Biologics in SLE: Towards new approaches




Abstract


In recent years the use of biologic therapies in the management of systemic lupus erythematosus (SLE) has increased, and a number of clinical trials have highlighted both the potential and the pitfalls in the development of such agents. Many investigators reported that the off-label use of rituximab seemed promising in patients with refractory disease, but randomised trials with this agent failed. Likewise, the theoretical appeal of the co-stimulation blocker abatacept could not be confirmed in two clinical trials. Various considerations and post hoc analyses nonetheless suggest that these two biologics might have a role in the treatment of SLE.


The anti-B-lymphocyte stimulator (anti-Blys) antibody belimumab demonstrated efficacy and safety in two large randomised trials and became the first approved biologic for lupus. Use in clinical practice has increased slowly, in part, due to uncertainty over which patients should be treated with this agent and in what stage of the disease. Finally, several other biologic agents are currently in advanced stages of clinical development for SLE. The overall picture that emerges is one of optimism that advances in SLE therapy will be realised through the targeted use of an increasing number of biologics.


Introduction


While biologic therapies have brought about dramatic changes in the treatment of rheumatoid arthritis, spondyloarthropathies and various non-rheumatological diseases, the development of biologics for the treatment of systemic lupus erythematosus (SLE) has met with several noted setbacks, causing some even to question the feasibility of such development. However, recent successful trials of biologics in SLE have demonstrated the feasibility of this and have encouraged optimism about these developments.


There is little doubt that major advances in the treatment of SLE are both possible and necessary. Thus, while glucocorticoids, antimalarials, non-steroidal anti-inflammatory drugs and various conventional disease-modifying antirheumatic and cytotoxic drugs are part of the armamentarium in the management of SLE, major limitations in the efficacies achieved in clinical practice, and the sometimes devastating long-term side effects of these therapies (most importantly from glucocorticoids), make it clear that novel therapeutic approaches are needed. Thus, the development of biologics for SLE has been pursued both by industry, developing entirely new therapeutics for this disease, and by academic clinicians, seeking to use biologics that are already approved for other diseases in the treatment of SLE as well.




Rituximab


The depleting anti-CD20 monoclonal antibody rituximab (MabThera and Rituxan, RTX) has been widely used in non-Hodgkin lymphoma and in rheumatoid arthritis, and several centres began reporting uncontrolled experiences with this biologic in the treatment of severe and refractory SLE, including groups of patients with lupus nephritis . At the Karolinska Institute, we reported encouraging results in 25 patients with severe refractory renal SLE when rituximab was combined as a ‘lymphoma course’ (375 mg m −2 per week for 4 weeks) with intravenous cyclophosphamide (CyX) . However, during the past decade, it has also become clear that mofetil mycophenolic acid (MMF) – which avoids the risk of premature ovarian failure – may be as efficacious as intravenous CyX in lupus nephritis, and it has gradually replaced this cytotoxic agent in most patients. Perhaps for this reason, the controlled trial that was performed with RTX in lupus nephritis (the LUNAR trial) tested the efficacy of RTX, as an add-on treatment to MMF, as a first-line therapy for newly diagnosed or newly relapsed lupus nephritis . In this trial, the predefined primary outcome (complete respondents at 1 year) was achieved slightly less often for RTX than for placebo. However, the total number of respondents was numerically higher for RTX than for placebo (56.9% for rituximab versus 45.8% for placebo); this difference did not achieve statistical significance, but it is possible that a true benefit of modest effect size was ‘missed’ because of the limited power of this study in 144 patients. Thus, while the negative results may have been due to the inefficacy of rituximab, there are other possible explanations. The time frame chosen for the primary end point of a clinical response at week 52 may be somewhat early in the case of lupus nephritis, and it is also possible that MMF does not combine well with rituximab. Indeed, Ramos-Casals et al. demonstrated in a systematic review of literature from 2002 to 2007 that higher therapeutic responses with rituximab were seen when it was combined with CyX rather than with MMF. Moreover, the ‘rheumatoid arthritis course’ that was used in both trials, two infusions of 1000 mg rituximab separated by 2 weeks, may be somewhat less effective in SLE than the ‘lymphoma regimen’. Again, in the review cited above patients had better responses with the ‘lymphoma course’ than with the ‘rheumatoid arthritis course’ . Finally, it is important to realise that the LUNAR trial did not study refractory patients, the population on which most uncontrolled reports were based. A controlled trial of RTX in refractory patients is currently being planned in Europe and South America, the rituximab in lupus nephritis group (RING) trial.


Although most uncontrolled studies with RTX were performed in patients with lupus nephritis, other reports and case series have highlighted the potential of RTX in patients with various non-renal SLE manifestations, such as haematological lupus refractory to conventional treatment . Therefore, a randomised trial with RTX in non-renal lupus was expected to demonstrate efficacy. However, the EXPLORER trial, in which RTX was added to background therapy with antimalarials, immunosuppressives and glucocorticoids, failed to demonstrate any benefit versus placebo for the primary and for most of the secondary outcomes . Some post hoc analyses were able to discern a ‘signal’ for patients with arthritis and in some ethnic groups, but the overall conclusions from this trial are strongly negative. A solid interpretation of these findings is that rituximab is not a suitable treatment for moderate musculoskeletal, mucocutaneous or ‘general’ SLE. It is of interest that these disease manifestations tend to correlate poorly with SLE-related autoantibodies; for instance, anti-Ro (SSA) and anti-La (SSB) antibodies, which do associate with cutaneous lupus and photosensitivity, are not down-regulated following rituximab treatment. It should be noted, however, that the potential use of RTX in severe and refractory non-renal lupus manifestations, such as haematological disease or neuropsychiatric SLE, can hardly be assessed based on the EXPLORER study, in which such patients were represented only in very small numbers.


A very different approach to the use of RTX in SLE has recently come to the fore. Based on successes with glucocorticoid-free treatment regimens in the field of transplantation medicine, various investigators have begun to explore such regimens in SLE as well. Thus, in patients with newly diagnosed renal SLE, who had already been on glucocorticoids, Pepper et al. reported on the use of RTX followed by MMF and only minimal or no additional glucocorticoids. Many of these patients were able to reduce the dose of, or even eliminate, glucocorticoids altogether, with overall very good responses. More strikingly, Condon et al. reported very recently that most of their 50 patients with newly diagnosed lupus nephritis (WHO class III, IV or V) who were treated with a single pulse of methyl-prednisolone, two infusions of RTX and subsequent MMF had excellent results (and no glucocorticoid-related side effects!) after 1 year. In a small group of patients with non-renal SLE, somewhat encouraging results were also reported with a similar regimen . These extraordinary findings warrant further investigation, and a randomised controlled trial in lupus nephritis investigating such a minimal glucocorticoid treatment based on RTX and MMF is currently being planned, nicknamed RITUXILUP.


For clinicians in a practice setting, where off-label use with RTX is possible, it would seem reasonable to consider this option in patients with lupus nephritis who have failed conventional therapy and perhaps in some other severe lupus manifestations. It can be hoped that remaining uncertainties about patient selection, and about optimal dosing and combination, will gradually be resolved in upcoming trials and studies.




Rituximab


The depleting anti-CD20 monoclonal antibody rituximab (MabThera and Rituxan, RTX) has been widely used in non-Hodgkin lymphoma and in rheumatoid arthritis, and several centres began reporting uncontrolled experiences with this biologic in the treatment of severe and refractory SLE, including groups of patients with lupus nephritis . At the Karolinska Institute, we reported encouraging results in 25 patients with severe refractory renal SLE when rituximab was combined as a ‘lymphoma course’ (375 mg m −2 per week for 4 weeks) with intravenous cyclophosphamide (CyX) . However, during the past decade, it has also become clear that mofetil mycophenolic acid (MMF) – which avoids the risk of premature ovarian failure – may be as efficacious as intravenous CyX in lupus nephritis, and it has gradually replaced this cytotoxic agent in most patients. Perhaps for this reason, the controlled trial that was performed with RTX in lupus nephritis (the LUNAR trial) tested the efficacy of RTX, as an add-on treatment to MMF, as a first-line therapy for newly diagnosed or newly relapsed lupus nephritis . In this trial, the predefined primary outcome (complete respondents at 1 year) was achieved slightly less often for RTX than for placebo. However, the total number of respondents was numerically higher for RTX than for placebo (56.9% for rituximab versus 45.8% for placebo); this difference did not achieve statistical significance, but it is possible that a true benefit of modest effect size was ‘missed’ because of the limited power of this study in 144 patients. Thus, while the negative results may have been due to the inefficacy of rituximab, there are other possible explanations. The time frame chosen for the primary end point of a clinical response at week 52 may be somewhat early in the case of lupus nephritis, and it is also possible that MMF does not combine well with rituximab. Indeed, Ramos-Casals et al. demonstrated in a systematic review of literature from 2002 to 2007 that higher therapeutic responses with rituximab were seen when it was combined with CyX rather than with MMF. Moreover, the ‘rheumatoid arthritis course’ that was used in both trials, two infusions of 1000 mg rituximab separated by 2 weeks, may be somewhat less effective in SLE than the ‘lymphoma regimen’. Again, in the review cited above patients had better responses with the ‘lymphoma course’ than with the ‘rheumatoid arthritis course’ . Finally, it is important to realise that the LUNAR trial did not study refractory patients, the population on which most uncontrolled reports were based. A controlled trial of RTX in refractory patients is currently being planned in Europe and South America, the rituximab in lupus nephritis group (RING) trial.


Although most uncontrolled studies with RTX were performed in patients with lupus nephritis, other reports and case series have highlighted the potential of RTX in patients with various non-renal SLE manifestations, such as haematological lupus refractory to conventional treatment . Therefore, a randomised trial with RTX in non-renal lupus was expected to demonstrate efficacy. However, the EXPLORER trial, in which RTX was added to background therapy with antimalarials, immunosuppressives and glucocorticoids, failed to demonstrate any benefit versus placebo for the primary and for most of the secondary outcomes . Some post hoc analyses were able to discern a ‘signal’ for patients with arthritis and in some ethnic groups, but the overall conclusions from this trial are strongly negative. A solid interpretation of these findings is that rituximab is not a suitable treatment for moderate musculoskeletal, mucocutaneous or ‘general’ SLE. It is of interest that these disease manifestations tend to correlate poorly with SLE-related autoantibodies; for instance, anti-Ro (SSA) and anti-La (SSB) antibodies, which do associate with cutaneous lupus and photosensitivity, are not down-regulated following rituximab treatment. It should be noted, however, that the potential use of RTX in severe and refractory non-renal lupus manifestations, such as haematological disease or neuropsychiatric SLE, can hardly be assessed based on the EXPLORER study, in which such patients were represented only in very small numbers.


A very different approach to the use of RTX in SLE has recently come to the fore. Based on successes with glucocorticoid-free treatment regimens in the field of transplantation medicine, various investigators have begun to explore such regimens in SLE as well. Thus, in patients with newly diagnosed renal SLE, who had already been on glucocorticoids, Pepper et al. reported on the use of RTX followed by MMF and only minimal or no additional glucocorticoids. Many of these patients were able to reduce the dose of, or even eliminate, glucocorticoids altogether, with overall very good responses. More strikingly, Condon et al. reported very recently that most of their 50 patients with newly diagnosed lupus nephritis (WHO class III, IV or V) who were treated with a single pulse of methyl-prednisolone, two infusions of RTX and subsequent MMF had excellent results (and no glucocorticoid-related side effects!) after 1 year. In a small group of patients with non-renal SLE, somewhat encouraging results were also reported with a similar regimen . These extraordinary findings warrant further investigation, and a randomised controlled trial in lupus nephritis investigating such a minimal glucocorticoid treatment based on RTX and MMF is currently being planned, nicknamed RITUXILUP.


For clinicians in a practice setting, where off-label use with RTX is possible, it would seem reasonable to consider this option in patients with lupus nephritis who have failed conventional therapy and perhaps in some other severe lupus manifestations. It can be hoped that remaining uncertainties about patient selection, and about optimal dosing and combination, will gradually be resolved in upcoming trials and studies.




Abatacept


The T-cell co-stimulation blocker abatacept is a molecular construct based on coupling two cytotoxic T-lymphocyte antigen 4 (CTLA4) molecules to an immunoglobulin G (IgG) frame. The molecule CTLA4 is a naturally occurring regulator of T-cell activation, the natural role of which is believed to be the prevention of excessive T-cell activation. It does so by interfering with the T-cell activating signal that is mediated through the binding of B7 (CD80/86) on the antigen-presenting cell with CD28 on the T-cell. Abatacept is approved for use in rheumatoid arthritis and has also been studied in other auto-immune conditions. Animal data raised hopes that abatacept would be effective in SLE, and two robust randomised clinical trials have been performed with this biologic. The first of these, done in non-renal SLE, failed to demonstrate efficacy . In this trial, patients were recruited if they had a flare in one of the three organ systems: mucocutaneous, musculoskeletal (mostly arthritis) or serositis (pleurisy and pericarditis). The patients were all started on glucocorticoids at 30 mg prednisone day −1 or equivalent, followed by a taper, and in addition they were randomised to receive abatacept or placebo. The primary outcome of this trial was the proportion of patients who flared following tapering of glucocorticoids; the idea was to first bring the SLE under control with the higher glucocorticoid dose started at randomisation, and then taper the glucocorticoids, anticipating a relapse which could then perhaps be prevented by abatacept. But after 1 year there was no difference in this regard: around 80% of patients had flared in both groups. Severe flares (defined as a new BILAG A) were somewhat less frequent in the abatacept group (41% versus 55%), and several patient-reported outcomes were significantly better with abatacept. These results must therefore be interpreted with some caution. It is of course possible that abatacept is, in fact, without benefit in SLE, but it also seems possible that a true benefit was ‘missed’ in this trial due to one of several reasons. One important clue to the latter possibility is the demonstration that agreement between different physician experts on the presence or absence of mild/moderate flares is very poor, as is the agreement between experts and various flare instruments . By contrast, expert physicians’ assessments of severe flares were quite consistent and corresponded well to various instruments. Therefore, the discrepancy between the overall-flare and severe-flare results in the abatacept trial suggests that perhaps the method by which all-flare data were collected was just not sufficiently accurate for this clinical trial setting.


A second randomised controlled trial with abatacept was also done, this time in lupus nephritis. This trial was recently reported in an abstract form, and the results were again negative . Patients with biopsy-proven active proliferative lupus nephritis were randomised to abatacept 30 mg kg −1 versus 10 mg kg −1 versus placebo, added to MMF plus a high-dose prednisone followed by a taper. Time to complete renal remission, which was the primary outcome of this trial, showed no difference between the groups. However, it was noted that the proportion of patients who in fact achieved a complete renal remission was exceedingly small in this trial for all groups, around 3–5%. Understandably, a clinical trial analysis that is based on a very restrictive definition of response has very little likelihood of demonstrating a difference. Based on this consideration, a reanalysis was done of the data set from this clinical trial using several other definitions of complete or partial renal response , all of which had been used in at least one prior clinical trial. This reanalysis led to some remarkable results. For example, when applying the definition of response that had been used in the LUNAR trial, a response rate was achieved in the abatacept group of more than 20% versus only 6% in placebo, and a strong positive effect was also seen using other definitions. Clearly, choosing the best possible outcome in clinical trials is absolutely critical to their success. For trials in lupus nephritis, it might be of interest that a set of lupus nephritis outcomes was proposed several years ago based on solid data sets involving hundreds of real patients ; somewhat disappointingly, these criteria have not been used in clinical trials to date. It remains a possibility that these data-driven outcomes would have performed better in clinical trials than various ad hoc outcomes, often derived rather more from expert consensus than from actual patient material.


The abatacept trial in lupus nephritis was based in part on observations made in the NZB/NZW murine lupus model, where efficacy for CTLA4-Ig had been demonstrated. However, it may be important to note that in this model the most striking efficacy was obtained when CTLA4-Ig was combined with CyX . An investigator-initiated trial named ACCESS using this same combination is currently underway.


Abatacept may be available to clinicians as an ‘off-label’ treatment for SLE. Despite the negative trials, it might be reasonable to consider this in patients with severe lupus-related arthritis and/or at risk of severe flares. It is hoped that further studies of this interesting agent can and will be performed.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Nov 11, 2017 | Posted by in RHEUMATOLOGY | Comments Off on Biologics in SLE: Towards new approaches

Full access? Get Clinical Tree

Get Clinical Tree app for offline access