Background Biological therapies have been a great advance in Rheumatoid Arthritis (RA) treatment. However, up to a third of the RA patients do not respond to a given biologic and a method to predict response is not yet available. Therefore, the only approach to select effective biological drugs is by trial and error. In addition, assessment of response requires months of treatment making this approach very inefficient and detrimental for patients and healthcare systems. Thus, biomarkers for prediction of anti-TNF response at an early stage are sorely needed. A potential source for these biomarkers is the analysis of serum proteins as shown by the report of six plasma proteins (only two of them were identified) detected with the SELDI-TOF technology1, which are potential biomarkers of response to anti-TNF therapy in RA patients.
Objectives To search for biomarkers of infliximab response by means of a very precise and sensitive quantitative proteomics technology applied to serum samples of patients with RA taken before the beginning of treatment.
Methods Donors with RA according to ACR criteria that never had received biologics were enrolled in the study. Serum was collected and stored at -80°C before the beginning of treatment. Donors were treated with infliximab and clinical response, assessed by the EULAR criteria, was determined six months later. Eight samples showing extreme responses were selected: 4 with good clinical response and 4 with no response. Each of the serum samples was depleted of the six most abundant serum proteins and, after tryptic digestion, they were differentially labeled with the iTRAQ (Isobaric Tags for Relative and Absolute Quantification) isotopic labels. The peptide mixture was fractionated by means of 2D-LC (strong cation exchange chromatography and reversed phase chromatography), on-line coupled with an Orbitrap Velos mass spectrometer. ProteinPilot 4.0 software was used for the identification and relative quantification of proteins.
Results Obtained mass spectra were compared with the human protein database, enabling the identification of 315 different proteins that were present in all the samples. Most of these proteins were of medium and low abundance in serum. A total of 289 proteins could be quantified in all samples using the iTRAQ differential labeling and 17 of them showed concentration differences between responders and no-responders (p<0.05). They were specially enriched in binding proteins (43%), followed by enzimatic and transport proteins. None of the 17 differential proteins has previously been described as a biomarker for anti-TNF response prediction; therefore they represent new valuable candidates.
Conclusions A new panel of serum protein biomarkers for the prediction of infliximab response was identified by means of a quantitative proteomics approach using isotopic labeling. These proteins should be validated on a larger population in order to assess the predictive value of the panel. Once validated, this panel could be useful for the classification of RA patients before the beginning of treatment, so that the most likely to show lack of response could be treated with an alternative drug without delay.
Trocmé C, et al. Ann Rheum Dis. 2009;68:1328-33
Disclosure of Interest None Declared