Arthritis & Rheumatism, Volume 60,
October 2009 Abstract Supplement

The 2009 ACR/ARHP Annual Scientific Meeting
Philadelphia October 16-21, 2009.

Rheumatoid Factor Interferes with Multiplex-Based ELISA in Patients with Rheumatoid Arthritis

Todd1,  Derrick J., Knowlton2,  Nicholas, Karlson1,  Elizabeth W., Shadick3,  Nancy A., Weinblatt1,  Michael E., Schur1,  Peter H., Roubenoff4,  Ronenn

Brigham and Women's Hospital, Boston, MA
Oklahoma Medical Research Foundation, Oklahoma City, OK
Brigham and Women's Hospital, Harvard Medical School, Boston, MA
Biogen-Idec Pharmaceuticals, Cambridge, MA
Millennium Pharmaceuticals, Cambridge, MA
OK Med Research Foundation, Oklahoma City, OK


Heterophilic antibodies such as rheumatoid factor (RF) potentially interfere with immunoassays. Multiplex-based ELISA arrays allow serum to be tested for multiple analytes. Data derived from these arrays are published with increasing frequency in RA research. We sought to answer whether serum RF in patients with RA might interfere with multiplex ELISA arrays.


Serum was obtained from 9 patients with RA. Aliquots of each sample were then RF-depleted (RD) or mock-depleted (MD) by affinity absorption against human IgG-conjugated or unconjugated sepharose, respectively. All absorptions were incubated vol:vol for >4 hr at 4°C. RF levels were measured by nephelometry. Using Searchlight™ multiplex ELISA, RD and MD samples were tested in duplicate for 16 analytes: A-SAA, E-selectin, IFNg, ICAM1, IL1ra, IL2R, IL6, IL7, MMP1, MMP3, MMP9, RANTES, TIMP1, TIMP2, TNFRI, and TNFRII (PerBio/ThermoFisher, Woburn, MA). RF interference (RFI) was defined as >2-fold false elevation of analyte signal in the non-RF-depleted sample (MD:RD >2).


RF ranged from 13–680 IU in MD samples and was undetectable in all RD samples (<10 IU). For 16 analytes tested in 5 patients with RF <100 IU, the mean MD:RD ratio was 1.47, and RFI occurred in 15/80 (19%) tests. In 4 patients with RF >100 IU, the mean MD:RD ratio was 3.61, and RFI was present in 43/64 (67%) tests. RFI for MMP1 (Table 1) was representative of that observed for A-SAA, ICAM1, IL1ra, IL6, IL7, MMP3, MMP9, TIMP2, TNFR1, and TNFRII. Table 1: MMP1 concentration in mock-depleted (MD) and RF-depleted (RD) samples in 9 patients with RA as measured by Searchlight™ multiplex ELISA. RF levels in MD samples are also shown.

PatientRF (IU)MD ± SEM (ng/ml)RD ± SEM (ng/ml)MD:RD Ratio
1133.5 ± 0.52.3 ± 0.11.5
2199.0 ± 1.34.8 ± 0.31.9
3285.3 ± 0.44.1 ± 0.41.3
4763.1 ± 0.84.0 ± 0.10.8
5766.7 ± 0.64.2 ± 0.11.6
613914.9 ± 2.93.4 ± 0.04.4
71785.3 ± 1.62.9 ± 0.11.8
835312.0 ± 3.11.2 ± 0.010.2
968022.7 ± 7.50.9 ± 0.024.7


In RA patients, serum RF interferes with a portion of analyte measurements in multiplex-based ELISA arrays. In a separate series of experiments, RFI was also present in samples tested by Ray Biotech protein arrays as well as Luminex assays from BioRad, Millipore, and Invitrogen. Additional studies are needed to determine whether RFI is assay-specific or analyte-specific, and whether blocking agents can dampen interference. These findings are important to consider when interpreting data generated by multiplex arrays in patients with RA.

To cite this abstract, please use the following information:
Todd, Derrick J., Knowlton, Nicholas, Karlson, Elizabeth W., Shadick, Nancy A., Weinblatt, Michael E., Schur, Peter H., et al; Rheumatoid Factor Interferes with Multiplex-Based ELISA in Patients with Rheumatoid Arthritis [abstract]. Arthritis Rheum 2009;60 Suppl 10 :746
DOI: 10.1002/art.25826

Abstract Supplement

Meeting Menu