The concept of systemic lupus erythematosus (SLE) as a relapsing and remitting condition is deeply embedded in how clinicians perceive this disease. This perception arises from observations of periods of low disease activity interspersed with episodes of more active disease, which are commonly referred to as “flares.” Although the concept of flare in SLE is widely accepted in general terms, how to define and measure flare in clinical research and clinical practice is a topic of debate that remains unresolved. In this issue of The Journal of Rheumatology, Rogers et al bring a further dimension to this debate, with respect to the differences between clinicians’ and patients’ understanding of the concept of flare.1

So first, what do clinicians mean by flare? In an attempt to standardize this, especially for research purposes, a consensus definition of flare was developed under the auspices of the Lupus Foundation of America as “a measurable increase in disease activity involving new or worse clinical signs and symptoms and/or laboratory measurements. This must be considered clinically significant by the assessor with usually at least some consideration of a change or an increase in treatment.”2 The emphasis here is on increase, differentiating flare from the concept of active disease through the addition of change over time. Indeed, the consensus document includes the statement, “a flare is an increase in disease activity as compared to a previous assessment.”2 Importantly, only clinicians participated in this consensus process.

For research purposes at least, flare is generally captured in registries and clinical trials using instruments such as the Safety of Estrogens in Lupus Erythematosus National Assessment–SLE Disease Activity Index (SELENA-SLEDAI) Flare Index (SFI)3 or by measuring increases in activity using the British Isles Lupus Activity Group (BILAG) index.4 Both of these include the concept of the intention to escalate treatment, such that the clinician’s intent is part of the definition of flare. Although clinically intuitive, this does render such instruments both more complex and more susceptible to variation in clinician responses, rather than being a simple objective change in disease activity. Again, the patient perspective is not considered in these measures of flare.

Patient perspectives are increasingly, though belatedly, receiving attention in SLE measurement, with efforts such as a global program to devise new primary outcome measures for SLE trials having prominent patient involvement.5 Indeed, patient-focused drug development is a major emphasis of regulatory agencies such as the US Food and Drug Administration, whereby effects of disease and treatment on how a patient “feels, functions, or survives” have gained primacy.6 How do we bring the established construct of flare in line with contemporary standards of incorporating the patient perspective?

In their study in this issue of The Journal of Rheumatology, Rogers et al undertook a qualitative study of patients and clinicians to shed light on this topic.1 These authors previously developed the model of Type 1 and Type 2 manifestations of SLE, wherein Type 1 manifestations reflect measurable immunological and/or inflammatory disease evident in one or more organ systems, and Type 2 manifestations include pain, fatigue, depression, and “brain fog,” which have a relationship to inflammation that is less clear. They selectively recruited 42 patients with SLE in 4 subgroups, with Type 1 manifestations, Type 2 manifestations, both, or neither. Type 1 and Type 2 manifestations were measured using the SLEDAI and the Polysymptomatic Distress Score (PSD), respectively, and standardized interviews were performed. Patients were most likely to describe flare as incorporating joint pain, fatigue, and skin symptoms; followed by swelling, myalgia, mood disturbance, and flu-like symptoms; and by brain fog, diffuse pain, and weakness. Importantly, patients mostly expressed the duration of flares as spanning only a few days, and all except the patients who had neither Type 1 nor Type 2 manifestations reported fatigue as a major feature.

In contrast, the 13 surveyed North American rheumatologists’ definition of flare was quite distinct, with a focus on measurable increases in organ-based disease activity, in line with the aforementioned consensus definition.2 How is this so different? We and many others have reported on the wide discrepancy between clinician-assigned disease activity measures and patient-reported outcome measures, showing they have weak or no numerical correlation.7 Indeed, the chief concerns of patients and clinicians are so divergent that in a previous survey, there was almost no overlap between the highest-ranked concerns of patients and those of their clinicians.8 Other studies have shown that patient-reported flares occur more frequently (85% in the previous year) than clinician-detected flares (0.19-1.2 flares annually per patient),9-12 again signifying that the word flare holds different meaning for these 2 groups. Patients with the highest rate of self-reported flares (> 7 in 1 year) sought urgent or emergency care more frequently but were not hospitalized more often.10,11 As Rogers et al noted,1 differences such as these create the potential for a divide to arise between patients and their treating clinicians, eroding trust and medication adherence, and harming clinical outcomes. How can this be reconciled?

Findings such as these provide valuable opportunities to validate patients’ experiences while also educating them about the possible biological basis of how they are feeling. From the perspective of a patient with SLE, flare is a valid experience, whether or not it is triggered by an immunological cause and associated with organ inflammation. Fluctuating and interacting triggers of patient-reported flares, be they immunological, lifestyle, psychological, hormonal, or other, are all contributors to poor quality of life (QOL) for a patient; however, not all patient-reported flares will require a change in immunosuppression or alter the disease trajectory. With the goal being the improvement of a patient’s QOL, the healthcare team should aim to identify and address all the factors that underlie patient-reported flares. Rheumatologists have the insight to effectively contribute to this, if they can be mindful of the whole picture, including the patient experience.

An important point of convergence between patient- and clinician-reported flares may be the association with stress. Patients reported this as a frequent precursor of their flares. Several studies have also illustrated an association between the onset of SLE, which by definition is a flare, and severe stress, such as posttraumatic stress disorder.13 Regardless of the characteristics, moving from remission to a state of active disease must, in the end, have a physiological basis, whether immunological or otherwise.

The immunological basis of SLE flare is a research field ripe for exploration. Why does the inflammatory response to the presence of autoimmunity worsen from time to time? More recent work has highlighted the important role of natural suppressors of immunity in maintaining healthy immune homeostasis. This raises the possibility that it is the disruption of this natural active suppression that results in flare—a physiological parallel to the disruption of therapeutic suppression that occurs when patients are nonadherent to their medications, or in response to withdrawal of glucocorticoids.14 An informative example is found in the biology of the protein glucocorticoid-induced leucine zipper (GILZ), a negative regulator expressed across the immune system at rest that tonically suppresses B cell activation,15 Th17 cell activation,16 and type I interferon (IFN) activation,17 all implicated in the pathogenesis of SLE. In fact, IFN itself inhibits GILZ in SLE,18 and increased serum concentrations of IFN are clearly linked to flare rates in SLE,19 suggesting an IFN-disruptible natural feedback loop for maintenance of immune homeostasis.

How such an immune-centric model applies to the construct of flare as recognized by patients requires further study. Opportunities exist to understand the basis and nature of patient-reported flares, and the discrepancy between patient- and clinician-reported SLE flares in particular yields important questions for the research community to unpick. How are patient flares temporally related to clinically defined SLE flares? Can we predict when one will lead to the other? Importantly, can patient QOL deterioration herald oncoming clinical flare and offer an opportunity for preventive action?

Deeply clinically annotated multi-omic biomarker studies may give actionable data on risk factors for flare. However, improving the identification of clinical phenomena mapped against such data to include the patient perspective is key. Perhaps the answer lies partly in semantics. The authors of this editorial consider the “medical” definition of flare to be synonymous with relapse, whereby the patient’s immune system fails, for whatever reason, to maintain a state of remission. Should the clinician community move to using the term relapse to define time-dependent increase in measurable inflammatory disease activity, leaving the term flare for patients to describe those characteristics important to them, as highlighted by Rogers et al?1 Moving toward separate terminology to described patient-assigned flares and clinician-assigned relapses will be helpful for addressing the biological associations that will guide future interventions for flare (relapse) prevention. In several recent trials, reduced rates of physician-assigned flare (relapse) were observed in response to belimumab or anifrolumab, suggesting that the construct of flare (relapse) can be addressed therapeutically.20-22

The puzzle of patient-reported flares comprises many pieces, and if we can assemble some of these through investigation and education, we can start to put the whole picture together. Building a foundation of evidence for the underlying causes of patient-defined flares as well as clinically evident relapses will allow the field to move toward reducing the effect of both on patients. Future work building on the important study by Rogers et al1 may reveal when a patient-reported spike in symptoms signifies a more sinister SLE-specific biological change to come, instructing a change of course for the treating clinician. This requires that we know what we are measuring.

EFM is supported by the National Health and Medical Research Council of Australia and Lupus Research Alliance. SAJ is supported by the National Health and Medical Research Council of Australia.

The authors declare no conflicts of interest relevant to this article.

See SLE flares, page 488