Response: Improving classification in CVID

We thank A. Giovannetti and coauthors for shedding some light on the T-cell side of common variable immunodeficiency (CVID). We absolutely agree, as already stated in our review,¹  that the analysis of alterations in the T-cell compartment will be of additional value for the classification of CVID. This has been nicely proven by the same group²  in their 2007 paper.

However, when we initialized the EUROClass trial³  in 2004 “in order to test and possibly improve and unify the current classification schemes” there were 2 major schemes of immunologic phenotyping of CVID, the Freiburg⁴  and the Paris⁵  classifications. Both were based on the analysis of circulating B cells. A B-cell basis for classification was chosen not because the authors felt that CVID is a B-cell intrinsic disorder, but because it is a B-cell targeted disorder because the hypogammaglobulinemia originates from a disturbed differentiation of B cells into immunoglobulin-producing plasma cells. Therefore the classification schemes try to identify steps affected by the altered differentiation in order to subgroup patients into those with severe overall B-cell deficiency, with altered naive, and/or with altered memory compartment. Thus, all patients with severe germinal center defect are characterized by nearly absent class-switched memory B cells. The best examples are inducible costimulator (ICOS) deficiency^6,7  and CD40 ligand deficiency,⁸  interestingly 2 intrinsic T-cell defects with no effect on T-cell homeostasis. At this time none of the known genetic defects causing CVID can be identified by T-cell phenotyping, while B-cell phenotyping is helpful in 3 of the 4 defined genetic defects. Giovannetti et al²  nicely demonstrated a correlation of the decrease in circulating naive CD4 T cells with a worse clinical course as well as Freiburg class Ia. We and others⁹  have observed similar changes; indeed we found a strong correlation of the relative expansion of CD21^low B cells and CD45R0⁺ memory CD4 T cells. The underlying reason for the B- and T-cell dysregulation remains, however, in most cases speculative and may well lie outside the T- and B-cell compartments. For example, viral infections have been postulated to play a role in the subgroup of CVID patients with poor outcome and altered T-cell homeostasis.¹⁰  In the cohort of CVID patients analyzed, persistent herpes virus infection correlated with the expansion of activated CD8 effector cells and an inverted CD4/CD8 ratio. Nevertheless, the work by Giovannetti et al demonstrates that CD45RA⁺ CD4 T-cell lymphopenia is a potential surrogate marker for worse outcome, and further work needs to confirm these data.

To combine B- and T-cell analysis into a common classification scheme, a large cohort of CVID patients needs to be analyzed under standardized conditions. For this purpose, we have initiated an international effort to define the immune phenotype of patients with primary immunodeficiency disorders, which will be launched this year by the European Society for Immunodeficiencies (ESID) clinical working party.