Interventions for cutaneous disease in systemic lupus erythematosus
Lupus erythematosus is an autoimmune disease with significant morbidity and mortality. Cutaneous disease in systemic lupus erythematosus (SLE) is common. Many interventions are used to treat SLE with varying efficacy, risks, and benefits.
To assess the effects of interventions for cutaneous disease in SLE.
We searched the following databases up to June 2019: the Cochrane Skin Specialised Register, CENTRAL, MEDLINE, Embase, Wiley Interscience Online Library, and Biblioteca Virtual em Saude (Virtual Health Library). We updated our search in September 2020, but these results have not yet been fully incorporated.
We included randomised controlled trials (RCTs) of interventions for cutaneous disease in SLE compared with placebo, another intervention, no treatment, or different doses of the same intervention. We did not evaluate trials of cutaneous lupus in people without a diagnosis of SLE.
Data collection and analysis
We used standard methodological procedures expected by Cochrane. Primary outcomes were complete and partial clinical response. Secondary outcomes included reduction (or change) in number of clinical flares; and severe and minor adverse events. We used GRADE to assess the quality of evidence.
Sixty‐one RCTs, involving 11,232 participants, reported 43 different interventions. Trials predominantly included women from outpatient clinics; the mean age range of participants was 20 to 40 years. Twenty‐five studies reported baseline severity, and 22 studies included participants with moderate to severe cutaneous lupus erythematosus (CLE); duration of CLE was not well reported. Studies were conducted mainly in multi‐centre settings. Most often treatment duration was 12 months. Risk of bias was highest for the domain of reporting bias, followed by performance/detection bias. We identified too few studies for meta‐analysis for most comparisons. We limited this abstract to main comparisons (all administered orally) and outcomes. We did not identify clinical trials of other commonly used treatments, such as topical corticosteroids, that reported complete or partial clinical response or numbers of clinical flares.
Complete clinical response
Studies comparing oral hydroxychloroquine against placebo did not report complete clinical response.
Chloroquine may increase complete clinical response at 12 months' follow‐up compared with placebo (absence of skin lesions) (risk ratio (RR) 1.57, 95% confidence interval (CI) 0.95 to 2.61; 1 study, 24 participants; low‐quality evidence).
There may be little to no difference between methotrexate and chloroquine in complete clinical response (skin rash resolution) at 6 months' follow‐up (RR 1.13, 95% CI 0.84 to 1.50; 1 study, 25 participants; low‐quality evidence).
Methotrexate may be superior to placebo with regard to complete clinical response (absence of malar/discoid rash) at 6 months' follow‐up (RR 3.57, 95% CI 1.63 to 7.84; 1 study, 41 participants; low‐quality evidence).
At 12 months' follow‐up, there may be little to no difference between azathioprine and ciclosporin in complete clinical response (malar rash resolution) (RR 0.83, 95% CI 0.46 to 1.52; 1 study, 89 participants; low‐quality evidence).
Partial clinical response
Partial clinical response was reported for only one key comparison: hydroxychloroquine may increase partial clinical response at 12 months compared to placebo, but the 95% CI indicates that hydroxychloroquine may make no difference or may decrease response (RR 7.00, 95% CI 0.41 to 120.16; 20 pregnant participants, 1 trial; low‐quality evidence).
Clinical flares were reported for only two key comparisons: hydroxychloroquine is probably superior to placebo at 6 months' follow‐up for reducing clinical flares (RR 0.49, 95% CI 0.28 to 0.89; 1 study, 47 participants; moderate‐quality evidence). At 12 months' follow‐up, there may be no difference between methotrexate and placebo, but the 95% CI indicates there may be more or fewer flares with methotrexate (RR 0.77, 95% CI 0.32 to 1.83; 1 study, 86 participants; moderate‐quality evidence).
Data for adverse events were limited and were inconsistently reported, but hydroxychloroquine, chloroquine, and methotrexate have well‐documented adverse effects including gastrointestinal symptoms, liver problems, and retinopathy for hydroxychloroquine and chloroquine and teratogenicity during pregnancy for methotrexate.
Evidence supports the commonly‐used treatment hydroxychloroquine, and there is also evidence supporting chloroquine and methotrexate for treating cutaneous disease in SLE. Evidence is limited due to the small number of studies reporting key outcomes. Evidence for most key outcomes was low or moderate quality, meaning findings should be interpreted with caution. Head‐to‐head intervention trials designed to detect differences in efficacy between treatments for specific CLE subtypes are needed. Thirteen further trials are awaiting classification and have not yet been incorporated in this review; they may alter the review conclusions.
Cora W Hannon, Collette McCourt, Hermenio C Lima, Suephy Chen, Cathy Bennett
Plain language summary
What are the benefits and risks of different treatments for skin disease in people with systemic lupus erythematosus (an autoimmune disease that affects the whole body)
Why is this question important?
Systemic lupus erythematosus (SLE; also known as ‘lupus’) is a disease in which the body's immune (defence) system mistakenly attacks healthy tissue in many parts of the body. It affects 7.5 million people worldwide. Around 70% of affected people develop skin problems such as rash on the nose or cheeks. Often, SLE also causes pain in joints and muscles and extreme tiredness. Symptoms can improve temporarily, or they can worsen suddenly (flares). In severe cases, SLE can cause life‐threatening damage to the heart, lungs, brain, or kidneys.
There is no cure for SLE. However, there are treatments designed to improve symptoms. In particular, there are a range of options for treating skin problems.
• Medicines that can be taken by mouth (orally), applied as creams, or given as injections.
• Therapies to help people cope with their skin problems, such as talking therapies.
• Other approaches, including herbal medicine, light therapy, or make‐up.
To find out which treatments work best for people with SLE, and to compare adverse (unwanted) effects, we reviewed the evidence from research studies.
How did we identify and evaluate the evidence?
We searched the medical literature for studies that compared any treatment for skin disease in SLE against:
• a placebo (dummy) treatment;
• no treatment;
• another treatment; or
• a different dose of the same treatment.
We compared the results and summarised the evidence from all the studies. Finally, we rated our confidence in the evidence based on factors such as study methods and sizes and the consistency of findings across studies.
What did we find?
We found 61 studies that included 11,232 people (mostly women) and investigated 43 different treatments. Most treatments lasted one year, and people were followed for up to 48 months.
Here we report the main findings of our review on the effects of five different oral medicines: hydroxychloroquine, chloroquine, methotrexate, ciclosporin, and azathioprine.
Disappearance of skin problems
We do not know if hydroxychloroquine is better or worse than placebo at making skin problems disappear because no studies reported information about this.
The evidence suggests that:
• chloroquine may be better at making skin problems disappear after 12 months than placebo (1 study, 24 people);
• when we compare methotrexate and choloroquine, there may be little to no difference in how often they make skin rashes disappear after six months (1 study, 25 people);
• methotrexate may be better for making skin rashes disappear after six months than placebo (1 study, 41 people); and
• there may be little to no difference in how often skin problems disappear after 12 months between ciclosporin and aziathropine (1 study, 25 people).
Partial disappearance of skin problems (at least 50% improvement in the skin condition)
It is unclear if hydroxycholoroquine is better or worse than placebo at making skin problems disappear at least partially after 12 months. This is because the evidence is too imprecise (1 study, 20 pregnant women).
No other studies have examined how treatments affect the partial disappearance of skin problems.
The evidence suggests that after six months, fewer flares probably occur with hydroxychloroquine than with placebo (1 study, 47 people).
It is unclear if flares are more, or less, likely to occur after 12 months with methotrexate compared to placebo (1 study, 86 people).
No other studies have reported information on how treatments affect flares.
Evidence is often imprecise, and whether treatments lead to more or fewer adverse events than placebo or other treatments is not clear.
We found limited data for adverse events, and reports were discrepant, but hydroxychloroquine, chloroquine, and methotrexate have well‐known adverse effects including stomach and liver problems. Hydroxychloroquine and chloroquine can cause eye problems, and methotrexate can cause serious harm to a developing baby if taken during pregnancy.
We do not know how treatments affect other aspects of disease severity or quality of life. This is because studies did not report information on this.
What does this mean?
When compared against a placebo, studies in people with SLE show that:
• fewer flares probably occur with hydroxychloroquine; and
• methotrexate and chloroquine may be better at making skin problems disappear.
Information about adverse effects is limited.
How up‐to‐date is this review?
The evidence in this Cochrane Review is current to June 2019.
Cora W Hannon, Collette McCourt, Hermenio C Lima, Suephy Chen, Cathy Bennett
Implications for practice
We emphasise that clinicians should use caution in translating the findings of this review into clinical practice. Available evidence consists mainly of small randomised controlled trials (RCTs) for many of the interventions, resulting in imprecision and considerable uncertainty. We found only limited‐quality evidence, so we are unable to draw firm conclusions regarding the effects of interventions for cutaneous disease in systemic lupus erythematosus (SLE). The included studies assessed a range of interventions, and there were few opportunities to combine results in meta‐analyses.
Key interventions with potential benefit compared with placebo derived from low‐ to moderate‐quality quantitative evidence were as follows: methotrexate, dehydroepiandrosterone (DHEA), and chloroquine (complete clinical clearance), and hydroxychloroquine (clinical flare).
Treatments with possible benefit compared with placebo derived from mainly low‐quality evidence (narrative data) reported by study investigators were as follows: anifrolumab, baricitinib, belimumab, BIIB059, CC‐220, cosmetic camouflage, nifedipine, SM101, ustekinumab, and ultraviolet light (UVA)‐1 phototherapy.
Treatments without demonstrated benefit compared with placebo were intravenous abatacept, oral fish oil, oral ginsenosides, subcutaneous lulizumab pegol, oral nicardipine, oral prednisone, repository corticotropin gel (subcutaneous injections), intravenous sirukumab, and oral zileuton.
Treatments with more possible harms than benefits included copper and bortezomib. Oral copper did not demonstrate efficacy and demonstrated some evidence of serious harms including six withdrawals from the study and three hospital admissions (one for stroke and two for chest infection) based on narrative reports from authors. Although bortezomib was reported to result in improved skin in one patient, a very high rate (87.5%) of severe side effects led to discontinuation of the medication in other patients (narrative report).
In terms of achieving complete clinical response, we found the following results.
- Methotrexate or chloroquine may increase this outcome when compared with placebo (both low‐quality evidence).
- Little to no difference may be seen between ciclosporin versus azathioprine, or methotrexate versus chloroquine, in achieving this outcome (both low‐quality evidence).
This outcome was not reported for the comparison hydroxychloroquine versus placebo.
In terms of achieving partial clinical response, we found the following results.
- When compared with placebo, hydroxychloroquine may increase this outcome, but the 95% confidence interval (CI) indicates that hydroxychloroquine may make no difference to or may decrease this outcome (low‐quality evidence).
This outcome was not reported for the following comparisons: oral methotrexate versus placebo; oral chloroquine versus placebo; oral ciclosporin versus oral azathioprine; and oral methotrexate versus oral chloroquine.
In terms of reducing the number of clinical flares, we found the following results.
- Hydroxychloroquine probably increases this outcome when compared with placebo (moderate‐quality evidence).
- There may be no difference between methotrexate and placebo in effects on this outcome, but the 95% CI indicates that there may be more or fewer flares with methotrexate (moderate‐quality evidence).
This outcome was not reported for the following comparisons: oral chloroquine versus placebo; oral ciclosporin versus oral azathioprine; and oral methotrexate versus oral chloroquine.
We found limited data for adverse events, and reports were discrepant, but hydroxychloroquine, chloroquine, and methotrexate have well‐documented adverse effects including gastrointestinal manifestations, liver problems, and retinopathy for hydroxychloroquine and chloroquine, and teratogenicity during pregnancy for methotrexate.
Studies awaiting classification and numerous Ongoing studies may alter the conclusions of this review, especially given the fortunate development that more studies than ever before are reporting skin‐specific outcomes such as the Cutaneous Lupus Disease Area and Severity Index (CLASI).
Implications for research
This review highlights that most RCTs for SLE have not reported skin‐specific outcomes. Researchers are encouraged to collect and report skin‐specific outcomes given the importance of such information to patients and clinicians. Studies have shown that skin disease is more important to participants than internal organ damage (Tebbe 1997). When possible, dermatologists should be included in future SLE research teams to help with the collection and analysis of meaningful skin‐specific outcomes. The CS‐COUSIN core outcome initiative is a resource that future SLE trial authors can consult to select and report uniform skin‐specific outcome measures (CS‐COUSIN 2019). The use of uniform skin‐specific outcome measures, with aligned timing of outcome assessment, will increase the possibility that future systematic reviews and meta‐analyses can pool and analyse meaningful and homogenous data.
Different general systemic lupus activity measures have been compared for their relevance to the skin, and Parodi 2000 concluded that the most reliable general SLE activity measurement instrument for the skin is the SLAM score (integument domain). However, none of the general SLE activity measurement instruments was found to be entirely satisfactory for the purposes of dermatologic research, and for this reason, a skin‐specific disease activity index called the Cutaneous Lupus Area and Severity Index (CLASI) was developed (Albrecht 2007; Bonilla‐Martinez 2008; Krathen 2008). Skin‐specific composite outcome measures such as CLASI and dermatology‐specific quality of life measures should be used more often in studies to provide more patient‐centred data and because skin outcomes are extremely important to patients yet are relatively understudied. Wider use of standardised and validated dermatology quality of life (QoL) scores for trial participants diagnosed with this often disabling condition would provide data to aid treatment decisions. For example, dermatology QoL scores would help investigators to monitor change in dosage or drug or cessation of therapy.
Treatments may have effects of differing magnitude or direction for each of the different manifestations, requiring careful consideration and weighing of risks and benefits depending on individual preferences. Future studies need to take into account these potential complexities by measuring relevant outcome variables.
Systemic oral or intravenous steroids are often considered standard of care in SLE. More RCTs on the optimal use of systemic steroids and ideal combinations with steroid‐sparing agents would be useful for clinicians in practice.
Although studies were performed in a wide range of countries around the world including multiple countries in North America, South America, Europe, Asia, and Australia, only three RCTs included participants in Africa (Clowse 2015; Merrill 2016; Merrill 2018), and two RCTs included participants in the Middle East (Merrill 2016; Merrill 2018). Only one RCT involved the paediatric population (Lima 2016).
In addition, virtually no studies have looked at CLE in SLE treatment and outcomes over much longer periods of time. Longer‐term studies of CLE in SLE should be designed, so that the extent of relief from cutaneous symptoms from participants' perspectives (symptoms, quality of life) and safety and efficacy should be included in the design of such studies. We do recognise that long‐term studies may be difficult to perform for reasons including expense and attrition, with fewer participants remaining in the study over long periods of time.
Future research studies do not necessarily need to compare interventions against placebo but could be head‐to‐head intervention trials that would help clinicians better understand the efficacy and trade‐offs of different interventions.
It is known that cutaneous SLE disease subgroups are clinically heterogeneous and correspond to different prognoses and responses to treatment. For example, lupus‐non‐specific skin symptoms are associated with higher SLE disease severity than lupus‐specific skin symptoms (Parodi 2000; Zecevic 2001). Therefore, combining data on subgroups of cutaneous SLE may not always be appropriate. Studies also need to be designed to detect possible differences in efficacy depending on population subsets of those with cutaneous lupus. Ideally as many details about the exact subsets of cutaneous SLE should be reported in clinical studies, namely, details about lupus‐specific versus lupus‐non‐specific disease and specific types of skin involvement within these categories (such as oral ulcers versus alopecia) should be included to allow subgroup analysis.
In general, the quality of studies could be enhanced by improving reporting and following recommended experimental design in the areas of allocation, methods, blinding of assessors, and intention‐to‐treat analysis. Strand 2004 summarises some of the concepts related to the design of studies in SLE. It is also recommended that researchers perform a formal sample size calculation to ensure that trials are adequately powered that any meaningful effects can be detected and highlighted. Authors of SLE trials are encouraged to publish their data on CLE. If their typical speciality journals have word or space limitations, study authors could consider submitting the data to dermatology‐oriented journals as a separate paper, because dissemination of this collected information would be valuable for dermatology researchers, clinicians, and patients. Finally, authors of negative studies should be encouraged to publish, so that data are available for clinical interest and for possible meta‐analysis in the future.