In February 2009, La Jolla Pharmaceutical Company announced that owing to a lack of efficacy it would stop developing Riquent™ (abetimus sodium), its primary candidate for the treatment of SLE.1 In the final study, Riquent™ was evaluated within an international phase III trial of 943 patients (ASPEN [Abetimus Sodium in Patients with a History of Lupus Nephritis]),2 representing one of the largest interventional trials in SLE to date. Similar to previous studies of Riquent™, ASPEN was a double-blind, placebo-controlled safety and efficacy study designed to evaluate whether this treatment could delay the time to renal flare or other major SLE flares, or reduce proteinuria in patients with a history of renal disease. Riquent™ seemed to offer some promise after several early trials provided impressive safety data and exploratory analyses suggested potential efficacy in a subset of patients.3,4,5 In the end, however, the drug development program was abruptly terminated after an interim efficacy analysis for ASPEN indicated that it would be futile to continue the study.1

SLE is a chronic, autoimmune/inflammatory disease, thought to involve complex genetic traits that increase the risk for immunologic imbalance in response to a diverse array of environmental triggers. The clinical course of the disease is characterized by protean manifestations, unpredictable flares and a variable prognosis.6 Although treatments for SLE have improved significantly in the past 20 years, a cure for this disabling, sometimes life-threatening disease remains out of reach. The currently available medications are often inadequate to control symptoms, and aggressive dosing is hampered by the risk of infections and other potentially serious adverse effects.7 In the past 15 years, multiple immune targets have been identified, and some tested in randomized, controlled trials of SLE, but all without success.

Riquent™ was originally formulated as a B cell toleragen, with the goal of crosslinking B-cell receptors specifically on B cells recognizing double-stranded DNA (dsDNA), thereby selectively eliminating reactivity to dsDNA. Although this mechanism has not been proven in humans, substantial evidence supports a reliable, dose-dependent decrease in anti-dsDNA antibodies in patients with SLE treated with Riquent™.3,4,5 Furthermore, in contrast to other drug trials in SLE, in the later trials of Riquent™ all patients were required to demonstrate the presence of the autoantibody biomarker that was relevant to the known biologic action of the treatment. Unfortunately, the agent failed to completely eliminate these antibodies in most patients, with mean decreases of, at best, below 50% of baseline values, using a sensitive Farr assay.3,4,5 As the doses used in those trials (200–400 mg per week) proved to be safe and well tolerated, it was hypothesized that raising the dose might increase the effect the anti-dsDNA antibodies, and perhaps improve the overall efficacy of Riquent™ without undue risks. By 2008, a preliminary study suggested that much higher doses of Riquent™ might be well tolerated by patients with SLE,8 and, in a protocol amendment to the ASPEN trial, a 900 mg dose was added. Nonetheless, the interim analysis failed to show benefit, and the trial and clinical development of the agent were terminated.

Is it possible that a subset of patients might respond to Riquent™...?

Is it possible that a subset of patients might respond to Riquent™, and if so, how might they be identified? Early data had suggested that the affinity of patient anti-dsDNA antibodies to Riquent™ might affect clinical outcomes. An intriguing consideration, therefore, was the possibility that in some patients factors other than antibody levels might underlie resistance to autoantibody depletion with lower doses of Riquent™. However, later human trials did not continue investigations of antibody–drug affinity, so the validity of this approach was never confirmed.

In the ASPEN trial, subgroups of patients were assigned to receive the low-dose and high-dose drug schedules without regard to antibody levels or binding variables. This approach calls into question the power of the study to compare clinical outcomes between patients on placebo and those who obtained effective anti-dsDNA antibody binding or biologic coverage by Riquent™. The concept of effective biologic coverage by a therapeutic agent could be critical in SLE. Certainly, as with any treatments for this disease, one dose will not suit all patients, and the clinical basis for selecting a given dose for a given individual is, at best, empiric.

A major impediment to clinical trials in SLE is the heterogeneity of the disease. Advances in genetic characterizations of patients have revealed a remarkable diversity of inherited risk factors. The combinations of immune abnormalities are unpredictable, and the underlying pathophysiology can be markedly different even in patients with similar clinical findings. These variables have proved challenging in the setting of clinical trials that enroll patients with disparate disease characteristics and apply outcome measures to all, regardless of heterogeneity, in an attempt to distil a simple answer from an exceedingly complex situation. Unfortunately, this approach often results in uninterpretable findings. However, as some progress has been made in parsing the intricate immune abnormalities associated with SLE, reliable patterns of disease have emerged within subsets of patients that could provide valuable clues towards simplifying the problem.9,10 Nonetheless, many patients have partial expression of various, overlapping disease features, underscoring the need for a better understanding of the anticipated impact of a specific targeted therapy in individual subsets of SLE.

A further fundamental problem with clinical development in SLE is a lack of biological depth in phase II safety and efficacy trials. Longer, early-phase studies with repetitive testing of patterns of subtle immune perturbations before, during and after treatment might be necessary to optimize patient selection and to begin to tackle the formidable dosing and administration variables that could be impeding successful outcomes in current pivotal trials of SLE.

The heterogeneity of SLE and the absence of reliable biomarkers raise the question of whether it is now time to stop the headlong dash into phase II and III trials before learning more about an agent and the responses of subsets of patients with SLE. To accomplish this, it will be necessary to apply the most sophisticated biology possible before determining optimal patient selection and treatment protocols and to define effective biologic coverage for each agent early in the course of clinical trials. In the absence of a step-by-step, data-driven plan, any optimism that a new treatment will be approved for lupus in the next 50 years must remain guarded.