Table 1 Overview of studies, concerning patient characteristics, used tissues and experimental set-up
Number of patients (% female)TissueArraySpecifics of analysisPower estimationReference
RA: n = 29 (75%). Average disease duration: 12 years. No information about ACR criteria was given.PBMCsU95Av2 (12626 cDNAs)*3 criteria for definition of differentially expressed genes in comparison of patients with RA and controls: (1) p<0.001 by an unpaired Student’s t test; (2) difference in expression of 100 signal units or greater when comparing the means of two groups; (3) greater than 1.4-fold change in mean gene expression between the two groups.264 patients and 264 controls are needed to detect differences (1.4-fold change) with a power of 90%Batliwalla et al7
RA: n = 8 (63%). Mean age 38.25 years. Mean disease duration: 1.7 years. All patients were DMARD naïve at enrolment, stable dose of NSAIDs and/or prednisolone was allowed.B cellsGenome-scale oligonucleotide library containing 21 329 human genes†Genes signal intensities >10 SD above mean of the distribution of random negative controls, and induced or repressed threefold or greater in RA versus controls were catalogued and used for further analysis.12 patients and 12 controls are needed to detect differences (threefold change) with a power of 90%Szodoray et al8
RA: n = 15 (87%). No information on disease duration provided. 14 patients received DMARD medication, 4 received prednisone.RA synovial tissuecDNA microarray (24 000 cDNAs) representing a random set of genes (custom made; not further specified)Transcripts varied in abundance at least twofold from their median level in at least four tissue samples. Comparison of statistical differences in gene expression were generated by SAM. q-values <5 were considered significant44 patients and 44 controls are needed to detect differences (twofold change) with a power of 90%Van der Pouw Kraan et al5
Early RA: n = 11 (82%). Average disease duration 1.1 year. Established RA: n = 8 (100%). Average disease duration of 10.5 years.PBMCsGene Filters release 1 (4329 cDNAs)‡Gene expression data were filtered to include only genes that showed significant variability (3 SD) in the clustering analyses.Number of analysed patients is sufficient to detect differences with 90% powerOlsen et al6
RA: n = 4 (unrelated individuals). Families: 1 parent with RA, 1 unaffected daughter (NS).PBMCsGene Filters GF-211 (4329 cDNAs)§Difference in expression intensity between two groups determined by Chen test:11 statistical analysis on the basis of a specified confidence intervals (99% for all tests).Using the number of samples that were analysed, the minimal fold change in expression that could be detected with a power of 90% is 4.2Maas et al3
Early RA: n = 17, Average disease duration 1.1 years. Established RA: n = 9. Average disease duration of 10.5 years.No information about gender was providedPBMCsGene Filters GF-211 (4133 cDNAs)‡Microarray data were statistically analysed using the SAM algorithm.12 SAM uses the SD of repeated gene expression measurements to assign a score to each gene.Using the number of samples that were analysed, the minimal fold change in expression that could be detected with a power of 90% is 2.0Liu et al4
RA: n = 33 (76%) DAS28 ⩾5.1. Average disease duration: 11–12 years. Patients were resistance to at least one DMARD (MTX included). Infliximab administration: baseline, weeks 2 and 6, and every 8th week thereafter. Change of DAS28 = 1.2 after 3 months was considered as responder (EULAR criteria13).PBMCs (collected at baseline and after 3 months of treatment)cDNA microarray with 12 000 cDNA probes for 10 000 non-redundant genes (custom made)A transcript was considered to be expressed if at least two hybridisations provided a positive signal. Normalised values with abundance differences in two or more comparisons between two samples, using a funnel-shaped confidence interval (p<0.05).Using the number of samples that were analysed, the differences in expression as defined in the study could be detected with a power of 90%Lequerré et al9
RA: n = 10 (80%), median age: 54 years. All patients were treated with MTX, 5 patients received prednisolone. Infliximab administration: baseline, 2 and 6 weeks. Disease activity assessment (DAS28): at baseline and after 12 weeks of treatment. EULAR criteria were used to determine response.13RA synovial tissue (taken before and after a median of 9 weeks of treatment. Biopsies were taken from the site of inflammation)HUM 30k cDNA array containing 30 000 spots (custom made)Moderate t test was used to identify differentially expressed genes. Probability of falsely identifying differentially expressed genes was controlled using a false discovery rate approach with q-value (analogous to a p-value) assigned to each gene. Cut-off was set at q = 0.025: expected proportion of false positives should be less than 0.025% among differentially expressed genesUsing the number of samples that were analysed, the differences in expression as defined in the study could be detected with a power of 90%Lindberg et al10
  • ACR, American College of Rheumatology; DAS28, Disease Activity Score 28; DMARDs, disease-modifying antirheumatic drugs; IDDM, insulin dependent diabetes mellitus; IL-1, interleukin 1; LPS, lipopolysaccharide; MS, multiple sclerosis; MTX, methotrexate; NS, not specified; NSAIDs, non-steroidal anti-inflammatory drugs; PBMC, peripheral blood mononuclear cells; PMA, phorbol mystrate acetate; RA, rheumatoid arthritis; SAM, Significance Analysis of Microarrays.

  • All patients met the ACR criteria unless indicated otherwise.43

  • *Affymetrix Inc., Santa Clara, CA, USA.

  • †Corning UltraGAPS, Big Flats, NY, USA.

  • ‡Research Genetics/Invitrogen, Carlsbad, California, USA.

  • §Research Genetics, Huntsville, Alabama, USA.