Boross et al. (1) recently described a C57BL/6 FcγRIIB-deficient mouse, usefully confirming FcγRIIB's contribution to autoimmunity initially ascribed to it by observations in deficient mice made on the more problematic B6/129 background (2) and in transgenic mice (3). In introducing the topic, they raised doubts about the genetic evidence linking FCGR2B to systemic lupus erythematosus (SLE) in humans. This is not justified by the data they reference. They quoted a study (4) in which the association between the 232T polymorphism and SLE “… could not be confirmed with an African–American or Caucasian–American cohort.” As the effect of the 232T polymorphism is seen only in homozygotes, these studies are underpowered and inconclusive; for example, for Caucasians, the odds ratio for the FCGR2B effect was 1.42, but the 95% confidence intervals were 0.32–6.91 (4). In fact, subsequent studies and a meta-analysis show that this is one of the strongest associations with SLE in Caucasians (odds ratio, 2.26; 95% confidence interval, 1.49–3.41) (5, 6). They then suggested a recent large “… genome-wide association study in women of European ancestry with SLE identified susceptibility variants in 15 different genes but not in the Fcγr2b gene” (7). This lack of association is because neither rs1050501, nor any other FCGR2B single nucleotide polymorphism, is represented on the Illumina HumanHap 300 array used in the study quoted. The Fc receptor region is complex, and extensive gene homology and copy number variation (6) have resulted in single nucleotide polymorphisms failing quality control and thus being excluded from typing arrays. Finally, they state that “… a strong SLE association in Caucasians of a ‘gain of function’ mutation in FcγRIIB, resulting in increased promoter activity, has been reported” (8). Another study has reported that a haplotype overlapping that described in Ref. 8 is also associated with SLE, but with reduced FcγRIIB expression (9). The association of variants in the promoter of human FCGR2B and SLE thus still needs clarification.
In summary, the genetic evidence associating FCGR2B with SLE in humans is strong. It is important that genetic data are analyzed critically before they are cited; inappropriate conclusions drawn from underpowered studies can perpetuate spurious uncertainty in a field. It is also vital to analyze details of arrays used in genome-wide association studies carefully when interpreting an apparent lack of association; their coverage is not comprehensive and may be particularly patchy in complex regions of immunological interest.
