Response to vaccination against different types of antigens in patients with chronic lymphocytic leukaemia


Dr Marjatta Sinisalo, Department of Internal Medicine, Tampere University Hospital, PO Box 2000, FIN-33521 Tampere, Finland. E-mail:


We investigated responses to vaccination against pneumococcal polysaccharide, Haemophilus influenzae b (Hib) conjugate and tetanus toxoid antigens in 31 patients with chronic lymphocytic leukaemia (CLL) and 25 controls. While in the control group all antibody responses against different antigens were highly significant, in the patient group clear evidence for responsiveness was detected only in the case of Hib polysaccharide antigen. Certain CLL patient subgroups showed low reactivity against tetanus toxoid antigen. In conclusion, plain polysaccharide vaccines seem to be ineffective in patients with CLL. Conjugate vaccines, in turn, are immunogenic and may offer protection against infections caused by encapsulated bacteria in these patients. Further studies concerning an optimal vaccination scheme and clinical efficiency are warranted.

Chronic lymphocytic leukaemia (CLL) is the most common type of leukaemia in the western world. It is a clonal disorder characterized by the accumulation of small, mature-like, but functionally incompetent lymphocytes that are able to escape apoptosis. Infections are the major cause of mortality in patients with CLL and Streptococcus pneumoniae is the most important pathogen. Susceptibility to infections is often associated with immunodeficiency, the nature of which is complex and poorly understood; hypogammaglobulinaemia, IgG subclass deficiencies, functional abnormalities in T cells and different cytokines, an impaired capacity to produce specific immunoglobulins, and neutropenia are all conceivable components of the immunodeficiency seen in patients with CLL (Molica, 1994).

Vaccination is a straightforward option in raising immunity and reducing infections. However, immune responses to pneumococcal polysaccharide and influenza vaccines have been weak in patients with CLL, especially in those with advanced disease stage (Shaw et al, 1960; Jacobson et al, 1988; Gribabis et al, 1994). Conjugation of polysaccharides to protein carriers has rendered the polysaccharide antigen more immunogenic (Goldblatt, 2000). The efficacy of this type of conjugate vaccine has not, to our knowledge, been studied in patients with CLL.

In this study, we investigated the antibody responses to vaccination against different types of antigens, including those of pneumococcal polysaccharide, Haemophilus influenzae b (Hib) conjugate, and tetanus toxoid in patients with CLL.

Patients and methods

Patients and controls The study population consisted of a cohort of 31 consecutive patients with CLL (22 men and nine women), aged 66 years (median, range 48–80 years), referred to the CLL outpatient clinic of Tampere University Hospital. Informed consent was obtained from all patients. The diagnosis and staging of CLL were based on standard clinical, morphological and immunophenotyping criteria, as described elsewhere (Aittoniemi et al, 1999). All patients had the B-cell phenotype. Disease stage according to Binet's classification was A in 14, B in eight and C in nine patients. The median duration of the disease was 3 years (range 0·5–14 years). At the time of vaccination, seven patients were on chemotherapy and 10 received steroids. Severe infections had been recorded in five patients and moderate infections in eight patients within a period of 2 years prior to vaccination (Aittoniemi et al, 1999). Hypogammaglobulinaemia (S-IgG < 6·7 g/l) was detected in 13 patients. Splenectomy had been performed in one case. For control purposes, 25 immunologically apparently healthy age- and sex-matched subjects with no haematological malignancy were included in the study.

Vaccines Pneumococcal polysaccharide (Pnu-Immune 23, Wyeth Pharma, Munster, Germany), Hib conjugate (HibTITER, Wyeth Manufacturing, UK) and tetanus–diphtheria toxoid (Tetanus-d-rokote, National Public Health Institute, Helsinki, Finland) vaccines in 0·5 ml doses were used.

Vaccination and sampling schedules If any of these vaccines had been given to the study subject within the last 5 years, no repeat vaccination was undertaken in order to avoid strong local reactions. All injections were given intramuscularly at separate sites during the same visit. Venous blood samples were taken before and 4 weeks after the vaccination. Post-vaccination blood samples were not obtained from two patients and one patient had been treated with intravenous gammaglobulin before the second blood sample was taken. These three patients were excluded from the study. Of the patients with CLL, 27 were included in the pneumococcal, 28 in the Hib and 18 in the tetanus part of the study. Of the controls, 25 were accepted for the pneumococcal and Hib parts, and 15 for the tetanus part of the study.

Determination of antibodies IgG anti-capsular polysaccharide antibodies to pneumococcal serotypes 1, 3, 6B, 14, 19F and 23F, and IgG antibodies to Hib capsular polysaccharide were determined by enzyme immunoassay (EIA), as described elsewhere (Käyhty et al, 1995; Kurikka et al, 1995). The total pneumococcal anti-capsular polysaccharide antibody level was calculated as a sum of serotype-specific antibody levels. For the determination of IgG antibodies to tetanus toxoid, a slightly modified double antigen EIA was used (Kristiansen et al, 1997). As diphtheria toxoid worked as a carrier in the Hib conjugate preparation, specific antibody levels against diphtheria toxoid were not evaluated.

Statistical analysis Pre-vaccination levels and proportionate changes between groups were compared using the Mann–Whitney U-test or the Kruskall–Wallis test, and the significance in antibody responses was evaluated using the Wilcoxon matched-pairs test. Proportionate changes had been calculated by dividing each patient's post-vaccination antibody level by that of prevaccination.


Pre-vaccination antibody levels and proportionate changes against different antigens in patients with CLL and controls are shown in Table I. The prevaccination antibody level against Hib polysaccharide antigen was lower in patients with CLL than in controls. The only significant response in the CLL patient group was observed in the case of Hib polysaccharide antigen: six (21%) out of 28 patients developed a protective antibody level (≥ 1 μg/ml), and the post-vaccination antibodies were at the protective level in 54% of the patients (Käyhty et al, 1983). In controls, all responses were significant and more intense than those observed in patients with CLL.

Table I.   Medians and quartiles of antibody levels before vaccination, and proportionate changes in patients with CLL and controls.
 Pre-vaccination level median (quartiles)Proportionate change median (quartiles)
Vaccine antigenCLLControlsP-value*CLLControlsP-value*
  • *

    Mann–Whitney U-test.

  • The significance of proportionate increase between pre- and post-vaccination samples, Wilcoxon matched-pairs test, P < 0·05; P < 0·001.

Pneumococcal polysaccharide
10·1 (2·92–24·6)9·41 (6·03–15·0)0·9121·00 (0·80–1·35)4·68 (3·25–7·70)< 0·001
 Serotype 10·77 (0·28–3·42)1·13 (0·42–1·41)0·9561·00 (0·84–1·36)3·81 (2·93–5·52)< 0·001
 Serotype 30·63 (0·27–2·39)1·14 (0·63–1·42)0·5330·93 (0·79–1·19)4·34 (2·49–7·93)< 0·001
 Serotype 6B1·02 (0·33–2·64)1·40 (0·68–3·06)0·5210·99 (0·75–1·88)2·41 (1·30–4·25)< 0·001
 Serotype 141·46 (0·45–2·87)0·53 (0·31–1·27)0·0941·00 (0·83–1·29)6·82 (1·37–21·6)< 0·001
 Serotype 19F2·40 (0·77–8·30)2·31 (1·39–3·81)0·7841·06 (0·73–1·40)3·61 (2·46–10·1)< 0·001
 Serotype 23F1·56 (0·33–3·91)1·70 (0·73–2·91)0·8980·90 (0·75–1·12)3·60 (1·54–7·27)< 0·001
Haemophilus influenzae conjugate
0·48 (0·22–3·06)1·54 (1·02–3·71)0·0221·58 (0·99–5·47)42·1 (11·9–105)< 0·001
Tetanus toxoid (IU/ml)0·12 (0·01–1·03)0·19 (0·01–3·61)0·5471·00 (0·88–2·00)3·33 (1·39–72·1)0·005

Individual antibody responses against different types of antigens in patients with CLL are illustrated in Fig 1. When comparing the CLL patient subgroups with respect to Binet's class, total serum IgG and prevaccination antibody levels, the proportionate antibody response [median (quartiles)] to tetanus toxoid antigen was somewhat higher in those with Binet's class A [A: 2·00 (1·00–6·25), B: 1·00 (1·00–2·25) and C: 0·85 (0·79–0·88); P = 0·031] and in those with normal serum IgG levels [IgG ≥ 6·7 g/l: 2·00 (1·00–2·25) and IgG < 6·7 g/l: 0·88 (0·83–1·00); P = 0·021]. A weak, albeit statistically significant, difference was also detected against pneumococcal polysaccharide antigen in patients with high prevaccination antibody levels (> median) compared with those with low levels [1·17 (1·00–1·36) and 0·86 (0·74–1·27) respectively; P = 0·042].

Figure 1.

 Antibody responses to different types of antigens in patients with CLL. *expressed as a sum of serotype-specific antibody levels.


Antibody responses against different vaccination antigens were weak or lacking in patients with CLL compared with those observed in controls. This finding is in concordance with the results of earlier studies concerning the efficiency of pneumococcal polysaccharide, influenza and toxoid vaccines in patients with CLL (Shaw et al, 1960; Jacobson et al, 1988; Gribabis et al, 1994). The most significant response in CLL patients was detected against Hib polysaccharide antigen, the structure of which had been modified to a thymus-dependent form by conjugating it to a protein carrier (Goldblatt, 2000). Despite the relatively small number of patients in the tetanus toxoid vaccination group, low reactivity to this antigen could also be demonstrated in certain subgroups. Plain polysaccharides as thymus-independent antigens seemed to be ineffective in patients with CLL. Conjugate vaccines have proved to be highly immunogenic in patients with Hodgkin's disease (Chan et al, 1996). However, their efficacy has not previously been studied in patients with CLL. Our finding indicates that the Hib conjugate vaccine is more immunogenic than vaccines with plain polysaccharide antigen.

The response to Hib polysaccharide antigen was clearly lower in CLL patients than in controls. However, the relatively high proportion of patients with a protective antibody level points to the potential efficiency of conjugate vaccines against infections in these patients. The situation may be even better than antibody determinations alone would indicate: in contrast to plain polysaccharide vaccines, the conjugate vaccines are able to induce immunological memory, and protection has also been attributed to this circumstance (Goldblatt, 2000). As antibody responses may be further amplified by different vaccination strategies or adjuvants (Jurlander et al, 1995; Chan et al, 1996), efficacy may be further improved using an optimal vaccination scheme.

In conclusion, vaccine with plain polysaccharide antigen seemed to be ineffective in patients with CLL. Hib conjugate vaccine, for its part, was immunogenic even in patients with advanced disease stage or with hypogammaglobulinaemia. This immunogenicity was at least equal to or even better than that of a conventional toxoid protein vaccine. Judging by these findings, conjugate vaccines offer a better basis in the battle against infections caused by encapsulated organisms in patients with CLL. The next obvious step would be to study the immunogenicity and safety of pneumococcal conjugate vaccines in these patients. Further studies concerning an optimal vaccination scheme and clinical efficiency are warranted.


This study was supported by grants from the Medical Research Fund of Tampere University Hospital and the Finnish Cancer Organization. We thank Raija Kytölinna, Leena Pankko, Hannele Lehtonen and Päivi Paalanen for their assistance.