Bilateral orbital myositis in a patient with rheumatoid arthritis on treatment with tofacitinib: a report and review of literature

Orbital myositis is a rare inflammatory condition affecting the extraocular muscles of the eyes. It has also been linked to systemic autoimmune diseases. We present a case of orbital myositis in a 57-year-old male undergoing treatment for rheumatoid arthritis (RA) with tofacitinib, a Janus kinase inhibitor (JAK). Prompt administration of intravenous steroids led to rapid symptom improvement. To date, only six published cases have documented the association between RA and orbital myositis. This is the first description of orbital myositis occurring during treatment with the anti-inflammatory drug tofacitinib, an increasingly used disease-modifying anti-rheumatic drug (DMARD). We review the literature and emphasize the importance of ongoing vigilance regarding adverse events linked to tofacitinib.

POS0033 GENETIC INVESTIGATION OF TUMOUR NECROSIS FACTOR INHIBITOR IMMUNOGENICITY IN PATIENTS WITH RHEUMATOID ARTHRITIS

Background

Rheumatoid arthritis (RA) is a chronic inflammatory disease that primarily affects the synovial joints. Tumour Necrosis Factor inhibitor (TNFi) therapy has transformed the clinical management of RA. However, monoclonal antibody derived TNFi is associated with development of immunogenicity and subsequent loss of therapeutic effects. Previous studies have observed associations between certain HLA alleles and TNFi immunogenicity. For example HLA-DQA1 and HLA-DRB1 have been associated with immunogenicity in inflammatory bowel disease 1,2 and RA 3,4, respectively.

Objectives

The aims of this study were to identify associations between HLA alleles and immunogenicity to TNFi in an observational cohort of RA patients and to replicate findings from previous studies.

Methods

Anti-drug antibody titres were measured using radioimmunoassay in serum samples from RA patients participating in Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate (BRAGGSS). An anti-drug antibody titre of ≥12 AU/mL following six months on treatment was used to define positive immunogenicity. Genotype data were generated using Illumina HumanCoreExome Arrays. Standard quality control (QC) was applied prior to HLA imputation using SNP2HLA software before low minor allele frequency markers were removed. Logistic regression was used to study the association between HLA alleles and immunogenicity, whilst the omnibus test was applied to amino acid positions; sex and concurrent conventional synthetic DMARD use were included as a covariate in all the models.

Results

In total, 445 RA patients were analysed, 377 patients (70 immunogenicity events) were underdoing adalimumab therapy and 68 certolizumab (30 immunogenicity events) therapy. Following QC, 162 HLA alleles and 361 amino acids positions were available for analysis. The strongest HLA allele association was observed for HLA-DQA1*03 when all patients were analysed (OR = 0.61; 95% CI = 0.43 – 0.86; p-value = 5e-3). The amino acids positions 187 (p-value = 5e-3) and 26 (p-value = 5e-3) within the HLA-DQA1 gene were significantly associated with immunogenicity events. When both drugs were analysed separately, they produced similar effect size for HLA-DQA1*03 association; patients treated with adalimumab (OR = 0.59; 95% CI = 0.38 – 0.88; p-value = 1e-2) and certolizumab (OR = 0.52; 95% CI = 0.24 – 1.1; p-value = 1e-1). Another strong association was found in HLA-DRB1*04 (OR = 0.62; 95% CI = 0.44 – 0.88; p-value = =7e-3) and the amino acid position of 180 (p-value = 7e-3) and 33 (p-value = 7e-3) of HLA-DRB1 gene. Additionally, the similar protective effect between the two presented alleles suggested possibility of linkage disequilibrium, upon investigation the r2 between the 2 alleles is 0.69.

Conclusion

The current study increases the evidence for association between immunogenicity development with HLA-DQA1 and HLA-DRB1 alleles in patients receiving monoclonal antibody derived TNFi therapy. Further well powered studies are now required to determine the utility of HLA markers as a potential tool to aid the clinical management of RA.

References

[1]Sazonovs, A. et al. HLA-DQA1*05 Carriage Associated With Development of Anti-Drug Antibodies to Infliximab and Adalimumab in Patients With Crohn’s Disease. Gastroenterology158, 189–199 (2020).

[2]Billiet, T. et al. Immunogenicity to infliximab is associated with HLA-DRB1. Gut64, 1344–1345 (2015).

[3]Liu, M. et al. Identification of HLA-DRB1 association to adalimumab immunogenicity. PLoS One13, e0195325 (2018).

[4]Rigby, W. et al. HLA-DRB1 risk alleles for RA are associated with differential clinical responsiveness to abatacept and adalimumab: data from a head-to-head, randomized, single-blind study in autoantibody-positive early RA. Arthritis Res. Ther.23, 245 (2021).

Disclosure of Interests

Chuan Fu Yap: None declared, Nisha Nair: None declared, Kimme Hyrich Speakers bureau: Abbvie, Grant/research support from: Pfizer and BMS, Anthony G Wilson: None declared, John Isaacs Speakers bureau: Abbvie, Gilead, Roche, UCB, Grant/research support from: GSK, Janssen, Pfizer, Ann Morgan Speakers bureau: Roche, Chugai, Consultant of: GSK, Roche, Chugai, AstraZeneca, Regeneron, Sanofi, Vifor, Grant/research support from: Roche, Kiniksa Pharmaceuticals, Anne Barton Grant/research support from: I have received grant funding from Pfizer, Galapagos, Scipher Medicine and Bristol Myers Squibb., Darren Plant: None declared

AB0011 DNA METHYLATION AS A BIOMARKER OF TOCILIZUMAB RESPONSE IN RHEUMATOID ARTHRITIS PATIENTS

Background

Tocilizumab (TCZ) is a disease-modifying antirheumatic biologic drug, which targets the IL-6 signalling pathway and is effective in ameliorating disease activity in rheumatoid arthritis (RA). However, approximately 50% of patients do not respond adequately to TCZ and some patients report adverse events. Considering there is growing evidence that DNA methylation is implicated in RA susceptibility and response to some biologics (1, 2), we investigated DNA methylation as a candidate biomarker for response to TCZ in RA.

Objectives

To identify differential DNA methylation signatures in whole blood associated with TCZ response in patients with RA.

Methods

Epigenome-wide DNA methylation patterns were measured using the Infinium EPIC 850k BeadChip (Illumina) in whole blood-derived DNA samples from patients with RA. DNA was extracted from blood samples taken pre-treatment and following 3 months on therapy, and response was determined at 6 months using the Clinical Disease Activity Index (CDAI). Patients who had good response (n=10) or poor response (n=10) to TCZ by 6 months were selected. Samples from secondary poor responders (n=10) (patients who had an improvement of CDAI and were in remission at 3 months, followed by a worsening of CDAI at 6 months) were also analysed. Differentially methylated positions (DMPs) were identified using linear regression, adjusting for gender, age, cell composition, smoking status, and glucocorticoid use.

Results

In the pre-treatment samples, 20 DMPs were significantly associated with response status at 6 months (unadjusted p-value <10-6), whilst in the 3 month samples, 21 DMPs were associated with response. One DMP, cg03121467, was significantly less methylated in good responders compared to poor responders in the pre-treatment samples. This DMP is close to EPB41L4A and may play a role in β–catenin signalling. Interestingly, cg10136146 was significantly less methylated in secondary poor responders compared to both good and poor responders in the 3 month samples. This DMP maps close to CD81, which plays a role in mediating the development and activation of B and T lymphocytes.

Conclusion

These preliminary results provide evidence that DNA methylation patterns may predict response to TCZ. Further regional and pathway analyses is in progress and validation of these findings in other larger data sets is required.

References

[1]Liu,Y., Aryee,M.J., Padyukov,L., Fallin,M.D., Hesselberg,E., Runarsson,A., Reinius,L., Acevedo,N., Taub,M., Ronninger,M., et al. (2013) Epigenome-wide association data implicate DNA methylation as an intermediary of genetic risk in rheumatoid arthritis. Nat. Biotechnol., 31, 142–147.

[2]Plant,D., Webster,A., Nair,N., Oliver,J., Smith,S.L., Eyre,S., Hyrich,K.L., Wilson,A.G., Morgan,A.W., Isaacs,J.D., et al. (2016) Differential Methylation as a Biomarker of Response to Etanercept in Patients With Rheumatoid Arthritis. Arthritis Rheumatol. (Hoboken, N.J.), 68, 1353–60.

Disclosure of Interests

Nisha Nair: None declared, Darren Plant: None declared, John Isaacs Speakers bureau: Abbvie, Gilead, Roche, UCB, Grant/research support from: GSK, Janssen, Pfizer, Ann Morgan Speakers bureau: Roche/Chugai, Consultant of: GSK, Roche, Chugai, AstraZeneca, Regeneron, Sanofi, Vifor, Grant/research support from: Roche, Kiniksa Pharmaceuticals, Kimme Hyrich Consultant of: AbbVie, Grant/research support from: Pfizer, BMS, Anne Barton Grant/research support from: I have received grant funding from Pfizer, Galapagos, Scipher Medicine and Bristol Myers Squibb., Anthony G Wilson: None declared

AB0345 THERAPEUTIC DRUG LEVELS TO ACHIEVE GOOD EULAR RESPONSE IN PATIENTS WITH RHEUMATOID ARTHRITIS RECEIVING ADALIMUMAB: RESULTS FROM THE BIOLOGICS IN RHEUMATOID ARTHRITIS GENETICS AND GENOMICS STUDY SYNDICATE (BRAGGSS) COHORT

Background

Rheumatoid arthritis (RA) is a systemic inflammatory disease often treated with biologic disease-modifying anti-rheumatic drugs (bDMARDs) such as Adalimumab (ADL), a tumour-necrosis factor inhibitor (TNFi). However, it is known that about a third of patients do not respond to ADL treatment. Previous studies have reported associations between poor response, decreased serum drug levels (SDLs) and poor adherence, but a therapeutic SDL has not been defined nor applied in clinical practice.

Objectives

To assess median ADL SDLs in RA European Alliance of Associations for Rheumatology (EULAR) good vs non/moderate responders, and to determine cut-off SDLs associated with a “Good” response in fully adherent RA patients.

Methods

In a prospective observational cohort study, patients with RA were treated with ADL. At baseline, 3-, 6-, and 12-months patients had 4-component DAS28 scores, self-reported treatment adherence data and SDLs measured. Median drug levels and receiver-operator characteristics (ROC) curves were used to compare SDLs between responders and non-responders, and to establish cut-off SDLs in self-reported fully adherent patients. Serum drug levels were measured using a sandwich ELISA produced by Progenika Biopharma. Patients were considered fully adherent if they self-reported never having altered, forgotten or omitted any dose of their biologic drug at follow-up. Between group comparisons were assessed using Fisher’s exact test, with a threshold for significance set at p<0.05. Statistical analyses were performed in R Version 4.1.0 and RStudio Version 1.4.1106.

Results

A total of 283 RA patients taking ADL were included in the analysis. Baseline characteristics are shown in Table 1. Of these patients 93 (32.9%) self-reported being fully adherent to treatment at 3 months follow-up and had SDLs measured.

Table 1.

Baseline characteristics of patient cohort with RA taking ADL (n=283)

CharacteristicnMissing (%)Age at baseline, median years (IQR)58 (51, 64)0Disease duration, median years (IQR)7 (3, 16)0Female Sex, n (%)206 (73)0BMI, median (IQR)27.4 (23.7, 31.9)0Smoking Status132 (46)Current, n (%)57 (38)-Ex, n (%)32 (21)-Non, n (%)62 (41)-On concurrent DMARD(s)1 (0.4)No, n (%)34 (12)-Yes, n (%)248 (88)-Baseline DAS Score, median (IQR)5.61 (5.18, 6.14)On MTX at baseline38 (13)No, n (%)44 (18)Yes, n (%)201 (82)

In 93 fully adherent RA patients taking ADL at 3 months, good EULAR responders had significantly higher SDLs compared to non/moderate EULAR responders (p=0.0234). In 47/93 (50.5%) fully adherent good responders median SDL at 3 months was 10.94mg/L (IQR 7.75 to 12.0), whereas in 46/93 (49.5%) non/moderate responders, median SDL at 3 months was 9.014 (IQR 6.96 to 11.1).

ROC analysis (see Figure 1) reported a 3-month non-trough ADL SDL cut-off of 7.5mg/L in fully adherent RA patients which discriminated Good EULAR responders compared to non/moderate responders with an AUC of 0.63 (95% CI 0.52 – 0.75), 39.1% specificity, and 80.9% sensitivity.

Figure 1.

ROC curve analysis: EULAR non/moderate vs good responders with 3 month ADL SDLs.

Conclusion

In keeping with previous work, SDLs were higher in adherent compared with non-adherent patients, but this is the first study to demonstrate that SDLs are higher in fully adherent good EULAR responders compared with non/moderate responders. Based on our methods, cut-offs of 7.5mg/L for ADL may be useful targets in clinical practice to achieve good EULAR response.

References

[1]Jani M, Chinoy H, Warren RB, Griffiths CEM, Plant D, Fu B, et al. Clinical Utility of Random Anti–Tumor Necrosis Factor Drug–Level Testing and Measurement of Antidrug Antibodies on the Long-Term Treatment Response in Rheumatoid Arthritis. Arthritis & Rheumatology. 2015;67(8):2011-9.

[2]Pouw MF, Krieckaert CL, Nurmohamed MT, van der Kleij D, Aarden L, Rispens T, et al. Key findings towards optimising adalimumab treatment: the concentration-effect curve. Ann Rheum Dis. 2015;74(3):513-8.

Disclosure of Interests

None declared

OP0088 STRATIFIED MEDICINE: GENETIC PREDICTORS OF RADIOGRAPHIC OUTCOME IN RHEUMATOID ARTHRITIS

Background

Rheumatoid arthritis (RA) displays great heterogeneity between patients for susceptibility to developing erosions. Genetic variations within the HLA-DRB1 gene (the shared epitope (SE) and polymorphisms coding for Valine at position 11) have been consistently associated with both susceptibility and radiographic outcome in RA.(1) However, associations of non-HLA markers are much less conclusive. Most studies looking outside the HLA have been candidate gene studies and very few have been replicated in independent cohorts.

Objectives:

1. Identify all single nucleotide polymorphisms (SNPs) outside the HLA that have ever been associated with radiographic outcome in RA

2. To perform a replication study to determine which of these are associated with radiographic outcome in the Norfolk Arthritis Register (NOAR), the worldwide largest prospective cohort with genetic and radiographic outcome data.

Methods

A systematic literature search was conducted as shown in the Figure 1.

Figure 1.

Flow chart for systematic review.

The Norfolk Arthritis register (NOAR) is a large primary care-based inception cohort of patients diagnosed with inflammatory polyarthritis. Patients were recruited at baseline from 1989 and followed up prospectively for up to 20 years with serial X-rays. Genome-wide genotyping was performed on the Illumina Human/Infinium Core Exome array and imputed with Minimac4 to the Haplotype Reference Consortium panel. Quality control resulted in 7.5 million SNPs available in each patient. SNPs identified from the literature were extracted and tested for an association with the presence of erosions (as a longitudinal binary variable) using a generalized estimating equation (GEE) model in STATA/IC 14.0.in NOAR.

Results

A total of 2119 participants (2440 radiographs) were identified with both genetic and radiographic data available. 66.2% of these patients were female and 33.3% were anti-CCP positive. Median age of onset was 54.5 and 74.9% satisfied the American College of Rheumatology (ACR) 1987 criteria for rheumatoid arthritis.

A total of 113 different non-HLA SNPs associated with radiographic outcome in RA were identified from the literature. Of these, 102 were successfully identified within NOAR and 91 were deemed to be independent SNPs based on R2 of 0.6. 14 SNPs were found to be significantly associated with the presence of erosions within NOAR (Table 1).

Table 1.

SNPs found to be associated with radiographic severity within NOAR. *Significant results only; Dominant models used (Odds ratios displayed in relation to minor alleles)

GeneChromosomeSNP (single nucleotide polymorphismOdds ratio (95% CI)P valueIL2RB2rs7437771.23 (1.01, 1.05)0.0398IL154rs68211710.82 (0.67, 1.00)0.0451IL45rs22432501.36 (1.08, 1.70)0.0094FOX036rs122120670.75 (0.58, 0.97)0.0278OPG8rs20736180.79 (0.64, 0.98)0.0295TRAF19rs107601301.33 (1.06, 1.65)0.0118TRAF19rs108184881.32 (1.06, 1.64)0.0141TRAF19rs29001801.32 (1.07, 1.61)0.0079IL4r16rs18050101.25 (1.01, 1.56)0.0393IL4r16rs18050111.31 (1.03, 1.66)0.0260LGALS917rs37639591.28 (1.03, 1.59)0.0260SOST17rs47929091.34 (1.09, 1.65)0.0052LILRA319rs1032940.80 (0.65, 0.98)0.0334MMP920rs119083520.70 (0.57, 0.85)0.0005

Conclusion

113 non-HLA SNPs have been previously reported to be associated with radiographic outcome in RA. Of these, only ~15% also showed an association in NOAR, the largest cohort with genetic and radiographic outcome data worldwide. Interestingly, rs2243250, a SNP located on chromosome 5 (IL4), previously found to be associated in a small Egyptian cohort, has been replicated in NOAR.(2) Current work consists of assessing the added clinical value of a genetic risk score based on HLA and non-HLA markers in predicting radiographic outcome when combined with clinical/serological/demographic markers.

References

[1]S. Viatte et al., JAMA313, 1645-1656 (2015).

[2]Y. M. Hussein, A. S. El-Shal, N. A. Rezk, S. M. Abdel Galil, S. S. Alzahrani, Cytokine61, 849-855 (2013).

Disclosure of Interests

None declared.

POS0509 DEVELOPMENT OF A MULITNOMIAL PREDICTION MODEL OF TREATMENT RESPONSE TO ETANERCEPT IN A MULTI-CENTRE COHORT OF PATIENTS WITH ESTABLISHED RA

Background

Treatment response in rheumatoid arthritis (RA) is assessed through EULAR response groups of good, moderate, and poor response. Clinical prediction models from the literature typically frame this as a binary model, to differentiate poor from good and moderate responders. Here, we develop a multinomial model, to predict each group separately, after 3 months on the anti-TNF drug Etanercept (ETN).

Objectives

Develop and validate a multinomial prediction model of treatment response to ETN in RA, based on baseline clinical covariates.

Methods

We identified patients treated with ETN or biosimilars (N = 778) from the Biologics in RA Genetics and Genomics Study Syndicate (BRAGGSS). Response groups were derived from the CRP based 4C-DAS28 at baseline and 3 month follow up, yielding 310 good, 320 moderate, and 148 poor responders. A multinomial logistic regression model was fitted, using good responders as reference category. Multiple imputation by chained equations was used to impute missing data, and models were internally validated via bootstrapping. We report model accuracy, as well as calibration, and compare effect sizes across response groups. Table 1shows the baseline statistics, and odds ratios for the included covariates.

Table 1.

Baseline covariate statistics and odds ratios (in bold: significant at p < 0.05); HADS: Hospital Anxiety and Depression Scale

VariableMean (± SD)ORModerate [95% CI]pORPoor [95% CI]por % YesSwollen Joint8.84450.980.350.948e-3Count (SJC)(± 5.20)[0.95 1.02][0.89 0.98]Tender Joint14.68771.076e-61.050.01Count (TJC)(± 6.74)[1.04 1.10][1.01 1.08]General Health74.74291.000.60.981e-3Visual Analog Scale (GHVAS)(±17.79)[0.99 1.01][0.97 0.99]CRP19.07391.000.220.990.26(±25.07)[1.00 1.01][0.98 1.00]BMI30.30351.000.481.000.41(±23.28)[0.99 1.01][0.99 1.01]Age of47.33301.010.121.020.06onset(±13.86)[1.00 1.03][1.00 1.04]Disease9.94011.000.840.990.45duration(±10.35)[0.98 1.02][0.96 1.02]HAQ1.60851.480.022.951e-6(± 0.65)[1.06 2.08][1.91 4.54]HADS-Anxiety8.08681.040.191.060.12(± 4.54)[0.98 1.10][0.99 1.13]HADS-Depression7.38411.060.120.970.55(± 4.02)[0.99 1.13][0.89 1.06]Concurrent81.49%0.412e-40.520.03DMARD[0.26 0.66][0.28 0.94]Female78.66%1.390.121.110.71[0.92 2.10][0.65 1.87]Seropositive77.89%0.540.020.470.01[0.33 0.89][0.26 0.86]1st Biologic90.62%1.060.860.480.03[0.55 2.06][0.24 0.94]

Results

Adjusted for optimism, the multinomial model achieves an accuracy of 50.7% (IQR: 50 – 51.3%), with calibration slopes of 0.574 (IQR: 0.569 - 0.579) and 0.534 (IQR: 0.525 - 0.544) for moderate and poor response, respectively. Figure 1 shows a comparison of odds ratios (OR) for the different outcome groups. The Health Assessment Questionnaire (HAQ) score is the biggest driver of both moderate and poor response. Previous biologic treatment also predicts poor but not moderate response. Compared to the multinomial model, a binary model, that discriminates poor from moderate and good responders, underestimates the effect size of HAQ.

Figure 1.

Odds ratios of FIRSTBIO and HAQ for moderate and poor response. Size of crosses indicate 95% confidence intervals.

Conclusion

The model predicts EULAR response groups moderately well but is poorly calibrated, which can partly be explained by the generally higher sample size requirement of multinomial modelling. In the multinomial model, moderate and poor response is largely driven by the same covariates, which leads to blurred boundaries between good and poor responders, when response groups are merged to create a binary problem. Future research should consider the most appropriate model choice to describe data, including the use of multinomial instead of binomial models. More research and bigger sample sizes are required to improve on this multinomial model.

Disclosure of Interests

Michael Stadler: None declared, Stephanie Ling: None declared, Nisha Nair: None declared, John Isaacs Speakers bureau: Abbvie, Gilead, Roche, UCB, Grant/research support from: GSK, Janssen, Pfizer, Kimme Hyrich Speakers bureau: Abbvie, Grant/research support from: Pfizer and BMS, Ann Morgan Speakers bureau: Roche/ Chuga, Consultant of: GSK, Roche, Chugai, AstraZeneka, Regeneron, Sanofi, Vifor, Grant/research support from: Roche, Kiniksa Pharmaceuticals, Anthony G Wilson: None declared, Darren Plant: None declared, John Bowes: None declared, Anne Barton Grant/research support from: Pfizer, Galapagos, Scipher Medicine, and Bristol Myers Squibb.

AB0337 BASELINE C-REACTIVE PROTEIN PREDICTS ADHERENCE TO ADALIMUMAB THERAPY AT 3 MONTHS IN AN OBSERVATIONAL COHORT OF PATIENTS WITH RHEUMATOID ARTHRITIS

Background

Adherence to biologic treatment in rheumatoid arthritis (RA) is often self-reported and little is known about the predictors of adherence to biologic medications. Many studies have reported the predictors of adherence to be linked to psychological factors. A systematic review [1] identified several predictors of adherence to methotrexate in RA patients with the strongest predictors related to psychological factors including beliefs in medication necessity and absence of low mood. Mild disease activity was also found to be a significant predictor of adherence from this study. It is unknown whether similar factors will predict adherence in an established cohort of patients with RA starting biologic therapy.

Objectives

To investigate levels of self-reported adherence to adalimumab treatment and identify the contribution of demographic, physical and psychological factors to medication adherence in an RA cohort.

Methods

Patients with RA who were commencing on adalimumab were recruited through the Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate (BRAGGSS), a large UK multicentre prospective observational cohort study. Demographics, baseline clinical and psychological measures including illness and medication beliefs were collected. Self-reported adherence, defined as the patient has never stopped, altered, missed, forgot to take, or took a lower dose than prescribed of adalimumab, were recorded at 3 months. Potential baseline predictors of adherence to adalimumab therapy were determined using logistic regression analyses.

Results

202 patients were included; 76% female, median (IQR): age 59 (52-67) years, pre-treatment DAS28-CRP score 5.6 (5.1-6.1) and disease duration 5 (2-15) years. During the first 3 months following commencement of adalimumab, 176 (87%) patients reported full adherence. Univariable analyses found that high baseline C-reactive protein (CRP) [odds ratio (OR) 1.04 per mg/L, 95% CI 1.01, 1.09] was associated with adherence to adalimumab at 3 months. However, there were no associations identified from the psychological variables and this includes perceived necessity towards medication [OR 0.92, 95% CI 0.79, 1.05], hospital depression score [OR 0.94, 95% CI 0.84, 1.06] and hospital anxiety score [OR 0.97, 95% CI 0.88, 1.08].

Conclusion

These findings suggest that the psychological measures were less able to predict adherence to adalimumab therapy. The high percentage of adherence during the first three months of therapy may limit power to detect small effects in this cohort. Further research to investigate whether psychological variables correlate with drug levels as an alternative surrogate for adherence and to consider including other biological agents with a longer follow-up timeline are needed.

High baseline CRP levels were associated with adherence. This finding suggests active disease with higher levels of inflammation in RA may be a factor for adherence in patients who are commencing biologic therapy.

References

[1]Hope, H. F., Bluett, J., Barton, A., Hyrich, K. L., Cordingley, L., & Verstappen, S. M. M. (2016). Psychological factors predict adherence to methotrexate in rheumatoid arthritis; findings from a systematic review of rates, predictors and associations with patient-reported and clinical outcomes. RMD Open, 2(1), e000171. https://doi.org/10.1136/rmdopen-2015-000171

Disclosure of Interests

Adlan Wafi Ramli: None declared, Nisha Nair: None declared, Kimme Hyrich Consultant of: AbbVie, Grant/research support from: Pfizer, BMS, John Isaacs Speakers bureau: Abbvie, Gilead, Roche, UCB, Grant/research support from: GSK, Janssen, Pfizer, Ann Morgan Speakers bureau: Roche/Chugai, Consultant of: GSK, Roche, Chugai, AstraZeneca, Regeneron, Sanofi, Vifor, Grant/research support from: Roche, Kiniksa Pharmaceuticals, Darren Plant: None declared, Anthony G Wilson: None declared, Anne Barton Grant/research support from: I have received grant funding from Pfizer, Galapagos, Scipher Medicine and Bristol Myers Squibb.

In premenopausal women with idiopathic osteoporosis, lower bone formation rate is associated with higher body fat and higher IGF-1

We examined serum IGF-1 in premenopausal IOP, finding relationships that were opposite to those expected: higher IGF-1 was associated with lower bone formation and higher body fat, and lower BMD response to teriparatide. These paradoxical relationships between serum IGF-1, bone, and fat may contribute to the mechanism of idiopathic osteoporosis in premenopausal women.

INTRODUCTION

Premenopausal women with idiopathic osteoporosis (IOP) have marked deficits in bone microarchitecture but variable bone remodeling. We previously reported that those with low tissue-level bone formation rate (BFR) are less responsive to teriparatide and have higher serum IGF-1, a hormone anabolic for osteoblasts and other tissues. The IGF-1 data were unexpected because IGF-1 is low in other forms of low turnover osteoporosis-leading us to hypothesize that IGF-1 relationships are paradoxical in IOP. This study aimed to determine whether IOP women with low BFR have higher IGF-1 and paradoxical IGF-1 relationships in skeletal and non-skeletal tissues, and whether IGF-1 and the related measures predict teriparatide response.

METHODS

This research is an ancillary study to a 24 month clinical trial of teriparatide for IOP. Baseline assessments were related to trial outcomes: BMD, bone remodeling.

SUBJECTS

 Premenopausal women with IOP(n = 34); bone remodeling status was defined by baseline cancellous BFR/BS on bone biopsy.

MEASURES

 Serum IGF-1 parameters, compartmental adiposity (DXA, CT, MRI), serum hormones, and cardiovascular-risk-markers related to fat distribution.

RESULTS

As seen in other populations, lower BFR was associated with higher body fat and poorer teriparatide response. However, in contrast to observations in other populations, low BFR, higher body fat, and poorer teriparatide response were all related to higher IGF-1: IGF-1 Z-score was inversely related to BFR at all bone surfaces (r =  - 0.39 to - 0.46; p < 0.05), directly related to central fat (p = 0.05) and leptin (p = 0.03). IGF-1 inversely related to 24 month hip BMD %change (r =  - 0.46; p = 0.01).

CONCLUSIONS

Paradoxical IGF-1 relationships suggest that abnormal or atypical regulation of bone and fat may contribute to osteoporosis mechanisms in premenopausal IOP.

The Association of Heart/Vascular Aging with Mild Cognitive Impairment in a Rural Multiethnic Cohort: The Project FRONTIER Study

BACKGROUND

Cardiovascular disease (CVD) and Alzheimer's disease and related dementias (ADRD) disproportionately affect rural communities. Identifying strategies to effectively communicate CVD risk to prevent these conditions remains a high priority.

OBJECTIVE

We assessed the relation between predicted heart/vascular age (PHA), an easily communicated metric of CVD risk, and mild cognitive impairment (MCI), an early manifestation of ADRD.

DESIGN, SETTING, PARTICIPANTS

Data were from 967 rural West Texas residents aged ≥40 years without CVD at baseline (2009-2012) enrolled in Project FRONTIER, an ongoing, multi-ethnic cohort study on cognitive aging.

MEASUREMENTS

MCI was diagnosed using the standardized consensus review criteria. PHA was calculated using the Framingham CVD risk equation. High excess PHA (HEPHA) was defined as the difference between PHA and chronological age >5 years. Logistic regression models were used to calculate odds ratios (OR) and 95% confidence intervals (CI).

RESULTS

At baseline, the mean age of participants (70% women and 64% Hispanics) was 55 years. Almost 13% had MCI and 65% had HEPHA. After adjusting for socio-demographic and health factors, HEPHA was positively associated with MCI (OR=2.98; 95%CI: 1.72-5.15). Among participants without MCI at baseline who returned for follow-up exam after three years (n=238), a three-year negative change in PHA was seemingly associated with reduced odds for MCI (OR=0.98; 95%CI: 0.96-1.01).

CONCLUSIONS

In this study, PHA was positively associated with MCI, with improvement in CVD risk profile seemingly related to reduced odds for MCI. PHA may provide a low-cost means of communicating CVD risk in rural settings to prevent both CVD and ADRD.

Using Disorder to Overcome Disorder: A Mechanism for Frequency and Phase Synchronization of Diode Laser Arrays

Noise and disorder are known, in certain circumstances and for certain systems, to improve the level of coherence over that of the noise-free system. Examples include cases in which disorder enhances response to periodic signals, and those where it suppresses chaotic behavior. We report a new type of disorder-enhancing mechanism, observed in a model that describes the dynamics of external cavity-coupled semiconductor laser arrays, where disorder of one type mitigates (and overcomes) the desynchronization effects due to a different disorder source. Here, we demonstrate stabilization of dynamical states due to frequency locking and subsequently frequency locking-induced phase locking. We have reduced the equations to a potential model that illustrates the mechanism behind the misalignment-induced frequency and phase synchronization.

P–460 Impact of various cancers on semen parameters in a tertiary onco-fertility unit in India

Study question

This study evaluated differences in semen parameters in male cancer patients in our ethnic population who banked their sperms prior to cancer treatment

Summary answer

We found significant differences in semen concentration, motility and morphology between different types of cancers, especially testicular cancers

What is known already

Impaired spermatogenesis and abnormal semen parameters in cancer patients has been noted, however certain cancer types are more damaging than others. In testicular cancer, spermatogenesis impairment is more quantitative than qualitative with sperm morphology being the most affected parameter. Among non testicular cancers, lymphoma cases usually show the most significantly impaired semen parameters

Study design, size, duration

We conducted a retrospective study analyzing semen parameters in 49 cancer patients between October 2014 to January 2020 who presented to the onco-fertility unit, Max Multispeciality Hospitals, New Delhi.

Furthermore, we did our analysis based on total of 101 samples and were broadly divided into testicular (37 samples) and non testicular cancers (64 samples). Patients who had previously received any form of cancer treatment including chemotherapy or radiotherapy were not included in this study

Participants/materials, setting, methods

Testicular Cancer(TC) group was further subcategorized into Seminoma and Non Seminoma groups whereas Non Testicular Cancer (NTC) group was subcategorized into Lymphoma and Non Lymphoma groups. Semen was collected by masturbation and analysis was performed in keeping with the WHO criteria. Statistical analyses was performed using SPSS software. p values &lt;0.05 were considered to indicate statistical significance.

Main results and the role of chance

In Testicular cancer (TC), 92% samples (34/37) had abnormal semen parameters whereas only 24.4% samples (22/64) were abnormal in Non Testicular cancer (NTC). Additionally, there were significant differences in sperm concentration, motility and morphology between TC and NTC groups.

Individually,

TC: Oligozoospermia was seen in 73% (27/37) with subdivision between Seminoma and Non Seminoma groups being 81.3% (13/16) and 61.9% (13/21).

Asthenozoospermia was seen in 86.5% (32/37) samples with subdivision between Seminoma and Non Seminoma groups being 87.5% (14/16) and 81% (17/21).

Teratozoospermia was seen in 59.5% (22/37) samples with subdivision between Seminoma and Non Seminoma groups being 75% (12/16) and 42.86% (9/21).

Combined OATS observed in 59.5% (22/37) samples with subdivision between Seminoma and Non Seminoma groups being 75% (12/16) and 42.86% (9/21)

NTC: Oligozoospermia was seen in 18.8% (12/64) samples with subdivision between Lymphoma and Non Lymphoma groups being 26.92% (7/26) and 26.32% (10/38).

Asthenozoospermia was seen in 32.8% (21/64) samples with subdivision between Lymphoma and Non Lymphoma groups being 34.62% (9/26) and 34.21% (13/38).

Teratozoospermia was seen in 17.2% (11/64) samples with subdivision between Lymphoma and Non Lymphoma groups being 26.9% (7/26) and 23.68% (9/38).

Combined OATS observed in 17.2% (11/64) samples with subdivision between Lymphoma and Non Lymphoma groups being 26.9% (7/26) and 23.68% (9/38).

Limitations, reasons for caution

Study was conducted in a single institution with lesser overall number of patients. Duration, staging and grading of cancers were also not individually assessed, which could be a further limiting factor.

Wider implications of the findings: Testicular cancers, especially seminomas, have the most severe effect upon semen parameters. Among NTC patients, lymphomas have the worst impact. Knowing the varying effect of different cancers on semen parameters in our ethnic population helps ART specialists and oncologists to appropriately modify patient counseling and improve fertility outcomes.

Trial registration number

RMO13019

POS0357 MiRNAs CORRELATE WITH IMPROVEMENT IN DISEASE ACTIVITY IN PATIENTS WITH RHEUMATOID ARTHRITIS ON TUMOUR NECROSIS FACTOR INHIBITORS

Background:

Tumour necrosis factor inhibitors (TNFi) although effective in the treatment of rheumatoid arthritis (RA), show a variable response rate. Therefore, there is a need to identify treatment response predictors to inform therapy selection in order to practise precision medicine. MicroRNAs (miRNAs) are endogenous, single-stranded, non-coding RNAs that can alter gene expression by regulating messenger RNA translation. There is evidence for miRNA involvement in RA pathogenesis and they may serve as a useful biomarker of treatment response.

Objectives:

To identify miRNAs associated with response to TNFi in RA.

Methods:

Biologic naïve patients were selected from the Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate (BRAGGSS), a prospective multi-center UK study investigating treatment response biomarkers to TNFi with a primary outcome measure of change in DAS28 scores. Patients were stratified into European League Against Rheumatism (EULAR) good or non-responders based on their 3 or 6-month DAS28-CRP score.

Pre-treatment and 3-month post-treatment serum samples were substrates for miRNA profiling, which was conducted by FIRALIS using the HTG EdgeSeq miRNA whole transcriptome V2 targeted sequencing assay. Linear modelling using R package limma compared miRNA expression at (i) pre-treatment and at three-months, in EULAR good-responders and non-responders (ii) longitudinal change in expression from pre-treatment to three-months in EULAR good and non-responders.

A literature search was conducted to identify miRNAs associated with RA as a diagnostic and/or treatment response predictor. Data on these miRNAs were extracted from the miRNAs identified in the serum samples. A correction for multiple testing was applied to statistical tests.

Results:

A total of 54 patients were analysed; of these, 35 (65%) were female, median disease duration [inter-quartile range] was 6 years [2 – 14] (n=51), and 44/51 (86%) patients were on a concomitant disease modifying anti-rheumatic drug. Of the 54 patients, 39 (72%) were classified as EULAR good-responders and 15 (28%) as non-responders. 1880 miRNAs were detected in the serum samples. 64 miRNAs were identified to be associated with RA from the literature, of which, 26 were identified in the serum samples tested.

No difference in pre-treatment or three-month miRNA levels was seen comparing EULAR good-responders and non-responders (FDR p<0.05). There was a significant differential expression of four miRNAs at 3-months in good-responders compared with pre-treatment levels; miR-125a-3p (downregulated, p-value 0.002), miR-149-3p (upregulated, p-value 0.004), miR-766-3p (downregulated, p-value 0.008), miR-146b-5p (upregulated, p-value 0.006). No significant differences were observed between 3-months and baseline in non-responders.

Conclusion:

Although no pre-treatment miRNAs were associated with TNFi response, changes in the levels of four miRNAs were detected at 3-months compared to baseline in EULAR good-responders. Future work involves validation of these samples in a larger patient cohort and analysing miRNA levels at 6 and 12 months. Replication and validation of these results in larger studies are required to analyse the role of miRNAs in stratifying EULAR good-responders from non-responders at three-months, and as treatment response predictors to TNFi in RA.

Acknowledgements:

Joint last-author: Dr. Darren Plant

Disclosure of Interests:

Trixy David: None declared, Nisha Nair: None declared, James Oliver: None declared, Eric Schordan: None declared, Hüseyin Firat: None declared, Kimme Hyrich Consultant of: consultancy/honoraria from AbbVie, Grant/research support from: Pfizer, UCB, BMS, Ann Morgan: None declared, Anthony G Wilson: None declared, John D Isaacs Speakers bureau: consultancy/speaker fees from AbbVie, Gilead, Roche, UCB, Consultant of: consultancy/speaker fees from AbbVie, Gilead, Roche, UCB, Grant/research support from: Pfizer, Darren Plant: None declared, Anne Barton: None declared

Anterior tibial artery pseudoaneurysm

A pseudoaneurysm, or false aneurysm, is a haematoma that is formed secondary to a leaking hole in an artery. This haematoma is contained by surrounding fascia. In contrast, a true aneurysm contains all three layers of vessel wall, namely intima. Pseudoaneurysms are scarce and can arise consequential of numerous iatrogenic influences, including but not limited to, blunt or penetrating trauma, orthopedic procedures like tibial nailing or ankle arthroscopy, and sports injury. A thorough history taking focusing on the recent history of trauma or instrumentation and clinical examination should raise the suspicion of a pseudoaneurysm. In doubtful cases, imaging modalities such as an ultrasound and doppler examination of the lower limb can be utilized to confirm the diagnosis. Our case was a 37-year-old gentleman presented with progressive swelling in the anterior aspect of his left leg for the past two weeks. The patient had a atypical presentation, with absence of classic signs of a pseudoaneurysm such as a pulsatile mass, absence distal pulses or a thrill or bruit. However, these injuries albeit rare can be sinister and prompt diagnosis is critical, so that pertinent treatment can be delivered. Our case highlights the importance of sonographic approaches for suspected vascular injuries.

OP0127 TRANSCRIPTIONAL ACTIVITY OF SURVIVIN CONTRIBUTES TO MATURATION AND FUNCTION OF THE INTERFERON-GAMMA PRODUCING T CELLS IN RHEUMATOID ARTHRITIS

Background:

Interferon gamma (IFNg) signalling and downstream effects make important contribution in pathogenesis of rheumatoid arthritis (RA). Here, we propose a mechanism by which oncoprotein survivin participates in development of IFN-dependent repertoire of T cells in RA patients.

Objectives:

We study the role of survivin in the phenotype of CD4 T cells of RA patients.

Methods:

CD4 cells of RA patients and healthy controls were purified from blood, activated and subjected to RNAseq, ChIPseq with antibodies to survivin (BIRC5) was performed on CD4+ cells. Histone H3 ChIPseq was performed using antibodies to H3K27ac, H3K27me3 and H3K4me3. Statistical analysis was performed In R-studio using the Bioconductor package DESeq2, clustering using Spearman and Ward.D2.

Results:

Unsupervised clustering of CD4 samples by expression of 48 core Th cell markers identified subsets of CD28hiCD27hiIFNnegcentral memory cells (Tcm), CD28loCD27loIFNloeffector memory cells (Tem) and CD28nullCD27nullIFNhiterminal effector cells (Tte). Tte cells showed classical features of Th1 cells including high levels of TBX21, TNFa and IL32 and signs of activation in IFN signalling machinery. Interestingly, they combined the features of peripheral Tregs CD25hiFoxp3hiIKZF2negand IL10 producing cells together with type 1 regulatory cells, which rely on transcription factors BATF and IRF1 for the differentiation and produced high amounts of perforin and granzyme B. Importantly, Tte CD4 cells had also high transcription of BIRC5 (p=1.15e-18).

To study if BIRC5 is a part of IFN signalling, CD4 cells were cultured with survivin inhibitor YM155 and activated with IFNg. RNAseq analysis revealed 2033 (FC<2.0, n=336) differentially expressed genes in the IFN stimulated cultures. Interestingly, a substantial number of these IFN-dependent genes was significantly reduced in the survivin-deficient cultures and included among others CD28, FoxP3, IKZF2, ICOS, BATF, PRDM1, CXCR3, IRF4 and IRF8. Analysis of the peak sequences identified enrichment for composite motifs for IRFs (ETS:IRF, p1.0e-124; bZIP:IRF, p=1.0e-640), indicating that survivin is important for IFNg signalling. Numerically, the peaks containing ETS:IRF motifs were most prevalent and identified in total within 49.7% of all survivin-ChIP peaks. Frequent was co-localisation of the IRF:bZIP and IRF:ETS motifs within the survivin peaks. Among the IRF motifs dominated those suitable for IRF1 (p=1,0e-127) and IRF8 (p=1,0e-84). However, the DNA binding motifs of these two are alike.

Encouraged by the survivin ChIPseq results, we wanted to know its relation to histone marks. We observed that 50% of survivin peaks containing both IRF:bZIP and IRF:ETS motifs are co-localized with the H3K27ac marks. In total, 16 of 48 core Th cell markers used for patients clustering were identified by survivin ChIPseq, co-localized with IRF composite motifs and histone marks. They were also dependent of survivin for expression.

Conclusion:

his study showed that survivin binds to DNA and regulates the core gene expression contributing to maturation and function of the IFNg producing Th1 cells.

References:

-

Disclosure of Interests:

Malin Erlandsson: None declared, Karin ME Andersson: None declared, Nisha Nair: None declared, Anastasius Damdimopoulos: None declared, Sofia Töyrä Silfverswärd: None declared, Rille Pullerits: None declared, Anne Barton Consultant of: AbbVie, Maria I Bokarewa: None declared

THU0005 VARIABILITY OF DNA METHYLATION IS A DRIVER OF LYMPHOCYTE DYSREGULATION IN EARLY RHEUMATOID ARTHRITIS

Background:

DNA methylation patterns differ between leukocyte subsets and mediate the impact of environmental exposures on the molecular and functional phenotype of immune cells. Besides differences in mean methylation of CpG positions amongst patients with immune mediated diseases, recent evidence indicates variability of site-specific DNA methylation also contributes to pathogenesis1,2.

Objectives:

To seek evidence of altered DNA methylation patterns in RA, controlling for systemic inflammation and immunotherapy use.

Methods:

Patients with confirmed clinical diagnoses were enrolled from the Northeast Early Arthritis Cohort (NEAC). CD4+and CD19+lymphocytes were isolated from fresh blood by positive selection prior to therapeutic immune modulation. Methylation was quantified in cell subset-specific DNA (Infinium MethylationEPIC BeadChip, Illumina)3. Differentially methylated positions and regions (DMPs, DMRs) between RA and non-RA patients were identified (linear modelling, filtering on 5% pairwise difference in mean DNA methylation, and DMRcate package). Next, to identify instances where methylation variance differed between comparator groups, Bartlett’s test was performed using the iEVORA package, which accounts for outlier values4. Findings were controlled for technical confounders and subject to multiple test correction (FDR). A validated hypergeometric test was used to annotate enriched pathways.

Results:

After sample- and probe-level quality control, CD4+ and B lymphocyte specific data were respectively available for 45 and 49 RA patients, and 64 and 81 disease controls matched for systemic inflammation (CRP, ESR). No DMPs were identified in either cell type at FDR < 0.05 and Δβ ≥0.05. Only following relaxation of multiple test correction was it possible to identify DMRs in either cell type, most notably encapsulating 10 CpGs relatively hypomethylated at the promoter of the endosome protein-encodingRUFY1gene in CD4+ lymphocytes of RA patients (Δβ = 0.076). By contrast, striking evidence for differential variation in DNA methylation was observed at 291 and 601 CpGs of CD4+ and B lymphocytes, respectively (exemplars depicted in Figure 1). Only 15 of these differentially variable positions (DVPs) were common to both cell types. Pathway analysis highlighted potential functional consequences of DVP associations; for example, RA-specific hypervariability implicates prostaglandinsignalling in CD4+ lymphocytes.

Conclusion:

We highlight a role for altered variability in DNA methylation during the molecular pathogenesis of RA, and emphasise the importance of its study in relevant cell subsets.

References:

[1]Paul DSet al. Nature Communications 7, 13555 doi: 10.1038/ncomms13555 (2016).

[2]Webster AP et al. Genome Medicine 10, 64 (2018)doi:10.1186/s13073-018-0575-9.

[3]Clark AD et al. Journal of Allergy and Clinical Immunology 2019; doi: 10.1016/j.jaci.2019.12.910

[4]Teschendorff AE et al. Nature Communications 2016; 7:12.

Disclosure of Interests:

Alexander Clark: None declared, Najib Naamane: None declared, Nisha Nair: None declared, Amy Anderson: None declared, Nishanthi Thalayasingam: None declared, Julie Diboll: None declared, Anne Barton Consultant of: AbbVie, Stephen Eyre: None declared, John D Isaacs Consultant of: AbbVie, Bristol-Myers Squibb, Eli Lilly, Gilead, Janssen, Merck, Pfizer, Roche, Louise Reynard: None declared, Arthur Pratt Grant/research support from: Pfizer, GlaxoSmithKlein

THU0022 DIFFERENTIAL DNA METHYLATION AS A PREDICTOR OF TOCILIZUMAB RESPONSE IN RHEUMATOID ARTHRITIS PATIENTS

Background:

Tocilizumab (TCZ) is a biological disease-modifying antirheumatic drug that blocks IL-6 signalling and is effective in ameliorating disease activity in rheumatoid arthritis (RA). However, approximately 50% of patients do not respond adequately to TCZ and some patients report adverse events. Considering there is growing evidence that DNA methylation is implicated in RA susceptibility and response to some biologics (1, 2), we investigated DNA methylation as a candidate biomarker for response to TCZ in RA.

Objectives:

To identify differential DNA methylation signatures in whole blood associated with TCZ response in patients with RA.

Methods:

Epigenome-wide DNA methylation patterns were measured using the Infinium EPIC BeadChip (Illumina) in whole blood-derived DNA samples from patients with RA. DNA was extracted from blood samples taken pre-treatment and following 3 months on therapy, and response was determined at 6 months using the Clinical Disease Activity Index (CDAI). Patients who had good response (n=10) or poor response (n=10) to TCZ by 6 months were selected. Samples from secondary poor responders (n=10) (patients who had an improvement of CDAI and were in remission at 3 months, followed by a worsening of CDAI at 6 months) were also analysed. Differentially methylated positions and regions (DMPs/DMRs) were identified using linear regression, adjusting for gender, age, cell composition, smoking status, and glucocorticoid use. Gene Set Enrichment Analysis (GSEA) was used to identify significant pathways associated with response and Functional Epigenetic Module analysis of interactome hotspots in regions of differential methylation.

Results:

20 DMPs were significantly associated with response status at 6 months in the pre-treatment samples. Another 21 DMPs were associated with response in the 3 month samples. Within good responders, 10 DMPs showed significant change in methylation level between pre-treatment and the 3 month samples (unadjusted P-value <10-6). One DMP, cg03121467, was significantly less methylated in good responders compared to poor responders in the pre-treatment samples. This DMP is close toEPB41L4Aand thought to have a role in β–catenin signalling. GSEA of DMRs in non- and secondary non- responders identified histone acetyltransferase pathways and included theKAT2Agene, which is a repressor of NF-κB. Additional analysis of interaction hotspots of differential methylation identified significant interactions withSTAMBPandPTPN12associated with response status.

Conclusion:

These preliminary results provide evidence that DNA methylation patterns may predict response to TCZ. Validation of these findings in other larger data sets is required.

References:

[1]Liu,Y., Aryee,M.J., Padyukov,L., Fallin,M.D., Hesselberg,E., Runarsson,A., Reinius,L., Acevedo,N., Taub,M., Ronninger,M.,et al.(2013) Epigenome-wide association data implicate DNA methylation as an intermediary of genetic risk in rheumatoid arthritis.Nat. Biotechnol.,31, 142–147.

[2]Plant,D., Webster,A., Nair,N., Oliver,J., Smith,S.L., Eyre,S., Hyrich,K.L., Wilson,A.G., Morgan,A.W., Isaacs,J.D.,et al.(2016) Differential Methylation as a Biomarker of Response to Etanercept in Patients With Rheumatoid Arthritis.Arthritis Rheumatol. (Hoboken, N.J.),68, 1353–60.

Disclosure of Interests:

Nisha Nair: None declared, Darren Plant: None declared, John Isaacs Consultant of: AbbVie, Bristol-Myers Squibb, Eli Lilly, Gilead, Janssen, Merck, Pfizer, Roche, Ann Morgan Grant/research support from: I have received a grant from Roche Products Ltd to establish a registry for GCA patients treated with tocilizumab., Consultant of: I have undertaken consultancy work for Roche, Chugai, Regeneron, Sanofi and GSK in the area of GCA therapeutics., Speakers bureau: I have presented on tocilizumab therapy for GCA and glucocorticoid toxicity on behalf of Roche products ltd., Kimme Hyrich Grant/research support from: Pfizer, UCB, BMS, Speakers bureau: Abbvie, Anne Barton Consultant of: AbbVie, Anthony G Wilson: None declared

0501 Development and Initial Evaluation of Web-Based Cognitive Behavioral Therapy for Insomnia “Nitecapp” in Rural Dementia Caregivers: A Mixed-Methods Study

Introduction

Informal caregivers (CGs) of persons with dementia frequently experience insomnia. The time consuming and unpredictable schedule of CGs, and associated emotional/physical exhaustion emphasize the need for brief, easily accessible interventions to treat insomnia. Internet-based behavioral insomnia interventions hold promise, particularly for rural CGs who have limited access to traditional in-person treatments. This study aimed to 1) translate an efficacious 4 session cognitive behavioral therapy for insomnia (CBT-I) to web-based “NiteCAPP” for dementia caregivers, and 2) conduct NiteCAPP usability testing/evaluate acceptability of content and features.

Methods

NiteCAPP is an online CBT-I that incorporates guided delivery through weekly therapist moderator feedback. A stepwise approach was implemented in order to explore user needs and validate NiteCAPP content in a focus group of rural dementia caregivers (n=5) and primary care providers (PCPs; n=5). Participants conducted usability testing and provided ratings of program content (1-least favorable to 5-most favorable) regarding ease of use, amount of information, website maintaining interest, adequate font size, videos maintaining interest/easy to understand/helpful. Participants also indicated whether they had at home internet access, method of internet access, and provided open ended feedback on NiteCAPP. Feedback transcripts were compiled and analyzed independently (C.S.M., A.F.C.) through deductive content analysis. Topics mentioned frequently were categorized and merged into common themes during consensus meeting, and NiteCAPP was subsequently adapted.

Results

Average ratings for NiteCAPP features were high, ranging from 4.1/5 to 4.7/5 across all items. All participants had access to internet through both phone and computer. No barriers to use identified. Feedback themes were largely positive (e.g., comprehensive written material, promotes independence, excellent visual tools for therapy moderator feedback, good pacing, use of visual contrast). Negative themes for improvement/adaptation included adding font size options, a light/dark mode, tab with all videos, reducing amount of scrolling, adding a glossary of terms.

Conclusion

Rural dementia CGs and PCPs evaluated NiteCAPP as easy to use with acceptable features and program content and no barriers to access. Improvement themes were used to adapt NiteCAPP. Next steps are to evaluate feasibility and preliminary efficacy of NiteCAPP in rural dementia CGs with insomnia.

Support

none

Health gains, costs and cost-effectiveness of a population-based screening programme for abdominal aortic aneurysms

BACKGROUND

Abdominal aortic aneurysm (AAA) rupture carries a high fatality rate. AAAs can be detected before rupture by abdominal ultrasound imaging, allowing elective repair. Population-based screening for AAA in older men reduces AAA-related mortality by about 40 per cent. The UK began an AAA screening programme offering one-off scans to men aged 65 years in 2009. Sweden has a similar programme. Currently, there is no AAA screening programme in New Zealand. This cost-utility analysis aimed to assess the cost-effectiveness of a UK-style screening programme in the New Zealand setting.

METHODS

The analysis compared a formal AAA screening programme (one-off abdominal ultrasound imaging for about 20 000 men aged 65 years in 2011) with no systematic screening. A Markov macrosimulation model was adapted to estimate the health gains (in quality-adjusted life-years, QALYs), health system costs and cost-effectiveness in New Zealand. A health system perspective and lifetime horizon was adopted.

RESULTS

With New Zealand-specific inputs, the adapted model produced an estimate of about NZ $15 300 (€7746) per QALY gained, with a 95 per cent uncertainty interval (UI) of NZ $8700 to 31 000 (€4405 to 15 694) per QALY gained. Health gains were estimated at 117 (95 per cent UI 53 to 212) QALYs. Health system costs were NZ $1·68 million (€850 535), with a 95 per cent UI of NZ $820 200 to 3·24 million (€415 243 to €1·65 million).

CONCLUSION

Using New Zealand's gross domestic product per capita (about NZ $45 000 or €22 100) as a cost-effectiveness threshold, a UK-style AAA screening programme would be cost-effective in New Zealand.

Abstract P5-14-07: Accuracy of psychosocial assessments in an online surgical decision aid developed for early breast cancer patients with resource and educational constraints

Background: Women with early breast cancer routinely face a choice between breast conservation therapy and mastectomy, and assume agency through shared decision making. However, for women with lower socioeconomic power or education, barriers such as access to understandable information, involvement of family in decision making, and a decreased sense of autonomy inhibits this agency. To better empower this population, a simple to understand, online, self-administered, conjoint analysis based decision aid called “Navya Patient Preference Tool” (PPT) is developed to be used outside the physician encounter. PPT is unique in its incorporation of several psychological scales that assess potential confounders of participation in shared decision making.

Methodology: This is a pre-planned analysis of the reliability and validity of the psychological scales used in all three arms of an IRB approved randomized controlled trial to assess PPT. Women with operable node negative breast cancer eligible for BCT or MRM at one of Asia's largest academic tertiary cancer centers were eligible. PPT trial consists of an initial conjoint analysis questionnaire analyzing implicit preferences for breast conservation given to the intervention arms. The following psychological scales were given to all patients regardless of randomization: Autonomy Preference Index (API), Traditional-Egalitarian Gender Roles (TEGR), Caregiving Role, Brief Resiliency Scale (BRS), Appearances Scale, and Decisional Conflict Scale (DCS). Cronbach's alpha as a measure of internal reliability for all scales, and correlations of scores with known demographic trends as a measure of external validity are calculated.

Results: Of the 102 patients enrolled, 30 completed PPT in English, 39 in Hindi, and 33 in Marathi, (vernaculars). 69/102 were in middle and lower socioeconomic groups (Kuppuswamy Index). 53/102 had completed less than high school education. Internal reliability of all scales were high, with Cronbach's alpha above 0.7: API 0.74, TEGR 0.78, Caregiving 0.7, BRS 0.7, Appearance 0.84. DCS was highly reliable at 0.91, and is the primary outcome measure for the RCT. Correlations in the dataset met those expected in real world data, suggesting external validity. For e.g., education was inversely correlated with traditional gender roles on TEGR (R -0.4, p &lt;0.01), and positively correlated with resilience on BRS (R 0.228, p &lt;0.05). Individual scale items that are unrealistic were not chosen by any of the 102 respondents (e.g.,. My doctor should not participate in my medical decisions), substantiating nuanced reading. 85% of patients “Strongly Agreed” on a 1-5 Likert scale that “The survey questions were easy to understand” (mean score 1.18/5. SD 0.4).

Conclusions: Women with limited education and low socioeconomic status complete the online, self administered PPT outside of a physician encounter, with high internal reliability and external validity. Decision Aids such as Navya PPT, which account for psychosocial confounders of agency, have the potential to benefit women otherwise marginalized from shared decision making.

Citation Format: Joshi S, Ramarajan L, Ramarajan N, Srivastava G, Begum F, Deshpande O, Tondare A, Nair N, Parmar V, Gupta S, Badwe RA. Accuracy of psychosocial assessments in an online surgical decision aid developed for early breast cancer patients with resource and educational constraints [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P5-14-07.

More Robust Monitoring for Continuous Quality Improvement in Screening Programmes

Background and context: The National Screening Unit in the Ministry of Health is responsible for the safety, effectiveness, and quality of five formal screening programs in New Zealand: breast screening, cervical screening, bowel screening, newborn metabolic screening, and newborn hearing screening. Currently, each screening program has a set of standards against which performance is assessed. Each program is monitored through a range of avenues: formally reported indicators, audits, contract reporting, case reviews, etc. Aim: We aimed to review the existing monitoring across all the screening programs to: (a) Clarify what existing measures served what purpose, and for which audience (b) Review what was being captured through different monitoring avenues, to assess gaps and duplications (c) To better align our monitoring approach across all the screening programs. Strategy/Tactics: For each measure, we used a formal template to justify its existence, covering multiple aspects including rationale, lifespan, dimension of quality, screening pathway step, targets and associated evidence, thresholds for corrective action, and “owner” responsible for triggering quality improvement. We also mapped the screening program standards to various monitoring avenues to assess gaps and duplications. Program/Policy process: This process was timed to coincide with the need for developing monitoring for a new screening program (bowel screening), and revising monitoring for an established screening program undergoing a change (cervical screening switching to HPV testing as a primary screen). Outcomes: *anticipated outcomes in italics, will be completed by time of Congress* This process resulted in an overarching monitoring framework, which functions as a blueprint for screening-program-specific monitoring plans. What was learned: Screening programs have different monitoring requirements depending on what phase they are in (i.e., new versus changing versus established). Having clearly documented rationale ensures that every measure is justified and has an “evidence trail”. Having clearly documented audiences, thresholds and responsibilities for corrective action ensures that monitoring contributes to continuous quality improvement, rather than monitoring for monitoring's sake.

Abstract P4-10-02: Validation of a clinical informatics system for online multidisciplinary expert opinions: Mapping treatment recommendations to the NCCN resource-Stratified framework

Background: Most cancer patients in Low and Middle Income Countries (LMIC) cannot afford effective, expensive, evidence based therapies. Therefore, oncologists must tailor treatment plans to individual resource constraints. To support this, NCCN has created a Resource-Stratified Framework® (NCCN-RSF), which is an evidence-based four-tier prioritization scheme. Further, only a fraction of patients in LMIC have ready access to oncologists. In India, there are only ˜1600 oncologists for 1.8 million patients. To bridge this gap, Navya's clinical informatics based mobile ExpertApp combines learning from evidence, prior tumor board decisions, patient resource constraints, and quick review from TMC NCG oncologists to recommend tailored treatment plans to patients via an online expert opinion service. 11865 patients in 22 LMIC have reached out to receive an online expert opinion through Navya (ASCO 2017). This study maps Navya to NCCN-RSF as an evidence-based index for resource-sensitive treatment selection.

Methods: All breast cancer patients who received an online expert opinion from TMC NCG Navya between July 1st 2014 and April 30th 2017 were included. Navya systematically gathered information on patient resource constraints (such as affordability for Trastuzumab). Navya recommendations (breast and nodal surgery, radiation site and fractionation, drug and dose density etc.) were mapped to NCCN-RSF resource tiers (Basic, Core, Enhanced, Parent guideline). Reasons were categorized for Navya recommendations not present in NCCN-RSF.

Results: 616 patients (36.3% metastatic), mostly from India, received 1203 recommendations. At the specific treatment protocol level, 88.3% of Navya recommendations mapped with at least one NCCN-RSF resource tier (Table 1). 78.5% mapped to the Enhanced tier. Only 8.6% of recommendations mapped to Parent guidelines, and did not require tailoring for resource constraints. Fewer than 2% recommendations mapped to Core and none to Basic. 11.7% recommendations were not present in NCCN-RSF, for minor reasons such as substitution of a drug within the same class (35.8%) (e.g., Epirubicin for Adriamycin), dose dense protocols (14.3%) (e.g., 3 weekly Paclitaxel vs weekly Paclitaxel), and recommending Trastuzumab for less than a year for patients unable to afford year long therapy (14.3%), currently not included in NCCN-RSF.

Table 1- Mapping Navya to NCCN RSFNCCN RSF TiersHIGH LEVEL: Multimodality treatment and sequencing (1203)INTERMEDIATE: Within modality treatment categories (1188)GRANULAR: Specific treatment protocols (1140)E.g.Neoadjuvant vs Adjuvant ChemoAnthracycline vs TaxaneHypofractionation vs Standard XRTAt least one Tier98.8%±0.696%±1.188.3%±2Enhanced94.4%±1.391%±1.778.5%±2.7Core1.9%±5.61.2%±5.71.2%±5.8Parent NCCN2.4%±5.63.8%±5.68.6%±5.5

Conclusion: Navya's treatment recommendations are sensitive to resource constraints and map to peer reviewed and evidence based NCCN RSF, primarily at the Enhanced tier. Navya's clinical informatics based online service scales access to resource constrained treatment selection for large numbers of patients in LMIC without easy access to oncologists.

Citation Format: Badwe RA, Gupta S, Feldman N, Pramesh CS, Ramarajan N, Srivastava G, Nair N, Anderson BO. Validation of a clinical informatics system for online multidisciplinary expert opinions: Mapping treatment recommendations to the NCCN resource-Stratified framework [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P4-10-02.

Tezacaftor-Ivacaftor in Residual-Function Heterozygotes with Cystic Fibrosis

BACKGROUND

Cystic fibrosis is an autosomal recessive disease caused by mutations in the CFTR gene that lead to progressive respiratory decline. Some mutant CFTR proteins show residual function and respond to the CFTR potentiator ivacaftor in vitro, whereas ivacaftor alone does not restore activity to Phe508del mutant CFTR.

METHODS

We conducted a randomized, double-blind, placebo-controlled, phase 3, crossover trial to evaluate the efficacy and safety of ivacaftor alone or in combination with tezacaftor, a CFTR corrector, in 248 patients 12 years of age or older who had cystic fibrosis and were heterozygous for the Phe508del mutation and a CFTR mutation associated with residual CFTR function. Patients were randomly assigned to one of six sequences, each involving two 8-week intervention periods separated by an 8-week washout period. They received tezacaftor-ivacaftor, ivacaftor monotherapy, or placebo. The primary end point was the absolute change in the percentage of predicted forced expiratory volume in 1 second (FEV₁) from the baseline value to the average of the week 4 and week 8 measurements in each intervention period.

RESULTS

The number of analyzed intervention periods was 162 for tezacaftor-ivacaftor, 157 for ivacaftor alone, and 162 for placebo. The least-squares mean difference versus placebo with respect to the absolute change in the percentage of predicted FEV₁ was 6.8 percentage points for tezacaftor-ivacaftor and 4.7 percentage points for ivacaftor alone (P<0.001 for both comparisons). Scores on the respiratory domain of the Cystic Fibrosis Questionnaire-Revised, a quality-of-life measure, also significantly favored the active-treatment groups. The incidence of adverse events was similar across intervention groups; most events were mild or moderate in severity, with no discontinuations of the trial regimen due to adverse events for tezacaftor-ivacaftor and few for ivacaftor alone (1% of patients) and placebo (<1%).

CONCLUSIONS

CFTR modulator therapy with tezacaftor-ivacaftor or ivacaftor alone was efficacious in patients with cystic fibrosis who were heterozygous for the Phe508del deletion and a CFTR residual-function mutation. (Funded by Vertex Pharmaceuticals and others; EXPAND ClinicalTrials.gov number, NCT02392234 .).