Failure to achieve lupus low disease activity state (LLDAS) six months after diagnosis is associated with early damage accrual in Caucasian patients with systemic lupus erythematosus

Background The aim was to assess the attainability and outcome of the lupus low disease activity state (LLDAS) in the early stages of systemic lupus erythematosus (SLE). Methods LLDAS prevalence was evaluated at 6 (T1) and 18 (T2) months after diagnosis and treatment initiation (T0) in a monocentric cohort of 107 (median disease duration 9.7 months) prospectively followed Caucasian patients with SLE. Reasons for failure to achieve LLDAS were also investigated. Multivariate models were built to identify factors associated with lack of LLDAS achievement and to investigate the relationship between LLDAS and Systemic Lupus International Collaboration Clinics (SLICC)/Damage Index (SDI) accrual. Results There were 47 (43.9%) patients in LLDAS at T1 and 48 (44.9%) at T2. The most frequent unmet LLDAS criterion was prednisolone dose >7.5 mg/day (83% of patients with no LLDAS at T1). Disease manifestations with the lowest remission rate during follow up were increased anti-double-stranded DNA (persistently present in 85.7% and 67.5% of cases at T1 and T2, respectively), low serum complement fractions (73.2% and 66.3%) and renal abnormalities (46.4% and 28.6%). Renal involvement at T0 was significantly associated with failure to achieve LLDAS both at T1 (OR 7.8, 95% CI 1.4–43.4; p = 0.019) and T2 (OR 3.9, 95% CI 1.4–10.6; p = 0.008). Presence of any organ damage (SDI ≥1) at T2 was significantly associated with lack of LLDAS at T1 (OR 5.0, 95% CI 1.5–16.6; p = 0.009) and older age at diagnosis (OR 1.05 per year, 95% CI 1.01–1.09; p = 0.020). Conclusion LLDAS is a promising treatment target in the early stages of SLE, being attainable and negatively associated with damage accrual, but it fit poorly to patients with renal involvement.


Background
Systemic lupus erythematosus (SLE) is a chronic multisystem disease, characterized by wide heterogeneity in clinical presentation, course and responsiveness to therapy [1,2]. Despite significant advances in the understanding of the pathophysiology and optimization of medical care [3,4], patients with SLE have an agestandardized mortality rate of 3.2 per 1 million people [5] and carry a risk of progressive organ damage accrual and reduced health-related quality of life [6,7].
The treat-to-target (T2T) approach, successfully applied in rheumatoid arthritis and other non-rheumatic conditions, is a therapeutic strategy aiming to improve disease outcomes through the achievement of a pre-specified goal [8]. An international task force has recently suggested applying the T2T strategy in SLE, recommending that the treatment target should be remission or, where this cannot be reached, the lowest possible disease activity [9]. Many different definitions of remission, none of them generally accepted, are currently available in the literature and many controversial issues have to be clarified, such as its real attainability as a treatment target in clinical practice [10][11][12]. On the one hand, a framework for definition of remission in SLE (DORIS) has been recently set up by an international expert group [12]. On the other hand, a composite definition of minimal acceptable disease activity, the lupus low disease activity state (LLDAS), has been proposed by the Asia-Pacific Lupus Collaboration (APLC) [13]. LLDAS is based on the following criteria: (1) SLE Disease Activity Index 2000 (SLEDAI-2K) ≤4, with no activity in major organ systems (renal, central nervous system, cardiopulmonary, vasculitis, gastrointestinal, haemolytic anaemia fever); (2) no new lupus disease activity compared with the previous assessment; (3) Safety of Estrogen in Lupus Erythematosus National Assessment (SELENA)-SLEDAI physician global assessment (PGA) ≤1; (4) current prednisolone (or equivalent) dose ≤7.5 mg daily; and (5) well-tolerated standard maintenance doses of immunosuppressive drugs and approved biological agents [13].
Early damage was demonstrated to be associated with lower 10-year survival, such that patients with initial damage have a fourfold higher mortality rate compared to those with no early damage [14]. In order to be considered a valid treatment target, LLDAS should be proved as being protective against damage accrual in the early SLE stages. However, LLDAS prevalence and its association with damage have been investigated in a few longstanding SLE cohorts with no homogeneous disease duration [13,15,16] and therefore no data are currently available on its potential role as an early treatment target.
This study primarily aimed to assess the frequency of LLDAS achievement and its association with early damage accrual in a homogeneous cohort of Caucasian patients with SLE prospectively assessed during the first 18 months of treatment after diagnosis. The secondary aim was to identify the main reasons for failure to achieve LLDAS.

Patients
Data from patients included in the Cagliari (Italy) SLE cohort [17] between 1 January 2006 and 31 December 2016 were used for this study. Inclusion criteria were: (a) SLE diagnosed according to the revised 1997 American College of Rheumatology (ACR) criteria [18]; (b) starting the first treatment for SLE at enrolment in the Cagliari cohort; (c) moderate to severe disease activity (SLEDAI-2K ≥6) at baseline; (d) at least quarterly visits during the study interval; and (e) age ≥18 years.
Retrospective analysis of prospectively collected data was performed, reviewing medical records and clinical files. Baseline was fixed at the time of diagnosis, which corresponded to treatment initiation (T0). No specific recommendations are currently available on the appropriate time to assess the target achievement in a T2T strategy for SLE [9]. In order to assess LLDAS as a goal for initial treatment we set the primary study endpoint at 6 months (T1), according to the average induction therapy duration recommended for severe disease manifestations, such as glomerulonephritis [19]. Eighteen months (T2) was considered as an appropriate time to evaluate the effect of maintenance treatment and the early damage accrual [14,20].

Patient assessment
Demographic factors including gender, age at onset and diagnosis were collected. Disease onset was defined as the time of appearance of the first clinical classification criterion. Disease duration was the time period between disease onset and diagnosis. The number of 1997 ACR criteria presented from disease onset up to the baseline was recorded. At baseline, antinuclear antibodies (ANA) (tested by indirect immunofluorescence (IIF), using Hep2 cell substrate, positivity was defined as a titre >1:160), anti-Ro/SSA, anti-La/SSB, anti-Sm, anti-RNP, lupus anticoagulant, anticardiolipin (IgG nv <12 GPL; IgM nv <12 MPL) and anti-B2-glicoprotein1 (IgG nv <12 GPL; IgM nv <12 MPL) antibodies were assessed. Anti-dsDNA antibodies (by Farr assay; nv <10 IU/mL) and C3 (nv 90-180 mg/dL), C4 (nv 10-40 mg/dL) serum complement fractions were quarterly tested.
At each visit, disease activity was assessed using the SLEDAI-2K score and the PGA (0-3) [21]. Damage accrual was assessed at 18 months by the SDI [22] and the possible attribution to corticosteroids was done according to a previous definition [23]. Ongoing use and new prescription of corticosteroids, anti-malarial and immunosuppressant drugs (azathioprine, methotrexate, mycophenolate mofetil, cyclosporine A, cyclophosphamide or rituximab) were assessed at every visit. Average daily dose of prednisolone (or equivalent) was recorded for each patient at every study follow-up visit.

Data analysis
Normally and non-normally distributed variables were summarized using mean ± standard deviation (SD) and median with interquartile range (IQR), respectively. Univariate and multivariate analysis were performed to evaluate potential factors associated with failure to achieve LLDAS. Univariate analysis was performed using the two-sample Student t test or Mann-Whitney test for quantitative variables, and the Chi-square test or Fisher's exact test for qualitative variables. Variables with a p value <0.1 on univariate analysis were included in a multivariate model for stepwise logistic regression. The odds ratio (OR) with 95% confidence interval (95% CI) was calculated.
To assess the association between lack of LLDAS achievement and damage development, a logistic regression model was created. Occurrence of any damage (SDI ≥1) at 18 months was included as the dependent variable, whereas age, disease duration, male gender, average daily steroids dosage, renal involvement, use of anti-malarial or immunosuppressant drugs [24] and failure to achieve LLDAS at T1 and at T2 comprised the independent variables. Statistical significance was set at a p value <0.05. MedCalc® statistical software (Mariakerke, Belgium) was used.

Patients
Overall, 178 new patients joined the Cagliari (Italy) SLE cohort during the study interval. The study cohort consisted of 107 (60.1%). The relevant features of Caucasian patients with SLE who fulfilled the inclusion criteria for study enrolment are summarized in Table 1. Regarding excluded patients, 31 were diagnosed elsewhere and were already on treatment at the time of enrolment, 25 had no quarterly follow up, 7 were younger than 18 years and 6 had SLEDAI-2K <6 at baseline.

Persistently active SLEDAI items
Individual SLEDAI-2K items were grouped into neurologic, vasculitis, arthritis, renal, myositis, mucocutaneous, serositis, low complement, increased anti-dsDNA and haematologic disorders. SLEDAI-2K items with worse remission rates were high anti-dsDNA concentration, complement consumption and renal abnormalities. Increased anti-dsDNA levels were still present in 85.7% and 67.5% of patients at T1 and T2 respectively, low complement in 73.2% and 66.3% of patients and renal disorders in 46.4% and 28.6% of patients (Fig. 3).

Discussion
In our study cohort 44% and 45% of patients were in LLDAS 6 and 18 months after treatment initiation, respectively. Despite a seemingly overall stable LLDAS rate, the dynamic nature of this condition was demonstrated. In fact, 14% of subjects first achieved LLDAS between the 6 th and 18 th month, whereas 13% of patients who were in LLDAS after 6 months of treatment lost LLDAS within the following 12 months. Golder et al. found that in a in a cross-sectional assessment, 44% of 1846 patients with SLE and a mean disease duration of 8.6 years did not achieve LLDAS [15]. Franklyn et al., in an SLE cohort with mean disease duration of 7.1 years, observed that 88% of patients achieved LLDAS at least once and that 38% of patients had this status for more Fig. 3 Frequency (percentage) of manifestations still identified at 18 months (T2) and respective remission rate. Purple, frequency (percentage) of manifestations recorded at T0 and still present at T2. Dark blue, manifestations that were in remission at 6 months (T1). Light blue, manifestations that were recorded as in remission for the first time at T2. Below the bars the global remission rate is reported than half of the follow-up duration [13]. These data suggest that the minimal disease activity, according to the LLDAS criteria, would be a more applicable goal in a T2T strategy than remission, according to the current definition [12,25], especially in early disease stages [10]. Actually, Wilhem et al. observed that the median time to achieve durable remission, classified in four subtypes according to the principle proposed by the DORIS group, ranged between 1.8 and 11.0 years [10].
In this study, the most frequent reason for failure to achieve LLDAS 6 months after therapy initiation was daily prednisolone dosage >7.5 mg (83% of no LLDAS). In the study by Golder et al., prednisolone dosage was an unmet criterion in 57% of patients who did not achieve LLDAS [15]. In a study aiming to test the construct validity of LLDAS compared to expert opinion, disagreement was mainly due to low disease activity classification of patients with prednisolone daily dosage  ≥7.5 mg (up to 10 mg/day), suggesting that such a higher dose was deemed acceptable by several experts who were interviewed [26]. Nevertheless, the lowest corticosteroid dose to be considered safe has not been definitely identified. Prednisolone >5 mg/day was reported to be associated with greater risk of osteoporosis, infections, cataracts and metabolic and cardiovascular disorders [27][28][29]. Tharmer et al. observed that the risk of damage accrual in patients with SLE did not substantially increase with prednisolone cumulative doses <180 mg/month (equivalent to 6 mg/day) [30]. In our cohort, damage was definitely attributable to steroid use in 40% of cases. Gladman et al. reported that 58% and 80% of the damage accrued in the first year after diagnosis and in later disease, respectively, could be described as "possibly or definitely" steroid-related [23]. In contrast, a larger proportion of early damage was related to early inflammatory organ damage [24]. In our study we found that failure to achieve early LLDAS (at 6 months) and older age at diagnosis, but not corticosteroid daily dose, were independently associated with presence of damage after 18 months. Two further retrospective studies reported that patients with LLADS had a reduced risk of damage accrual in at least half of observations [13,16]. Supported by our data and literature evidence on damage development, we consider 7.5 mg/ day an acceptable cutoff to define low disease activity during initial treatment, but A lower cutoff should be targeted to minimize risk of steroid-related damage during maintenance therapy in patients with SLE. Renal involvement and serological disorders, consisting of low complement and increased anti-dsDNA, had the lowest remission rate in our cohort. However, renal involvement at baseline was the most important factor associated with failure to achieve LLDAS at 6 months (OR 7.8, 95% CI 1.4-43.4; p = 0.019) and 18 months (OR 3.9, 95% CI 1.4-10.6; p = 0.008) from treatment initiation. In a cross-sectional study, increased anti-dsDNA (OR 0.65, 95% CI 0.53-0.81; p < 0.001), renal involvement (OR 0.60, 95% CI 0.48-0.75; p < 0.001), low complement (OR 0.52, 95% CI 0.40-0.67; p < 0.001) shorter disease duration (OR 0.31, 95% CI 0.19-0.49; p < 0.001) and history of discoid rash (OR 0.66, 95% CI 0.49-0.89; p = 0.006) were identified as negatively associated with LLDAS [15]. These findings suggest that renal involvement and serological abnormalities are associated with a lower likelihood of LLDAS achievement.
Whether or not serological disorders should be considered in the definition of minimal disease activity it is a matter of debate [26]. Persistently increased anti-dsDNA and low complement may be part of the serological active clinically quiescent (SACQ) disease pattern [31], with some evidence suggesting a proportion of patients with SACQ disease can spend years without flares [32] whilst other may relapse [33]. However, increasingfluctuating anti-dsDNA level rather than absolute values can predict disease flares [34]. Thus, in order to enhance the association between minimal acceptable disease activity and long-term outcomes, it would be conceivable to consider rising levels of anti-dsDNA antibodies rather than their steady positivity in the LLDAS definition.
This study has some limitations. First, the relatively small sample size may have hampered the study results. Nevertheless, by enrolling consecutively diagnosed patients at the time of treatment initiation and following them up prospectively, we added some novel information on LLDAS as a potential treatment target. Second, the retrospective design of the study prevented us from testing LLDAS criterion validity by comparing it with other treatment targets such as the SLE Responder Index [35].

Conclusion
In conclusion, LLDAS is a promising treatment target in SLE, being attainable and negatively associated with damage accrual in the early stages of disease. However, it seems to poorly fit with the heterogeneity of clinical presentation in patients with SLE, mostly in those with renal involvement. On the one hand, the poor LLDAS attainability in patients with renal involvement could be due to greater severity and lasting disease activity in the renal domain. On the other hand, it could be due to the unsuitability of LLDAS in capturing the minimal disease activity in patients with renal involvement. Finally, greater consensus should be reached on the definition of serological disorders and prednisolone dose to be included in a unique definition of minimally acceptable disease activity.