Abatacept ameliorates both glandular and extraglandular involvements in patients with Sjögren’s syndrome associated with rheumatoid arthritis: Findings from an open-label, multicentre, 1-year, prospective study: The ROSE (Rheumatoid Arthritis with Orencia Trial Toward Sjögren’s Syndrome Endocrinopathy) and ROSE II trials

ABSTRACT

Objective

To clarify the efficacy and safety of intravenous abatacept for glandular and extraglandular involvements in Sjögren’s syndrome (SS) associated with rheumatoid arthritis (RA).

Materials and methods

We performed an open-label, prospective, 1-year, observational multicenter study (ROSE and ROSE II trials). The primary endpoint was the remission rate as measured by SDAI at 52 weeks. The secondary endpoints included the changes in the Saxon’s test, Schirmer’s test, ESSDAI and ESSPRI. Adverse events and adherence rates were also analyzed.

Results

68 patients (36 in ROSE and 32 in ROSE II, all women) were enrolled. SDAI decreased significantly from 23.6 ± 13.2 at baseline to 9.9 ± 9.5 at 52 weeks. Patients with SDAI remission increased from 0 (0 weeks) to 19 patients (27.9%) at 52 weeks. Saliva volume increased significantly at 24 weeks. Tear volume increased significantly at 52 weeks. Both ESSDAI and ESSPRI were significantly decreased at 12 weeks, and these responses were maintained up to 52 weeks. The rate of adherence to abatacept over the 52-week period was 83.8%. Twenty-two adverse events occurred in 15 patients.

Conclusion

Abatacept ameliorated both glandular and extraglandular involvements, as well as the systemic disease activities and patient-reported outcomes based on composite measures, in SS associated with RA.

Sjögren’s syndrome, rheumatoid arthritis, abatacept

Introduction

Sjögren’s syndrome (SS) is an autoimmune disease characterized pathologically by lymphocytic infiltration into the exocrine glands (including the salivary and lacrimal glands) and clinically by dry mouth and dry eyes. SS is classified into primary SS, which is not associated with any other well-defined connective tissue disease (CTD), and secondary SS, which is associated with other well-defined CTD [1]. Importantly, 14.5% of patients with rheumatoid arthritis (RA) have been reported to have associated secondary SS [2].

With regard to the treatment for SS proposed by currently published European League Against Rheumatism (EULAR) recommendations [3] and clinical practice guideline for SS 2017 in Japan [4], topical therapy and pharmacologic stimulation with muscarinic agonists against glandular involvements, as well as corticosteroids and immunosuppressants for extraglandular involvements such as in the joints, skin, lungs, kidneys, and nervous system based on disease activities, have been recommended. Although some biologic agents, including rituximab and abatacept have been suggested to improve active and intractable extraglandular manifestations of SS, the effectiveness of these drugs against glandular involvements and of the currently established composite measures such as EULAR Sjögren’s Syndrome Disease Activity Index (ESSDAI) and EULAR Sjögren’s Syndrome Patient-Reported Index (ESSPRI) has not been confirmed [5]. For abatacept, three open-label prospective studies [6–8] revealed its effectiveness against glandular and extraglandular involvements of primary SS, as well as of the ESSDAI and ESSPRI. However, findings of two recently published randomized controlled trials (RCTs) [9, 10] showed that abatacept could not significantly improve the glandular manifestations, ESSDAI, and ESSPRI of primary SS as compared with the placebo, while some biologic markers, including immunoglobulin, rheumatoid factor (RF), and complement, were significantly changed by abatacept when compared with the placebo.

On the other hand, it was reported that RA associated with SS showed higher RA disease activity [2], more progressive joint destruction [11], and more resistance against TNF inhibitors [12, 13] in comparison with isolated RA without SS. Thus, RA associated with SS seems to have a higher risk in terms of difficult to treat RA (D2T RA) [14, 15], which has been the focus of a recent research agenda by rheumatologists.

We previously revealed that intravenous abatacept could ameliorate SS manifestations, including salivary and lacrimal secretions, in addition to RA manifestation among SS associated with RA in a previous open-label, prospective, 1-year, observational multicentre study named the ROSE (Rheumatoid Arthritis with Orencia Trial Toward Sjögren’s Syndrome Endocrinopathy) trial [16, 17]. A therapeutic study targeting SS associated with RA might be clinically important from the following two points of view: first, it could contribute to the establishment of a new treatment strategy for SS. Second, it could provide new insight into the management of D2T RA mentioned above.

We performed an open-label, prospective, 1-year, observational, multicentre study (ROSE II trial) to confirm the efficacy and safety of intravenous abatacept against both SS and RA involvements in patients with SS associated with RA, targeting many more cases and centres in Japan. In this report, we present the combined results of the ROSE and ROSE II trials.

Materials and methods

Patients

Patients aged 20 years or older with a diagnosis of RA according to the American College of Rheumatology (ACR) 1987 [18] or ACR/EULAR 2010 [19] criteria and of SS according to the 1999 Japanese Ministry of Health criteria for the diagnosis of SS [20] and the 2002 revised version of the European criteria proposed by the American-European Consensus Group [1] who presented with sicca symptoms were eligible for this study. The 1999 Japanese Ministry of Health criteria for the diagnosis of SS include the presence of two or more of the following four clinicopathologic findings: (1) lymphocytic infiltration of the salivary or lacrimal glands, (2) dysfunction of the salivary glands, (3) keratoconjunctivitis sicca (KCS), and (4) presence of anti-SS-A or SS-B antibodies [20]. The patients were followed up at the Departments of Rheumatology of 13 hospitals in Japan (University of Tsukuba Hospital, University of Occupational and Environmental Health Hospital, Nagasaki University Hospital, Hokkaido University Hospital, Kanazawa University Hospital, Keio University Hospital, Nihon University Hospital, Juntendo University Hospital, Saitama Medical University Hospital, Kyoto University Hospital, National Hospital Organization Osaka Minami Medical Center Hospital, Hyogo College of Medicine Hospital, and Kurashiki Medical Center Hospital). Approval for this study was obtained from the local ethics committee of each study site, and signed informed consent was obtained from each participant (approval number from the University of Tsukuba: H23-29, certification date of approval: 28 July 2011 for the ROSE trial; H26-69 and 4 August 2014 for the ROSE II trial). This study was registered at the University Hospital Medical Information Network (UMIN)-Clinical Trials Registry (CTR) (UMIN-ID: UMIN000005724 for the ROSE trial; UMIN000016273 for the ROSE II trial).

Patients with contraindications for abatacept (e.g. hypersensitivity, severe infection, and hepatitis B virus infection); aged older than 75 years or younger than 20 years; with leukopenia (leukocyte count ≤3000/mm²); with severe liver or kidney disease; with severe hematologic disorders; negative for both anti-SS-A and SS-B antibodies and positive for anticentromere antibody were excluded from this study. Patients who were pregnant, nursing, or wanted to become pregnant and those being treated with palliative therapies for dryness, including with cevimeline, anethole trithione, and pilocarpine within the last 4 weeks, were also excluded. Patients who were considered unsuitable for this study by their attending physician were also excluded.

Medications

The dosing regimen approved for the treatment of RA was used in this study. The weight-adapted dose of abatacept (500 mg for patients weighing <60 kg, 750 mg for those weighing ≥60 kg) was administered intravenously at Weeks 0, 2, and 4, every 4 weeks, over a period of 1 year. Other disease-modifying antirheumatic drugs, corticosteroids, and nonsteroidal antiinflammatory drugs were allowed to be used during the 1 year treatment period according to the clinical judgment of the attending physician.

Analysis of abatacept efficacy

The ROSE trial and ROSE II trial were each designed as an open-label, 1-year, prospective, observational study. For RA manifestations, the number of tender and swollen joints among 28 joints, the physicians’ global visual analogue scale (VAS), the patients’ global VAS, the Simplified Disease Activity Index (SDAI), serum C-reactive protein, erythrocyte sedimentation rate, RF, and anticyclic citrullinated peptide (anti-CCP) antibody were assessed at Weeks 0 (baseline), 4, 12, 24, and 52. For SS manifestations, patients’ VAS for dry mouth, dry eye, and parotid pain; physicians’ VAS for dry mouth, KCS, and general condition; saliva volume measured using the Saxon’s test; tear volume measured using the Schirmer’s test; anti-SS-A/SS-B antibody; and serum IgG level were examined at Weeks 0, 12, 24, and 52. Established composite measures for SS such as the ESSDAI [21–24] and the ESSPRI [22, 23, 25] were also assessed in the ROSE II trial.

The primary endpoint was the percentage of patients who achieved clinical remission as assessed using the SDAI at 52 weeks. The secondary endpoints included changes in the SDAI, Saxon’s test, Schirmer’s test, IgG, ESSDAI, and ESSPRI.

Analysis of safety of abatacept

Adverse events (AEs) during the 52-week study period were analysed at each visit. The types of AEs, onset, use of corticosteroids, treatment for AEs, hospital admission, cessation of abatacept use, association with abatacept, and outcome were recorded. We also recorded the rate of adherence to abatacept over the 52-week period and the causes of abatacept discontinuation.

Statistical analysis

Data were expressed as means ± SDs. Differences between measures taken at baseline and those taken after treatment with abatacept were examined for significance using the Wilcoxon signed-rank test. Differences between groups were examined using the Mann–Whitney U test for continuous variables. Categories of disease activity assessed using the SDAI and ESSDAI were examined by the Cochran–Armitage Test. Probability values less than .05 were considered to denote the presence of a significant difference. The rate of adherence to abatacept was analysed using the Kaplan–Meier method. Deficits of data were compensated for by the use of the last observation carried forward (LOCF) method.

Results

Clinicopathologic features at baseline

Sixty-eight patients (36 in ROSE and 32 in ROSE II, all women) were enrolled in this study. The baseline clinicopathologic features of the 68 patients are summarized in Table 1. The patients’ mean age was 56.1 ± 12.5 years, and their RA disease duration was 118.8 ± 136.8 months. The diseases in most of the patients were assessed as Stage I or II and Class 1 or 2 according to the Steinbrocker classification. The RA disease activity assessed using the SDAI was 23.6 ± 13.2 (remission: 0 cases, low: 10 cases, moderate: 33 cases, high: 25 cases). The positivities of RF and anti-CCP antibody were 85.3% and 79.4%, respectively.

Table 1.

Baseline clinicopathologic features of the 68 enrolled patients.

Age, years	56.1 ± 12.5
Sex, M/F	0/68
RA disease duration, mo	118.8 ± 136.8
RA stage, I/II/III/IV	19/28/7/14
RA functional class, 1/2/3/4	21/40/6/1
SDAI (remission/low/moderate/high)	23.6 ± 13.2 (0/10/33/25)
IgG, mg/dL (n = 67)	1727.7 ± 529.5
RF-positive (n = 68) RF-titer, IU/ml (n = 67)	58 (85.3%) 195.6 ± 316.8
Anti-CCP antibody-positive (n = 63)	50 (79.4%)
Anti-SS-A antibody-positive (n = 67)	58 (86.6%)
Anti-SS-B antibodies (n = 64)	15 (23.4%)
Organ involvement
Interstitial lung disease (n = 68)	7 (10.3%)
Others (n = 68)	13 (19.1%)
Saxon’s test, mg/2 min (n = 66)	2015.1 ± 1695.4
Schirmer’s test, mm/5 min (n = 52)	5.0 ± 6.0
Greenspan LSG grade, 0/1/2/3/4/unknown)	2/11/4/18/20/13
ESSDAI (n = 32) (low/moderate/high)a	9.5 ± 4.6 (4/21/7)
ESSPRI (n = 32)a	5.7 ± 2.0
Concomitant methotrexate (n = 68)	42 (61.8%), mean dose: 9.4 ± 3.5 mg/week
Concomitant corticosteroid (n = 68)	34 (50.0%), mean dose: 5.5 ± 2.7 mg/day of equivalent PSL
Previous biologics (n = 68)	17.6% (bio-switch 12, bio-naïve 56) (including overlap)
IFX	7
ETN	6
ADA	2
TCZ	3
GLM	2
CER	1

Left column: n represents the number of patients tested.

a

Assessed only in ROSE II trial-enrolled patients.

RA, rheumatoid arthritis; SDAI, Simplified Disease Activity Index; LSG, labial salivary gland; ESSDAI, EULAR Sjögren’s Syndrome Disease Activity Index; ESSPRI, EULAR Sjögren’s Syndrome Patient-Reported Index; PSL, prednisolone; IFX, infliximab; ETN, etanercept; ADA, adalimumab; TCZ, tocilizumab; GLM, golimumab; CER, certolizumab.

For SS manifestations, saliva volume assessed using the Saxon’s test was 2015.1 ± 1695.4 mg/2 min (n = 66); tear volume assessed using the Schirmer’s test was 5.0 ± 6.0 mm/5 min (n = 52); and Greenspan grading of the labial salivary gland (LSG) biopsy was Grade 0 in 2 patients, Grade 1 in 11 patients, Grade 2 in 4 patients, Grade 3 in 18 patients, Grade 4 in 20 patients, and unknown in 13 patients. The positivities of anti-SS-A and anti-SS-B antibodies were 86.6% and 23.4%, respectively. Fifteen anti-SS-B antibody-positive cases were all positive for anti-SS-A antibody. Thus, the present study did not include any patients with positive for isolated anti-SS-B antibody. In the ROSE II trial-enrolled patients, the ESSDAI was 9.5 ± 4.6 (n = 32), and the ESSPRI was 5.7 ± 2.0 (n = 32).

Twelve of the 68 patients (17.6%) were previously treated with biologics other than abatacept, whilst 56 patients were biologics-naïve. Forty-two of the patients (61.8%) were treated with concomitant methotrexate (mean dose: 9.4 ± 3.5 mg/week), and 34 of the patients (50.0%) were treated with concomitant corticosteroid (mean dose: 5.5 ± 2.7 mg/day of equivalent prednisolone) at baseline. None of the patients received secretagogues such as cevimeline and pilocarpine from baseline to the end of the study period.

Collectively, the enrolled patients had secondary SS with relatively preserved secretory functions and moderate systemic disease activity in addition to moderately active long-standing RA.

Effectiveness of abatacept against RA involvement

The disease activity of RA assessed using the SDAI significantly decreased after treatment with abatacept, from 23.6 ± 13.2 (0 weeks, baseline) to 9.9 ± 9.5 (52 weeks) (p < .05). Significant reduction in the SDAI relative to baseline (p < .05) was noted at 4 weeks and was maintained over the 52-week period (Figure 1(a)). Patients with clinical remission, as assessed using the SDAI, increased from 0 patients at 0 weeks to 19 patients (27.9%) at 52 weeks (primary endpoint); patients with low disease activity also increased, from 10 patients (14.7%) at 0 weeks to 46 patients (67.6%) at 52 weeks (Figure 1(b)). In contrast, the number of patients with moderate or high disease activity, as assessed using the SDAI, decreased from 58 patients (85.3%) at 0 weeks to 22 patients (32.4%) at 52 weeks (Figure 1(b)). The number of patients with remission or low disease activity according to the SDAI was significantly increased (p < .05), whereas those with moderate or high disease activity was significantly decreased (p < .05) after 4 weeks as compared with baseline (Figure 1(b)). These observations confirmed the effectiveness of abatacept against RA manifestations in patients with SS with RA, which could become D2T RA.

Figure 1.

Effects of abatacept on RA involvement. (a) Efficacy of abatacept treatment on the SDAI in 68 patients. Data deficits were compensated for by use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean SDAI at each time point. (b) Efficacy of abatacept treatment on disease activity as assessed using the SDAI in 68 patients over the 52-week period. Data deficits were compensated for by use of the LOCF method.

*p < .05 vs 0 weeks (baseline), Cochran–Armitage test. The numbers represent the cases in each category.

Effectiveness of abatacept against SS involvement

Saliva volume as assessed using the Saxon’s test significantly increased after treatment with abatacept, from 2015.1 ± 1695.4 mg/2 min at 0 weeks to 2219.1 ± 1758.5 mg/2 min at 12 weeks and to 2311.3 ± 1804.4 mg/2 min at 24 weeks (n = 66, p < .05) (Figure 2(a)). However, saliva volume slightly decreased to 2177.3 ± 1739.9 mg/2 min at 52 weeks, resulting in no significant difference between the measurements at baseline and those at 52 weeks (Figure 2(a)). Importantly, in patients whose Greenspan LSG biopsy grading was Grade 0, 1, or 2 (n = 16), saliva volume assessed using the Saxon’s test was significantly increased at 12 weeks, and the response was maintained up to 52 weeks as compared with baseline (p < .05) (Figure 2(b)). On the other hand, in patients whose Greenspan LSG biopsy grading was Grade 3 or 4 (n = 37), saliva volume was also significantly increased at 12 and 24 weeks but had returned at 52 weeks to the baseline level (p < .05) (Figure 2(b)).

Figure 2.

Effects of abatacept on SS involvement. (a) Effects of abatacept treatment on saliva volume assessed using the Saxon’s test in 66 patients. Data deficits were compensated for by use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean Saxon’s test value at each time point. (b) Comparison of effects of abatacept treatment on saliva volume assessed using the Saxon’s test between patients with Greenspan LSG biopsy grading of 0, 1, or 2 (n = 16) and those with grading of 3 or 4 (n = 37). Data deficits were compensated for by use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean Saxon’s test value at each time point. (c) Effects of abatacept treatment on tear volume as assessed using the Schirmer’s test in 52 patients. Data deficits were compensated for by use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean Schirmer’s test value at each time point. (d) Effects of abatacept treatment on serum IgG levels in 67 patients. Data deficits was compensated for by use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean serum IgG level at each time point. (e) Effects of abatacept treatment on serum RF levels in 67 patients. Data deficits was compensated for by the use of the LOCF method.

*p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean serum RF level at each time point.

Tear volume measured using the Schirmer’s test significantly increased after treatment with abatacept, from 5.0 ± 6.0 mm/5 min at 0 weeks to 5.9 ± 7.1 mm/5 min at 24 weeks and to 5.6 ± 6.3 mm/5 min at 52 weeks (n = 52, p < .05) (Figure 2(c)).

Treatment with abatacept resulted in significant reductions in serum IgG levels, from 1727.7 ± 529.5 mg/dL at 0 weeks to 1587.5 ± 457.2 mg/dL at 12 weeks (n = 67, p < .05), and this significant reduction in IgG was maintained over the 52-week period (Figure 2(d)). No patients showed hypogammaglobulinemia (<500 mg/dL) during the study period. Abatacept also significantly reduced serum RF levels from 195.6 ± 316.8 IU/ml at 0 weeks to 151.5 ± 232.6 IU/ml at 12 weeks and 168.0 ± 326.8 IU/ml at 24 weeks (n = 67, p < .05), while serum RF levels significantly re-increased to 221.6 ± 618.8 IU/ml at 52 weeks compared with baseline (n = 67, p < .05) (Figure 2(e)).

In the ROSE II trial-enrolled 32 patients, the ESSDAI was significantly improved at 12 weeks, and the improvement was maintained up to 52 weeks as compared with baseline (p < .05) (Figure 3(a)). The number of patients with high disease activity as assessed using the ESSDAI was significantly decreased at 12 weeks, whilst those with low disease activity was significantly increased, and the response was maintained over the 52-week period as compared with baseline (p < .05) (Figure 3(b)). For each ESSDAI domain, disease activities in the constitutional, lymphadenopathy, glandular, articular, and biologic domains were reduced at 52 weeks as compared with baseline (Figure 3(c)). Moreover, the ESSPRI was also significantly improved at 12 weeks, and the improvement was maintained up to 52 weeks as compared with baseline in these 32 cases (p < .05) (Figure 4(a)). Among the ESSPRI components, only the pain component was significantly improved at 12 weeks and was maintained up to 52 weeks as compared with baseline (p < .05), whereas the fatigue did not significantly change, and the dryness was significantly and transiently improved at 12 weeks only (p < .05) (Figure 4(b–d)).

Figure 3.

Effects of abatacept on ESSDAI in ROSE II trial-enrolled patients. (a) Effects of abatacept treatment on the ESSDAI in the 32 ROSE II trial-enrolled patients. Data deficits were compensated for by use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean ESSDAI value at each time point. (b) Effects of abatacept treatment on disease activity as assessed using the ESSDAI (low: <5; moderate: 5–13; high: ≥14) in the 32 ROSE II trial-enrolled patients over the 52 week period. Data deficits were compensated for by use of the LOCF method.

*p < .05 vs 0 weeks (baseline), Cochran–Armitage test. The numbers represent the cases in each category. (c) Effects of abatacept treatment on each ESSDAI domain in the 32 ROSE II trial-enrolled patients. Data deficits were compensated for by use of the LOCF method.

Figure 4.

Effects of abatacept on ESSRPI in ROSE II trial-enrolled patients. (a) Effects of abatacept treatment on the ESSPRI in the 32 ROSE II trial-enrolled patients. Data deficits were compensated for by the use of the LOCF method. *p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean ESSPRI value at each time point. (b–d) Effects of abatacept treatment on each ESSPRI component, including dryness (B), fatigue (C), and pain (D) in the 32 ROSE II trial-enrolled patients. Data deficits were compensated for by use of the LOCF method.

*p < .05 vs 0 weeks (baseline), Wilcoxon signed-rank test. The numbers represent the mean ESSPRI components at each time point.

These results suggest the effectiveness of abatacept for secretory dysfunction and immunologic abnormalities, as well as for systemic disease activities and patient-reported outcomes, in patients with SS with RA.

Adherence to and safety of abatacept

The rate of adherence to abatacept for 52 weeks was 83.8% (57/68) (Supplementary Figure S1). Eleven patients dropped out before the completion of the 52-week study. The reasons for the dropout were inadequate effects in five patients, AEs in three patients, and other reasons (hospital transfer, economic reasons, and no visiting) in three patients.

Twenty-two AEs were recorded in 15 of the 68 patients (22.1%) over the 52-week period. Of those 22 AEs, 9 events (40.9%) were infections (2 urinary tract infections, 1 infectious cornea ulcer, 1 bronchitis, 1 herpes zoster, 1 sinusitis, 1 upper respiratory tract infection, 1 pharyngitis, and 1 suspected cellulitis of the left lower leg). Abatacept was discontinued owing to AEs in 3 patients. Most AEs (86.4%, 19/22 events) fully recovered within the observation period (Table 2).

Table 2.

Adverse events in 68 cases over the 52-week period.

Study	Case	Adverse events	Onset, weeks	Treatment required	ABT administration	Outcome
ROSE	1	Urinary tract infection	7	+	Cessation	Recovery
	2	Urticaria	0	+	Continuation	Recovery
	3	Skin rash	0	+	Discontinuation	Recovery
	4	Infectious cornea ulcer	24	+	Continuation	Recovery
	5	Compression fracture of lumbar spine	12	+	Cessation	Recovery
	6	Bronchitis	4	+	Cessation	Unknowna
	7	Herpes zoster	4	+	Continuation	Recovery
	8	Vomit and diarrhea	8	+	Cessation	Recovery
	9	Sinusitis	36	+	Discontinuation	No recovery
	10	Upper respiratory tract infection	22	+	Continuation	Recovery
		Pharyngitis	32	+	Continuation	Recovery
		Stomatitis	48	+	Continuation	Recovery
ROSE II	1	Endometriotic ovarian cyst	14	+	Continuation	Recovery
	2	Liver dysfunction	8	+	Continuation	Recovery
		Non-alcoholic steatohepatitis	28	+	Continuation	No recovery
	3	Urinary tract infection	36	+	Continuation	Recovery
		Loss of consciousness	37	−	Discontinuation	Recovery
	4	Fracture of right humerus	23	+	Cessation	Recovery
		Suspected cellulitis of left lower leg	48	+	Cessation	Recovery
	5	Stomatitis	3	+	Continuation	Recovery
		Ulcer of auricle	49	+	Cessation	Recovery
		Fever	50	+	Cessation	Recovery

a

Unknown because of hospital transfer; ABT, abatacept; +, Yes; −, No.

Discussion

In this multicentre, observational, and prospective study, we confirmed the effectiveness and safety of abatacept for patients with SS associated with RA, showing that both RA- and SS-related manifestations, including established composite measures were significantly improved by 52 weeks’ treatment with intravenous abatacept. From these results, we have revealed the following four clinically important findings for the therapeutic strategy of SS associated with RA.

First, abatacept improved the RA disease activity of patients with SS associated with RA who had a higher risk for D2T RA [14, 15]. In comparison with isolated RA without SS, SS associated with RA could become D2T RA for several reasons, such as showing higher RA disease activity [2], more progressive joint destruction [11], and more resistance against TNF inhibitors [12, 13]. Regarding resistance against one of the TNF inhibitors, infliximab, among anti-SS-A antibody-positive RA, we previously proposed 3 possible mechanisms [12]: higher frequency of human antichimeric antibody (HACA), seroconversion of antinuclear antibody (ANA) positivity, and lower serum TGF-ß levels. Anti-SS-A antibody, which was detected in 33–74% of primary SS patients [26] but in 3–15% of RA patients [27], has two target antigens, Ro52kDa and Ro60kDa proteins [28]. Ro52kDa protein, also called tripartite motif-containing 21 (TRIM21), could function as an intracellular Fc receptor, resulting in the activation of innate immune signaling, as well as the limiting of immune signaling via autophagy of activated dimeric IFN regulatory factor-3 (IRF3) [29]. Interestingly, a recent study showed that anti-Ro52 (TRIM21) antibody-positive patients with systemic lupus erythematosus (SLE) had enhanced differentiation of plasmablasts when compared with negative SLE patients and that anti-Ro52 (TRIM21) antibody might contribute to the pathogenesis of SLE and SS and B-cell differentiation via suppression of TRIM21 function [30]. Collectively, these observations suggest that RA associated with SS, which frequently has anti-SS-A (Ro52, TRIM21) antibody, could have enhanced B cells and plasmablast differentiation, causing higher HACA and ANA production after infliximab treatment. Abatacept reportedly reduced serum levels of RF and anticitrullinated peptide antibody in RA patients, and these reductions correlated with the clinical responses [31]. Thus, abatacept seems to be an appropriate therapeutic strategy for RA manifestation in SS associated with RA because it could inhibit B-cell activation and antibody production via suppression of T cells, which might contribute to resistance against TNF inhibitors, especially against infliximab.

Second, abatacept increased salivary and lacrimal secretion in patients with SS associated with RA. In contrast to our results, the results from two recently published RCTs showed that abatacept could not significantly improve glandular manifestations [9, 10]. One of the reasons why abatacept could improve the secretory function of SS associated with RA in the present study might be relatively preserved secretory salivary and lacrimal functions. Importantly, saliva volume was increased and was maintained over the 52-week period after initiation of abatacept in patients with mildly infiltrated LSG, whereas saliva volume returned at 52 weeks to baseline levels in patients with advanced infiltrated LSG. Moreover, a previous study showed that abatacept increased saliva secretion in primary SS patients, when adjusting for disease duration, indicating that treatment early in the disease course is more effective than at a later stage [6]. Thus, early intervention for SS, which has relatively preserved secretory function and mild infiltration and survival of glandular tissues in LSG, might be necessary to obtain and maintain the adequate secretory improvement by abatacept. For lacrimal glands involvements of SS, the histopathology of lacrimal glands has not been adopted by current 2016 ACR/EULAR classification criteria for primary SS [32, 33]. In the present study, we did not examine the pathological features of lacrimal glands. Thus, we could not analyse the association between the effect of abatacept on lacrimal secretion and the disease stage of SS in lacrimal glands such as the early disease of SS where lacrimal inflammation was active without fibrosis or the late disease with advanced fibrosis. However, we could speculate that intervention by abatacept might be more effective on lacrimal function in the early disease of SS than in the late disease, as same as salivary function. Regarding SS associated with RA, RA was reported to precede SS in the majority of patients [2]. Therefore, we have a chance to diagnose early SS among patients with RA who are followed up and treated at rheumatology clinics. Unfortunately, we did not analyse whether SS preceded RA, or RA preceded SS in the enrolled cases. If RA patients develop secondary SS during RA treatment, abatacept might be an alternative therapy that could be beneficial for both RA and SS.

Third, in the present study, we clarified, for the first time, that abatacept could improve the composite disease activity of SS assessed using the ESSDAI and the patient-reported outcomes based on the ESSPRI among patients with SS associated with RA. To date, many biologics including abatacept [9, 10], rituximab [34–36], infliximab [37], etanercept [38], tocilizumab [39], anti-ICOS ligand monoclonal antibody [40], ianalumab [41], and baminercept (lymphotoxin β receptor fusion protein) [42] have failed to obtain improvement in the ESSDAI and ESSPRI in current RCTs. Iscalimab (anti-CD40 monoclonal antibody) is the sole biologic that has been reported to significantly improve the ESSDAI when compared with placebo in an RCT targeting primary SS [43]. Why is the effectiveness of biologics for SS manifestations limited? The following reasons have been proposed [44]. First, because of the chronicity and slowly progressive nature of the disease, the underlying pathogenic mechanisms are well established, and the disease is advanced when biologics are initiated. Second, in most clinical trials, the observation time was limited to either 24 or 48 weeks, a relatively short period to record a significant improvement in clinical manifestations in a chronic and slowly progressive disease. Third, the diversity of phenotypes and endotypes of the disease may have also interfered with the outcomes and primary endpoints of many clinical trials conducted to assess biologic treatments in SS. Lastly, we still do not understand all the pathophysiologic mechanisms involved in different phases of the disease. SS associated with RA might be a distinct phenotype for which abatacept could modify the pathogenic process, and biologics could be initiated earlier for these patients than for patients with isolated SS without RA, as described above. Importantly, the previous immunohistochemical studies comparing the composition of lymphocytic infiltrates in LSG between RA patients with sicca symptoms and primary SS patients reported that milder lesions, together with increased prevalence of dendritic cells and lower prevalence of CD4⁺cells in the RA-sicca subtype [45]. Therefore, we could suppose that abatacept might inhibit sialadenitis more successfully in patients with SS associated with RA than in primary SS, because of milder lesions in SS with RA. To confirm these speculations, RCTs of abatacept and other biologics targeted on not only primary SS but also secondary SS associated with RA would be needed.

Forth, abatacept was well tolerated among patients with SS associated with RA in the present study, showing that the rate of adherence to abatacept for 52 weeks was 83.8%. However, 22.1% of enrolled patients experienced AEs of which 40.9% were infections. Although many infectious events were common and not specific for patients with SS associated with RA, one case developed infectious cornea ulcer, which could be related with dry eye and KCS.

In conclusion, in patients with SS associated with RA, abatacept ameliorated both the glandular and the extraglandular involvements, as well as the systemic disease activities and patient-reported outcomes based on composite measures. To confirm the effectiveness of abatacept on SS and RA manifestations, RCTs especially targeting patients with SS associated with RA are needed.

Acknowledgements

We thank F. Miyamasu, Medical English Communications Center, University of Tsukuba, for critical reading of the manuscript. This work was supported by Health and Labour Sciences Research Grants for research on intractable diseases (Research Team for Autoimmune Diseases) from the Ministry of Health, Labour and Welfare of Japan.

Supplementary data

Supplementary data is available at Modern Rheumatology online.

Conflict of interest

None declared.

Funding

This study was supported in part by Bristol-Myers Squibb (BMS) and Ono Pharmaceutical Co., Ltd. H.T. has received speaking fees and/or honoraria from BMS. Y.T. has received speaking fees and/or honoraria from Gilead, Abbvie, Behringer-Ingelheim, Eli Lilly, Mitsubishi-Tanabe, Chugai, Amgen, YL Biologics, Eisai, Astellas, BMS, Astra-Zeneca, received research grants from Asahi-Kasei, Abbvie, Chugai, Mitsubishi-Tanabe, Eisai, Takeda, Corrona, Daiichi-Sankyo, Kowa, Behringer-Ingelheim, and consultant fee from Eli Lilly, Daiichi-Sankyo, Taisho, Ayumi, Sanofi, GSK, Abbvie. A.K. and T.S. have received speaking fees and/or honoraria from BMS and received research grants from BMS and Ono Pharmaceutical Co., Ltd. T.A. reports grants from Astellas, Takeda Pharmaceutical, Mitsubishi Tanabe, Chugai Pharmaceutical, Daiichi Sankyo, Otsuka Pharmaceutical, Pfizer, and Alexion and personal fees from Mitsubishi Tanabe, Chugai Pharmaceutical, Astellas, Takeda Pharmaceutical, Pfizer, AbbVie, Eisai, Daiichi Sankyo, BMS, UCB Japan, Eli Lilly Japan, AstraZeneca, MEDICAL & BIOLOGICAL LABORATORIES, Ono Pharmaceutical Co., Ltd., Novartis, and Nippon Boehringer Ingelheim, outside the submitted work. Y.K. has received speaker fee from BMS. T.T. has received speaker fee from BMS and received research grants from Ono Pharmaceutical Co., Ltd. M.T. has received speaking fees and/or honoraria from BMS. N.T. has received speaker fees and/or consulting fees from AbbVie, Astellas, BMS, Eisai, Eli Lilly, Janssen, Kyowa Kirin, Mitsubishi-Tanabe, and Novartis. I.M. has received speaking fees and/or honoraria from BMS and received research grants from BMS. T.S. has received speaking fees and/or honoraria from BMS and received research grants from BMS and Ono Pharmaceutical Co., Ltd.

Author contributions

All the authors contributed to the design of the study, collection of the data, and writing of the manuscript, and all agree to accept equal responsibility for the accuracy of this paper’s contents.