Overcoming the Challenge of Rheumatoid Factor Interference in Immunogenicity Assays for Human Monoclonal Antibody TherapeuticsLaura Kelly1, Amruta Kale1, Cheikh Kane2, Joe Bower1 and Lin Yang1; 1Immunochemistry Services, Covance Laboratories Inc., Chantilly, VA; 2Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, CT

Presented at WRIB 2017

IntroductionBridging anti-therapeutic antibody (ATA) assays are vulnerable to matrix factors that can bridge the conjugated therapeutic and generate false positive signals. Rheumatoid factor (RF) and autoimmune antibodies associated with rheumatoid arthritis (RA) pose this problem. The object of this study was to develop a sensitive, drug tolerant and robust assay to detect ATA against therapeutic protein X (TPX) in RA patients. The TPX is a humanized monoclonal antibody for the treatment of RA.

Method▶ A bridging electrochemiluminescence (ECL) ATA assay

was validated in human plasma using a three tier approach: screen, confirm and titer.

▶ Acidified samples are added to a mixture which contains Tris base, biotin-labeled TPX and Sulfo-TAG-labeled TPX.

▶ This mixture was then allowed to complex with ADA overnight. The labeled drug-ADA complexes are transferred to a streptavidin plate where they are captured.

▶ The plate is washed to remove any non-specific bound complexes, and read buffer is added. The plate is read on the MSD Sector™ Imager 6000.

▶ Results from initial validation demonstrated normal plasma assay sensitivity of 18.1 ng/mL, a multiplicative normalization factor (NF) of 1.5, and individual signals ranging from 37 to 81 RLU.

▶ RA plasma however had a sensitivity of 333.3 ng/mL, a NF of 8.5 and individual signals ranging from 69 to 3050 RLU.

▶ This suggested strong interaction between RF and RA within the Fc region of the labeled TPX generating false positive bridges.

▶ Multiple reagents were evaluated to block the generation of false positive bridges in RA individuals.

▶ These reagents were tested with RH positive RA individuals with low, mid and high ATA signals obtained from BioreclamationIVT, along with normal pool (NC) and spiked positive controls (PCs).

Results

Figure 1. Design of immunogenicity assay.

Figure 2. Proposed mechanism of RF interference.

Conclusions▶ The ability to overcome interference of RF is vitally important for accurate

determination of true ATA response. ▶ Commercially available whole molecule IgG demonstrated significant

reduction of false positive bridges and allowed for a simple and easily accessible reagent to overcome this challenge.

ReferencesAraujo, J., Zocher,M., Wallace, K., Peng, K., Fischer, S., Increased rheumatoid factor interference observed during immunogenicity assessment of an Fc-engineered therapeutic antibody, Journal of Pharmaceutical and Biomedical Analysis, 55(2011) 1041-1049.Edelman, G. Kunkel, H., Franklin, E., Interaction of the rheumatoid factor with antigen-antibody complexes and aggregated gamma globulin, Published July 21, 1958. Rockefeller Instutute for Medical Research. Jem.rupress.org.

-100

400

900

1400

1900

2400

2900

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30

Comparison of Signal Responses of Plasma Samples from Normal Subjects and RA Patients

NHS Individual Sample RA Individual Sample

Resp

onse

(RLU

)

Figure 3. Comparison of signal responses of plasma samples from normal subjects and RA patients. Unusually high signals in the RA population with this ATA assay suggested a strong interaction between RF and the Fc region of the labeled TPX. RFs recognize epitopes on the Fc region of IgG molecules forming bridges. RFs are usually associated with the IgM class, but may also include IgG and IgA.

0

1000

2000

3000

4000

5000

6000

0 100 200 300 400 500 600

Resp

onse

(RLU

)

RF (IU/mL)

Correlation between RF Value and ADA Assay Response

Figure 4. Correlation between RF values and ADA assay response. RF values were obtained from Bioreclamation for each individual. A direct correlation of RF values and ATA signal response was not observed. As depicted in Figure 4, individuals with high RF values demonstrate both high and low signal in the ATA.

0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0

5000 2500 1250 625 313 156 78 39 20 10 0

S/N

Rati

o

Assay Sensitivity in Human Normal Plasma with and without Human IgG Addition to Mastermix

No IgG Addition

1000 μg/mL IgG Addition

S/N=2

Antibody Concentration (ng/mL)

Figure 10. Assay sensitivity with and without human IgG addition to Mastermix. The addition of whole molecule Human IgG did not affect the assay sensitivity or the dynamic signal range of the assay.

0.0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

500 ng/mL PC + 100μg/mL Drug

500 ng/mL PC + 0μg/mL Drug

250 ng/mL PC + 100μg/mL Drug

250 ng/mL PC + 0μg/mL Drug

Drug Tolerance in Human Normal Plasma with and without Human IgG Addition to Mastermix

No IgG Addition

1000 μg/mL Human IgG Addition

S/N

Ratio

Figure 13. Drug Tolerance with and without human IgG addition to Mastermix. No impact on drug tolerance was observed due to the addition of IgG. Both 500 ng/mL and 250 ng/mL screen positive with addition of 100 µg/mL drug. Estimated multiplicative cut point is 1.153.

0

200

400

600

800

1000

1200

1400

1600

1800

2000

Elimination of RF Interference by Anti-Human IgM

0 μg/mL Anti-Human IgM 50μg/mL Anti-Human IgM100μg/mL Anti-Human IgM 250μg/mL Anti-Human IgM2xNC

Resp

onse

(RLU

)

Figure 5. Elimination of RF interference by anti-human IgM. Anti-human IgM antibody was tested at 0, 50, 100 and 250 µg/mL. This reagent reduced false positive signal in some samples, however several samples showed an increased response (Sample 3 and 30). Additionally, sample 23 increased its signal with higher concentrations of anti-IgM.

0200400600800

100012001400160018002000

Comparison of Anti-Human IgM and Anti-Human IgM (Fab)

0 μg/mL Anti-Human IgM (Fab)250 μg/mL Anti-Human IgM250 μg/mL Anti-Human IgM (Fab)2xNC

Resp

onse

(RLU

)

Figure 6. Comparison of anti-human IgM and anti-human IgM (Fab). Anti-human-IgM Fab reduced signal in three of the RA samples tested, however most samples showed either no change or an increase in signal with addition of Fab. Sample 9 saw the most improvement with Fab addition. Overall, whole antibody to IgM performed better than Anti-IgM Fab.

0

200

400

600

800

1000

1200

1400

1600

Elimination of RF Interference by Human IgG Fc and RFRR

0 μg/mL Human IgG Fc100 μg/mL Human IgG Fc250 μg/mL Human IgG FcRFRR2xNC

Resp

onse

(RLU

)

Figure 7. Elimination of RF interference by human IgG Fc and RFRR. Rheumatoid Factor Removal Reagent (RFRR) per manufacturer, should remove all IgGs (up to 15 mg/mL) and IgM specific to RF. Addition of this reagent eliminated all signal response from PCs, however 4 out of 13 RA samples still produced high signals. Human Fc was tested at 100 and 250 µg/mL and lowered false signals. Higher concentrations could not be tested due to the low starting concentration of human Fc material.

0

500

1000

1500

2000

2500

3000

3500Elimination of RF Interference by Human IgG

0 μg/mL human IgG

500 μg/mL Human IgG

1000 μg/mL Human IgG

2xNC

Resp

onse

(RLU

)

Figure 8. Elimination of RF interference by human IgG. IgG aggregates are cited in the literature to immobilize RF factors. In place of aggregates, whole molecule human IgG (and human IgG Fc) were tested. Commercially available whole molecule IgG was tested over a range from 50 to 2000 µg/mL. The above figure illustrates results from 500 and 1000 µg/mL. Whole molecule IgG significantly dropped the signal for RA individuals.

0

200

400

600

800

1000

1200

1400

1600

1800

Elimination of RF Interference by Human IgG and/or Anti-Human IgM

No addition of Human IgG or Anti-human IgM

1000 ug/mL Human IgG

100 ug/mL Anti-human IgM

1000 ug/mL Human IgG and 100 ug/mL Anti-human IgM

2xNC

Resp

onse

(RLU

)

Figure 9. Elimination of RF interference by human IgG and/or anti-human IgM. The combination of both whole molecule IgG and anti-human IgM (whole antibody) demonstrated the best combination to reduce false positive signal. The addition of anti-IgM antibody could remove response from anti-IgM antibodies to the TPX, therefore the IgG and anti-IgM combination was not selected for the assay.

Figure 11. Selectivity in RA plasma samples. There is a significant improvement of assay sensitivity and method selectivity in RA samples with the addition of IgG.

Figure 12. Selectivity in normal plasma samples. No impact on assay selectivity results in normal samples was observed due to the addition of IgG.