Predictors of flare in SLE

Patients with systemic lupus erythematosus (SLE) can experience periods of both low- and high-disease activity, this includes SLE flares.¹ Management of SLE flares can involve increasing  glucocorticoid or immunosuppressant dose; however, the risk of adverse events (AEs) is high. Identifying patients at particular risk of flares is essential to reduce damage accrual and side effects, ensuring effective preventative treatment.¹

Here, we discuss two studies assessing the risk and predictors of flares in patients with SLE.

Study design

Across both retrospective studies, 520 participants were enrolled (Kwon et al., n = 228; Cunha et al., n = 292).^1,2

The first study, by Kwon et al., included patients with SLE diagnosed between 2009 and 2019, with SLE confirmed according to the 2012 Systemic Lupus International Collaborating Clinic classification criteria.¹ Patients entered the observation period at the time of SLE diagnosis and remained under observation until last follow-up or the first flare, whichever came first.¹

The second study, a cohort study by Cunha et al., enrolled patients with SLE in a low disease activity state who fulfilled either the American College of Rheumatology 1997/2019, 2012 Systemic Lupus International Collaborating Clinic, or European League Against Rheumatism 2019 classification criteria.² Further inclusion criteria were at least two clinic visits less than 12 months apart within the designated study period from 2017 to 2020.²

Results

Patient characteristics at baseline, which were consistent across both studies, are shown in Table 1. The majority of patients were female and were undergoing treatment with hydroxychloroquine.^1,2

Table 1. Baseline patient characteristics*

SLE, systemic lupus erythematosus. *Adapted from Kwon, et al.1 and Cunha, et al.2
Characteristic, % (unless otherwise stated)	Study 1 (N = 228)	Study 2 (N = 292)
Mean age at diagnosis, years	38.3	33.3
Female	92.1	87.3
Clinical manifestation
Arthritis	—	48.5
Mucocutaneous	55.3	57.2
Hematological	64.5	79.8
Renal	18	36
Neurological	—	5.8
Musculoskeletal	45.6	—
Serositis	9.2	—
Neuropsychiatric	7.5	—
Vasculitis	2.6	—
Ongoing medication
Hydroxychloroquine	93.4	92.8
Immunosuppressants		34.6
Glucocorticoid	100.0	—
Cyclophosphamide	14.9	—
Mycophenolate                              mofetil	29.8	—
Azathioprine	32.5	—
Methotrexate	9.6	—
Cyclosporin	5.3	—
Tacrolimus	6.1	—
Oral prednisone	-	29.8

Occurrence of flares

In the first study, 78.5% of patients experienced a flare, with 40.4% experiencing a severe flare in a median observation period of 1.9 years.¹ In the second study, SLE flare was measured by Safety of Estrogen in Lupus Erythematosus National Assessment (SELENA) Flare Index (r-SFI), SLE Disease Activity Score (SLE-DAS), and SLE Disease Activity Index 2000 (SLEDAI-2K); 28.4%, 24.7%, and 13.4% of patients experiencing a flare according to these measurements, respectively.²

Risk and predictors of flare

Univariate and multivariate analysis of the study by Kwon et al. revealed that positivity for anti-dsDNA antibody and anti-Sm antibody was significantly associated with a higher risk of SLE flare.¹ Multivariate analysis also revealed that anti-dsDNA antibody positivity was significantly associated with the risk of severe flare. The risk of experiencing flare events was also higher in patients with double positivity (97.2%) compared with single positivity for anti-dsDNA or anti-Sm (79.5% and 85.3%, respectively) and double negativity (66.3%).¹ The incidence of flare based on antibody positivity is shown in Table 2.

Table 2. Occurrence of flare based on antibody positivity*

Ab, antibody; ds, double stranded; Sm, smith. *Adapted from Kwon, et al.1
Flare event, %	Anti-dsDNA Ab			Anti-Sm Ab
	+ (N = 114)	− (N = 114)	p value	+ (N = 70)	− (N = 158)	p value
Patients with any flare	85.1	71.9	0.016	91.4	72.8	0.002
Patients with severe flare	52.6	28.1	<0.001	52.9	34.8	0.010
patients with mild-to-moderate flare	32.5	43.9	0.076	38.6	38.0	0.932

In the study by Cunha et al., irrespective of which flare definition was used, multivariate analysis revealed that treatment with immunosuppressants, positivity for anti-U1RNP antibodies, and SLE-DAS score at baseline were all predictors of flare.² According to the r-SFI or SLEDAI definitions, treatment with glucocorticoids was also found to be predictive of flare. In addition, patients in remission with no disease activity who had not been treated with glucocorticoids had a lower risk of flare compared with patients in low disease activity state and not in remission.²

Conclusion 

These findings demonstrate that presence of anti-dsDNA, anti-Sm, and anti-U1RNP antibodies, as well as other treatment-related factors, can increase the risk of flare in patients with SLE.^1,2 Furthermore, patients with double positive anti-dsDNA and anti-Sm antibodies were at an increased risk for flare.¹

The authors noted several limitations of these studies. In the first study, possible variation to the antibody titers over time was not considered,¹ while predictors seen at baseline in the second study were not evaluated for changes over the study period.² Also, both studies were single center.^1,2

Prevention of flares should remain a priority of SLE treatment. Monitoring of patients for antibodies linked to higher flare risk and increased monitoring of patients receiving treatment with immunosuppressants could assist in the management and prevention of flares. Additional studies will help to determine other predictors of flare in SLE.