POS0269 RESULTS OF ONE YEAR OBSERVATIONAL EXTENSION OF THE BRIDGE-PMR STUDY, A RANDOMIZED DOUBLE-BLIND PLACEBO CONTROLLED TRIAL WITH RITUXIMAB IN POLYMYALGIA RHEUMATICA

Background

Glucocorticoids (GC) are the cornerstone of treatment in Polymyalgia rheumatica (PMR) [1]. However, they are associated with considerable toxicity and inefficacy in part of the patients. Rituximab (RTX) was effective for PMR in a 21-week randomized controlled trial (RCT), however results from longer follow-up is still absent [2].

Objectives

To assess, in a randomized double blinded fashion, the clinical and GC-sparing effects one year after RTX.

Methods

In the BRIDGE-PMR, an RCT of 38 recently diagnosed and 9 relapsing PMR (2012 EULAR/ACR classification criteria) patients recruited from the Sint Maartenskliniek, patients were randomly allocated in a 1:1 ratio and treated with 1x 1000mg RTX / placebo (PCB) iv, identical pre-medication and an accelerated GC tapering protocol. After the 21-week study, patients were assessed in a double blinded prospective extension study up to one year after infusion. The primary outcome at one year was between group difference in GC-free remission (PMR-activity score < 10). Analysis was performed with Fischer’s exact test and a two-tailed p-value < 0.05 was considered significant. Secondary outcomes were proportion of relapsing patients during the extension, proportion of patients with CRP > 5mg/l during the extension, cumulative GC dose, DMARD use, EQ-5D score, and adverse events (AE).

Results

The proportion of patients in GC-free remission after one year was significantly higher in the RTX group (48%, 11/23) compared to the PCB group (17%, 4/24), with an absolute difference of 31% (95%-CI 6-56), a relative risk of 2.9 (95%-CI 1.1-7.7), p=0.03. The secondary outcomes showed statistically significant differences in RTX versus PCB in median GC cumulative dose: 1595 versus 2302 mg (p = 0.04) and median PMR-AS: 6 versus 15 (p = 0.02) (Table 1 and Figure 1). No differences were seen in other secondary outcomes.

Table 1.

Primary and Secondary Outcomes for Rituximab Versus Placebo Treatment One Year After Infusion

Placebo [n=24]Rituximab [n=23]p-valueRemission, number (%)10 (42%)15 (65%)0.15GC-free remission, number (%)4 (17%)11 (48%)0.03Cumulative GC dose 0-52 weeks, in mg2302 (1595 - 2881)1595 (1275 – 2260)0.04Cumulative GC dose 21-52 weeks, in mg959 (91 – 1442)160 (0 - 902)0.10Relapse patient 21-52 weeks, number (%)*14 (58%)12 (52%)0.77PMR-AS**15.25 (7.75 - 22.5)6.3 (4.7 - 12.1)0.02CRP serum level, in mg/L3.5 (2 - 5)3 (1 - 4)0.29physicians’ VAS, 0-102 (0.2 - 3.7)1 (0 - 2)0.08Morning stiffness, in minutes25 (4 - 60)10 (0 - 30)0.06VAS pain, 0-103 (1.45 - 6.4)1.8 (0.7 - 5)0.16EQ5D-5L, score at week 52#0.71 (0.65 – 0.77)0.71 (0.63 – 0.77)0.87EQ5D-5L, change week 21-52#0 (-0.03 - 0.09)0.07 (-0.05 - 0.10)0.56Methotrexate use, number (%)4 (17%)2 (9%)0.67Adverse events, total, % of patients8, 26%6, 33%0.75

Notes. * Relapse was defined as therapy intensification, based on either a) an increase in oral prednisolone, b) adding intramuscular methylprednisolone, or c) starting or switching a DMARD due to treatment inefficiency ** Remission is based on the PMR-AS, calculated by CRP +VASp +VASph + (MST *0.1) + EUL, and a PMR-AS < 10 was considered remission or low disease-activity. # Number of patients for placebo versus RTX were n=23 versus n=23 respectively at week 52, and for comparison of change between week 21-52 total number of patients were n=21 versus n=23 respectively

Figure 1.

Cumulative GC Dose 0-52 Weeks, in mg

Conclusion

Efficacy of 1x1000 mg RTX in PMR was maintained up to 1 year follow-up, while also demonstrating a GC sparing effect. A larger trial, also assessing effect of on demand retreatment, is needed to confirm our results, and provide insight in which patients most likely benefit from RTX.

References

[1]Dejaco C, et al. Ann Rheum Dis 2015;74(10):1799-807. doi: 10.1136/annrheumdis-2015-207492

[2]Marsman DE, et al. The Lancet Rheumatology 2021;3(11):e758-e66. doi: 10.1016/S2665-9913(21)00245-9

Disclosure of Interests

None declared

POS0343 RITUXIMAB IS SUPERIOR TO PLACEBO IN POLYMYALGIA RHEUMATICA

Background:

Corticosteroids remain the cornerstone of polymyalgia rheumatica treatment, but their use has several disadvantages such as long treatment duration and glucocorticoid-related adverse events.1,2 Data on evidence based effective glucocorticoid-sparing agents are negative or absent.2

Objectives:

Because B-cells may be involved in the pathogenesis of polymyalgia rheumatica, we evaluated the efficacy of rituximab in polymyalgia rheumatica.

Methods:

In a 21-week double-blind placebo controlled exploratory study, 47 polymyalgia rheumatica patients (recently diagnosed n=38 / relapsing on prednisolone ≥7.5mg/day n=9) fulfilling the 2012 EULAR/ACR criteria, were randomized 1:1 to intravenous rituximab 1 x 1000 mg (n=23) or placebo (n=24), with a 17-week long glucocorticoid co-treatment. Primary outcome was glucocorticoid-free remission at week 21. Secondary outcomes were glucocorticoid ≤5mg/day and adverse events. Several post-hoc analyses were done for robustness of results.

Results:

Glucocorticoid-free remission was achieved in 48% (rituximab) versus 21% (placebo), one-sided 95%-CI 4% to 100%; p=0.049, and glucocorticoid ≤5mg/day in 100% versus 54% (one sided 95%-CI 20% to 100%; p=0.005). Post-hoc analysis showed efficacy mainly in recently diagnosed patients: glucocorticoid-free remission in 58% versus 21% (one-sided 95%-CI 10% to 100%; p=0.02); glucocorticoid ≤5mg/day in 100% versus 47% (one-sided 95%-CI 29 to 100%; p<0.001). No significant differences were observed regarding other outcomes (Table 1), except for less morning stiffness after rituximab.

Table 1.

Outcomes at week 21*

Rituximab (n=23)Placebo (n=24)Difference, one sided 95%-CIOne sided p-valueCumulative glucocorticoid dose in mg1356 (151)1406 (189)- ∞ to 340.16Median CRP in mg/l (median, IQR) ‡3 (1 to 5)- 1.5 (-9 to 2)2 (1 to 12) [21]-0.16Change from baseline (median, IQR)[22]- 5 (-13 to 0)[21]Mean ESR in mm/hour19 (11) [21]16 (13) [22]- ∞ to 170.79Change from baseline-7 (25) [21]-12 (19) [21]Relapsing patients during follow-up – no (%)†7 (30)8 (33)- 100 to 200.54Morning stiffness in minutes (median, IQR)Change from baseline (median, IQR)5 (0 to 30) -60 (-120 to -5)30 (9 to 90)-20 (-60 to 0)-0.02Rate ratio rituximab versus placebo (95% CI)Any adverse event §1361480.9 (0.8 to ∞)Serious adverse event(s) ††10Infections ‡ ‡17141.2 (0.7 to ∞)Infusion related complaints ¶1033.3 (1.2 to ∞)

*Values are means, SD, unless specified otherwise. Numbers of observations is indicated between brackets []. No correction for type I error.

† Judged by research physician.

‡Wilcoxon rank sum (Mann-Whitney) test

§ All adverse events that occurred during the study period were included in the analyses. Adverse events numbers are number of events, not number of patients with events. Safety outcomes were compared by chi-squared test (cumulative incidences). No correction for type I error was performed. All adverse events were graded according to Common Terminology Criteria for Adverse Events version 5.0, grade range 0–5;higher scores indicate worse events.

†† was A pulmonary embolism occurred in one patient

‡ ‡ Labelled by the research physician. No serious infections ≥ grade 3 occurred.

¶ Labelled by research physician. No serious infusion related complaints ≥ grade 3 occurred.

CRP denotes C-reactive protein, IQR interquartile range, ESR erythrocyte sedimentation rate, CI confidence interval

Conclusion:

Rituximab is superior to placebo in combination with 17-week glucocorticoid-treatment to achieve glucocorticoid free remission in polymyalgia rheumatica. The largest effect was seen in recently diagnosed polymyalgia rheumatica patients (funding: Sint Maartenskliniek; Dutch trial number NL7414).

References:

[1]González-Gay MA et al. Polymyalgia rheumatica. Lancet 2017;390(10103):1700–12.

[2]Dejaco C et al. 2015 recommendations for the management of polymyalgia rheumatica. Ann Rheum Dis. 2015 Oct 1;74(10):1799–807.

Figure 1.

Mean Polymyalgia Rheumatica Activity Score at each visit.

Disclosure of Interests:

Diane Marsman: None declared, Nathan den Broeder: None declared, Frank van den Hoogen: None declared, Alfons den Broeder Consultant of: Expert witness fee adalimumab biosimilar litigation for Boehringer/Fresenius, Amgen, Merck. Editorial work education for Abbvie,Novartis., Grant/research support from: Research Grants to employer SMK from Abbvie/Sanofi/Pfizer/Novartis or Lilly, Aatke van der Maas: None declared

AB0230 STATINS TO PREVENT RHEUMATOID ARTHRITIS: INCONCLUSIVE RESULTS OF THE STAPRA TRIAL

Background:

Persons at high risk for developing rheumatoid arthritis (RA) may benefit from a low-risk pharmacological intervention aimed at primary prevention. Statins are safe and widely-used drugs; previous studies demonstrated disease-modifying effects of statins in RA patients1as well as an association between statin use and a decreased risk of RA development2.

Objectives:

We designed a multi-center, randomized, double-blind, placebo-controlled trial to investigate if atorvastatin use for 3 years could prevent arthritis.

Methods:

Persons at high risk for RA, defined by the presence of arthralgia and anti-citrullinated protein antibody (ACPA) concentration >3xULN or both ACPA and rheumatoid factor (RF), were randomized to atorvastatin 40 mg daily or placebo for 3 years. Eligible participants were ≥18 years, had no indication for lipid-lowering therapy and had no clinical synovitis. The primary endpoint was development of clinical arthritis. Our goal was to include 220 patients, based on an anticipated 30% risk reduction by atorvastatin. Analysis was by intention-to-treat.

Results:

189 patients were screened, 175 were eligible, but only 67 persons were included of whom 62 were randomized (figure 1). The main reason for non-inclusion was unwillingness to use study medication (n=58, 54%). Inclusion was stopped after 38 months due to the low inclusion rate. Analyses were performed 1 year after inclusion stop. Mean follow up was 18 (0-36) months. Mean age was 48 years and 74% of participants were female. 14 persons (23%) developed clinical arthritis: 8/31 (26%) in the atorvastatin group and 6/31 (19%) in the placebo group (HR 0.8, 95% CI 0.3-2.2) after a median period of 7.5 (IQR 5.3-21.8) months (atorvastatin) and 4 (0-14.8) months (placebo). In the atorvastatin group, 17 persons completed the study according to protocol, 6 dropped out and 8 continued follow-up after prematurely stopping study medication. In the placebo group, 16 persons completed the study according to protocol, 11 dropped out and 4 continued follow-up after prematurely stopping study medication. Median duration of study medication use was 9 (6-26) months (atorvastatin group) and 8 (3-17) months (placebo group).

Conclusion:

The results of this trial are inconclusive due to severe difficulties with patient inclusion and low treatment adherence. The difficulty to enter and retain participants in this prevention trial is highly relevant given the current interest in treating RA in an ever earlier phase. At-risk individuals’ perceptions should be taken into account when designing preventive trials and will be important in optimizing acceptance and adherence to preventive treatment. Currently we are finalizing research into the motivation and barriers for participation in different primary prevention trials of RA and the willingness to initiate different types of preventive treatment in individuals in the at-risk phase of RA.

References:

[1]McCary et al. Lancet. 2004; 19;363(9426):2015-21

[2]Chodick G et al. PLoS Med. 2010;7(9):e1000336

Disclosure of Interests:

Laurette van Boheemen: None declared, S.A. Turk: None declared, M.H. van Beers - Tas: None declared, W.H. Bos Grant/research support from: abbvie, sanofi, roche, celgene, ucb, novartis, Speakers bureau: abbvie, Sanofi, eli lilly, Diane Marsman: None declared, E.N. Griep: None declared, M. Starmans: None declared, C.D. Popa: None declared, A.M. van Sijl: None declared, Maarten Boers: None declared, Michael Nurmohamed Grant/research support from: Not related to this research, Consultant of: Not related to this research, Speakers bureau: Not related to this research, Dirkjan van Schaardenburg: None declared

OP0072 IDENTIFYING POLYMYALGIA RHEUMATICA RELAPSE AND ITS ASSOCIATIONS IN A RETROSPECTIVE COHORT

Background:

Polymyalgia rheumatica (PMR) relapse occurs frequently when tapering glucocorticoids (GC)1. Some risk factors for varying definitions of relapse have been identified, although with conflicting results2. Identification of relapse and its associations can help identify patients in need of tighter follow up or additional medication.

Objectives:

To identify PMR GC-taper related relapse (proportion 1 and 2 years after starting treatment and per year of treatment) and candidate predictors, for a future prediction model.

Methods:

In a retrospective cohort of new PMR patients, visiting our hospital from April 2008 – January 2018, all visits > 30 days after starting GC treatment and with > 2.5mg oral prednisolone were used to identify substantial relapses. Relapse was defined in two ways: rheumatologist judgement (RJ) and treatment intensivation (TI). Agreement between RJ and TI at visits was assessed. TI relapse was used going forward for treatment based prediction. The proportion of relapsers after 1 and 2 years (cumulative incidence) and the amount of relapses per year of treatment (incidence rate (IR)), were assessed. Unadjusted associations with candidate predictors, present when starting GC treatment, were assessed using logistic and Poisson regression respectively.

Results:

Data from 417 patients was used (figure 1). Relapse occurred at 405 and 325 (of 2455) visits based on RJ and TI respectively. TI relapse (cumulative incidence) after 1 and 2 years was 134 (32%) and 184 (44%) and IR was 0.35 per patient year. Unadjusted significant associations for the cumulative incidence were CRP and ESR at baseline, and symptom duration before treatment (table 1), but only CRP and ESR were significantly associated with yearly IR.

Table 1:

Associations between predictors and relapse after 1 and 2 years of treatment and the amount of relapses per year of treatment.

Within first yearWithin first 2 yearsIncidence ratePredictorsORCIORCIIRRCIAge –years1.000.981.031.000.981.021.000.991.01Sex –ref. Male1.440.952.201.380.942.051.380.791.27Medical history –ref. No- inflammatory- malignancy- cardiovascular1.180.691.680.750.360.071.861.352.951.000.961.740.650.531.001.531.733.031.000.961.740.950.640.891.601.351.66Smoking –ref. no, n=336Stopped0.850.491.461.160.701.921.160.961.74Yes0.810.4121.570.980.531.820.980.671.40Symptom duration before baseline –per week, n=4110.980.960.9980.980.9640.9970.980.981.00Clinical disease severity –score 0-80.930.821.070.930.821.050.930.871.02(Suspected) presence of peripheral arthritis –ref. No0.730.401.330.670.381.160.670.551.16Presence of systemic symptoms –ref. No0.780.511.190.840.571.250.840.791.28CRP –per mg/L, n=3631.011.001.011.011.001.011.011.001.01ESR –per mm/h, n=3961.021.011.031.021.011.031.021.001.01Hb –per mmol/L, n=3161.030.751.390.990.741.320.990.861.25APR –ref. normal, n=3760.990.412.400.790.341.830.790.631.68

Ref., reference category; Hb, haemoglobin; APR, acute phase reactants; CI, 95% confidence interval

Conclusion:

PMR relapse while tapering GC occurs frequently, and some – although weak – associations were found. Longer symptom duration before treatment decreased chance of relapse, but did not increase the amount of relapses per year of treatment, potentially indicating a more self-limiting disease course. A uniform definition of relapse and identifying further predictors for a potential prediction model is needed to focus GC sparing agents for patients.

References:

[1]Mackie SL et al. Can the prognosis of polymyalgia rheumatica be predicted at disease onset? Results from a 5-year prospective study. Rheumatology (Oxford). 2010;49(4):716–22.

[2]Dejaco C et al. Current evidence for therapeutic interventions and prognostic factors in polymyalgia rheumatica: A systematic literature review informing the 2015 EULAR/ACR. ARD. 2015;74: 1808–17.

Disclosure of Interests:

None declared

AB0505 SEASONAL INFLUENCE IN PMR: NOT ONLY SUMMER, BUT WINTER IS COMING TOO

Background:

The cause for polymyalgia rheumatica (PMR) is currently unknown. Disease onset may be triggered by a combination of genetic predisposition and environmental factors such as infection.1In different regions of Denmark a simultaneous peak incidence of giant cell arthritis and PMR occurred together with epidemics of Mycoplasma pneumoniae, Chlamydophila pneumoniae and Parvovirus B19.1A seasonal epidemics pattern for PMR would be supporting evidence for an infectious cause.1However, the current evidence of seasonal effect on the occurrence and disease severity of PMR is limited and show conflicting results.2,3

Objectives:

To evaluate whether there is a seasonal effect on the risk of developing PMR in the Netherlands

Methods:

We retrospectively collected data on patient-, disease,- and treatment characteristics from newly diagnosed PMR patients (clinical diagnosis) who visited our outpatient clinic during April 2012 and September 2017. Exclusion criteria was other concomitant inflammatory rheumatic disease. Based on the onset of PMR (start symptoms, not time of diagnosis) patients were grouped per month. Descriptive statistics were used [mean (SD), median (p25-p75) or n (%) as appropriate]. The Chi square goodness of fit test was used to determine whether the incidence of onset of symptoms was different between months of the year.

Results:

In total 448 patients were included and 55 % were female and mean age was 66 years. Other baseline characteristics are described in table 1. The chi-square goodness of fit test to determine whether there was a peak incidence in months was p=0.06. As shown in figure 1 the incidence of onset PMR symptoms is higher in December-January, April through June with a peak in August. The April-June peaks coincides with incidence peaks of Mycoplasma pneumoniae infections and possibly Parvovirus B19 in spring and summer, the December-January peak coincides with Parvovirus B19 infections.4,5

Table 1.

Baseline characteristics (n=448)

CharacteristicFemale (%)247 (55)Age, years (SD)66 (8.6)PMR symptoms before diagnosis, weeks (IQR)10 (6-16)Neck pain (%)205 (46)Bilateral shoulder pain/stiffness (%)412 (91)Bilateral hip pain/stiffness (%)380 (85)Morning stiffness>45 min (%)233 (52)Peripheral arthritis (%)35 (8)Systemic symptoms* (%)199 (44)Elevated ESR mm/hour and / or CRP mg/l309 (87)ESR mm/hour(IQR)37 (26-51)CRP mg/l (IQR)30 (15-54)

*Fever, night sweats, weight loss, anorexia

** ESR n= 428; CRP n=396

Conclusion:

No definitive seasonal effect was found on risk of developing PMR, although a bimodal seasonal pattern compatible with the proposed respiratory infections is suggested.

References:

[1]González-Gay MA et al.Polymyalgia rheumatica. Lancet. 2017 Oct Oct 7.

[2]Narváez J et al.JRheumatol. 2000 Apr;27(4):953-7. Lack of association between infection and onset of polymyalgia rheumatica.

[3]F. Perfetto et al. seasonal pattern in the onset of polymyalgia rheumatic. Abstract EULAR 2005.

[4]Chen ZR et al. Epidemiology of community-acquired Mycoplasma Pneumoniae among hospitalized Chinese children.Hippokratia. 2013;17(1):20–26.

[5]Enders M et al. Current epidemiological aspects of human parvovirus B19 infectionEpidemiol Infect. 2006;135(4):563–569.

Disclosure of Interests:

None declared

Healthy ageing: the natural consequences of good nutrition-a conference report

Many countries are witnessing a marked increase in longevity and with this increased lifespan and the desire for healthy ageing, many, however, suffer from the opposite including mental and physical deterioration, lost productivity and quality of life, and increased medical costs. While adequate nutrition is fundamental for good health, it remains unclear what impact various dietary interventions may have on prolonging good quality of life. Studies which span age, geography and income all suggest that access to quality foods, host immunity and response to inflammation/infections, impaired senses (i.e., sight, taste, smell) or mobility are all factors which can limit intake or increase the body's need for specific micronutrients. New clinical studies of healthy ageing are needed and quantitative biomarkers are an essential component, particularly tools which can measure improvements in physiological integrity throughout life, thought to be a primary contributor to a long and productive life (a healthy "lifespan"). A framework for progress has recently been proposed in a WHO report which takes a broad, person-centered focus on healthy ageing, emphasizing the need to better understand an individual's intrinsic capacity, their functional abilities at various life stages, and the impact by mental, and physical health, and the environments they inhabit.