Article Text

FRI0035 Prediction of TNF Inhibitor Response in Rheumatoid Arthritis Patients Using Single Cell Network Profiling of Intracellular Immune Signaling
  1. J. Ptacek1,
  2. R. Hawtin1,
  3. B. Louie1,
  4. E. Evensen1,
  5. J. Cordeiro1,
  6. B. Mittleman1,
  7. M. Atallah1,
  8. A. Cesano1,
  9. G. Cavet1,
  10. C.O. Bingham III2,
  11. S.S. Cofield3,
  12. J.R. Curtis3,
  13. M.I. Danila3,
  14. R. Furie4,
  15. M.C. Genovese5,
  16. M.C. Levesque6,
  17. L.W. Moreland6,
  18. P.A. Nigrovic7,
  19. J.R. O'Dell8,
  20. W.H. Robinson5,
  21. N.A. Shadick7,
  22. E.W. St. Clair9,
  23. C. Striebich10,
  24. G.M. Thiele8,
  25. P.K. Gregersen4,
  26. S.L. Bridges Jr.3
  1. 1Nodality, Inc., South San Francisco
  2. 2Johns Hopkins University, Baltimore, MD
  3. 3University of Alabama Birmingham, Birmingham, AL
  4. 4The Feinstein Institute for Medical Research and North Shore-LIJ Health System, Manhasset, NY
  5. 5Stanford University, Stanford, CA
  6. 6University of Pittsburgh, Pittsburgh, PA
  7. 7Brigham and Women's Hospital/Harvard University, Boston, MA
  8. 8University of Nebraska, Lincoln, NE
  9. 9Duke University, Durham, NC
  10. 10University of Colorado Denver, Denver, CO, United States


Background Biomarkers predictive of drug efficacy are lacking in rheumatoid arthritis (RA) and would be useful in clinical practice and clinical trials. Single cell network profiling (SCNP) is a multiparametric flow cytometry-based assay that measures induced changes (phosphorylation) in intracellular signaling proteins, providing a functional measure of pathway activity and immune networking in multiple cell subsets without physical separation.

Objectives Induced signaling was measured in specific subsets of monocytes, B and T cells from RA patients (pts) initiating new treatment, and analyzed to build models to predict treatment response.

Methods PBMCs from RA pts (n=87) starting TNF inhibitors (TNFi) were examined by SCNP of 42 nodes (combinations of modulator and intracellular readout) within 21 immune cell subsets. A subset of ∼200 RA pts from the Treatment Efficacy and Toxicity in Rheumatoid Arthritis Database and Repository (TETRAD) were studied. Blood samples were collected before treatment with TNFi (adalimumab, etanercept, infliximab, golimumab). Clinical data included DAS28 and EULAR response criteria at baseline, 3, 6, and 12 months. For the 53 evaluable patients, ordinal logistic regression and multivariate modeling were performed to identify signaling profiles associated with response to TNFi.

Results Immune cell subsets from RA pts before TNFi treatment exhibited heterogeneity in induced intracellular signaling. Of note, T cell receptor (TCR) and IFNα modulation produced cell subset-specific signaling profiles that were associated with response at 3 months. Specifically, in CD4CD45RA+ (naive and effector) T cells, phosphorylation of CD3ζ after stimulation through the TCR (TCR→p-CD3ζ) was weakest in pts that had a good EULAR response to TNFi (p=0.04). In contrast, phosphorylation of STAT3 after stimulation with IL-6 (IL-6→p-STAT3) in naive CD4+ T cells was weakest in autoantibody-positive pts with no response (p=0.01). Signaling nodes modulated by IFNα, TNFα, and IL-6 were combined to construct models to predict response and compared to models generated with standard clinical variables, including age, sex, and DAS28. Models utilizing cell signaling capacity of samples had greater performance with an internal cross-validated AUROC of 0.75 compared to 0.45 for clinical models.

Figure 1.

Range of performance observed over 500 models in bootstrapping.

Conclusions This is the first evidence that measurement of peripheral blood immune cell function can: 1) identify patients likely to respond to TNFi, and 2) reveal the biology associated with TNFi response or lack thereof. SCNP has revealed predictive biomarkers that, once replicated in future studies, may enable patient stratification in clinical practice and clinical trials.

Disclosure of Interest J. Ptacek Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., R. Hawtin Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., B. Louie Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., E. Evensen Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., J. Cordeiro Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., B. Mittleman Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., M. Atallah Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., A. Cesano Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., G. Cavet Shareholder of: Nodality, Inc., Employee of: Nodality, Inc., C. Bingham III: None declared, S. Cofield: None declared, J. Curtis: None declared, M. Danila: None declared, R. Furie: None declared, M. Genovese: None declared, M. Levesque: None declared, L. Moreland: None declared, P. Nigrovic: None declared, J. O'Dell: None declared, W. Robinson: None declared, N. Shadick: None declared, E. W. St. Clair: None declared, C. Striebich: None declared, G. Thiele: None declared, P. Gregersen: None declared, S. L. Bridges, Jr.: None declared

DOI 10.1136/annrheumdis-2014-eular.2057

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.