1. Introduction
Systemic lupus erythematosus (SLE) is a rheumatic condition that is challenging to diagnose and treat, mainly owing to its immense heterogeneity of clinical symptoms and complexity with regard to comorbidity burden, often necessitating interdisciplinary care, with the goal being the best possible quality of life and long-term outcomes [1][2]. With the premise that preventing is better than restoring, early diagnosis and treatment initiation is imperative [3], a need of particular urgency given that up to 10% of SLE patients develop life-threatening conditions or complications, such as end-stage kidney disease [4][5]. Patient-reported outcome measures (PROMs) receive increasing attention within the lupus research community, especially PROMs addressing HRQoL [6]. Even though this signifies a shift of the current paradigm towards increasing patient participation in their care, more distance has to be bridged before PROMs are an integral part of the evaluation in clinical practice.
2. PROMs in the Treat-to-Target Context
Remission of systemic symptoms and organ manifestations was identified by the treat-to-target for SLE initiative (T2T/SLE) as one of the most important targets in the management of patients with SLE
[7]. Several definitions of remission were used in clinical trials and observational studies of SLE
[8]. The Definitions Of Remission In SLE (DORIS) is an international task force consisting of expert rheumatologists, nephrologists, dermatologists, clinical immunologists, and patient representatives, who jointly proposed a set of remission definitions in response to the T2T/SLE research agenda
[9]. Later, the group decided on one prevailing remission definition that incorporated a physician-reported global assessment (PhGA), the SLE Disease Activity Index 2000 (SLEDAI-2K)
[10] after suppression of the serological descriptors (anti-double-stranded (ds)DNA and complement levels), the current daily glucocorticoid dose, and restrictions regarding medication allowance. Practically, however, some patients’ individual situations make it hard for them to achieve this stringent target. In such cases, one should aim for low disease activity (LDA). Again, several definitions of LDA have been proposed in the literature, with the Lupus Low Disease Activity State (LLDAS)
[11] being the most commonly used.
In addition, several recent clinical trials of SLE have used composite indices to define response to treatment, e.g., the SLEDAI Responder Index (SRI)
[12] or the British Isles Lupus Assessment Group (BILAG)-based Combined Lupus Assessment (BICLA)
[13]. Notably, none of the proposed definitions of remission, LDA, or response to treatment incorporate self-reported patient evaluation, which may constitute a pitfall of the definitions. Indeed, the DORIS task force discussed the issue of the non-inclusion of a PROM and partwise argued that the PhGA should incorporate the patient’s perspective by paying careful attention to the patient’s symptoms and experience
[14]. However, even the judgment of a very well-trained physician might not replace a PROM, not only owing to the multidimensionality of patient-perceived health experience (
Figure 1), but also since this notion would constitute a direct contradiction to the definition of a PROM; patient-reported outcomes are directly recorded by patients, without interpretation by their clinicians, and are additional markers in the assessment of treatment impact
[15][16]. In the definition of remission in rheumatoid arthritis (RA), the patient’s global assessment (PtGA) was chosen as one of the four Boolean criteria because of its ability to show a large sensitivity to change and its discriminatory validity between active drug and placebo in clinical trials
[14]. However, it is not easy to address the question of whether PtGA is an ideal PROM to complement the current definition of remission. In fact, a debate has been ongoing for a decade with regard to the appropriateness of PtGA to be included in the remission definition in RA, at least with the currently used cut-offs, since substantial proportions of patients score their disease activity higher than their rheumatologists
[17][18][19]. Due to the lack of evidence-based alternatives to PtGA, a summary report of an Outcome Measures in Rheumatology (OMERACT) interest group recently stated that currently, no better tools for representing the patients’ perspective are available, thus the definition should be kept as is
[20]. It is, however, worth noting that the differences in heterogeneity and complexity between SLE and RA are also reflected in discrepant longitudinal patterns of self-reported health experience
[21] and direct comparisons or extrapolations of psychometric properties of PROMs between the two diseases may be misleading.
Figure 1. Illustration of the multilateral impact across disease facets and health-related quality of life in people living with systemic lupus erythematosus.
The issue of imperfect agreement in the perception of disease severity between patients and physicians was also addressed in the field of SLE. While LDA is coupled with an overall favorable HRQoL experience, at least when assessed with the LLDAS
[22], results from other investigations are conflicting. One study used a Systemic Lupus Activity Questionnaire (SLAQ) score < 6 as the cut-off for LDA as perceived by the patients and found that only one-quarter of patients who were classified as being in LLDAS fulfilled the definition of SLAQ score < 6
[23]. In the same study, Medical Outcomes Survey Short Form 36 (SF-36) component summary scores and Functional Assessment of Chronic Illness Therapy Fatigue scale (FACIT-F) scores were higher among patients in LLDAS who reported SLAQ scores < 6 than among patients in LLDAS who reported SLAQ scores ≥ 6
[23]. This not only underscores that the physician’s perspective does not entirely represent the patient’s perception of HRQoL, but also highlights that LLDAS alone may not be a sufficient target in the management of people with SLE. The inclusion of HRQoL measures in treatment evaluation processes was supported by the T2T statements and received additional weight through findings indicating that poor HRQoL is associated with increased mortality
[7][24]. To this end, it is important to clarify that this discussion is not intended to devalue the PhGA or the rheumatologists’ ability to perform adequate clinical judgment; in fact, PhGA scores have shown good correlates with mental health, overall disease activity, and flares
[25].
3. The Matter of Not Only Optimal Choice but Also Optimal Use of PROMs
The OMERACT IV consensus conference
[26] propounded disease activity, HRQoL, medication side-effects, and organ damage as the four core outcomes for SLE clinical trials in that priority order. In light of accumulating evidence of the discordance in perceptions of disease activity between physicians and patients with SLE
[27], PROMs are increasingly used in SLE clinical trials
[6]. The SF-36
[28] and FACIT-F
[29] were reviewed for their psychometric properties with regard to the extent to which they comply with the US Food and Drug Administration (FDA) guidance
[30] and were suggested as secondary endpoints to support the labeling of novel therapies for SLE
[31]. Changes in scores in various SF-36 domains and FACIT-F have shown an ability to discriminate between verum drug (belimumab) and placebo in the BLISS-52 and BLISS-76 clinical trials
[32]. In the same analysis, changes in EQ-5D utility index scores
[33] did not exhibit discriminative ability. However, in light of satisfactory psychometric properties of EQ-5D for SLE patients, particularly in terms of validity and reliability
[34], a recent study investigated the discriminative ability and known-group validity of EQ-5D full health state (FHS), i.e., a utility index score of 1, and found a remarkably robust ability of EQ-5D FHS to discriminate between drug and placebo, and between responders and non-responders in the BLISS-52 and BLISS-76 clinical trials
[35]. This not only illustrates the need for determining optimal PROMs in SLE but also supports the notion that optimal use of the currently available ones may be even more important. For example, the differential ability of PROMs to capture changes in the different SLE disease patterns, i.e., persistently quiescent, persistently relapsing-remitting, and persistently active disease
[36], comprises one of the many questions framing the future research agenda.