Lupus Nephritis Response Criteria in Lupus Activity Indices
Glucocorticoids and intermittent intravenous cyclophosphamide (IVC) have long been considered the standard induction agents to treat LN. Steinberg and Decker reported the success of cyclophosphamide when compared with azathioprine (AZA) or placebo for inducing remission over the course of a 10-week period in a randomized trial of 38 patients during the mid-1970s. Patients were evaluated with respect to change in creatinine clearance, proteinuria, and urine sediment. Amongst the treatment regimens studied by Austin and colleagues, the authors reported reduced rates of ESKD in a sample of mostly Caucasian patients who received IVC and glucocorticoids when compared to glucocorticoid monotherapy. LN has remained a research focus in an effort to identify more effective and less toxic treatment strategies. Table 2 includes a summary of important randomized controlled trials for the treatment of LN from the last 15 years. Although these studies have influenced the care of patients, none has led to FDA approval for an agent in the treatment of LN. Studies have varied in inclusion of patients by race and ethnicity, geographic region, size, duration of follow-up, and chosen primary and secondary endpoints.
Studies can be divided into two types: studies of induction or maintenance of remission. An induction trial compares two treatments with respect to efficacy in achieving disease remission, whereas maintenance studies compare therapies with respect to limiting the frequency of flares. Common endpoints in an induction trial are measures of disease activity. As demonstrated in Table 2, there is no standardized definition of 'complete remission'. The definitions for remission of proteinuria vary from less than 0.3 g/day to less than 3 g/day to an improvement of more than 50%. In studies that have included a measurement of eGFR, definitions of remission differ from comparisons made to the baseline value versus comparisons made to normal values. Trials of maintenance therapy focus on 'treatment failure' as the primary endpoint. These studies tend to be of longer duration, which affords them the opportunity to invoke hard endpoints such as patient survival, the need for renal replacement therapy, the occurrence of flare, or progressive kidney disease.
The trial that assessed efficacy and safety of adding abatacept to mycophenolate mofetil (MMF) highlights the need to define endpoints carefully. The definition of 'complete response' chosen by those investigators was likely too restrictive as it included a composite measure that required maintenance of eGFR, minimal proteinuria, and inactive urinary sediment over the 52-week treatment period. This may be one reason why the response rate among all participants was much lower than expected.
It is unethical to deny study participants effective treatment, and therefore investigators must decide on a steroid dosing strategy in the trial design. As the examples in Table 2 illustrate, defining the dose and type of glucocorticoid to be used for induction is important but not standardized across trials. In addition, clinical trials must provide instructions for a taper and specify how to treat a flare to avoid confounding due to different cumulative steroid exposure between groups. Clear guidance on the use of medications, especially non-steroidal anti-inflammatory drugs (NSAIDs), angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers, is also an essential component of trial design for LN.
The studies done to determine efficacy of MMF for remission of LN illustrate the distinction between superiority and non-inferiority trials. The studies by Chan and colleagues and Ginzler and colleagues were non-inferiority trials. Based on the success of MMF in these trials, the Aspreva Lupus Management Study Group trial was designed as a superiority trial, and numerous sites around the globe participated. Comparable rates of patients responded to treatment in the two arms; however, MMF failed to demonstrate superiority, and therefore this was considered a negative trial.
Four major clinical trials to attempt to improve treatment of LN have recently been completed or are under way (Table 3). These include the following:
All four clinical trials use glucocorticoids as standard-of-care therapy. The ACCESS trial and BLISS-LN trial allow IVC remission induction therapy as another option. The ACCESS trial has completed 1-year follow-up data collection on the 134 participants, and the interim data have been published as abstracts. The remaining three clinical trials are led by pharmaceutical companies and are attempting to recruit hundreds of patients from many national and international sites.
Each trial is designed with a primary, composite endpoint of complete renal response; however the criterion used in the endpoint definition varies across studies. Aside from the ACCESS trial, each study includes microscopic review of the urine sediment, namely looking for RBC casts, as a part of the composite endpoint. This may be questioned upon trial completion given the variability in an investigator's ability to reliably and uniformly assess the sediment at each site. While a multi-faceted definition of complete renal response reflects the approach used in clinical care, it is likely difficult to achieve in a research setting, especially in a diverse group of patients recruited from centers around the world.
Renal Endpoints in Clinical Trials of Lupus Nephritis: Past and Present
Completed Trials
Glucocorticoids and intermittent intravenous cyclophosphamide (IVC) have long been considered the standard induction agents to treat LN. Steinberg and Decker reported the success of cyclophosphamide when compared with azathioprine (AZA) or placebo for inducing remission over the course of a 10-week period in a randomized trial of 38 patients during the mid-1970s. Patients were evaluated with respect to change in creatinine clearance, proteinuria, and urine sediment. Amongst the treatment regimens studied by Austin and colleagues, the authors reported reduced rates of ESKD in a sample of mostly Caucasian patients who received IVC and glucocorticoids when compared to glucocorticoid monotherapy. LN has remained a research focus in an effort to identify more effective and less toxic treatment strategies. Table 2 includes a summary of important randomized controlled trials for the treatment of LN from the last 15 years. Although these studies have influenced the care of patients, none has led to FDA approval for an agent in the treatment of LN. Studies have varied in inclusion of patients by race and ethnicity, geographic region, size, duration of follow-up, and chosen primary and secondary endpoints.
Studies can be divided into two types: studies of induction or maintenance of remission. An induction trial compares two treatments with respect to efficacy in achieving disease remission, whereas maintenance studies compare therapies with respect to limiting the frequency of flares. Common endpoints in an induction trial are measures of disease activity. As demonstrated in Table 2, there is no standardized definition of 'complete remission'. The definitions for remission of proteinuria vary from less than 0.3 g/day to less than 3 g/day to an improvement of more than 50%. In studies that have included a measurement of eGFR, definitions of remission differ from comparisons made to the baseline value versus comparisons made to normal values. Trials of maintenance therapy focus on 'treatment failure' as the primary endpoint. These studies tend to be of longer duration, which affords them the opportunity to invoke hard endpoints such as patient survival, the need for renal replacement therapy, the occurrence of flare, or progressive kidney disease.
The trial that assessed efficacy and safety of adding abatacept to mycophenolate mofetil (MMF) highlights the need to define endpoints carefully. The definition of 'complete response' chosen by those investigators was likely too restrictive as it included a composite measure that required maintenance of eGFR, minimal proteinuria, and inactive urinary sediment over the 52-week treatment period. This may be one reason why the response rate among all participants was much lower than expected.
It is unethical to deny study participants effective treatment, and therefore investigators must decide on a steroid dosing strategy in the trial design. As the examples in Table 2 illustrate, defining the dose and type of glucocorticoid to be used for induction is important but not standardized across trials. In addition, clinical trials must provide instructions for a taper and specify how to treat a flare to avoid confounding due to different cumulative steroid exposure between groups. Clear guidance on the use of medications, especially non-steroidal anti-inflammatory drugs (NSAIDs), angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers, is also an essential component of trial design for LN.
The studies done to determine efficacy of MMF for remission of LN illustrate the distinction between superiority and non-inferiority trials. The studies by Chan and colleagues and Ginzler and colleagues were non-inferiority trials. Based on the success of MMF in these trials, the Aspreva Lupus Management Study Group trial was designed as a superiority trial, and numerous sites around the globe participated. Comparable rates of patients responded to treatment in the two arms; however, MMF failed to demonstrate superiority, and therefore this was considered a negative trial.
Ongoing Clinical Trials
Four major clinical trials to attempt to improve treatment of LN have recently been completed or are under way (Table 3). These include the following:
The ACCESS trial (Abatacept and Cyclophosphamide Combination: Efficacy and Safety Study), sponsored by the National Institute of Allergy and Infectious Diseases through the Immune Tolerance Network, assesses the efficacy of abatacept (a fusion protein composed of the Fc region of IgG1 fused to the extracellular domain of CTLA-4 that prevents T-cell activation) versus placebo in the treatment of proliferative LN (class III or IV with/without class V) with background therapy of Eurolupus IVC (500 mg IVC every 2 weeks for six doses) followed by maintenance with AZA.
The ALLURE (Advancing Leading-Edge Lupus Research) trial also assesses the efficacy of abatacept with background therapy of MMF.
The BLISS-LN (Belimumab International Lupus Nephritis Study) trial assesses the efficacy of belimumab—a human monoclonal antibody that inhibits the B-cell survival factor called B-cell activating factor (BAFF; also known as B-lymphocyte stimulator or BLyS) to prevent B-cell survival—with background therapy of Eurolupus IVC or MMF per investigator choice followed by MMF maintenance.
The ATLAS (Adjuvant Tamoxifen: Longer Against Shorter) trial assesses the efficacy of BIIB023—a humanized monoclonal antibody that inhibits tumor necrosis factor-related weak inducer of apoptosis (TWEAK) to reduce tissue inflammation—with background therapy of MMF.
All four clinical trials use glucocorticoids as standard-of-care therapy. The ACCESS trial and BLISS-LN trial allow IVC remission induction therapy as another option. The ACCESS trial has completed 1-year follow-up data collection on the 134 participants, and the interim data have been published as abstracts. The remaining three clinical trials are led by pharmaceutical companies and are attempting to recruit hundreds of patients from many national and international sites.
Each trial is designed with a primary, composite endpoint of complete renal response; however the criterion used in the endpoint definition varies across studies. Aside from the ACCESS trial, each study includes microscopic review of the urine sediment, namely looking for RBC casts, as a part of the composite endpoint. This may be questioned upon trial completion given the variability in an investigator's ability to reliably and uniformly assess the sediment at each site. While a multi-faceted definition of complete renal response reflects the approach used in clinical care, it is likely difficult to achieve in a research setting, especially in a diverse group of patients recruited from centers around the world.