INTRODUCTION

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

Primary immunodeficiencies are one of the diagnoses to be considered in patients with increased susceptibility to respiratory infection. Of these immunodeficiencies, those with predominant antibody defects such as common variable immunodeficiency (1, 2), selective immunoglobulin G (IgG) subclass deficiency (1, 3, 4), and selective antibody deficiency with normal immunoglobulins (1, 5), may present initially in adult life. An early diagnosis is highly important because in most cases immunoglobulin therapy is indicated, thereby improving long-term prognosis (1). Though the measurement of serum immunoglobulins and IgG subclasses greatly helps in indicating the possible presence of these entities, it is insufficient to establish a definitive diagnosis as low levels have been described in healthy adults (6). It is therefore necessary to demonstrate a defect in specific antibody response, particularly to polysaccharide antigens such as the pneumococcal or Haemophilus influenzae type b (Hib) vaccine, to establish a definitive diagnosis and indicate immunoglobulin therapy (1), which is costly and not without risk (7, 8). However, the present lack of reference values of specific antibody response against these antigens in adults, together with a uniform normal response criterion, renders its interpretation difficult (9).

We previously standardized an enzyme-linked immunosorbent assay (ELISA) to quantify IgG and its subclasses against the 23-valent pneumococcal vaccine and established a criterion for a normal response to the vaccine in healthy adults (12, 13). However, the results showed that not every healthy adult responded to the vaccine (12). The aim of this study was to evaluate the Hib-conjugated vaccine in addition to the pneumococcal vaccine in the diagnosis of humoral immunodeficiency in adults.

	METHOD

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

Study Population

This was a cross-sectional study of the specific antibody response to a conjugated Hib vaccine in a group of healthy adults and a group of patients with humoral immunodeficiencies.

Fifty-nine healthy adult volunteers (30 women; age range: 20-59 yr; mean age: 32 yr) were studied. None had a history of recurrent or severe infections, acute or chronic pulmonary disease, primary or secondary immunodeficiency, autoimmune systemic disorder, or any disease or intake of drugs that might affect antibody production. They had not suffered any infection during the month preceding the vaccination. Forced spirometry, white cell count, creatinine, transaminases, and serum total immunoglobulin levels were within normal range.

Twenty-two patients with humoral immunodeficiencies characterized by defective antibody formation (17 men; age range: 17-64 yr; mean age: 32 yr) were also studied. Eighteen suffered from common variable immunodeficiency, two immunodeficiency with thymoma, and two X-linked agammaglobulinemia. The diagnosis was made according to the diagnostic criteria of the World Health Organization (WHO) immunodeficiency group (1). Twelve of these patients were receiving intravenous immunoglobulin therapy.

This study was approved by the Ethics Committee of the Vall d'Hebron Hospitals.

Immunization

All healthy adults and patients with humoral immunodeficiencies were immunized intramuscularly in the buttock with a single injection of Hib-conjugated vaccine composed of 15 µg of Hib polyribosylribitol phosphate (Hib-PRP) linked to an outer membrane protein complex of Neisseria meningitidis (PedvaxHIB; Merck Sharp & Dohme, West Point, PA). Twenty of the healthy adults and nine of the patients were also immunized intramuscularly in the deltoids with 0.5 ml of PNU-Immune 23 polyvalent pneumococcal vaccine (Lederle Laboratories Division, Pearl River, NY).

Blood samples were drawn immediately before and 21 d after vaccination. These samples were centrifuged at 3,000 rpm for 15 min and the sera obtained were stored in aliquots frozen at 20° C until testing. In patients receiving intravenous immunoglobulins, the treatment was suspended 1 mo prior to vaccination and resumed after the second blood extraction.

Immunoglobulin Quantification

Serum total IgG, IgA, and IgM levels were measured by kinetic nephelometry. IgG subclasses were measured by ELISA (3, 14).

Specific IgG and IgG Subclasses to H. influenzae Type B

Specific IgG, IgG₁, and IgG₂ to Hib were determined by ELISA based on a previous method described by Rodrigo and coworkers (3, 12) using an antigen of Hib oligosaccharides conjugated to human serum albumin (HbO-HA) kindly donated by Dr. P. Anderson (University of Rochester Medical Center, Rochester, NY). Results were expressed as µg/ml using a reference serum of 70, 30.9, and 16.1 µg/ml of specific IgG, IgG₁, and IgG₂ against Hib, respectively, from the Center of Biological Evaluation and Review (CBER), U.S. Food and Drug Administration (FDA). Within-run coefficients of variation (CV) were 9.1%, 12.2%, and 9.5%, and day-to-day CV were 12%, 14%, and 11.3% for specific IgG, IgG₁, and IgG₂, respectively. The minimal amount of antibody detectable by this assay was 0.027, 0.05, and 0.1 µg/ml for specific IgG, IgG₁, and IgG₂, respectively. To study the linearity of sera compared with reference serum, we performed titrations of five sera with anti-Hib IgG antibody levels between 5 and 132 µg/ml, IgG₁ between 1.6 and 18 µg/ml, and IgG₂ between 4.6 and 65.4 µg/ml. The curves obtained from the sera and reference serum were plotted for linear regression analysis.

Specific IgG to Pneumococcal Vaccine

Specific IgG to pneumococcal vaccine was measured by ELISA using the pneumococcal vaccine as antigen, as previously described (12).

Specificity of the Antibodies to Hib Vaccine

Inhibition experiments. Increasing concentrations of soluble PRP, HbO-HA, and human albumin, ranging from 0 to 15 µg/ml, were incubated with 1 ml of a 1:50 dilution of a pool of sera with specific IgG concentration of 86.70 µg/ml for 2 h at 37° C. After centrifugation, the amount of specific IgG to Hib in the supernatant was measured by the same ELISA.

Cross-reactivity between anti-Hib and anti-S. pneumoniae antibodies. Two pools of sera with specific IgG concentrations to Hib of 7.2 and 86.7 µg/ml were incubated with increasing concentrations of pneumococcal vaccine and PRP ranging from 0 to 15 µg/ml, and the same ELISA was carried out.

Purified IgG antibodies. IgG-purified antibody preparations were obtained by affinity chromatography with protein A-Sepharose CL-4B (Pharmacia Diagnostics, Sweden). A sera pool (IgA 212 mg/dl, IgM 136 mg/dl, IgG 932 mg/dl, IgG₁ 588 mg/dl, IgG₂ 492 mg/dl, IgG₃ 98 mg/dl, IgG₄ 72 mg/dl, and specific IgG to Hib 278 µg/ml) was loaded into the column and the fractions were collected from the maximum peak (higher than 0.200 A₂₈₀) of the two elutes obtained. IgG IgA, IgM, IgG subclasses and specific IgG to Hib were determined in each elute.

Statistical Analysis

Antibody titers, as well as changes in titers, were transformed to their natural logarithms to meet the assumption of normality and were expressed as the mean and 95% confidence interval (CI). Whenever the difference between post- and preimmunization titers was negative or the ratio was < 1, the lowest value of the variable was assigned. An arbitrary value, corresponding to the lower limit of the two-tailed 90% probability interval of postimmunization specific IgG, IgG₁, and IgG₂ of the healthy adults, was defined as the minimum significant increase for an adequate response. All subjects showing an increase in specific antibody titers equal to or greater than this value were considered to be responders. In the case of pneumococcal vaccine this value was 395 U/ml, as described elsewhere (12).

Statistical significance of differences among mean antibody values between the study populations was determined by Student's unpaired t test. The level of significance was p < 0.05.

	RESULTS

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

Linearity of the ELISA Method for Anti-Hib Antibodies

No significant differences in the slopes of the lines were obtained by analysis of covariance (ANCOVA, F_3.23 = 0.96 for IgG, F_5.35 = 1.83 for IgG₁, and F_4.29 = 1.07 for IgG₂). We may therefore assume that the behavior of the reference and test sera was similar.

Specificity of the Antibodies to the Hib Vaccine

Effect of absorption with the PRP, HbO-HA, and human albumin. Results of the specificity study of antibodies with soluble PRP or HbO-HA added to the sera showed a decrease in absorbance values of the ELISA to background levels, indicating effective competition with the coated HbO-HA antigen for antibody binding. Soluble HA added to the serum did not decrease absorbance.

Cross-reactivity experiment. Soluble pneumococcal polysaccharide added to the sera did not decrease the absorbance of specific IgG to Hib before or after incubation with pneumococcal vaccine.

Purified IgG antibodies. Elution patterns obtained from affinity chromatography were as follows: in the first peak the results were IgG 33 mg/dl, IgA 105 mg/dl, IgM 30 mg/dl, IgG₁ < 5 mg/dl, IgG₂ < 5 mg/dl, IgG₃ 20 mg/dl, IgG₄ < 5 mg/dl, and specific IgG to Hib < 0.027 µg/ml. The results in the second peak were IgG 630 mg/dl, IgA 7 mg/dl, IgM 20 mg/dl, IgG₁ 360 mg/dl, IgG₂ 250 mg/dl, IgG₃ < 5 mg/dl, IgG₄ 35 mg/dl, and specific IgG to Hib 193.8 mg/dl.

Antibody Response to the Hib Vaccine

Concentrations of specific total IgG, IgG₁, and IgG₂ against Hib in the healthy adults pre- and postimmunization are shown in Table 1. To define the response in this population, we calculated the two-tailed 90% probability interval (PI) of the log-transformed titers of the total IgG, IgG₁, and IgG₂ in postimmunization sera. The lower PI limit was 2.28 µg/ml for total IgG, 0.74 µg/ml for IgG₁, and 0.34 µg/ml for IgG₂. We considered as a responder a person who suffered an increase in antibody titers equal to or greater than these values.

Fifty of the 59 (85%) healthy adults met the response criteria for specific total IgG. When the specific response of IgG₁ and IgG₂ was studied, three of the nonresponders with specific total IgG responded with specific IgG₁ and IgG₂, three with specific IgG₁, and one with specific IgG₂, but two (3.4%) presented no response with any of the subclasses tested (Figure 1).

View larger version (9K): [in this window] [in a new window]   Figure 1.   Distribution of the specific total IgG, IgG1, IgG2, antibody responses against Hib in the healthy adult population. The number and percentage of individuals meeting response criteria for IgG and its subclasses are expressed.

None of the 22 patients with humoral immunodeficiencies met the response criteria for specific total IgG. Specific IgG concentrations against Hib-PRP in these patients were significantly lower than in the healthy adults, both in pre- and postimmunization titers as well as in the increases (p < 0.0001, p < 0.0001, and p < 0.0001, respectively) (Table 1).

Antibody Response to the Pneumococcal Vaccine

Five of the 20 healthy adults (25%) who received the pneumococcal vaccine failed to respond with specific total IgG. None of the nine patients with humoral immunodeficiencies vaccinated with the pneumococcal vaccine responded with specific total IgG.

Antibody Response to Both Vaccines

All 20 healthy adults (100%) who received the two vaccines responded to at least one with specific IgG. None of the nine patients with humoral immunodeficiencies who received the two vaccines responded to either.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHOD RESULTS DISCUSSION REFERENCES

Uniform guidelines on the interpretation of normal antibody response remain to be adequately defined, and over- or underdiagnosis of defects in response may occur. The fold-rise in antibody titers is often used as a measure of the response (15, 16), but may be misleading, as it depends on the concentration of antibodies before immunization. This implies that a higher absolute increase is required in patients with higher preimmunization levels than those with lower levels. In this study we used an arbitrary value as the minimum significant increase (MSI) to consider response to the vaccine as normal. MSI corresponds to the minimum concentration of postimmunization antibodies reached by 95% of the healthy adults. This value is not influenced by preimmunization antibody titers; it may be applied in the study of response to different vaccines (12), and permits direct comparisons between different populations.

The results of this study would indicate that not all healthy adults respond to one vaccine, even with the study of specific IgG subclasses. The possibility of none responding to two vaccines, however, is considerably reduced. On the other hand, none of the patients with humoral immunodeficiencies studied responded to either vaccine. In our opinion, therefore, the use of only one vaccine as a diagnostic test of a defective antibody formation may be insufficient and would require the study of another antigen. When primary immunodeficiency disease is suspected in patients with recurrent pyogenic sinopulmonary infections, the sequential study of serum immunoglobulins, IgG subclasses, and specific IgG antibody responses to at least two polysaccharide antigens would, in the proposed strategy, permit more selective diagnostic criteria to be established.

Antibody response to Hib vaccine among healthy adults varies widely. Comparison of results in different studies is difficult owing to differences in response criteria, population studied, type of vaccine used, and technical aspects. Most studies have been conducted in children (17), who may be affected by a physiological delay in antibody response maturation or may have lower previous exposure to the antigen. Pure polysaccharide and conjugated vaccines, both with different concentrations of the polysaccharide, have been used in studies elsewhere (15, 16, 18). The polysaccharide of the vaccine was conjugated to a carrier protein, converting it into a thymus-dependent antigen with the aim of stimulating the antibody response in children under 2. The conjugated vaccine elicits a greater antibody response than the nonconjugated form in adults (18, 21), but whether it still maintains many of the characteristics of a polysaccharide response remains unclear (1). In this respect, Mäkelä and coworkers (18) showed that anti-Hib antibodies induced by the conjugated vaccine had, in adults, the same IgG subclass composition as the ones induced by the polysaccharide. Furthermore, although the IgG response to polysaccharide antigens in adults has been described mainly as being mediated by IgG₂ (22, 23), later reports indicate that both IgG₁ and IgG₂ contribute to the antibody response (12, 18, 22, 24, 25), at times even with IgG₁ being the predominant isotype (12, 22, 26). This would suggest that the immunoregulatory pathways that define isotype restriction patterns are complex and no single pattern can be unequivocally defined (27).

The production of antibodies to unconjugated and conjugated vaccines in patients with common variable immunodeficiency is undetectable, as we show in our study. In contrast, some patients diagnosed with more selective immunodeficiencies, such as the selective deficiency of IgG₂, particularly in the young, may respond to the conjugated vaccine, but not to the polysaccharide (28, 29). The possibility of an immunoregulation defect has been suggested here (28). Therefore, the unconjugated vaccine might on occasions be insufficient to establish the diagnosis of antibody deficiency while the conjugated vaccine could complete the study and identify which are the true nonresponders who in the end will require immunoglobulin therapy.

In this study, we described an ELISA that uses Hib oligosaccharides covalently linked to HbO-HA as an antigen for quantification of specific IgG and its subclasses to Hib. Low background, better sensitivity than a radiolabeled antigen binding assay (30), and high specificity have been obtained with this method. In addition, no cross-reactions existed with other polysaccharide antigens such as Streptococcus pneumoniae. The calibrated human reference serum approved by the FDA and used in this study allows the results to be expressed in mass per volume units and is the standard used in many of the published studies. Different studies carried out to evaluate the total contents of IgG anti-Hib in the reference serum (CBER, FDA 1983) using different assays and calibration systems showed differences in their concentration, ranging from values of 21.9 to 60.9 µg/ml (10). Commercially available kits that permit measurement of specific antibodies to Hib are now available; nevertheless, each laboratory should establish its own reference values, since there are many technical and genetic variables that may affect the results.

We conclude that this ELISA method and Hib-conjugated vaccine may be useful for studying specific antibody response. Evaluation of the antibody response to both the Hib and pneumococcal vaccines may facilitate the diagnosis of humoral immunodeficiency and selection of patients to receive immunoglobulin therapy.