Abstract
Objective To assess whether Lupus Low Disease Activity State (LLDAS) attainment is associated with favorable outcomes in patients with recent onset systemic lupus erythematosus (SLE).
Methods Data from a 13-country longitudinal SLE cohort were collected prospectively between 2013 and 2020. An inception cohort was defined based on disease duration < 1 year at enrollment. Patient characteristics between inception and noninception cohorts were compared. Survival analyses were performed to examine the association between LLDAS attainment and damage accrual and flare.
Results Of the total 4106 patients, 680 (16.6%) were recruited within 1 year of SLE diagnosis (inception cohort). Compared to the noninception cohort, inception cohort patients were significantly younger, had higher disease activity, and used more glucocorticoids, but had less organ damage at enrollment. Significantly fewer inception cohort patients were in LLDAS at enrollment than the noninception cohort (29.6% vs 52.3%, P < 0.001), but three-quarters of both groups achieved LLDAS at least once during follow-up. Limiting analysis only to patients not in LLDAS at enrollment, inception cohort patients were 60% more likely to attain LLDAS (hazard ratio 1.37, 95% CI 1.16-1.61, P < 0.001) than noninception cohort patients and attained LLDAS significantly faster. LLDAS attainment was significantly protective against flare in both the inception and noninception cohorts. A total of 88 (13.6%) inception cohort patients accrued organ damage during a median 2.2 years of follow-up.
Conclusion LLDAS attainment is protective from flare in recent onset SLE. Significant protection from damage accrual was not observed because of low rates of damage accrual in the first years after SLE diagnosis. (ClinicalTrials.gov: NCT03138941)
Lupus Low Disease Activity State (LLDAS) is a validated treatment target in patients with systemic lupus erythematosus (SLE).1 In addition to undergoing multicenter prospective validation,2 LLDAS has been studied in numerous cohorts with consistent protective associations with reduced damage accrual, flare, and mortality, as well as improved quality of life reported.3-12
The natural history of SLE varies between patients from monophasic active disease, patients with periods of good disease control contrasted by flares, and patients who experience persistently active disease.13 Higher disease activity is an established predictor of irreversible organ damage,14,15 which in turn can lead to early mortality.16 In comparison to those with established disease, patients with newly diagnosed SLE tend to have higher disease activity levels but less irreversible damage,17 as damage typically accrues slowly. Given that existing damage is known to propagate further damage accrual,18 recently diagnosed patients afford a unique potential opportunity to prevent irreversible organ damage and halt progression to poor outcomes—such as flare, poor quality of life, and death—by achieving and maintaining a protective treatment target early in the disease course.
Studies assessing the protective effects of LLDAS attainment have been conducted in cohorts with variable disease durations. Whether the protective associations of LLDAS are present in recent onset disease and whether the magnitude of protection differs compared to patients with established disease is less well known, with only a few inception cohorts studied,11,19-21 and only one with a direct comparison to patients with more established disease.22
In this study, we compared LLDAS attainment and the protective effects thereof between inception and noninception cohort patients with SLE who are part of a large multinational prospective longitudinal cohort study.
METHODS
Study population. Data from patients across 13 countries participating in the Asia Pacific Lupus Collaboration (APLC),23 collected prospectively between 2013 and 2020, were used to conduct this analysis. All patients were adults aged ≥ 18 years and met either the 1997 American College of Rheumatology (ACR) Modified Classification Criteria for SLE or the Systemic Lupus International Collaborating Clinics (SLICC) 2012 Classification Criteria. All patients provided informed consent.
Research ethics approval. Each site obtained local ethics approval to participate in the APLC research activities. Storage of the central dataset and analyses of the pooled data have been approved by the Monash University Human Research Ethics Committee (MUHREC project ID 187c78). This study is registered with ClinicalTrials.gov: NCT03138941.
Data collection. Data were collected during patients’ routine clinical assessments using standardized paper or electronic data collection forms. Patient demographics (age, sex, ethnicity, dates of SLE symptom onset and diagnosis, smoking status, education level, SLE family history) and SLE classification criteria (ACR and SLICC) were collected at enrollment (baseline visit). Variables related to organ damage (SLICC/ACR Damage Index [SDI]) were collected at baseline and annually. Disease activity (Systemic Lupus Erythematosus Disease Activity Index 2000 [SLEDAI-2K], Safety of Estrogens in Lupus Erythematosus National Assessment–SLEDAI flare index, and physician global assessment [PGA; 0-3]), SLE-related medications including doses (antimalarials: hydroxychloroquine, chloroquine; immunosuppressants: mycophenolate, mycophenolic acid, azathioprine, cyclosporine, methotrexate, tacrolimus, leflunomide, cyclophosphamide; biologics: rituximab, belimumab), and routine laboratory results were recorded at each visit. Details of death (date and cause) were recorded in the database at time of death. A time-adjusted mean (TAM) SLEDAI-2K was calculated as a measure of disease activity over time.24 The TAM-PGA and TAM-prednisolone (PSL) dose were similarly calculated.
Key variables and definitions. Organ damage accrual and occurrence of flare episodes were the 2 primary outcomes assessed.
Patients who were recruited to the APLC cohort within ≤ 1 year of SLE diagnosis were classified as the inception (newly diagnosed) cohort.
LLDAS was defined as described.2 Briefly this requires SLEDAI-2K ≤ 4 with no major organ activity, absence of new SLEDAI-2K activity compared with the preceding visit, PGA (0-3) < 1, and PSL or equivalent ≤ 7.5 mg/day. Antimalarials and standard maintenance doses of immunosuppressants were allowed.
Percent time in LLDAS was calculated as the sum of all time intervals in LLDAS divided by the total length of follow-up, and then multiplied by 100, as described.2 Time intervals in LLDAS equaled the length of time between visits if a patient was in LLDAS at 2 consecutive visits, or half the length of time between visits if a patient was in LLDAS at 1 visit but not the other. Although proportions of cumulative time in LLDAS as low as 20% have been shown to be protective against adverse outcomes,2 a threshold of > 50% cumulative time in LLDAS (LLDAS-50) was used in line with reporting practices in other studies of this endpoint.25 In addition, we derived ≥ 3 month, ≥ 6 month, and ≥ 12 month sustained LLDAS variables if patients were in LLDAS continuously for at least 91 days, 182 days, or 365 days, respectively.
Gross domestic product (GDP) based on purchasing power parity per capita (simply referred to as GDP, representing the value of all final goods and services produced within a country divided by the average population) as estimated by the International Monetary Fund 2021 was used as a surrogate marker to account for regional variations.
Statistical analysis. Statistical analyses were performed using Stata V.17.0 (StataCorp). Continuous variables are described as median with IQR because of the skewed nature of data distribution, whereas categorical variables are described as frequency and percentage, stratified by inception vs noninception cohorts. Multivariate survival analysis was used to examine the longitudinal associations of LLDAS attainment and organ damage accrual and flare at subsequent visit, since several patients experienced recurrent flare or accrued damage > 1 time. The total time from baseline to the outcome event was fitted using the Prentice, Williams, and Peterson model to set up the data to allow multiple “failures” (ie, outcomes) per patient. Clustering was specified in regression models to account for intragroup correlation. Multivariable regression analyses were carried out to account for potential confounders. Validity of the statistical models was tested using diagnostic tools. The results from survival analyses are presented as hazard ratios (HRs) with corresponding 95% CIs. P values < 0.05 were considered statistically significant.
RESULTS
Demographic and clinical characteristics of the study cohort. From a total of 4106 patients enrolled, patients with unknown disease duration and/or missing LLDAS status were excluded, restricting the study population to 4100. The majority (92%) of patients were female and had a median (IQR) age of 29 (21.0-39.0) years at SLE diagnosis and 39 (30.0-50.0) years at enrollment. Patients were predominantly of Asian ethnicity (88.9%). Irreversible organ damage (SDI > 0) was present in 38.2% of patients at enrollment. The median (IQR) study duration of the whole cohort was 2.5 (1.0-5.1) years. During the study observation period, 85% (3484/4100) of patients used PSL, 78.9% (3236/4100) used antimalarial medications, and 70.4% (2887/4100) used immunosuppressants at least once, alone or in combination. Approximately 53.2% (2180/4100) of patients experienced ≥ 1 flare and 19.5% (717/3672) accrued organ damage (ΔSDI > 0).
Patient characteristics in inception vs noninception cohort. A total of 680 (16.6%) patients had a disease duration of < 12 months (inception cohort). We compared patient characteristics between inception and noninception cohorts (Table 1). In brief, patients in the inception cohort were significantly younger, but noninception cohort patients were diagnosed at a younger age, consistent with their longer duration of follow-up. A significantly higher proportion of inception cohort patients were from countries with GDP < Int$20,000. During the follow-up period, inception cohort patients had higher disease activity measured by TAM-SLEDAI-2K (P = 0.002) and TAM-PGA (P < 0.001) and were more likely to use glucocorticoids (GCs), antimalarials, and immunosuppressants compared to the noninception cohort; TAM-PSL dose was likewise higher in the inception cohort. Correspondingly, flares were more frequent in the inception cohort (Table 1 and Table 2). In contrast, inception cohort patients were less likely to accrue organ damage (P < 0.001). Thirty percent (201/680) of the inception cohort patients and 52.3% (1787/3420) of the noninception cohort were in LLDAS at recruitment (P < 0.001), and 74% of the inception cohort and 79.4% of the noninception cohort achieved LLDAS at least once (LLDAS-ever; P = 0.002; Table 1).
LLDAS attainment and outcomes of noninception and inception cohort patients. We next examined LLDAS attainment at annual timepoints following enrollment. Eighty-two percent (2806/3420) of noninception cohort patients had ≥ 1 year of follow-up data, 55.9% (1913/3420) had ≥ 2 years of follow-up data, 47.1% (1610/3420) had ≥ 3 years of follow-up data, 36.6% (1253/3420) had ≥ 4 years of follow-up data, and 29.6% (1013/3420) had ≥ 5 years of follow-up data. In the inception cohort, 76.5% (520/680), 52.2% (355/680), 38.1% (259/680), 22.5% (153/680), and 13.5% (92/680) had ≥ 1 year, ≥ 2 years, ≥ 3 years, ≥ 4 years, and ≥ 5 years of follow-up data, respectively (Table 2).
At the first year from enrollment annual assessment, the proportions of LLDAS-ever patients were similar in both inception and noninception groups (71.1% and 71.4%, respectively), but became higher at subsequent annual timepoints in the inception cohort compared to the noninception cohort. In contrast, proportions of patients with the majority of their cumulative time in LLDAS (LLDAS-50) and with sustained LLDAS were higher in the noninception cohort in the early years but gradually became similar across the 5 years of annual follow-up (Table 2).
Damage accrual was significantly more frequent in the noninception patients from year 1 to year 4 (Table 2). In contrast, more inception cohort patients experienced flares across all 5 years of observation (Table 2).
Time to first LLDAS attainment. We conducted a survival analysis (Cox regression model) comparing the survival probability of first attaining LLDAS in patients who were not in LLDAS at baseline between the inception and noninception cohorts. Inception cohort patients attained LLDAS significantly faster than noninception cohort patients (Figure 1). When the effect of potential confounders was ignored (ie, univariable Cox regression model), we observed that inception cohort patients were approximately 60% more likely to attain LLDAS than noninception cohort patients (unadjusted HR 1.60, 95% CI 1.40-1.82, P < 0.001). After adjusting for the confounding effects of age at enrollment, disease duration, and GDP (Supplementary Table S1, available with the online version of this article), the magnitude of effect was slightly reduced but remained statistically significant (adjusted HR 1.37, 95% CI 1.16-1.61, P < 0.001).
LLDAS attainment and organ damage accrual. We examined the longitudinal association of LLDAS attainment with irreversible organ damage accrual in the inception and noninception cohorts separately. We first examined these associations using data for all available visits, then restricted the analyses to patients with ≥ 1 year, ≥ 2 years, and ≥ 5 years of follow-up data. A statistically significant negative association between LLDAS attainment and organ damage accrual was observed in the noninception cohort (Table 3). In the noninception cohort, these observations were consistent in both univariable models (not shown) and multivariable models, accounting for patients’ age, GDP of their country, and for the presence of baseline organ damage (Table 3), and were consistent whether analyzing individual visits in LLDAS, sustained LLDAS, or cumulative LLDAS-50 (Table 3). Sustained LLDAS for 12 months provided the most significant protection against organ damage accrual, which was consistently observed in all 3 subcohorts with 1, 2, and 5 years of data (Table 3). In contrast, statistical significance was not observed for associations between LLDAS attainment and damage accrual in the inception cohort (Table 3).
LLDAS attainment and flare. LLDAS attainment was significantly protective against flare at subsequent visits in both inception and noninception cohorts (Table 4). Dose-dependent, independent protective associations of sustained LLDAS with flare were also observed in both inception and noninception cohorts. For instance, based on the results from patients with ≥ 5 years of data, the HR of flare occurrence subsequent to spending ≥ 12 months in LLDAS was 0.33 (95% CI 0.28-0.39) in the noninception and 0.38 (95% CI 0.22-0.64) in the inception cohorts (Table 4). These HRs are adjusted for patients’ age and GDP of their country.
Effect of LLDAS attainment within 6 months. In the subgroup analysis, we assessed the effect of attaining LLDAS within 6 months of cohort entry within the inception cohort patients. Compared to patients who did not achieve LLDAS, those inception cohort patients who attained LLDAS by 6 months were significantly less likely to experience flares by 12 and 24 months after enrollment (Figure 2). Statistical differences in damage accrual were not observed, because of low overall rate of damage accrual in the inception cohort.
DISCUSSION
In this study, we sought to determine the differences in LLDAS attainment between newly diagnosed patients with SLE compared to those with established disease, as well as determine associations of LLDAS with disease outcomes in these 2 groups. Inception cohort patients had higher disease activity, received more treatment, and required higher doses of GCs compared to patients with established disease. These findings are consistent with the described natural history of clinical disease in patients with SLE.17
Despite significantly fewer inception cohort patients being in LLDAS at study entry, newly diagnosed patients were able to catch up to their established disease counterparts over time, with nearly three-quarters of both groups having attained LLDAS at least once by 12 months of study follow-up. This demonstrates that LLDAS is an achievable treatment target in recently diagnosed patients starting with a baseline of higher disease activity and GC use. Further, among patients with active disease at baseline, inception cohort patients achieved LLDAS faster than active noninception patients. This is similar to a previous analysis from the Hopkins lupus cohort that found disease duration < 1 year to be associated with earlier LLDAS attainment.22 These findings suggest that patients with de novo disease may respond better to first-line immunosuppression options compared to patients with long disease durations who have active disease, at least some of whom are likely to be refractory to treatment. A previous analysis of medication use in SLE showed that patients with higher disease activity tended to receive more combinations of medications and have lower persistence of treatment with an immunosuppressive drug, consistent with more difficult-to-control disease in this group.25
We were not able to observe significant protective associations of LLDAS attainment with reduced organ damage accrual in the inception cohort. Only one-fifth of these patients had 5 years of follow-up at the time of analysis, compared to half of the patients in the noninception group. Further, disease duration is a significant predictor of the risk of organ damage accrual, as is preexisting organ damage, both of which reduce the likelihood of organ damage events in the inception cohort.26 In the SLICC SLE cohort, which recruits patients within 15 months of diagnosis, thus representing a larger group of inception patients, authors were able to demonstrate an association of modified LLDAS was predictive of a lower probability of organ damage accrual, including in a sensitivity analysis of patients with at least 5 years and 10 years of follow-up.19 Piga et al and Sharma et al were also able to demonstrate protective associations of LLDAS with organ damage accrual in recently diagnosed patients with SLE11,21; of note, Asian ethnicity has also been reported to be associated with lower rates of organ damage accrual, potentially limiting the ability to demonstrate this association in the current study.26
Inception cohort patients who attained LLDAS by 6 months from study entry experienced significantly fewer flares at the 12- and 24-month mark compared to inception cohort patients, who took longer to attain LLDAS. A study of a monocentric cohort of newly diagnosed patients with SLE found significantly lower rates of organ damage accrual in patients achieving and maintaining LLDAS by 6 months after treatment initiation,20 a finding we could not reproduce. Despite the differences in findings, given that disease flares are known to be associated with eventual damage accrual,27 the results of both studies support the clinical benefit of adopting treat-to-target strategies from the time of diagnosis. A small study from Northern Europe has found a protective association of LLDAS against mortality, specifically in an inception cohort of patients with SLE,11 further highlighting the need to control disease activity early.
This study has some limitations. The median follow-up of slightly > 2 years and consequent low number of organ damage events in the inception cohort may have hindered the ability to detect significant protective associations of LLDAS against irreversible damage in this subgroup. In addition, although multicenter and multiethnic, the majority of patients were of Asian ethnicity. Further, standards of care in countries that participate in this cohort may differ from those with better established healthcare systems.
In conclusion, in this study we have demonstrated that recently diagnosed patients with SLE have more active disease, more flares, and take more medications—including higher doses of GCs—than patients with established disease. Despite this, recently diagnosed patients attain LLDAS faster than active patients with established disease, and have a significantly reduced number of disease flares once LLDAS is attained, especially if attained by 6 months. These findings should encourage clinicians to actively pursue LLDAS as a clinical target state for newly diagnosed patients with SLE.
ACKNOWLEDGMENT
We thank all patients participating in the APLC cohort, and all data collectors for their ongoing support for APLC research activities.
Footnotes
The Asia Pacific Lupus Collaboration has received unrestricted project grants from AstraZeneca, BMS, Eli Lilly, Janssen, Merck Serono, and UCB to support data collection contributing to this work. SCB has been supported in part by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (NRF-2021R1A6A1A03038899).
V. Golder and R. Kandane-Rathnayake contributed equally to this work.
SVN has received consulting fees from Biogen and Boehringer Ingelheim; lecture/speaker fees from Janssen, Novartis, Pfizer, and GSK; conference registration fees from Pfizer; and payment from Biogen for participation on advisory boards. ZL has received consulting fees from Pfizer, Roche, Janssen, Abbott, AbbVie, BMS, MSD, Celgene, Eli Lilly, GSK, Novartis, and UCB Pharma, and has royalties with Pfizer, Roche, Janssen, Abbott, AbbVie, BMS, MSD, Celgene, Eli Lilly, GSK, Novartis, and UCB Pharma. SS has received consulting fees from Pfizer, AstraZeneca, and ZP Therapeutics. YK has received payment/honoraria from GSK, AstraZeneca, Sanofi, Pfizer Japan, Janssen Pharmaceutical, Chugai Pharmaceutical, Asahi Kasei Pharma, Astellas Pharma, and Mitsubishi Tanabe Pharma. MH has received payment for postmarketing surveillance from GSK; a research grant from Novartis Pharma; and honoraria for lectures from GSK, AstraZeneca, and Astellas Pharma. TT has received grants from Astellas, Asahi Kasei, Chugai, and Mitsubishi Tanabe, and consulting fees from Astellas and Chugai. KPLN has received advisory board participation fees from AbbVie. MN has a received research grant from Janssen, and consulting fees from AstraZeneca, Boehringer Ingelheim, GSK, and Janssen. AH has received a research grant from AstraZeneca; consulting fees from EUSA Pharma; and speaker/honoraria from Limbic, Novartis, Moose Republic, AbbVie, and Eli Lilly. EFM has received consulting fees from AstraZeneca, Biogen, BMS, Eli Lilly, Genetech, Janssen, Novartis, Servier, and EMD Serono Research. The remaining authors declare no conflicts of interest relevant to this article.
- Accepted for publication February 23, 2024.
- Copyright © 2024 by the Journal of Rheumatology
REFERENCES
DATA AVAILABILITY
Access to APLC pooled data is subject to the specific guidelines outlined in the APLC Data Access Policy (available on request). The APLC welcomes requests for aggregate (summary) data or to perform analyses of new research questions, and such requests can be submitted to the APLC Steering Committee through the APLC Project Manager.