Article Text

Download PDFPDF

24 A molecular signature based on ifn gene signature and serology defines two populations of patients with different baseline disease activity
  1. M Petri1,
  2. K Kalunian2,
  3. M Urowitz3,
  4. R Furie4,
  5. MA Morgan-Cox5,
  6. M Silk6,
  7. E Dow7,
  8. R Higgs5,
  9. S Watts5,
  10. D Isenberg8 and
  11. M Linnik9
  1. 1Johns Hopkins University School of Medicine, Rheumatology, Baltimore, USA
  2. 2University of California- San Diego School of Medicine, Rheumatology, San Diego, USA
  3. 3University of Toronto Faculty of Medicine, Rheumatology, Toronto, Canada
  4. 4Hofstra North Shore-LIJ School of Medicine, Rheumatology, Great Neck, USA
  5. 5Eli Lilly, Statistics, Indianapolis, USA
  6. 6Eli Lilly, Immunology, Indianapolis, USA
  7. 7Eli Lilly, Bioinformatics, Indianapolis, USA
  8. 8University College London, Rheumatology, London, UK
  9. 9Lilly Biotechnology Centre, Immunology, San Diego, USA


Background and Aims Randomised controlled trials in SLE have shown that response to treatment is influenced by baseline disease activity. The current investigation used objective molecular and biochemical baseline parameters to characterise SLE patients in two large multinational trials (n=2262 patients).

Methods Patients were categorised with four dichotomous baseline parameters. SLE(+) was defined by any of the following: IFN signature (high), anti-dsDNA (+), C3 (low) and/or C4 (low). SLE(-) required all of the following: IFN signature (normal), anti-dsDNA (-), C3 (normal) and C4 (normal).

Results Baseline RNA transcript data were available for 1747 of 2262 patients. When IFN status was combined with the serology criteria, 1500 (86%) were classified as SLE(+) and 247 (14%) were classified as SLE(-). At baseline, SLE(-) patients had significantly lower mean SLEDAI scores (8.3) compared to SLE(+) (10.7). Baseline SLEDAI <10 was observed in 72% of SLE(-) compared to 38% of SLE(+). The proportion on corticosteroids at baseline was 49% in SLE(-) compared to 78% in SLE(+); the proportion on immunosuppressants at baseline was 31% in SLE(-) compared to 44% in SLE(+). In the US, 22% were SLE(-) compared to 10% for Latin America, 7% for Europe, and 5% for ROW.

Conclusions A subset of clinical trial patients was identified using biochemical and molecular markers with high sensitivity for SLE. Seronegative SLE patients with normal IFN gene signature had lower disease activity and were taking less background medication at baseline, two factors which have been negatively associated with response to treatment in some previous trials.

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.