Anti-SARS-CoV-2 antibodies following vaccination are associated with lymphocyte count and serum immunoglobulins in SLE

OBJECTIVES

Patients with Systemic Lupus Erythematosus are known to have dysregulated immune responses and may have reduced response to vaccination against COVID-19 while being at risk of severe COVID-19 disease. The aim of this study was to identify whether vaccine responses were attenuated in SLE and to assess disease- and treatment-specific associations.

METHODS

Patients with SLE were matched by age, sex and ethnic background to healthcare worker healthy controls (HC). Anti-SARS-CoV-2 spike glycoprotein antibodies were measured at 4-8 weeks following the second COVID-19 vaccine dose (either BNT162b2 or ChAdOx1 nCoV-19) using a CE-marked combined ELISA detecting IgG, IgA and IgM (IgGAM). Antibody levels were considered as a continuous variable and in tertiles and compared between SLE patients and HC and associations with medication, disease activity and serological parameters were determined.

RESULTS

Antibody levels were lower in 43 SLE patients compared to 40 HC (p < 0.001). There was no association between antibody levels and medication, lupus disease activity, vaccine type or prior COVID infection. Higher serum IgA, but not IgG or IgM, was associated with being in a higher anti-SARS-CoV-2 antibody level tertile (OR [95% CI] 1.820 [1.050, 3.156] p = 0.033). Similarly, higher lymphocyte count was also associated with being in a higher tertile of anti-SARS-CoV-2 (OR 3.330 [1.505, 7.366] p = 0.003).

CONCLUSION

Patients with SLE have lower antibody levels following 2 doses of COVID-19 vaccines compared to HC. In SLE lower lymphocyte counts and serum IgA levels are associated with lower antibody levels post vaccination, potentially identifying a subgroup of patients who may therefore be at increased risk of infection.

The BILAG-2004 index is associated with development of new damage in SLE

OBJECTIVE

To determine whether BILAG-2004 index is associated with the development of damage in a cohort of SLE patients. Mortality and development of damage were examined.

METHODS

This was a multicentre longitudinal study. Patients were recruited within 12 months of achieving fourth ACR classification criterion for SLE. Data were collected on disease activity, damage, SLE-specific drug exposure, cardiovascular risk factors, antiphospholipid syndrome status and death at every visit. This study ran from 1 January 2005 to 31 December 2017. Descriptive statistics were used to analyse mortality and development of new damage. Poisson regression was used to examine potential explanatory variables for development of new damage.

RESULTS

A total of 273 SLE patients were recruited with total follow-up of 1767 patient-years (median 73.4 months). There were 6348 assessments with disease activity scores available for analysis. During follow-up, 13 deaths and 114 new damage items (in 83 patients) occurred. The incidence rate for development of damage was higher in the first 3 years before stabilizing at a lower rate. Overall rate for damage accrual was 61.1 per 1000 person-years (95% CI: 50.6, 73.8). Analysis showed that active disease scores according to BILAG-2004 index (systems scores of A or B, counts of systems with A and BILAG-2004 numerical score) were associated with development of new damage. Low disease activity (LDA) states [BILAG-2004 LDA and BILAG Systems Tally (BST) persistent LDA] were inversely associated with development of damage.

CONCLUSIONS

BILAG-2004 index is associated with new damage. BILAG-2004 LDA and BST persistent LDA can be considered as treatment targets.

P03 Not everything is lupus

Case report - Introduction

The association between malignancy and rheumatic diseases is well established.  Systemic lupus erythematosus (SLE) is no exception and is known to have associations with lymphoma and solid cancers of the lung, liver, and thyroid.  Non-Hodgkin’s lymphoma is the commonest malignancy with a standardised incidence ratio of 4.4—5.7.  However, they can present atypically, at any time, and can easily be confused with flares of SLE, so the rheumatologist must remain vigilant.  

We present a case of Sezary Syndrome masquerading as cutaneous lupus that presented 23 years after the initial diagnosis of SLE.

Case report - Case description

A 57-year-old black female had anti-nuclear antigen and anti-Ro antibody positive SLE.  In 2017, she attended with arthritis and a persistent severe malar and widespread lupus rash affecting her arms, chest, abdomen, and legs.  She was taking mycophenolate, hydroxychloroquine, and prednisolone.  She had previously experienced recurrent arthritis, oral ulcers, skin lesions, secondary Sjogren's Syndrome, episcleritis, leukopenia and Raynaud's phenomenon, and had found methotrexate and ciclosporin ineffective.  However, epratuzumab (anti-CD22 antibody), given during the EMBODY trial, had helped her symptoms. She was therefore given rituximab, which helped her arthritis but her cutaneous features got worse.

In March 2018 her dermatologist described a widespread erythema with adherent scales covering her back, abdomen and legs that had been preceded by vesicles and pustules on her chest, abdomen and thighs.  She was given emollients and topical corticosteroids.  A biopsy from her lower back showed abnormal and atypical CD4 positive T cells with cerebriform nuclei in the epidermis, follicular epithelium, and superficial dermis, with increased papillary dermal fibrosis. These features were felt to be in keeping with patch stage Mycosis fungoides.  She was referred to the skin lymphoma clinic where a scalp biopsy showed features of cutaneous lupus but a later thigh biopsy confirmed that she also had Mycosis fungoides. A staging CT confirmed axillary lymphadenopathy and blood tests showed Sezary cells, suggesting Sezary Syndrome.

As lupus prevented the use of UVB light therapy, she was given methotrexate but her skin rash progressed to Stage IV Mycosis Fungoides. She developed side effects with gemcitabine so moved to a cyclophosphamide, doxorubicin, vincristine and prednisolone chemotherapy regimen which helped her skin and SLE. However, her skin progressed again when this stopped. She had shown only a partial response to mogamulizumab (anti-CCR4 antibody) and so she is currently being considered for extracorporeal photopheresis as a last treatment option.

Case report - Discussion

A prolonged history of SLE with arthritis and mucocutaneous features made the diagnosis of Sezary Syndrome more complex in this case and may have delayed the diagnosis. However, the diagnosis was astutely reconsidered when rituximab failed to induce skin improvement. This prompted appropriate referrals to the dermatologist and thereafter for a skin biopsy. Coexisting cutaneous lupus on the scalp biopsy further complicated the diagnosis but strong clinical suspicion led to the confirmation of the diagnosis of Sezary Syndrome on further skin biopsy and blood tests.  

Mycosis fungoides is rare but it represents 60—70% of all cutaneous T-cell lymphomas. It typically presents in patients aged 50—60 with a chronic itch and fine scaling erythematous patches that can spread and thicken. Black patients can present at a younger age, they may have hypopigmented lesions, and the lesions may be more widespread and aggressive. Patients may have ‘B symptoms’ as occur in other lymphomas. The diagnosis is often delayed, requiring multiple biopsies.  Biopsies show cutaneous infiltration of malignant CD4 T-cells. The condition contrasts with cutaneous lupus as it typically affects sun-protected sites.

Sezary Syndrome can appear similar to Mycosis fungoides but typically presents with erythroderma and lymphadenopathy. Histology is also similar but can show acanthosis, dermal fibrosis, and cerebriform nuclear atypia as in this case.  Sezary cells have a cerebriform nucleus and can be detected in blood or on lymph node biopsy.  Unlike Mycosis fungoides which is an indolent condition with a 5-year survival of 87%, Sezary Syndrome is more aggressive with a 5-year survival less than 30%.  It is therefore important that cutaneous lymphoma be considered in any patient with autoimmune disease who has cutaneous features that behave atypically.

Case report - Key learning points

Patients with rheumatic diseases have an increased risk of developing malignancy. These may occur at any time and may be obscured by flares of the rheumatic disease.  As in this case, features may be similar to the rheumatic disease but rheumatologists must be vigilant and responsive to atypical patterns or atypical responses to treatment. Other specialties may also need support when investigating such patients as atypical features may be less apparent if they are unfamiliar with the rheumatic condition leading to symptoms being attributed to the rheumatic disease. As in this case, seek confirmation of the diagnosis, ideally with tissue histology, when the diagnosis remains uncertain.

Mycosis fungoides and Sezary Syndrome are rare but are thought to be associated with chronic antigen stimulation so may be more common in autoimmune disease, especially SLE. We highlight these conditions so that they may be considered by the rheumatologist in the differential diagnosis of any patient presenting with an itchy, scaling, erythematous rash.

OP0081 A CASE OF ATYPICAL MYCOBACTERIUM INFECTION COMPLICATING EXTRA-NODAL ROSAI-DORFMAN DISEASE IN A PATIENT WITH SYSTEMIC LUPUS ERYTHEMATOSUS (SLE)

Background:

We present the case of a 28 year old Black-British female with severe SLE requiring treatment with rituximab in 2012 due to persistent low-grade activity and severe episodes of pleuro-pericardial effusions, pancytopaenia, fever and weight loss

Her other background includes beta-thalassaemia trait and excision of calcific fibrotic tissue on bilateral anterolateral orbits in 2015.

In 2018 she reported an 18-month history of non-tender, non-fluctuant, slow growing left thigh mass with USS revealing a well demarcated subcutaneous complex cystic lesion of ~2x4x7cm. There was no preceding trauma or skin infection. Histology from a needle biopsy revealed diffuse histiocytosis with positive immunohistochemistry (ICH) for S100, CD68 and CD31, it was negative for CD1a, consistent with Extra-nodal Rosai-Dorfman disease (RDD).

She developed constitutional symptoms after reporting months of gradual weight loss with gradual ESR, CRP rise and leucocytosis. Her SLE symptoms were stable and given lack of SLE-specific symptoms; PET-CT was used to identify systemic RDD; the thigh mass showed strong FDG avidity along with a small focus of uptake in the small bowel, thought to be RDD related with no other areas of uptake.

She had ongoing ooze from the enlarging thigh lesion (5 x 26 x 15 cm), this was sent for MCS and AAFB; which isolated Mycobacterium avium. She was treated with rifampicin, ethambutol and clarithromycin resulting in improved thigh lesion, constitutional symptoms and inflammatory markers.

Objectives:

[1]To describe a rare associated complication of severe SLE and to educate and inform clinicians regarding possible masquerades of disease

[2]To education and inform about the approach to diagnosis of mycobacterium infection.

Methods:

Case report and literature review.

Results:

Mycobacterium infections rarely complicate RDD; to date, only one case report is published involving an HIV infected patient with RDD confirmed on LN biopsy presenting with splenomegaly and treated with oral corticosteroids (OCS) complicated by Mycobacterium avium complex and Salmonella enterica confirmed on bone marrow biopsy/culture, similar to our patient, he presented with constitutional symptoms and weight loss(2).

Mycobacterium can also mimic RDD, a case report has described a 74 year old with tender lymphadenopathy diagnosed with RDD on LN biopsy. She was treated with IV and OCS, but was unresponsive. A repeat LN biopsy and CT imaging revealed the presence of mycobacterium kansasii; her biopsy was positive for CD68/S100 throughout. Of note, she had high levels of anti-interferon autoantibodies and was diagnosed with adult-onset immunodeficiency syndrome(3).

Conclusion:

This case illustrates the need for a MDT approach for multi-system diseases such as SLE and RDD, and the need to consider atypical infections when blood tests are incongruent with clinical state.

References:

[1]Bruce-Brand C, Schneider JW, Schubert P. Rosai-Dorfman disease: an overview. J Clin Pathol. 2020 Nov;73(11):697-705. doi: 10.1136/jclinpath-2020-206733.

[2]Olmedo-Reneaum A, Molina-Jaimes A, Conde-Vazquez E, Montero-Vazquez S. Rosai-Dorfman disease and superinfection due to Salmonella enterica and Mycobacterium avium complex in a patient living with HIV. IDCases. 2020 Jan 14;19:e00698.

[3]King YA, Hu CH, Lee YJ, Lin CF, Liu D, Wang KH. Disseminated cutaneous Mycobacterium kansasii infection presenting with Rosai-Dorfman disease-like histological features in a patient carrying anti-interferon-γ autoantibodies. J Dermatol. 2017 Dec;44(12):1396-1400.

Image 1.Table 1.SLE Clinical HistoryDiagnosed 2006 (‘97 ACR Classification Criteria)Clinical -Polyarthritis -Glandular (lacrimal swelling) -Pericardial effusion/Pleural Effusion -MyositisSerological -Anti-nuclear antibody (ANA) -Anti-dsDNA -Anti-U1-RNP -Anti- SS-A/Ro -Lupus AnticoagulantPrevious SLE Treatment -Hydroxychloroquine (HCQ) -Methotrexate (MTX) -Azathioprine (AZA) -Rituximab (RTX)

Disclosure of Interests:

None declared

POS0106 BILAG-2004 LDA AND BST LDA ARE VALID TREAT TO TARGET IN SLE

Background:

Low disease activity state has been defined using SLEDAI and used as treatment target in SLE. However, there has not been any such definition using BILAG-2004 index (BILAG-2004).

Objectives:

This study was to determine if low disease activity state according to BILAG-2004 is valid for use as treatment target in SLE. We also assessed disease activity longitudinally using BILAG-2004 systems tally (BST). BST is an alternative way of representing BILAG-2004 scores that combines the flexibility and simplification of numerical scoring of BILAG-2004 with the clinical intuitiveness of BILAG-2004 structure.

Methods:

This was a prospective multi-centre longitudinal study in the UK of an inception cohort of SLE patients (recruited within 12 months of achieving 1997 ACR revised criteria for SLE). Data were collected on disease activity (BILAG-2004 and BILAG2004-Pregnancy Index during pregnancy), SLICC/ACR DI (SDI), cumulative drug exposure and death at every visit. This study ran from 1st January 2005 to 31st December 2017. Four low disease activity states (LDA) were defined using BILAG-2004: 1) BILAG-2004 LDA when all 9 systems had scores of C, D or E on assessment (no Grade A or B), 2) BST LDA when there was persistent score of C, D or E in all 9 systems between 2 consecutive visits (equivalent to 2 consecutive visits with BILAG-2004 LDA), 3) BILAG-2004 Remission when all 9 systems had scores of D or E on assessment and 4) Persistent Remission when there was persistent score of D or E in all 9 systems between 2 consecutive visits. Longitudinal analysis using Poisson regression with random effects model was used with development of new damage as the outcome of interest. Gender, cardiovascular risk factors, antiphospholipid syndrome status and most drugs (except hydroxychloroquine, glucocorticoids, mycophenolate and cyclophosphamide) were excluded from the model as they were not associated with development of damage in univariate analysis.

Results:

273 patients were recruited (91.2% female, 59.3% Caucasian, 17.2% African/Caribbean, 17.2% South Asian) with mean age at recruitment of 38.5 years (SD 14.8). 97.8% had no damage at recruitment (2.2% had SDI score of 1). Median follow-up was 73.4 months (range: 1.8, 153.8) with total follow-up of 1767 patient-years. There were 13 deaths and 114 new damage items occurred during follow-up. There were 6674 assessments with disease activity score: 319 assessments with Grade A activity in 95 patients (84.6% had only 1 system with grade A, range: 1 - 4) and 1704 assessments with Grade A or B activity in 239 patients (78.7% had only 1 system with Grade A or B, range: 1 - 5).

BILAG-2004 LDA was achieved in 74.5% of assessments (from 271 patients). BILAG-2004 Remission occurred in 28.2% of assessments (from 234 patients).

6401 observations with BST were available (1 observation derived from change in activity between 2 consecutive assessments) and 63.7% were in BST LDA. There was no observation with Persistent Remission between consecutive visits.

Table 1 summarises multivariate analysis which showed BILAG-2004 LDA to be inversely associated with damage. Similar results were obtained with BILAG-2004 Remission (RR 0.60 with 95% CI 0.38, 0.96) and BST LDA (RR 0.65 with 95% CI 0.43, 0.99). Cumulative drug exposure since recruitment for mycophenolate was protective against new damage (RR 0.99 with 95% CI 0.99, 0.99).

Table 1.VariableRelative Risk (95% CI) for New DamageEthnicityAfro-Caribbean1.22 (0.68, 2.18)South Asian1.81 (0.97, 3.38)Others2.22 (0.63, 7.85)Age at diagnosis1.06 (1.04, 1.08)Prior SDI score0.68 (0.43, 1.06)BILAG-2004 LDA0.60 (0.39, 0.94)Hydroxychloroquine since last visit (per g)0.99 (0.98, 0.99)Steroids since last visit (per 100mg)1.02 (1.01, 1.03)Cyclophosphamide since last visit (per g)1.67 (1.15, 2.41)

Conclusion:

BILAG-2004 LDA and BST LDA are valid treatment targets in SLE. BILAG-2004 Remission and Persistent Remission are uncommon, which make them unrealistic as a treatment target.

References:

[1]Yee C. S., et al. The BILAG-2004 systems tally – a novel way of representing the BILAG-2004 index scores longitudinally. Rheumatology (Oxford) 2012; 51[11]: 2099-2105.

Acknowledgements :

Versus Arthritis, Vifor Pharma

Disclosure of Interests:

Chee-Seng Yee Consultant of: Bristol Myers Squibb, ImmuPharma, Grant/research support from: Vifor Pharma, Caroline Gordon Speakers bureau: UCB, Consultant of: Center for Disease Control, Astra-Zeneca, MGP, Sanofi and UCB, Mohammed Akil: None declared, Peter Lanyon: None declared, Christopher John Edwards Consultant of: Glaxo Smith Kline, Roche, Grant/research support from: Glaxo Smith Kline, Roche, David Isenberg: None declared, Anisur Rahman: None declared, Lee-Suan Teh: None declared, Sofia Tosounidou: None declared, Robert Stevens: None declared, Ahtiveer Prabu: None declared, Bridget Griffiths: None declared, Neil McHugh: None declared, Ian N. Bruce: None declared, Yasmeen Ahmad: None declared, Munther Khamashta: None declared, Vernon Farewell: None declared

POS1452 DE NOVO LUPUS NEPHRITIS FOLLOWING THE INTRODUCTION OF BELIMUMAB

Background:

Belimumab inhibits the activity of the soluble cytokine BLyS (B lymphocyte stimulator) and is recommended for use in moderate refractory systemic lupus erythematosus (SLE). A recent randomised controlled trial reported that Belimumab improves the outcomes for patients with lupus nephritis when used with standard therapy.

Objectives:

We present two cases of lupus nephritis that developed in SLE patients without pre-existing renal disease shortly after commencing treatment with Belimumab.

Methods:

Both patients are Afro-Caribbean females with similar immunological profiles, including ANA, dsDNA, anti-Sm, anti- Ro and anti-RNP antibodies. The first was 59 years old with a long standing diagnosis of SLE since 1996 that had required previous Cyclophosphamide for neuropsychiatric lupus. She was most recently taking Hydroxychloroquine, Mycophenolate and Prednisolone having previously failed with Azathioprine twice. Belimumab was commenced in February 2020 due to worsening arthritis, mouth ulcers, pleuritis and systemic features associated with a significant rise in her dsDNA and low complement. Her SLEDAI score was 13. After six months of treatment she developed proteinuria for the first time. Her urine protein creatinine ratio (uPCR) was measured at 205mg/mmol and a subsequent renal biopsy revealed features of active Class IV and V lupus nephritis. Belimumab was changed to Rituximab. Mycophenolate was stopped due to persistent neutropenia and Tacrolimus was introduced instead. After 5 months treatment her uPCR is now 33mg/mmol.

The second patient was 37 years old with a recent diagnosis of SLE in 2018. She presented with inflammatory arthritis, oral and nasal ulcers, and cytopenia. She responded poorly to Azathioprine and was intolerant of Mycophenolate. Her most recent treatment was Hydroxychloroquine and Prednisolone. She was commenced on Belimumab in February 2020 due to active mucocutaneous and musculoskeletal features of lupus with a SLEDAI score of 12. The mucocutaneous features of lupus responded well but she developed proteinuria seven months later, and by November 2020 her uPCR was 149mg/mmol. Belimumab was switched to Rituximab and initially her uPCR continued to rise but it has now fallen to 43mg/mmol.

Results:

The occurrence of new lupus nephritis soon after the initiation of Belimumab monotherapy has been reported previously and our cases raise further concerns. A prospective observational study recently reported significantly increased rates of new lupus nephritis developing in patients receiving Belimumab in addition to standard care compared to those receiving standard treatment. An association between Belimumab and the development of de novo lupus nephritis has not yet been conclusively established but it would create a significant challenge in how Belimumab is used and consented for in SLE, especially if it becomes more widely used to treat lupus nephritis. The mechanism by which Belimumab may increase the risk of, or trigger, lupus nephritis is currently unclear but may result from increased activity in BLyS associated pathways following the blockade of the BLyS pathway.

Conclusion:

These two cases raise questions about the role of using Belimumab in patients at risk of developing lupus nephritis. We therefore recommend caution in its use and recommend active monitoring of renal parameters especially in patients with poor clinical and serological response to Belimumab.

References:

[1]Furie R, Rovin BH, Houssiau F, et al. Two-year, randomized, controlled trial of belimumab in lupus nephritis. N Engl J Med. 2020; 383(12):1117-28

[2]Sjöwall C, Cöster L. Belimumab may not prevent lupus nephritis in serologically active patients with ongoing non-renal disease activity. Scand J Rheumatol. 2014; 43(5):428-30

[3]Staveri C, Karokis D, Liossis SN. New onset of lupus nephritis in two patients with SLE shortly after initiation of treatment with belimumab. Semin Arthritis Rheum. 2017; 46(6):788-790

[4]Parodis I, Vital EM, Hassan SU, et al. De novo lupus nephritis during treatment with belimumab. Rheumatology. 2020; keaa796

Disclosure of Interests:

None declared

POS0111 DEVELOPMENT OF DAMAGE AND MORTALITY IN AN INCEPTION COHORT OF SLE PATIENTS

Background:

There had been very limited data on the development of damage and mortality in an inception cohort of SLE patients who were recruited very soon after diagnosis.

Objectives:

This study aimed to analyse the development of damage and death in an inception cohort of SLE patients recruited within 1 year of diagnosis with up to 13 years of follow-up.

Methods:

This was a prospective multi-centre longitudinal study in the UK of SLE patients recruited within 12 months of achieving 1997 ACR revised criteria for SLE. Data were collected on BILAG-2004, BILAG2004-Pregnancy Index (during pregnancy), SLICC/ACR DI (SDI), cumulative drug exposure and death at every visit. Information on cardiovascular risk factors and antiphospholipid syndrome status were also collected. This study ran from 1st January 2005 to 31st December 2017. Mortality and development of damage were analysed.

Results:

There were 273 patients recruited (91.2% female, 59.3% Caucasian, 17.2% African/Caribbean, 17.2% South Asian) with mean age at recruitment of 38.5 years (SD 14.8). 97.8% had no damage at recruitment (2.2% had SDI score of 1). Median follow-up was 73.4 months (range: 1.8, 153.8) with total follow-up of 1767 patient-years.

There were 13 deaths (4.8%): 76.9% female, 84.6% Caucasian, 15.4% South Asian, mean age 62.6 years (± SD 15.8) and mean disease duration 3 years (± SD 1.8). Causes of death were cancer in 5 (38.5%), infection in 3 (23.1%), ischaemic heart disease in 1 (7.7%) and unknown in 4 (30.8%).

114 new damage items in 83 patients occurred during follow-up. The distribution of damage was musculoskeletal (21, 18.4%), ophthalmic (18, 15.8%), neuropsychiatric (18, 15.8%), renal (14, 12.3%), malignancy (12, 10.5%), cutaneous (7, 6.1%), GIT (7, 6.1%), cardiac (6, 5.3%), pulmonary (4, 3.5%), diabetes mellitus (4, 3.5%) and vascular (3, 2.6%). The rate of development of damage appears to be higher in the first 3 years which subsequently stabilised (Table 1).

Table 1.

Incidence rate of development of damage over period of follow-up at 3 yearly intervals

Period of follow-up (year)Person-years at riskNumber of new items of damageIncidence rate, per 1000 person-years (95% CI)0 – 3753.46079.6 (61.8, 102.6)3 – 6534.03158.1 (40.8, 82.6)6 – 9321.21237.4 (21.2, 35.8)9 – 12152.5532.8 (13.6, 78.7)> 125.90-

Conclusion:

Mortality is uncommon during the first 12 years of follow-up for newly diagnosed SLE patients. However, development of damage appears to be higher in the first 3 years before stabilizing to a lower rate subsequently.

Acknowledgements:

Versus Arthritis, VIfor Pharma

Disclosure of Interests:

Chee-Seng Yee Consultant of: Bristol Myer Squibb, ImmuPharma, Grant/research support from: Vifor Pharma, Vernon Farewell: None declared, Mohammed Akil: None declared, Peter Lanyon: None declared, Christopher John Edwards Consultant of: Glaxo Smith Kline, Roche, Grant/research support from: Glaxo Smith Kline, Roche, David Isenberg: None declared, Anisur Rahman: None declared, Lee-Suan Teh: None declared, Sofia Tosounidou: None declared, Robert Stevens: None declared, Ahtiveer Prabu: None declared, Bridget Griffiths: None declared, Neil McHugh: None declared, Ian N. Bruce: None declared, Yasmeen Ahmad: None declared, Munther Khamashta: None declared, Caroline Gordon Speakers bureau: UCB, Consultant of: Center for Disease Control, Astra-Zeneca, MGP, Sanofi and UCB

POS0705 BILAG-2004 INDEX ACTIVE DISEASE PREDICTS DEVELOPMENT OF DAMAGE

Background:

BILAG-2004 Index (BILAG-2004) has undergone construct and criterion validity and is used to assess disease activity in SLE. However, its predictive validity has yet to be established.

Objectives:

This study was to determine if disease activity according to BILAG-2004 was predictive of development of damage in an inception cohort.

Methods:

This was a prospective multi-centre longitudinal study in the UK of an inception cohort of SLE patients (recruited within 12 months of achieving 1997 ACR revised criteria for SLE). Data were collected on disease activity (BILAG-2004 and BILAG2004-Pregnancy Index during pregnancy), SLICC/ACR DI (SDI), cumulative drug exposure and death at every visit. Information on cardiovascular risk factors (hypertension, diabetes mellitus, hypercholesterolaemia and smoking status) and antiphospholipid syndrome status were also collected. This study ran from 1st January 2005 to 31st December 2017. Longitudinal analysis using Poisson regression with random effects model was used to determine predictors of development of new damage. Death was not included in the analysis due to small numbers.

Results:

273 patients were recruited (91.2% female, 59.3% Caucasian, 17.2% African/Caribbean, 17.2% South Asian) with mean age at recruitment of 38.5 years (SD 14.8). 97.8% had no damage at recruitment (2.2% had SDI score of 1). Median follow-up was 73.4 months (range: 1.8, 153.8) with total follow-up of 1767 patient-years. Prevalence of risk factors during follow-up were: hypertension 23.1%, hypercholesterolaemia 35.5%, diabetes mellitus 5.5%, smoker or ex-smoker 44% and antiphospholipid syndrome 7%. There were 13 deaths and 114 new damage items occurred during follow-up.

There were 6674 assessments with disease activity score: 293 assessments with Grade A activity in 95 patients (92.4% had only 1 system with grade A, range: 1 - 4) and 1704 assessments with Grade A or B activity in 239 patients (78.7% had only 1 system with Grade A or B, range: 1 - 5).

Univariate analysis showed that gender, cardiovascular risk factors, antiphospholipid syndrome and most drug exposure (except hydroxychloroquine, glucocorticoids, mycophenolate and cyclophosphamide) were not associated with new damage (they were not included in the multivariate analysis).

Table 1 summarises multivariate analysis. Similar results were obtained when the disease activity variable was changed to Number of Systems with Grade A per assessment (RR 2.04 with 95% CI: 1.05, 3.94). Analysis using BILAG-2004 systems tally showed that persistent minimal disease was protective of development of damage (RR 0.74 with 95% CI: 0.57, 0.95). Cumulative drug exposure since recruitment for mycophenolate was protective against new damage (RR 0.99 with 95% CI 0.99, 0.99) but not cumulative drug exposure since last visit.

VariableRisk Ratio (95% CI) for New DamageEthnicity Afro-Caribbean1.21 (0.68, 2.17) South Asian1.81 (0.97, 3.36) Others2.37 (0.68, 8.20)Age at diagnosis1.06 (1.04, 1.08)Prior SDI score0.69 (0.44, 1.08)Constitutional A or Bunreliable estimate due to low numbersMucocutaneous A or B1.80 (1.04, 3.14)Neuropsychiatric A or B4.68 (1.68, 13.05)Musculoskeletal A or B0.76 (0.33, 1.73)Cardiorespiratory A or B0.35 (0.05, 2.59)GIT A or Bunreliable estimate due to low numbersOphthalmic A or Bunreliable estimate due to low numbersRenal A or B2.08 (0.99, 4.40)Haematological A or B4.37 (1.15, 16.65)Hydroxychloroquine since last visit (per g)0.99 (0.98, 0.99)Prednisolone since last visit (per 100mg)1.01 (1.00, 1.02)Cyclophosphamide since last visit (per g)1.42 (0.94, 2.14)

Conclusion:

Active disease (Grade A or B) according to BILAG-2004 index is predictive of development of new damage in SLE patients.

References:

[1]Yee C. S., et al. The BILAG-2004 systems tally – a novel way of representing the BILAG-2004 index scores longitudinally. Rheumatology (Oxford) 2012; 51[11]: 2099-2105.

Acknowledgements:

Versus Arthritis and Vifor Pharma

Disclosure of Interests:

Chee-Seng Yee Consultant of: Bristol Myers Squibb, ImmuPharma, Grant/research support from: Vifor Pharma, Vernon Farewell: None declared, Mohammed Akil: None declared, Peter Lanyon: None declared, Christopher John Edwards Consultant of: Glaxo Smith Kline, Roche, Grant/research support from: Glaxo Smith Kline, Roche, David Isenberg: None declared, Anisur Rahman: None declared, Lee-Suan Teh: None declared, Sofia Tosounidou: None declared, Robert Stevens: None declared, Ahtiveer Prabu: None declared, Bridget Griffiths: None declared, Neil McHugh: None declared, Ian N. Bruce: None declared, Yasmeen Ahmad: None declared, Munther Khamashta: None declared, Caroline Gordon Speakers bureau: UCB, Consultant of: Center for Disease Control, Astra-Zeneca, MGP, Sanofi and UCB

2. Two cases of anti-MDA5 positive dermatomyositis with rapidly progressive interstitial lung disease

Introduction

Melanoma differentiation-associated gene 5 (MDA5) is a myositis-associated autoantibody. It is increasingly being recognised that this antibody presents with typical skin lesions and the potential for a rapidly progressive interstitial lung disease but without muscle involvement. We present two cases of patients with MDA5 positive dermatomyositis who both developed rapidly progressive lung disease and despite intensive treatment passed away.

Case description

A previously well 48-year-old Caucasian man presented with a few months history of inflammatory arthritis, Raynaud’s and Gottron’s papules. He also described exertional breathlessness and was found to have fine inspiratory crackles bibasally. An HRCT scan showed cryptogenic organising pneumonia (COP). Inflammatory markers and creatine kinase were normal but an extended myositis panel showed positive anti-MDA5 antibodies consistent with a diagnosis of amyopathic dermatomyositis. He was started on cyclophosphamide as part of a research trial, given iloprost and sildenafil for worsening digital ischaemia and commenced on home oxygen. He was admitted with worsening shortness of breath after the second cycle of cyclophosphamide and found to have pneumocystis jirovecii (PCP) positive sputum. He developed pyrexia with a positive influenza A swab and increasing oxygen requirements requiring transfer to ITU. Despite further antibiotics, antivirals, steroids, IV immunoglobulin and rituximab infusion he deteriorated further and died 13 days later.

The second patient was a 45-year-old Asian man. He was initially seen by dermatology with alopecia and a scaly rash on his face, elbow and hands. He was then diagnosed with early inflammatory arthritis and commenced steroids and methotrexate. He developed skin ulceration and respiratory symptoms with a CT chest showing features of COP. He was ANA negative but anti-Ro and anti-Scl-70 positive. He was given antibiotics, methylprednisolone and switched to mycophenolate mofetil. Whilst abroad he was admitted to hospital, diagnosed with anti-MDA5 positive amyopathic dermatomyositis and given methylprednisolone and cyclophosphamide. Cyclophosphamide was continued on his return to the UK, with PCP prophylaxis but he also required home oxygen. He was admitted to hospital with increasing breathlessness and given further methylprednisolone and treatment dose co-trimoxazole. Two weeks later he deteriorated further and repeat CT scan showed a pneumomediastinum. He received antibiotics, antifungals and rituximab but died after three days on ITU.

Discussion

Although in both cases it was recognised the patients had some form of inflammatory condition the diagnosis of anti-MDA5 positive dermatomyositis took some time. The lack of muscle involvement is typical and means clinicians need to give more thought to the possible diagnosis particularly when patients present with skin lesions and look specifically for MDA5 antibodies. These cases also show how rapidly the lung disease can progress. Being aware that a patient is MDA5 positive gives important information to the clinical team regarding the potential prognosis and in these cases it has been questioned whether these concerns were entirely relayed to the patients and their families. High serum ferritin, ground-glass opacities in all six lung fields and worsening of pulmonary infiltrates during therapy have been suggested as further poor prognostic factors.

Both patients presented particular challenges in trying to decide whether their deterioration was due to infection in the context of immunosuppression, disease progression or both and consequently full infection screens were performed including bronchoscopies at various points. Given how unwell both patients were all available treatments were considered. Once it was recognised how rapidly the lung disease was progressing they both received cyclophosphamide and rituximab. IV immunoglobulin was requested for both patients but only agreed for the first patient as he had proven PCP pneumonia.

Key learning points

Anti-MDA5 positive dermatomyositis commonly presents with typical mucocutaneous lesions (such as cutaneous ulceration, alopecia and oral ulcers) which can differ from those seen in classical dermatomyositis. It is important to consider the possibility of anti-MDA5 positive dermatomyositis in a patient with skin abnormalities and a normal CK, and in such circumstances request an extended myositis screen ensuring MDA5 is included. Patients who are MDA5 positive and have lung involvement often have rapidly progressive interstitial lung disease. Prognosis is especially poor when patients are admitted to ITU and worse than patients with anti-synthetase syndrome. Spontaneous pneumomediastinum can be a feature when the outcome is almost always poor in a ventilated patient. Treatment options are limited, generally aggressive immunosuppression is recommended when there is lung involvement and induction therapy with cyclophosphamide or rituximab has been tried. There are emerging reports of JAK inhibitors being used in dermatomyositis and so this may be something to consider in this subset of difficult to treat patients.

Consideration should be given to vaccinating against influenza and pneumococcal infections as soon as possible, together with PCP prophylaxis and aggressive treatment of superadded infections. When appropriate patients and relatives should be made aware of the potentially poor prognosis. It is important to work closely with other specialities such as dermatology, respiratory and when necessary ITU. Due to the complexity and severity of this condition an MDT approach is recommended.

Conflict of interest

The authors declare no conflicts of interest.

9. Successful use of IL-6 pathway blockade to treat autoinflammation occurring in the context of a novel TNFRS1A gene mutation

Introduction

Autoinflammatory fever syndromes are rare and present significant diagnostic and treatment challenges. We present a case which illustrates some key concepts regarding the diagnosis and treatment of a patient with an autoinflammatory disorder, and also touches on the management of active autoinflammation in pregnancy.

Case description

A 29-year-old woman with a recently identified autoinflammatory disorder was referred to the local rheumatology service in September 2018. She reported having been symptomatic of fever, rashes and arthralgia since the age of 10, and TNF receptor 1 associated periodic syndrome (TRAPS) was suspected on the basis of an N71 deletion on axon 2 of the TNFRS1A gene when tested by a national reference centre 9 months previously.

At the time of presentation she was 25 weeks into her sixth pregnancy, the first with a new partner. She reported significant pregnancy morbidity having given birth to a daughter with multiple congenital abnormalities who unfortunately died two days after birth. She suffered 4 subsequent miscarriages at 5-7 weeks gestation with the same partner and underwent extensive genetic testing. At her initial review in the obstetric clinic, she was already receiving low-molecular-weight heparin with aspirin. Colchicine 1 mg tds, did not confer significant symptomatic benefit.

One month later, with careful counselling about pregnancy exposure to this biological treatment, IL-1 receptor antagonist (anakinra) therapy was instituted but discontinued after three weeks for generalised rash, as well as lack of efficacy manifesting in raised inflammatory markers. Infection was excluded during a subsequent hospital admission, and prednisolone treatment resulted in significant improvement in clinical course and acute phase response. The patient gave birth to a healthy infant at 37 weeks’ gestation.

In the postpartum period, a recurrence of symptoms was observed. IL-6 receptor antagonist (tocilizumab) treatment was commenced at 12 weeks postpartum but discontinued owing to reports of sore throat, cough, headache and fever on the day of the injection. The patient had also discontinued prednisolone on the day of the injection and tocilizumab was rechallenged with good symptomatic response, normalisation of inflammatory markers, and successful reduction in prednisolone dose.

Discussion

This case highlights some interesting points with relation to the treatment of autoinflammatory disorders refractory to IL-1 pathway blockade, and also, of the management of flares of autoinflammation during pregnancy.

Firstly, this lady failed to respond to IL-1 receptor antagonist anakinra which has been associated with efficacy in several reports for individuals with TRAPS, and has supplanted TNF blockade with agents such as etanercept for this condition. Only a handful of case reports describe successful IL-6 inhibitor administration for this condition, and this merits further study. This patient’s flare of autoinflammation was treated in the post-partum period, but the demographic characteristics of the autoinflammatory diseases are such that people of child-bearing age may be required to receive treatment in order to prevent pregnancy morbidity as a result of uncontrolled inflammation. IL-6 blockade has not to date been associated with adverse pregnancy outcomes, although this data requires extension and validation. 2 cases of renal agenesis have been reported in children born to mothers with anakinra, but the very small numbers of exposed parents warrants further examination of this observation and the careful study of ongoing pharmacovigilance data to explore this observation.

Secondly, success with IL-6 inhibition has been reported in a series of patients with familial Mediterranean fever, the most common autoinflammatory disorder, but this treatment does not feature in the most recent European guidelines for this condition. IL-6 inhibition is well-established in the treatment of adult-onset Still’s disease (AoSD) however, and this lady’s presentation shares important features with that condition. The existence of TNF-receptor 1 mutations has been reported in patients with AoSD, and it may be that this patient’s presentation is more akin to adult-onset Still’s disease and that her TNF receptor mutation is an incidental finding.

Key learning points

This report adds to the handful cases in which IL-6 blockade has been used successfully to treat the autoinflammatory manifestations of suspected TNF receptor 1 associated periodic syndrome (TRAPS) after the failure of IL-1 receptor blockade with anakinra.

Careful history and examination are required to differentiate between potentially overlapping symptoms of drug reaction, autoinflammation, and infection in patients with systemic autoinflammatory disorders (SAIDs). Once a history consistent with autoinflammatory flare was established, cautiously rechallenging with an IL-6 pathway inhibitor ensured this effective treatment was not discounted due to concerns over a drug reaction.

With a degree of commonality between the more common conditions in the SAID family, a detailed examination of the clinical phenotype is essential to the interpretation of genetic tests used for the diagnosis of TRAPS and related disorders.

Uncontrolled inflammatory disease is associated with pregnancy morbidity and very limited information is available about the short and long-term safety of treatments for autoinflammatory diseases in general. Current guidelines do not recommend the use of IL-6 inhibitor therapy in pregnancy.

Conflict of interest

The authors declare no conflicts of interest.

To what extent is NICE guidance on the management of rheumatoid arthritis in adults being implemented in clinical practice? A regional survey

Rheumatoid arthritis (RA) is a chronic disease associated with significant morbidity. The 2009 NICE guidance advises on the management of patients with RA. In this study, we undertook a survey to assess the implementation of the guidance into practice across the Midlands. In total, 19 rheumatology units participated, of which nine have designated early inflammatory arthritis clinics (EIAC). Data for 311 patients with RA attending clinics were collected during a two week period. The median time from symptom onset to first visit was four months. Of the patients, 95.6% were seen within 12 weeks of referral. Of those seen in EIAC, 75.9% had erosions documented on X-rays versus 49.4% of non-EIAC patients. In addition, 57.9% of patients were offered combination disease-modifying antirheumatic drugs (DMARD) therapy in EIAC, versus 30.4% in non-EIAC units. Monthly disease-activity scores were calculated more in patients attending EIAC than non-EIAC units (51.1% versus 25.4%). Based on our results, there is significant regional variation in implementation of the NICE guidance. In addition, patients with RA attending EIACs are more likely to receive a treat-to-target approach.