ACR Meeting Abstracts

ACR Meeting Abstracts

Home » Meeting Abstracts » ACR/ARHP Annual Meeting 2017

Abstract Number: 122

Poor Prognostic Factors at the Start of Methotrexate Therapy Are Not Associated with Worse Treatment Response: Results from the Rheumatoid Arthritis Medication Study

JM Gwinnutt1, Kimme L. Hyrich1, M Lunt1, Darren Plant1, M Brazil2, R Postema2, Anne Barton1 and Suzanne M Verstappen1, 1Manchester Academic Health Science Centre, Manchester, United Kingdom, 2Bristol-Myers Squibb, Uxbridge, United Kingdom

Meeting: ACR/ARHP Annual Meeting 2017

Date of first publication: September 18, 2017

Keywords: , , ,

Session Information

Date: Sunday, November 5, 2017

Title: Epidemiology and Public Health Poster I: Rheumatoid Arthritis

Session Type: ACR Poster Session A

Session Time: 9:00AM-11:00AM

Background/Purpose: As anti-citrullinated protein antibody positivity (+), RF+ and erosions are independently associated with poor outcomes in patients (pts) with RA, clinicians may use these prognostic factors in treatment decisions. However, it is not known whether pts with these poor prognostic factors respond equally well to conventional synthetic DMARD therapy. The aim of this analysis was to compare clinical and pt-reported outcomes (DAS28 [CRP], HAQ) of pts with RA with poor prognostic factors (RF+ and/or anti-cyclic citrullinated peptide antibody 2+ [anti-CCP2+] and erosions) vs those without over 1 year. Methods: Pts with RA in the UK starting MTX therapy were recruited to the Rheumatoid Arthritis Medication Study (RAMS), a 1-year, prospective cohort study. RF was determined from baseline blood samples, and data about anti-CCP status and erosions were obtained from medical notes. Pts who were anti-CCP2+ and/or RF+ and had erosions were classified as having poor prognosis (PP), while those not were classified as not having poor prognosis (NPP). For this analysis, pts were excluded if they were recruited >2 years following symptom onset or had missing data for any of the three prognostic factors. At baseline, 6-month and 12-month demographic and clinical data were recorded by a research nurse and pts completed the HAQ. The association between prognosis group and DAS28 (CRP) and HAQ at 6 and 12 months was assessed using linear and negative binomial regression, respectively. The association between prognosis group and longitudinal DAS28 (CRP) and HAQ was assessed using linear and negative binomial random effects models, respectively. Baseline age, sex and symptom duration were included in the models as covariates. Results: In total, 545 pts with RA were included (PP, n=79 [14.5%]; NPP, n=466 [85.5%]). At baseline, PP pts were older, but other characteristics were similar (Table). DAS28 (CRP) and HAQ scores were similar between groups at 6 months (median [interquartile range (IQR)] DAS28 [CRP]: PP, 3.03 [2.4, 4.1] vs NPP, 3.2 [2.3, 4.3], coefficient –0.11, 95% CI –0.52, 0.29; HAQ: PP, 0.88 [0.25, 1.50] vs NPP, 0.63 [0.13, 1.25], incidence rate ratio [IRR] 1.12, 95% CI 0.76, 1.64) and 1 year (median [IQR] DAS28 [CRP]: PP, 2.98 [2.25, 3.75] vs NPP, 3.00 [2.05, 4.05], coefficient 0.02, 95% CI –0.39, 0.43; HAQ: PP, 0.75 [0.13, 1.38] vs NPP, 0.63 [0.00, 1.25], IRR 1.12, 95% CI 0.76, 1.64). Similar results were seen when using longitudinal models to combine all the repeated measures (DAS28 [CRP]: coefficient –0.02, 95% CI –0.30, 0.26; HAQ: IRR 1.17, 95% CI 0.92, 1.48).

Conclusion: Treatment response is similar in pts with and without poor prognostic factors. These results suggest that, at this early stage of disease, baseline poor prognostic factors do not predict treatment response to conventional synthetic DMARDs, nor do they indicate that RA disease is reported as worse at 6 or 12 months in pts with vs without these poor prognostic factors.

Table. Patient Demographic and Clinical Data at Baseline

 

Total cohort (N=545) PP
(n=79) 		NPP
(n=466) 		p-value*
Age at onset, years 57 (47, 66) 60 (48, 71) 56 (47, 66) 0.03†
Female, n (%) 354 (65.0) 49 (62.0) 305 (65.5) 0.54‡
Symptom duration (months) 6 (4, 12) 7 (4, 11) 6 (4, 12) 0.93†
Smoking status, n (%)        Never        Ex-smoker        Current   201 (37.2) 208 (38.5) 131 (24.3)   31 (39.7) 30 (38.5) 17 (21.8)   170 (36.8) 178 (38.5) 114 (24.7)   0.83‡
SJC (28) 4 (2, 9) 4 (2, 9) 4 (2, 9) 0.72†
TJC (28) 6 (2, 12) 5.5 (2, 9) 6 (2, 12) 0.56†
CRP (mg/L) 5 (2, 16) 6 (3, 16) 5 (2, 16) 0.23†
DAS28 4.25 (3.25, 5.20) 4.30 (3.55, 4.80) 4.20 (3.25, 5.25) 0.96†
HAQ 0.88 (0.38, 1.50) 1.00 (0.38, 1.75) 0.88 (0.38, 1.50) 0.25†
RF status, n (%)        Positive        Negative   372 (68.3) 173 (31.7)   74 (93.7) 5 (6.3)   298 (64.0) 168 (36.1)   <0.001‡
Anti-CCP2 status, n (%)        Positive        Negative   344 (63.1) 201 (36.9)   66 (83.5) 13 (16.5)     278 (59.7) 188 (40.3)   <0.001‡
Erosions, n (%)        Positive        Negative   98 (18.0) 447 (82.0)   79 (100.0) 0 (0.0)   19 (4.1) 447 (95.9)   <0.001‡
Other csDMARD at baseline, n (%)        Hydroxychloroquine        Sulfasalazine     66 (12.3) 22 (3.9)     13 (16.5) 4 (5.1)     53 (11.6) 18 (3.9)     0.22‡ 0.61‡
Steroids at baseline, n (%) 104 (19.2) 14 (18.0) 90 (19.4) 0.76‡
Data are median (IQR) unless otherwise stated *p-values resulting from comparison of baseline score across the two prognosis groups  †Mann–Whitney U test ‡Chi-square test CCP=cyclic citrullinated peptide; IQR=interquartile range; NPP=patient group not having poor prognosis; PP=patient group with poor prognosis; csDMARD=conventional synthetic DMARD
   
Disclosure: J. Gwinnutt, None; K. L. Hyrich, None; M. Lunt, None; D. Plant, None; M. Brazil, Bristol-Myers Squibb, 1,Bristol-Myers Squibb, 3; R. Postema, Bristol-Myers Squibb, 1,Bristol-Myers Squibb, 3; A. Barton, Roche-Chugai, Celgene and Boehringer Engelheim, 5,Bristol-Myers Squibb, 2; S. M. Verstappen, None.

To cite this abstract in AMA style:

Gwinnutt J, Hyrich KL, Lunt M, Plant D, Brazil M, Postema R, Barton A, Verstappen SM. Poor Prognostic Factors at the Start of Methotrexate Therapy Are Not Associated with Worse Treatment Response: Results from the Rheumatoid Arthritis Medication Study [abstract]. Arthritis Rheumatol. 2017; 69 (suppl 10). https://acrabstracts.org/abstract/poor-prognostic-factors-at-the-start-of-methotrexate-therapy-are-not-associated-with-worse-treatment-response-results-from-the-rheumatoid-arthritis-medication-study/. Accessed .

ACR Meeting Abstracts - https://acrabstracts.org/abstract/poor-prognostic-factors-at-the-start-of-methotrexate-therapy-are-not-associated-with-worse-treatment-response-results-from-the-rheumatoid-arthritis-medication-study/