Arthritis & Rheumatism, Volume 62,
November 2010 Abstract Supplement

Abstracts of the American College of
Rheumatology/Association of Rheumatology Health Professionals
Annual Scientific Meeting
Atlanta, Georgia November 6-11, 2010.


Variability of Anti-CCP Kit Testing in Cohorts with and without RA.

Deane7,  Kevin D., Demoruelle10,  Kristen, Parish4,  Mark, Hilton4,  Whitney, Harrington4,  Lauren, Derber4,  Lezlie, Kolfenbach6,  Jason R.

Oak Park, IL
University of Colorado School of Medicine
University of Manitoba, Winnipeg, MB, Canada
University of Manitoba, Winnipeg, MB, Canada
University of Manitoba
University of Nebraska Medical Center, Omaha, NE
Benaroya Research Institute at Virginia Mason, Seattle, WA
Cedars-Sinai Medical Center, Los Angeles, CA
Division of Rheumatology, University of Colorado School of Medicine, Aurora, CO
N Shore Univ Hosp Rsch Ctr, Manhasset, NY
Univ of CO Schl of Med, Aurora, CO
Univ of Colorado School of Med, Aurora, CO
University of Nebraska Medical Center, Omaha, NE
University of Colorado Denver, Aurora, CO

Purpose:

Anti-CCP positive (CCP+) subjects without current symptoms of RA are important to identify in order to investigate the biology and evolution of RA development. However, it is unknown how clinically-available CCP tests identify CCP+ in individuals who do not currently have RA. Here we examine the sensitivity (SENS) and specificity (SPEC), prevalence and test agreement of 3 CCP kits in RA cases and currently healthy first-degree relatives (FDRs) of probands with RA.

Methods:

CCP testing was performed in a single laboratory using standard protocols for 3 ELISA kits: CCP2 (IgG) (Axis-Shield), CCP3 (IgG) and CCP3.1 (IgG, IgA) (INOVA). Sera from 2 established RA cohorts were tested with each kit: a Cree/Ojibway North American Native (NAN) population from Central Canada, and the RA population from the U.S.-based Studies of the Etiologies of RA [SERA] (Caucasian 85%). The SENS/SPEC of kits for RA were calculated compared to non-RA controls (NAN, N=100; SERA random blood donor controls, N=200). Sera from 2 FDR cohorts (NAN and SERA) were also tested with each kit. The NAN FDRs were selected on the basis of CCP2+ at least once during the course of their longitudinal follow-up in the study. SERA FDRs were randomly selected regardless of CCP+ from a cohort of ~1600 subjects. Kappa testing was performed to assess agreement of CCP tests.

Results:

See Table 1: The SENS of CCP for RA by any kit was higher in NAN RA cases compared to SERA cases, and the SENS of CCP+ in RA cases was highest using combined kit testing, although SPEC was lower. In NAN and SERA FDRs, CCP3 and 3.1 kits had higher prevalence of positivity compared to CCP2, with differences most pronounced in SERA FDRS: CCP2+(2%), CCP3+(4%) CCP3.1+(9%). Median levels of CCP were lowest in SERA FDRs.

Table 1. Anti-CCP Kit Test Results (Sensitivity/Specificity, Prevalence, and Titers) in Established RA and Healthy First-Degree Relative (FDR) Cohorts

CohortCCP2CCP3CCP3.1>=1 of 3 Kits Positive
NANRA (N = 112) vs Non-    
  RA Controls (N = 100)    
  SENS71%67%69%75%
  SPEC97%98%96%95%
SERA RA (N = 230) vs Non-    
  RA Controls (N = 200)    
  SENS54%63%65%67%
  SPEC99%95%94%91%
NAN FDR (N = 31)    
  Prevalence of CCP positivity13/31 (42%)15/31 (48%)14/31 (45%)16/31 (52%)
SERA FDR (N = 299)    
  Prevalence of CCP positivity7/299 (2%)12/299 (4%)26/299 (9%)31/299 (10%)
Median level (range) [levels calculated only in those positive]    
  NAN RA104 (>5–134)315 (38–401)366 (39–401)N/A
  NAN FDR107 (47–127)280 (25–401)327 (30–401) 
SERA RA78 (>5–201)296 (21–420)286 (23–406) 
  SERA FDR22 (>5–113)43 (21–401)28 (21–401) 
Abbreviations: NAN = North American Native cohorts; SERA = Studies of the Etiologies of Rheumatoid Arthritis cohorts; SENS = sensitivity; SPEC = specificity

See Table 2:

Test agreement (kappa) was lowest in SERA FDRs, although kappa improved when anti-CCP cutoffs for positivity were set at 3-times established kit cut-offs.

Table 2. Anti-CCP Kit Agreement, By Cohort

CohortCCP2 and 3 (kappa*)CCP2 and 3.1 (kappa*)CCP3 and 3.1 (kappa*)
NAN RA0.810.770.96
SERARA0.720.710.93
NANFDRs   
  Standard kit cut-off**0.740.800.94
  Cut-off >3× normal**0.870.871.0
SERAFDRs   
  Standard kit cut-off**0.190.150.55
  Cut-off >3× normal**0.440.440.39
*Cohen's kappa
**For CCP2 standard kit cut-off >5 units, ×3 = >15 units, for CCP3, 3.1 standard cut-off >20 units, ×3 = >60 units
Abbreviations: NAN = North American Native cohorts; SERA = Studies of the Etiologies of Rheumatoid Arthritis Cohorts

Conclusion:

In this study CCP kits differ in SENS/SPEC for RA, prevalence in FDRs and cohort-related kit agreement. In particular, in SERA FDRs the differential prevalence of CCP+ by kit and decreased kit agreement may indicate that CCP reactivity in these currently healthy individuals, who have perhaps early RA-specific autoimmunity, is fundamentally different than in established disease, possibly due to differences in autoantigen specificity or autoantibody isotypes. Further study is needed to understand the biologic factors behind these CCP results and the role of these CCP tests for identification of pre-clinical RA-specific autoimmunity.

To cite this abstract, please use the following information:
Deane, Kevin D., Demoruelle, Kristen, Parish, Mark, Hilton, Whitney, Harrington, Lauren, Derber, Lezlie, et al; Variability of Anti-CCP Kit Testing in Cohorts with and without RA. [abstract]. Arthritis Rheum 2010;62 Suppl 10 :1094
DOI: 10.1002/art.28861

Abstract Supplement

Meeting Menu

2010 ACR/ARHP