Antibody deficiency in bronchiectasis

To the Editors:

We have read with interest the article by van Kessel et al. 1 on impaired antibody response to the pneumococcal polysaccharides of Streptococcus pneumoniae in patients with bronchiectasis of unknown aetiology. Several aspects of this work are controversial and there is no unanimous agreement on some of their conclusions. The first aspect that merits attention is the lack of a control group to establish the normal total antibody response and the isotypes to the vaccine, and, consequently, the diagnostic criteria of impaired antibody response. There are no universal criteria for adequate antibody response to polysaccharides and each laboratory should establish its own 2. This is particularly important because of the therapeutic implications. Treatment of symptomatic patients with a lack of antibody response is performed with periodic administration of intravenous immunoglobulins (Ig) at a high cost, dependent on healthcare centres and potential side-effects 3. For this reason, strict diagnostic criteria must be established.

It is interesting to note that van Kessel et al. 1 used different criteria for the response with total antibodies to the serotypes 3, 4 and 9V, in which the post-vaccination titre should be >20 U·mL⁻¹ with at least a two-fold increase for two of the three serotypes, and to the isotypes IgA and IgG2 of the six serotypes studied, in which response was defined individually for each serotype as post-immunisation concentrations >50 U·mL⁻¹ and an isotype responder had to have a positive IgA and/or IgG2 response to more than four out of six of the serotypes. The reason for these criteria are not clear and should be based on the knowledge of the response of healthy individuals. With their criteria, a patient could potentially be classified as a reponder if he/she presented post-immunisation concentrations of IgA >50 U·mL⁻¹ to five of the six serotypes, but with low concentrations of IgG2 to all serotypes. The clinical meaning of the antibody response with the different isotypes is not equivalent. As an example, the clinical manifestations of patients with IgA deficiency in the form of respiratory infections are due, in great part, to the possible association of an IgG-subclass deficiency, particularly IgG2 deficiency 4, 5.

The usefulness of their response criteria can be validated, at least in part, with the clinical characteristics of the responder and nonresponder patients. There were very few, nonsignificant and probably not clinically meaningful differences between the group of responders and nonresponders. Only the number of resected segments was significantly different in both groups, but it would be interesting to see the number of patients requiring surgery in both groups. This result could be subjected to possible bias in a case in which a single severe patient requiring extensive surgery was included in the group of nonresponders 1.

The next query refers to the meaning of the lack of response to individual serotypes of S. pneumoniae. The immunological interest is clear, but the clinical meaning is controversial. It would be very informative to know the clinical characteristics of the four patients classified as nonresponders by showing post-vaccination total antibody titres below the threshold and comparing them with the remaining 22 patients 1. In contrast, the lack of response with a particular isotype of Ig to one or some individual serotypes can potentially be less harmful than the lack of response with total IgG or IgG2 to all serotypes considered together. This last condition may be a marker of a more profound impairment in antibody response and, if this impairment is also observed against conjugated polysaccharides such as the Haemophilus influenzae type B (Hib) vaccine in a patient with recurrent respiratory infections, it is a criterion for Ig replacement therapy 6. This lack of response to both vaccines is a characteristic of patients with a severe immune impairment 6; however, it is not at all clear that the lack of response restricted to one or more unconjugated polysaccharides deserves treatment with Ig. van Kessel et al. 1 mention that without the study of the response to different pneumococcal serotypes with both isotypes, IgA and/or IgG2, 11 out of 15 so-called nonresponders would have been missed, but there is no convincing clinical or immunological evidence that these individuals had a worse prognosis and were, therefore, candidates for Ig replacement therapy.

By studying a group of healthy adults and a group of patients with humoral immunodeficiencies characterised by defective antibody formation, Rodrigo and coworkers 6, 7 were able to establish a response criteria to both the pneumoccocal and the conjugated Hib vaccines. It was observed that not all healthy individuals responded adequately to either vaccine with all isotypes, but, conversely, no healthy subject presented a lack of response to both vaccines 6. In contrast, none of the patients with humoral immunodeficiencies responded to either vaccine. Therefore, it was shown that evaluation of the antibody response to both the conjugated and nonconjugated vaccine allows the diagnosis of the humoral immunodeficiency characterised by a lack of antibody response to polysaccharides to be established, and permits the selection of patients as candidates to receive Ig therapy.

By using these criteria based on lack of response to both vaccines, we could observe an antibody-production deficiency with normal IgG levels in 11% of a group of 107 patients with bronchiectasis of unknown aetiology. The nonresponders had a significantly higher incidence of otitis media, lower serum IgG2-subclass levels and lower pre-immunisation antibody titres to S. pneumoniae and Hib, and had recurrent pneumonia more frequently (in this latter case, differences were not significant due to the low number of nonresponders) 8, similar to the first case described by Ambrosino et al. 9.

We totally agree that a sizeable fraction of patients with bronchiectasis of unknown cause can now be classified as bronchiectasis associated with polysaccharide antibody-response deficiency 1, 8, 10, 11. However, before this diagnosis can be established and substitution therapy with immunoglobulins can be indicated, comprehensive immunological evaluation is mandatory. This evaluation should include analysis of antibody response to a conjugated and unconjugated vaccine, and response criteria must be defined based on the response of a healthy adult population 6, 7.