Measures of Adult Systemic Lupus Erythematosus: Disease Activity and Damage

Many instruments are available to assess disease activity and capture damage in patients with systemic lupus erythematosus (SLE). These indices are frequently used in clinical trials, and some are used in epidemiological studies and daily practice, because they help to guide clinical decisions and evaluate treatment response. SLE is a heterogeneous disease with variable presentations of different organ systems and an unpredictable disease course in the same and different patients. Because of disease heterogeneity and lack of a gold standard reflecting the complete disease spectrum, a combination of different measures in composite indices is necessary to follow disease activity. Several indices have been developed over the last 30 years with different levels of validation 1; however, there is no universal consensus about their use. Feasible and sensitive assessment of SLE in clinical trials, observational studies, and clinical settings remains a challenge to the rheumatology community. This review presents a summary of the indices most frequently used in the last 5 years for disease activity and damage measurement in patients with SLE. Selection of these indices was based on evidence assessing their psychometric properties and popularity of use. The updated versions are included here with the original versions reviewed previously 2. Four disease activity measures have been included, one of which is an organ/system assessment scale that assesses disease activity in individual organs or systems (the British Isles Lupus Assessment Group [BILAG] 2004 index); the other three are global scoring systems that provide an overall measure of activity (the Systemic Lupus Erythematosus Disease Activity Index 2000 [SLEDAI 2K], the Systemic Lupus Activity Questionnaire [SLAQ], and the Lupus Foundation of America Rapid Evaluation of Activity in Lupus [LFA-REAL]). It is inherent in the scoring of these indices that only items attributable to active SLE are recorded and not items attributable to damage or comorbid conditions. The global scoring systems, such as the SLEDAI 2K and SLAQ, are binary (ie, they record clinical features to be present or absent and do not distinguish clinical features that are improving, but not resolved, from those that are unchanged or worse). The BILAG index and LFA-REAL, on the other hand, are able to capture improvement or worsening of persistent clinical features. We have also included the Cutaneous Lupus Erythematosus Disease Area and Severity Index (CLASI), which grades cutaneous manifestations of SLE, and the Safety of Estrogens in Lupus National Assessment (SELENA)–Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) Flare Index (SFI), which is useful for better capturing flares and is increasingly used in therapeutic trials in SLE. Damage is irreversible by definition in patients with SLE and may be driven by diverse factors: inflammation from SLE itself, concomitant comorbidity, side effects of drugs, and infectious complications. Damage is an important outcome in SLE and is known to predict morbidity and mortality 3-5. Two indices have been included in this review to assess cumulative damage: the Systemic Lupus International Collaborating Clinics (SLICC)/American College of Rheumatology (ACR) Damage Index (SDI) and the Brief Index of Lupus Damage (BILD). Patient-reported outcomes (PROs) have gained popularity for use in SLE because these are unique in capturing patients' perspectives of their disease. A majority of PRO measures focus on measurement of overall health assessment and quality of life in SLE; however, few are available for measurement of disease activity and damage 6, 7. The SLAQ and BILD, included in this review, are PRO measures, and recently a PRO version of the LFA-REAL has been described. These measures offer advantages of being simple to use and applicable in large epidemiological studies. The European League Against Rheumatism (EULAR) Outcome Measures Library is a growing and freely available database of validated PRO measures (http://oml.eular.org/), providing unified access to PRO measures to improve research and enhance the use of these measures in clinical practice 8. When considering the use of PROs, it should be recognized that it might be hard for a patient (and sometimes even for a physician) to distinguish whether the cause of a clinical feature is due to activity or damage or even both. One of the aims of the disease measures is to gauge improvement over time, which is especially relevant in clinical trials to be able to measure the response from interventions effectively. Different responder indices have been developed that are either derivatives or combinations of existent indices. The SLE Responder Index 4 (SRI 4) and the BILAG-based Combined Lupus Assessment (BICLA), the two most commonly used responder indices, have been included and reviewed for this purpose. Numerous failed SLE therapeutic trials are blamed, in part, on the deficiencies in available disease activity measures, highlighting the importance of the instrument selected to measure disease activity or damage in SLE. Disease activity, damage, and responder indices, with their specific characteristics, are reviewed to help clinicians and researchers in the selection of the most appropriate measure according to the intended purpose of the study or clinical evaluation. The BILAG 2004 index is a transitional score to assess organ-/system-based activity due to lupus based on the physician's intent-to-treat premise. The BILAG 2004 index, published in 2005 9, is a revision of the classic BILAG index published in 1988 10. This index comprises specific manifestations across nine organs/systems: constitutional, mucocutaneous, neuropsychiatric, musculoskeletal, cardiorespiratory, gastrointestinal, ophthalmic, renal, and hematological. There are 97 items in the 9 different organs/systems and 4 additional items for calculating the glomerular filtration rate. Each item (assumed to be due to activity) is scored as follows: ND = not done, 0 = not present, 1 = improving, 2 = same, 3 = worse, and 4 = new, yes, or no. The recall period refers to the assessment of manifestations in the last 4 weeks compared with the previous 4 weeks. There are no additional costs if the measure is performed manually. However, this instrument includes 97 items that need to be scored according to a glossary, adding complexity, even for a trained investigator, to complete manually. The BILAG 2004 index form, glossary, and scoring scheme are available at https://www.ncbi.nlm.nih.gov/pmc/artic​les/PMC29​19194/. The BILAG 2004 index is completed by the physician and requires additional training on use of the glossary items and scoring scheme. Each item is scored as either ND or from 0 to 4. Scores are then converted into alphabetical scores (A-E) for each system either through a computerized system or manually after training. Scoring is based on the principle of the physician's intention to treat, in which category A (for activity) implies severe disease activity requiring systemic high-dose oral glucocorticoids (equivalent to prednisolone at > 20 mg/dl), systemic immunomodulators, or high-dose anticoagulation; category B (for beware) implies moderate disease activity requiring systemic low-dose oral glucocorticoids (equivalent to prednisolone at ≤ 20 mg/dl), intramuscular/intraarticular or soft tissue glucocorticoid injection, or topical glucocorticoids, topical immunomodulators, or antimalarials; and categories C (for contentment], D (for discount), and E (for never, ever active) imply mild disease, inactive disease but previously affected, and system never involved, respectively. Ten to 15 minutes in addition to completing a history and physical examination. Usually up to 30 to 60 minutes. Some items cannot be scored until laboratory test results are available (usually later that day or the following day). Ophthalmic manifestations usually need to be assessed by an ophthalmologist, and these items would need to be recorded after the response is received from the ophthalmologist. The BILAG 2004 system tally (BST) and simplified BST divide systems into six and three components, respectively, based on the level of activity and have been shown to have good area under the receiver operating characteristic curve in predicting an increase in therapy (>0.8) 11. The BILAG 2004 pregnancy index has been shown to be reliable for assessment of disease activity in pregnant patients with SLE, with a good level of agreement greater than 70% in all systems 12. The British Lupus Integrated Prospective System (BLIPS) can calculate the BILAG 2004 index score and also records other disease activity indices. A web-based program, known as iBLIPS, has been launched and is continuously updated for use in research studies and clinical trials 13. It has been widely used for more than 10 years. Assessment of the BILAG 2004 index in a sample of 250 patients with SLE revealed a new A or B score in 61.6% of patients over a 12-month period. An A flare was observed in 10.4% of patients, a B flare (in which a B score was preceded by a D or E score) in 26.0% of patients, and 25.2% of patients had a B score in a system in which a C score was previously recorded 14. Isenberg et al 9 conducted two real-patient exercises with eight adult patients with SLE and eight rheumatologists during the development of the BILAG 2004 index and demonstrated good reliability (intraclass correlation coefficient [ICC] >0.60) and high levels of physician-patient agreement (ratio of SD attributable to the physicians/the SD attributable to the patient: <0.40) across all organ systems except for the musculoskeletal system 15. Interrater reliability was also assessed in a multicenter study by Yee et al 16 using 97 patients with 2 exercises (E1 and E2). Several issues were identified in the glossary following E1, which was then updated and additional training provided to the raters. The overall ICC determined in E1 was 0.45 (95% confidence interval [CI] 0.31-0.58) and improved to 0.67 (95% CI 0.54-0.76) in E2 16. Construct and criterion validity of the BILAG 2004 index to assess disease activity in SLE was determined in a multicenter cross-sectional study using 369 patients with SLE. Increasing overall scores, using the BILAG 2004 index, were associated with increasing erythrocyte sedimentation rates, decreasing C3 and C4 levels, elevated anti–double-stranded DNA levels, and higher SLEDAI 2K scores (P < 0.01). Scores indicating active disease (overall BILAG 2004 index scores of A and B) were significantly associated with an increase in therapy (odds ratio 19.3; P < 0.01). The BILAG 2004 and classic BILAG indices have comparable sensitivity, specificity, positive predictive value, and negative predictive value 17. The ability to detect change was studied in a prospective longitudinal study in which the relationship between change in disease activity and change in therapy between two consecutive visits was analyzed using 1761 assessments from 347 patients with SLE. An increase in the overall score was associated with an increase in therapy (coefficient: 1.35; 95% CI 1.01-1.70), and a decrease in the overall score was associated with a decrease in therapy (coefficient: 0.44; 95% CI 0.16-0.71) 18. By using the BILAG 2004 index, a flare can be defined in terms of the number of systems scoring a new A or B score based on items recorded as 4 (new) or 3 (worse). On this basis, severe flare is defined as a score of A in any system, and moderate flare is defined as two B scores, whereas mild flare is a single new B score 14. The numerical scoring system facilitates comparisons with global indices by converting the assessments so that A = 12 points, B = 8 points, C = 1 point, and D/E = 0 points 15. The BILAG 2004 index has been extensively included in randomized clinical trials (RCTs). However, it is important to consider that exploratory RCTs, by their nature, involve selected patient populations and focus on short-term efficacy. For long-term efficacy and safety, registry data are more useful to evaluate biologic agents in real-world practice between other registries, as exemplified by the BILAG Biologics Register 19 and the Registry of Systemic Lupus Erythematosus Patients of the Spanish Society of Rheumatology (RELESSER) 20. The BILAG 2004 index is a comprehensive index that has been shown to be valid, reliable, and sensitive to change. The BILAG 2004 index measures disease activity in different organs/systems separately and can also measure partial improvement in a given organ system, in contrast with the SLEDAI. Administrative time burden is a caveat to the BILAG 2004 index, along with the need for special training to the raters. The BILAG 2004 index is useful in assessing flares between consecutive visits. The assessment of disease flare using the BILAG 2004 index was studied in 16 patients who were also rated by a panel of 16 rheumatologists, and the rate of complete agreement for any flare versus no flare was 81% (95% CI 55%-94%) for the BILAG 2004 index and 75% (95% CI 49%-90%) for Physician Global Assessment (PGA) 21. The BILAG 2004 index records each item as new, the same, worse, or improving in separate organs, which is relevant when assessing the effect of an intervention in a clinical trial. To measure global disease activity from lupus modeled on clinicians' global judgment. The SLEDAI 2K 22 is the 2002 version of the original SLEDAI, which was published in 1992 23. The SLEDAI 2K includes evaluation of specific manifestations in nine organ systems. The SLEDAI 2K contains 24 items, of which 16 are clinical and 8 are based solely on laboratory test results. A manifestation is recorded if it is present, regardless of severity or whether it has improved or worsened. In the original index, the variables rash, alopecia, mucus membrane lesions, and proteinuria were counted as active only if they represented their first occurrence or recurrence (to distinguish from chronic lesions). In contrast, the SLEDAI 2K scores the presence of any rash, alopecia, and mucosal ulcers and a new, recurrent, or persistent proteinuria (urine protein levels >0.5 gm/24 hours) to capture persistent disease activity. This index records disease manifestations occurring within the 10 days preceding assessment. No additional costs. The SLEDAI 2K can be obtained by contacting Drs. Dafna Gladman, MD, and Murray Urowitz, MD, at Toronto Western Hospital, 399 Bathurst Street, IE 410B, Toronto, Ontario, M5T 2S8, Canada. The SLEDAI 2K is completed by the physician. Weighting is used, resulting in individual item scores ranging from 1 to 8, which are simply added to give a global score ranging from 0 to 105. Remission, low disease activity (LDA), and high disease activity (HDA) have been defined for SLEDAI 2K scores. Remission has been defined as no clinical manifestation (with or without serologic manifestations), although patients could be taking antimalarial medications only. LDA has been defined as a clinical SLEDAI 2K score less than 3 (with or without positive serology results) based on the presence of only one clinical manifestation, with a score range of 1 to 2 for patients who could be taking antimalarial medications but not glucocorticoids or immunosuppressive drugs. HDA has been defined as an SLEDAI 2K score greater than 6 24. Similarly, lupus LDA state has been defined using the following criteria, the attainment of which is associated with improved outcomes in SLE: 1) SLEDAI 2K score less than or equal to 4, with no activity in major organ systems; 2) no new lupus disease activity compared with the previous assessment; 3) PGA less than or equal to 1; 4) a current prednisolone (or equivalent) dose less than or equal to 7.5 mg daily; and 5) well-tolerated standard maintenance doses of immunosuppressive drugs and approved biological agents 25. The SLEDAI 2K takes approximately 10 minutes to complete. A history and physical examination are needed. The SLEDAI 2K cannot be scored until laboratory test results are available, which may take a few days; however, clinical SLEDAI 2K scores without laboratory test results could be reported. There are simplified English and Spanish versions of the SLEDAI without the immunological tests, which makes the index cheaper to administer, such as the MEX-SLEDAI 26. The SELENA-SLEDAI was devised for use in the SELENA study, in which the definitions of some of the descriptors in the SLEDAI were modified to ensure that the descriptors captured ongoing disease activity 27. A modified version of the SLEDAI has been developed for use in pregnancy 28. The SLEDAI 2K score measured over the last 30 days has been described and found to be comparable to the SLEDAI 2K score measured over 10 days 29. Recently, the SLEDAI 2KG, a modification of the SLEDAI 2K, has been suggested to describe disease activity while accounting for glucocorticoid dose category, and to identify more responders compared with the SLEDAI 2K 30. The mean SLEDAI 2K score in the initial validation cohort of 960 patients with SLE was 10.4 (SD 8.7). On studying outcomes of remission and LDA of SLE using a cohort of 1783 patients with lupus observed between 1970 and 2015, 620 were found to fulfill the criteria for one of the three groups (LDA: 80 patients [12.9%]; remission: 191 patients [30.8%]; HDA: 349 patients [56.3%]) 24. The reliability of the SLEDAI 2K was demonstrated in a multicenter and multiethnic study of 93 patients with SLE, with an agreement for each of the items between 81.7% and 100%. Gladman et al 22 showed a high correlation between SLEDAI 2K and SLEDAI scores across all visits (r = 0.97; P = and indices The SLEDAI 2K demonstrated a of the same over a range of disease activity as the SLEDAI when compared assessment of disease activity In study of patients with SLE, evaluation of validity of indices showed a correlation coefficient of between the SLEDAI 2K and PGA (P < of SLE disease activity indices in a longitudinal study of patients with SLE showed that the BILAG 2004 index and the SLEDAI 2K have the with treatment change The SLEDAI 2K score by more than 3 when the assessed that the patient was a flare In a longitudinal cohort of patients with SLE with of mean PGA and SLEDAI 2K scores demonstrated a high correlation = P < although the SLEDAI 2K had to evaluate for change in disease activity, defined by a PGA change greater than or equal to points from Activity categories have been and a score greater than 5 is associated with a of therapy in more than of patients. the SLEDAI not capture improving or worsening in a given organ system and less sensitive to it has been included in RCTs, for in the recently published for the B in patients with systemic lupus erythematosus with the BILAG 2004 index, to evaluate the use of after B therapy in patients with SLE The SLEDAI 2K is completed from the laboratory test results) and requires which for in clinical and research version of the SLEDAI worsening of an feature or partial their in mean SLEDAI 2K scores, which measure change in the SLEDAI 2K score over time, have been shown to with and other major outcomes in SLE, presence of damage and disease The SLEDAI 2K is one of the most commonly used global disease activity measures in longitudinal observational studies and clinical The SLAQ is a assessment of disease activity in SLE that is useful for and large groups of patients in epidemiological studies. It is modeled on the disease activity index Systemic Lupus Activity using items to patient The SLAQ of on 24 specific of disease activity, patient assessment of and patient global assessment of their disease activity. The SLAQ has with response categories and for specific and patient assessment of flare and has a numerical score of 0 to 10 for The previous 3 was because it may be for the patients in an epidemiological study to No additional costs. in A of the by et al or is available and can be or through are and in a to the scoring system used in the and scores can range from 0 to for the 2. a score of 9 points or more on the SLAQ results in a of of positive predictive value of and negative predictive value of to predict a score of 3 points on the to 10 minutes. to 10 minutes because it not any laboratory and validation is available in and In a validation study of the SLAQ using the Lupus Outcome sample from of SLAQ scores were to be although a was with of the sample scores greater than The SLAQ demonstrated with a of in the previously longitudinal cohort et al during development of the SLAQ, studied it in 93 patients with SLE and showed high correlation between the SLAQ score and the score (r = P < However, between of items from to In the study, the SLAQ demonstrated a to moderate of response were and for those clinical worsening and predictive for the SLAQ range from to for disease activity The SLAQ is useful to identify patients with new or disease activity who need evaluation by a with a positive predictive value ranging from to for disease activity. It is less frequently used in and more frequently used in epidemiological studies. The SLAQ is the first disease activity measure developed for patients with SLE. It has been shown to with and other PRO measures, such as the Assessment of Index 3 but it with other disease activity measures In the SLAQ, the patient is to rate the presence and severity of lupus but to SLE cannot be at of disease activity, such as laboratory of or disease, might not be associated with are not The SLAQ is intended to be used as an initial to identify patients with new or disease activity who may need evaluation by a physician. The SLAQ is useful in research studies in which the of patients and from the are to assessment by The LFA-REAL is an and SLE disease activity measure for use as an outcome measure in clinical trials as as in real-world clinical It is as an version of PGA on completing a separate PGA for each active organ The LFA-REAL has six or more the first six assess the most commonly organs, and other can be added to record features that do not the six categories or to separately score each in organs with two or more Each is in with 0 = 1 = 2 = and 3 = The to score the that are active in a given patient on the given day of No additional costs. as a in the by et al Physician Scoring requires simple addition of of individual items, and the score from 0 to An at as a for the of to 5 minutes. to 5 minutes because it not any additional laboratory Recently, a PRO version of the LFA-REAL was with a recall period of 4 patients to evaluate the severity and of associated with their current lupus activity through a of studies on assessment of the of LFA-REAL PRO scores, of and are under Usually two to in are scored for an individual LFA-REAL scores were compared when scored by trained lupus clinical or clinicians at two separate visits. The ICC between the scores of and clinicians was for 1 (P < and for 2 (P < higher were in trained In the initial validation study of consecutive patients with SLE, the LFA-REAL score of each with the SLEDAI, and BILAG index scores, with correlation coefficient of and (P < for all of the LFA-REAL was and compared with standard SLE measures and recently to in LFA-REAL scores at each with change in PGA (r = change in SLEDAI scores (r = and change in BILAG index scores (r = (P < in LFA-REAL scores also with scores of the responder indices 4 and (r = and P < Evaluation is Evaluation is The LFA-REAL is not included in so The LFA-REAL can be by a in the of patients with lupus within in practice, it can be scored for most patients. it is to detect in activity using the LFA-REAL, these may not are they the most points for clinical There also may be measurement in using due to in these no data on important or are The data from the LFA-REAL can be on the of individual or overall disease activity, and it is to distinguish between different combinations of or severe each to to an overall It is to the course of disease over time in response to in trials and in clinical The is an outcome measure used to assess and cutaneous manifestations in SLE and cutaneous lupus erythematosus The has two scores. The activity score activity of the disease, whereas the damage score damage by the disease. It is as a in which and scores for major clinical for activity membrane and no The of and is recorded for different for damage and in different and are recorded as of and from 0 to 6, on the of The severity of for each area is by the within each specific Assessment on the given No additional as a in the by et al and and Physician Scores are in each of the two of activity and To increase the more has been given to such as the and Patients are whether has in and has for more than 12 which is to be the score is and severe disease, based on the physician's assessment of with A score of 0 to 9, 10 to and to 2 and 10 minutes. to 10 minutes because no laboratory results are A version of the was in with additional items the of the added to the original score