faster. However the numbers are small and the trial was only powered to look at relapse rates. When our study was designed, anti-PLA2R antibodies had not yet been discovered; hence we did not include anti-PLA2R levels in our assessment or end points. It is now known that anti-PLA2R levels are associated with disease activity and outcomes [23, 24]. We were able to retrospectively stain the available biopsies for PLA2R and this was found to be positive in 55 and 63% of patients in the TAC and the TAC/MMF group respectively. The percentage of anti-PLA2R positivity in our cohort is lower than in some other studies and perhaps this has masked some benefit of the MMF. It is possible that patients with high antibody titres, which are associated with resistant disease, could benefit more from the addition of MMF than those with a low or absent titre. In our study, 4 patients in the TAC group compared to only 1 in the TAC/MMF group did not achieve remission and although this was not statistically significant it is also possible that the addition of MMF to TAC may be of benefit to some patients. Again the availability of anti-PLA2R levels would help interpret these results and larger studies with specific subgroups are required to further investigate treatment options.

Importantly the combination of MMF with TAC appeared to be safe in this group of patients, with only 3 patients experiencing a serious adverse event in the TAC/MMF group. Infection and leukopenia were not a major issue in the combination arm. A novelty in our study was to monitor MPA levels during treatment for MN; MMF dose was titrated to achieve MPA levels between 1.5–3 mg/ml to ensure effective treatment and minimise side effects, and this may have contributed to the safety of the combination therapy, although nephrotoxicity due to CNI was an issue in 2 patients in the TAC group. Both these patients underwent renal biopsy for rising serum creatinine that showed tubular injury and hyalinosis consistent with CNI toxicity. In both treatment groups there was a non-significant rise in the serum creatinine from baseline to 2 years; in patients that were followed up for longer, the serum creatinine subsequently remained unchanged over time. Two patients progressed to end stage renal disease despite treatment with the trial medication or other therapies, and this demonstrates that existing treatments are not always effective, and more options should be available.

Recruitment was slow and the trial was designed over 10 years ago, without the current knowledge for the role of anti-PLA2R antibodies in MN or for the use of rituximab as first line treatment. Although randomised controlled trials (RCTs) are the most acknowledged methods of testing a treatment for a specific condition there are only a limited number of RCTs examining the efficacy of TAC alone in MN [15]. Therefore we believe that this RCT is important and the results relevant because they contribute to the evidence for the use of TAC and MMF as treatments for MN. Reporting a negative outcome is important in order to avoid using unnecessary or ineffective immunosuppression. The length of follow up is one of the strong points of the trial, and although this trial did not show a benefit from the addition of MMF to TAC it showed that TAC alone or in combination with MMF is safe in MN and CNI toxicity occurred in only two patients.

In summary, this study did not show a benefit from the addition of MMF to TAC for the treatment of MN either in rates of remission, time to remission or subsequent relapse. Although this study is relatively small it was appropriately powered to detect a benefit of MMF on relapse rate. TAC is effective in inducing remission and it is well tolerated in most patients but CNI toxicity does occur therefore close monitoring of patients is required. Although effective regimens for the treatment of MN are available there are still patients who do not respond or are unable to receive specific biological or cytotoxic agents; therefore it is possible that the combination of TAC with MMF could allow lower doses of tacrolimus to be used to achieve remission and this should also be investigated. Most patients who relapsed did so after discontinuation of treatment rather than during withdrawal; therefore it could be that low doses of tacrolimus may be adequate to maintain remission once this has been achieved. Both treatment arms were well tolerated and further studies are needed to investigate if the combination of TAC/MMF would benefit a subgroup of patients with a high anti-PLA2R antibody titre and more resistant disease. Anti-PLA2R antibodies need to be included in future studies and comparative studies between the anti-PLA2R positive and negative patients would be useful to further clarify which patients may benefit from specific regimens so that immunosuppression can be better tailored to individual patients.