HIV/AIDS
MAJOR ARTICLE
Response of Human Immunodeficiency Virus–Infected Patients Receiving Highly Active Antiretroviral Therapy to Vaccination with 23Valent Pneumococcal Polysaccharide Vaccine Maria C. Rodriguez-Barradas,1,2 Irene Alexandraki,1,2,a Tabinda Nazir,3,4,a Michael Foltzer,5 Daniel M. Musher,1,2 Sheldon Brown,3,4 and John Thornby1,2 1
Baylor College of Medicine and 2Veterans Affairs Medical Center, Houston Texas; and 3Mt. Sinai School of Medicine, New York, 4Veterans Affairs Medical Center, Bronx, and 5Bassett HealthCare, Cooperstown, New York
Whether highly active antiretroviral therapy (HAART) impacts responses to 23-valent pneumococcal polysaccharide vaccine (PV) is not known. Immunoglobulin G (IgG) levels for 6 capsular polysaccharides in human immunodeficiency virus (HIV)–infected patients who had received ⭓6 months of HAART were measured either after their first dose of PV (n p 46 ) or after revaccination (n p 41 ); control subjects had never received HAART and had received the first dose of PV (n p 38 ). There were no significant differences in pre- or postvaccination IgG levels among these groups but for 1 capsular polysaccharide. The 3 groups had significant postvaccination increases in IgG levels to all capsular polysaccharides. The control group had a greater number of 2-fold responses than did the combined HAART groups (P ! .05 ). Patients with a CD4 cell count of ⭓200 cells/mm3 had a greater number of 2-fold responses than did those with a CD4 cell count of !200 cells/mm3 (P ! .05). For revaccinated patients, postvaccination IgG levels were correlated with the CD4 cell count at the initial vaccination. The immunogenicity of PV among patients receiving long-term HAART is modest. It seems best to immunize HIV-infected patients early in the course of disease. HIV-infected subjects are highly susceptible to pneumococcal infection [1, 2], and the United States Public Health Service recommends that all HIV-infected per-
Received 24 September 2002; accepted 31 March 2003; electronically published 22 July 2003. Presented in part: 40th Meeting of the Interscience Conference on Antimicrobial Agents and Chemotherapy, Toronto, Canada, September 2000. Financial support: SmithKline Beecham Development Partners’ Junior Faculty Award Program in Infectious Diseases (M.C.R-B.) and Department of Veterans Affairs under Merit Review Funding. a Present affiliations: Department of Medicine, Division of Infectious Diseases, North Central Bronx Hospital, New York (T.N.); Schuyler Hospital, Montour Falls, New York (I.A.).
Reprints or correspondence: Dr. Maria C. Rodriguez-Barradas, Veterans Affairs Medical Center, Section of Infectious Diseases (MS 111G), 2002 Holcombe Blvd., Houston, TX 77030 (
[email protected]). Clinical Infectious Diseases 2003; 37:438–47 2003 by the Infectious Diseases Society of America. All rights reserved. 1058-4838/2003/3703-0018$15.00
438 • CID 2003:37 (1 August) • HIV/AIDS
sons be immunized with the 23-valent pneumococcal polysaccharide vaccine (PV) [3]. Because of the lack of data on the efficacy of pneumococcal vaccination for HIV-infected persons from prospective studies in developed countries, recommendations are based on the surrogate marker of antibody responses and on expert opinion. Initial studies that evaluated antibody responses to pneumococcal antigens among HIV-infected persons were conducted when HAART was not yet available [4–9]. HIV-infected patients with a CD4 cell count of !500 cells/mm3 were found to have an impaired IgG response to pneumococcal antigens [7, 10]. This defect in humoral immunity deficiency was not overcome by revaccination with a double dose of PV [11] or by vaccination with 1 or 2 doses of protein-conjugate pneumococcal vaccine [10, 12]. Immune reconstitution after initiation of HAART
reduces the incidence of opportunistic infections [13] and results in improved in vitro and in vivo responses to protein antigens [14, 15]. Whether HAART-mediated CD4 cell reconstitution also improves the antibody response to T cell–independent antigens, such as pneumococcal polysaccharides, is not known. Retrospective studies have shown a decreased incidence of bacterial pneumonia [16], recurrent pneumonia [13], and communityacquired bacteremia [17] associated with the use of HAART. One potential explanation for these observations would be the improved immunogenicity to pneumococcal polysaccharides among patients receiving current combination treatments. Few studies have examined antibody responses to pneumococcal capsular polysaccharides (CPSs) among patients receiving HAART [18–20]. Moreover, antibody responses to specific CPSs have not been studied among HIV-infected patients receiving long-term HAART. The objective of this study was to evaluate specific IgG responses to vaccination or revaccination with PV in HIV-infected subjects receiving long-term HAART.
METHODS Patients. HIV-infected patients with indications to receive PV [3] were enrolled from the outpatient clinics of the 3 participating health care centers (Houston Veterans Affairs Medical Center, Bronx Veterans Affairs Medical Center [New York], and Bassett HealthCare [Cooperstown, NY]). Patients needed to be 18–65 years of age, in stable clinical condition (i.e., no intercurrent illness in the prior 2 weeks), and receiving long-term HAART, defined as ⭓6 months of a regimen that included a protease inhibitor (PI) or a nonnucleoside reverse-transcriptase inhibitor (NNRTI) and 2 nucleoside reverse-transcriptase inhibitors (NRTI), or as any combination that included ⭓1 PI, a NNRTI, and/or ⭓1 NRTI. Eighty-seven patients who were receiving long-term HAART were enrolled. Forty-six of the patients were receiving PV for the first time (the HAART first PV group), and 41 were revaccinated (the HAART re-PV group); the latter patients had received PV before the initiation of HAART and ⭓5 years before the current vaccination. The control subjects were 38 HIV-infected patients who had never received HAART and who were receiving PV for the first time (the no-HAART group); 30 of these patients had participated in studies of PV before 1997, and serum samples had been stored at ⫺80C for the purpose of determination of antibody levels. Data on HIV-1 load were not available for those 30 patients. All patients had blood samples obtained immediately before and 4–12 weeks after vaccination. Informed consent was obtained from all subjects in this study. Specific anticapsular antibody determination. ELISA [21] was used to measure the levels of specific anticapsular IgG to
pneumococcal serotypes 3, 4, 6B, 8, 19F, and 23F in pre- and postvaccination serum samples. All serum samples were adsorbed with cell-wall polysaccharide (Statens Seruminstitut) to remove cross-reactive antipneumococcal antibodies. Every plate included paired pre- and postvaccination samples, positive laboratory reference standard serum with a known concentration of IgG to a specific CPS (based on the common reference serum 89-SF), and negative laboratory reference serum that contained no IgG to the individual polysaccharide being studied. Data are presented as micrograms of IgG per milliliter. ELISA readings for laboratory negatives were !0.5 mg/mL, and, therefore, such values are recorded as negative. Any reading of ⭓0.5 mg/ mL was considered to be a positive IgG result. For the purposes of statistical analysis, values of !0.5 mg/mL were recorded as 0.25. For individual CPSs, a 2-fold response was defined as a 2-fold postvaccination increase over prevaccination IgG levels. For those patients with a negative baseline ELISA reading (!0.5 mg/mL), a postvaccination level of ⭓1 mg/mL was considered to be a 2-fold response. Statistical analysis. Groups were categorized on the basis of treatment and vaccination status (no-HAART, HAART first PV, and HAART re-PV groups). The main outcome examined was the pre- to postvaccination increase in IgG levels. For analysis, IgG levels were loge-transformed. The variable used for sample size calculations was defined as the difference between the pre- and postvaccination average natural logarithms of the 6 serotypes. For simplicity, the hypothesis [post/pre(HAART)]/ [post/pre(no-HAART)] ⭓ 1.5 was considered equivalent to [log post ⫺ log pre](HAART) ⫺ [log post ⫺ log pre](no-HAART) ⭓ 0.405. Sample size calculations were based on the latter, because that is the way the data were to be analyzed. Calculations were based on a 2-tailed, 2-sample Student’s t test with a of 0.05. The SD for each group was assumed to be 0.6 on the basis of previous data. The hypothesized difference between groups was 0.405, and the power requirement was set to 80%. With these parameter values, the required sample size in each group was 38 subjects. Other outcomes examined were the number of serotypes to which patients had 2-fold responses (as defined above) and the pre- to postvaccination increase in the proportions of patients with positive IgG levels (as defined above). For some analyses, patients were stratified on the basis of CD4 cell count (!200 vs. ⭓200 cells/mm3) at the time of the most recent vaccination. Comparisons between pre- and postvaccination antibody levels were done using paired Student’s t test. Proportions were compared using Fisher’s exact test. Comparisons of continuous dependent variables among 12 groups were done using analysis of variance. Bonferroni’s correction was applied whenever pairwise comparisons were made among 3 groups. The KruskalWallis test was used for comparisons among ordinal dependent variables. Correlation coefficients were calculated for postvacHIV/AIDS • CID 2003:37 (1 August) • 439
cination IgG levels with each of the following variables: CD4 cell count at the time of current vaccination, CD4 cell count at the time of initial vaccination (for revaccinated patients for whom this value was available), nadir CD4 cell count, virus load at the time of current vaccination (for HAART recipients only), time from initiation of antiretroviral treatment to vaccination, time from initiation of antiretroviral treatment to revaccination, and time between first and second vaccination. Correlations with a chosen parameter were considered to be consistent if they were present for ⭓3 of the 6 pneumococcal CPSs tested. Correlation coefficients were calculated for each of the 3 groups and for all groups combined. SAS software (SAS Institute) was used for all statistical analysis. RESULTS Vaccination groups. A total of 125 HIV antibody–positive patients were included in the study; 87 were receiving HAART, and 38 were not. Forty-one of the patients receiving HAART underwent revaccination. The patients’ demographic and clinical data are presented in tables 1 and 2. HAART recipients were older, and the HAART first PV group included a higher proportion of African American patients (P ! .05). Revaccinated patients had received HAART for a longer period of time than their first-vaccination counterparts (P ! .05 ). A higher proportion of patients in the no-HAART group had a CD4 cell
count of !200 cells/mm3 (a nonsignificant difference, compared with the other 2 groups), but the HAART groups had nadir CD4 cell counts that were significantly lower than were those for the patients who never received HAART (P ! .05). For some analyses, patients were further stratified on the basis of their current CD4 cell count (!200 vs. ⭓200 cells/mm3; table 2). Among those with CD4 cell counts of ⭓200 cells/mm3, the patients receiving HAART were different from their counterparts who did not receive HAART: the nadir CD4 cell counts were significantly lower (P ! .05 ), whereas the CD4 cell counts at the time of vaccination were similar, denoting moreadvanced past immunosuppression and CD4 cell reconstitution with receipt of treatment. Most patients (55%–57%) receiving HAART with CD4 cell counts of ⭓200 cells/mm3 had a virus load of !400 RNA copies/mL at the time of vaccination, whereas the proportion was much lower for those with CD4 cell counts of !200 cells/mm3 (25%). Postvaccination CD4 cell counts and virus loads did not differ from prevaccination values (data not shown). No clinical events were associated with first receipt of PV or revaccination. Antibody levels. Pre- and postvaccination IgG levels to individual pneumococcal CPSs were similar among all 3 groups for all serotypes except for CPS 3 (table 3). The preto postvaccination increase in IgG levels to individual serotypes was significant for all serotypes within each group (P ! .01) and was not different among the 3 groups. A sig-
Table 1. Demographic and clinical characteristics of HIV-infected patients who received 23-valent pneumococcal vaccine (PV), categorized by treatment and vaccination status.
Characteristic Age, mean years SD Male sex
No HAART first PV group (n p 38) a
HAART first PV group (n p 46)
HAART re-PV group (n p 41)
42.0 8.9
47.1 8.1
46.8 8.4
37 (97)
45 (98)
37 (90)
8 (21)
26 (57)a
13 (32)
26 (68)
19 (41)
25 (61)
23 (61)
19 (41)
20 (49)
Ethnicity African American White Risk group Homosexual Injection drug user Otherb
5 (13)
7 (15)
8 (20)
10 (26)
20 (43)
13 (32) c
Duration of HAART, mean months SD (median)
—
23 14 (21)
32 13 (35)
CD4 cell count at current vaccination, mean cells/ mm3 SD (median)
274 252 (215)
352 236 (315)
366 229 (336)
Patients with a CD4 cell count of !200 cells/mm3 Nadir CD4 cell count, mean cells/mm3 SD (median)
18 (47) a
286 260 (240)
16 (35)
12 (29)
161 171 (80)
132 95 (120)
NOTE. Data are no. (%) of patients, unless otherwise indicated. HAART first PV, HAART recipients who underwent their first vaccination with PV; HAART re-PV, HAART recipients who underwent revaccination with PV; no-HAART first PV, patients who did not receive HAART who underwent their first vaccination with PV. a b c
Significantly different from the other 2 groups (P ! .05). Includes heterosexuals and persons for who the risk group was unknown. Significantly different from the other HAART group (P ! .01)
440 • CID 2003:37 (1 August) • HIV/AIDS
Table 2. Clinical characteristics of HIV-infected patients who received 23-valent pneumococcal vaccine (PV), categorized by treatment and vaccination status and stratified by CD4 cell count. Patient group, by CD4 cell count No-HAART first PV
Characteristic
HAART first PV
cells/mm3 (n p 18)
⭓200 cells/mm3 (n p 20)
!200
!200
cells/mm3 (n p 16)
HAART re-PV
⭓200 cells/mm3 (n p 30)
⭓200 cells/mm3 (n p 29)
!200
cells/mm3 (n p 12)
Treatment received 16 (89)
6 (30)
NA
PI and NRTI
NRTI
NA
NA
10 (63)
23 (77)
NA
10 (83)
NA
20 (69)
NA
NNRTI and NRTI
NA
NA
6 (37)
7 (23)
2 (17)
9 (31)
Duration of therapy, mean months SD (median) NRTI HAART Time between first vaccination and revaccination, mean months SD
19 17 (10)
13 6 (13)
NA
NA
NA
NA
25 16 (23)
21 12 (21)
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
274 230 (252)
107 56 (117)
483 185 (464)
111 51 (123)
472 185 (448)
57 51 (45)
219 186 (210)
65 55 (77)
160 95 (175)
31 15 (35)
32 13 (35)
77 16
87 23
CD4 cell count, mean cells/mm3 SD (median) At prior vaccination At current vaccination
69 56 (60)
446 214 (384)
Nadir
57 45 (60)
421 241 (370)
c
a
408 137 (448)
b
Virus load at current vaccination Mean RNA copies/mL !400 RNA copies/mL
NA NA
780 127
d
d
0
77 181 4 (25)
37
102 109
17 (57)
3 (25)
4 11 16 (55)
NOTE. Data are no. (%) of patients, unless otherwise indicated. HAART first PV, HAART recipients who underwent their first vaccination with PV; HAART re-PV, HAART recipients who underwent revaccination with PV; NNRTI, nonnucleoside reverse-transcriptase inhibitor; no-HAART first PV, patients who did not receive HAART who underwent their first vaccination with PV; NRTI, nucleoside reverse-transcriptase inhibitor; PI, protease inhibitor. a b c d
Data are for 7 patients. Data are for 20 patients. P ! .05, compared with patients in the HAART groups who had a CD4 cell count of ⭓200 cells/mm3. Data are for 8 patients.
nificantly better response was observed between the noHAART first PV and HAART re-PV groups when the mean increase for all CPSs tested was compared (P ! .05 ). Analysis performed after stratification by CD4 cell count (figure 1) showed significant increases in IgG levels for all 6 serotypes for all groups with CD4 cell counts of ⭓200 cells/mm3 and for patients in the no-HAART group with CD4 cell counts of !200 cells/mm3. However, the increase was not significant for 2 serotypes (CPS 4 and 6B) in the HAART first PV group and for 3 serotypes (CPS 4, 6B, and 23F) in the HAART rePV group among those with CD4 cell counts of !200 cells/ mm3. Among patients with a CD4 cell count of !200 cells/ mm3, patients in the HAART re-PV group had higher prevaccination levels to 4 serotypes than did those in the HAART first PV group (CPS 3, 4, 8, and 19F; P ! .05), and they had higher prevaccination levels to 1 serotype than did patients in the no-HAART group (CPS 4; P ! .05). Among patients with CD4 cell counts of ⭓200 cells/mm3, there were no observed differences in the pre- or postvaccination IgG levels. Two-fold responses. Patients were classified on the basis of the number of CPSs to which they had a 2-fold response
(table 4). Significantly more responses were noted for the noHAART group than for the combined HAART groups (P ! .05). The same analysis was applied for patients stratified by CD4 cell count; when results for all patients were pooled, significantly more responses were observed for patients with CD4 cell counts of ⭓200 cells/mm3 than for those with counts of !200 cells/mm3 (P ! .05). Proportion of patients with pre- and postvaccination positive IgG levels. The proportion of patients with postvaccination IgG levels of ⭓0.5 mg/mL was significantly higher than the proportion of patients with these levels before vaccination for all CPSs tested for all groups of patients with CD4 cell counts of ⭓200 cells/mm3 (figure 2). Among patients with CD4 cell counts of !200 cells/mm3, the increase was significant for 4 serotypes each in the no-HAART group and the HAART first PV group but for no serotypes in the HAART re-PV group. Because of its small size (n p 12), conclusions derived from this latter group’s results need to be taken with caution. When averaging the results for all serotypes, the mean postvaccination increase in the proportion of positive IgG levels was somewhat higher for the no-HAART and HAART first PV groups than for the HAART re-PV group (24.8%, 26.9%, and 17.6%, reHIV/AIDS • CID 2003:37 (1 August) • 441
Table 3. Pre- and postvaccination mean IgG levels to each pneumococcal capsular polysaccharide (CPS) for patients categorized according to treatment and vaccination status. Mean IgG level, mg/mL, by patient group
Period, CPS type
No HAART first PV (n p 38)
HAART first PV (n p 46)
HAART re-PV (n p 41)
0.94
0.47a
0.97
Prevaccination 3 4
0.57
0.65
0.72
6B
0.88
0.76
0.96
8
0.64
0.49
0.74
19F
1.11
1.03
1.70
23
0.56
0.63
0.79
0.76
0.65
0.94
3
2.87
1.00a
1.90
4
1.24
1.18
1.40
6B
3.12
1.40
1.95
8
2.37
1.86
1.57
19F
3.39
2.05
3.19
23
1.74
1.44
1.76
1.44
1.87
All Postvaccinationb
All
c
2.32
NOTE. All statistical analyses were done using the log values. Mean IgG values represent the antilog of the mean values of the log data. All CPS values represent the antilog of the mean across all 6 CPS log data. HAART first PV, HAART recipients who underwent their first pneumococcal vaccination; HAART re-PV, HAART recipients who underwent revaccination with pneumococcal vaccination; no-HAART first PV, patients who did not receive HAART who underwent their first vaccination with pneumococcal vaccination. a
Significantly different from the other 2 groups. Significant pre- to postvaccination increases for all serotypes among all groups (P ! .01). c The increase from pre- to postvaccination levels was significantly greater than for the HAART re-PV group (P p .03). b
spectively; P 1 .05); the mean increase in the proportion was also slightly higher among pooled patients with CD4 cell counts of ⭓200 cells/mm3 versus those with CD4 cell counts of !200 cells/mm3 (25.5% and 20.3%, respectively; P 1 .05). Correlation coefficients. Consistent correlations were observed only for the HAART re-PV group. For this group, postvaccination IgG levels were correlated with CD4 cell count at the time of initial vaccination for CPSs 3 (R p 0.40; P p .04), 4 (R p 0.38; P p .05), 6B (R p 0.42; P p .03), and 19F (R p 0.38; P p .05). No other consistent correlations were obtained.
DISCUSSION Current guidelines recommend prompt vaccination with PV for all patients as soon as HIV infection is diagnosed [1]. In
442 • CID 2003:37 (1 August) • HIV/AIDS
addition, guidelines recommend consideration of revaccination 5 years after initial vaccination, and they recommend consideration before that time if the CD4 cell count was !200 cells/ mm3 at the time of vaccination and has subsequently increased to ⭓200 cells/mm3 with treatment. No available data support either revaccination recommendation. In the United States, no prospective study has been performed with HIV-infected persons to evaluate the efficacy of PV in preventing pneumococcal disease. Retrospective studies have suggested that vaccination is 49%–78% effective in protecting against invasive pneumococcal disease [22–25]. Effectiveness seems to be higher for white patients [24] and for patients vaccinated at higher CD4 cell counts (⭓200 cells/mm3 [21] or ⭓500 cells/mm3 [25]). The only published prospective, placebo-controlled study of the efficacy of pneumococcal immunization in preventing pneumonia and pneumococcal disease was performed among Ugandan patients who were not receiving HAART, and the study yielded a surprising result of increased disease-related events (all-cause pneumonia and pneumococcal disease) among the vaccinated subjects in the first 6 months, although there were no differences between the groups after 6 months [26]. The incidence of pneumococcal disease, which is reported to be as much as 100-fold higher among HIV-infected patients than among HIV-uninfected, age-matched control subjects [2], seems to have decreased since the advent of the HAART era [13, 16, 17]. Case-control retrospective studies also have shown a protective effect of antiretroviral treatment, with HAART offering more protection than do regimens that do not contain PIs [25]. These observations suggest that HAART, in yet undefined way(s), enhances immune mechanisms involved in protection against pneumococcal disease. Because protection against pneumococcal disease is mediated largely by humoral responses, it seems reasonable to infer that HAART is positively affecting humoral responses to pneumococcal antigens. In the absence of prospective studies with clinical end points, the evaluation of the efficacy of pneumococcal immunization practices among HIV-infected patients is determined on the basis of surrogate markers. These markers include specific anticapsular IgG levels [21], qualitative characteristics of the IgG response [5, 11, 21, 27, 28], functional characteristics of the IgG response [29–31], and in vivo protective properties of the produced IgG [32]. Our study is limited to the quantification of pre- and postvaccination IgG levels to 6 different CPSs. We evaluated IgG responses to pneumococcal vaccine using 3 criteria: IgG levels, frequency of 2-fold increases, and conversion from negative to positive IgG levels (!0.5 to ⭓0.5 mg/mL, respectively). We used all 3 criteria to maximize the detection of any potential benefit associated with vaccination and revaccination and from receipt versus no receipt of HAART among this group of HIV-infected patients.
Figure 1. Mean pre- and postvaccination IgG levels for indicated groups and pneumococcal capsular polysaccharide (CPS) levels among HIV-infected patients receiving the 23-valent pneumococcal polysaccharide vaccine (PV). Mean IgG values represent the antilog of the mean values of the log data. Circles, patients who did not receive HAART who underwent their first PV immunization; triangles, HAART recipients who underwent their first PV immunization; squares, HAART recipients who underwent revaccination with PV. Filled and open symbols denote patients with CD4 cell counts of !200 and ⭓200 CD4 cells/mm3, respectively. *P ! .05, compared with prevaccination levels; **P ! .01, compared with prevaccination levels.
After vaccination, IgG levels to all serotypes tested increased significantly in the 3 vaccination groups. Except for a lower IgG level to CPS 3 in patients in the HAART first PV group, no differences in IgG levels among groups were noted to individual serotypes. The mean pre- to postvaccination increase was, however, significantly higher for patients in the no-HAART group than for those receiving HAART who underwent revac-
cination. In addition, patients in the no-HAART group were significantly more likely than were treated patients to have 2fold responses. Our study also showed that a CD4 cell count of ⭓200 cells/mm3 at the time of vaccination, regardless of prior nadir CD4 cell count or treatment status, influences responsiveness to pneumococcal vaccination. Overall, patients with a CD4 cell count of ⭓200 cells/mm3 had significant post-
HIV/AIDS • CID 2003:37 (1 August) • 443
Table 4. Rates of 2-fold responses to the indicated number of pneumococcal capsular polysaccharides (CPSs) in different patient groups. No. (%) of patients, by no. of CPSs Total no. of patients
0–1
2–4
5–6
Mean no. per patient
No-HAART first PV
38
15 (39)
14 (37)
9 (24)
2.66
HAART first PV
46
21 (46)
23 (50)
2 (4)
1.93
HAART re-PV
41
19 (46)
18 (44)
4 (10)
1.88
!200 cells/mm3
46
28 (61)
15 (33)
3 (7)
1.65
⭓200 cells/mm
79
27 (34)
40 (51)
12 (15)
2.42
Group
a
All patients, by CD4 cell count 3
b
NOTE. A 2-fold response was defined as a 2-fold postvaccination increase over prevaccination IgG levels. For those patients with a negative ELISA result (!0.5 mg/mL) at baseline, a postvaccination level of ⭓1 mg/mL was considered to be a 2-fold response. HAART first PV, HAART recipients who underwent their first pneumococcal vaccination; HAART re-PV, HAART recipients who underwent revaccination with pneumococcal vaccination; no-HAART first PV, patients who did not receive HAART who underwent their first vaccination with pneumococcal vaccination. a
Significantly more responses, compared with the combined HAART groups (P ! .05). Significantly more responses, compared with patients who had a CD4 cell count of !200 cells/mm3 (P ! .05). b
vaccination increases in IgG levels to more serotypes, a significantly higher frequency of 2-fold responses, and a nonsignificantly higher proportion of conversions from negative to positive IgG levels than did patients with a CD4 cell count of !200 cells/mm3. Only 1 study has shown a definitive impact of antiretrovirals on antibody response to PV; this study involved patients treated with zidovudine alone [6]. A recent study evaluating a conjugate vaccine, showed no impact of HAART on immune response [18]. Another recent study showed no impact of revaccination among HIV-infected patients compared with placebo, 50% of those patients were receiving HAART (mostly for !3 months) [19]. Our patients had received treatment for a prolonged course (the median duration was close to 2 and 3 years for the first-vaccination and revaccination groups, respectively). For those receiving vaccine for the first time or for those receiving revaccination, this prolonged course of HAART was not associated with better responses, compared with what was observed among those that never received HAART. Taking into account how much more immunosuppressed patients in the HAART group have been (on the basis of their nadir CD4 cell count), achieving a comparable response to those never on HAART may represent an improved response. Among the revaccinated patients, postvaccination IgG levels correlated with CD4 cell count at the time of initial vaccination, suggesting that this characteristic and not receipt of HAART is a more important factor for response to revaccination among patients receiving HAART. The fact that CD4 cell count at the time of initial vaccination may impact responses suggests a role for reconstitution of memory cells rather than reconstitution of naive cells in the responsiveness to revaccination with pneumococcal antigens. However, our study was not designed to 444 • CID 2003:37 (1 August) • HIV/AIDS
address this question. At least one-half of the patients receiving HAART who had a CD4 cell count of ⭓200 cells/mm3 had virus loads of !400 RNA copies/mL, compared with none of those in the no-HAART group for whom results were available; however, virus load had no impact on responses among patients who received HAART. Prevaccination IgG levels were similar among all 3 of the main groups studied, but a trend for a higher proportion of prevaccination positive levels (⭓0.5 mg/mL) was observed among the revaccination patients, compared with the other patients receiving HAART. Some—but not all—studies that have examined the persistence of IgG levels after pneumococcal immunization concluded that IgG levels decrease more rapidly among HIV-infected patients; most of those studies were done before the HAART era [9, 11, 33]. The only published study to have evaluated revaccination after 5 years among HIVinfected subjects found that prevaccination IgG levels to 4 individual CPSs were not different between patients who were never vaccinated and the revaccination group; those patients had not been receiving long-term HAART. The level of IgG required for protection and whether detectable levels suffice to confer some degree of protection are not known. In serum samples obtained from otherwise healthy young adults, detectable levels of antibody consistently protected mice against pneumococcal bacteremia [32]. It is possible that HAART may contribute to the persistence of specific anticapsular IgG levels, even if at low level, and that it may offer protection against pneumococcal disease through this mechanism. A prospective long-term study will be required to evaluate the impact of HAART on the persistence of specific IgG levels after pneumococcal immunization. In the evaluation of the impact of HAART on antibody re-
Figure 2. Percentage of patients with IgG levels of ⭓0.5 mg/mL for indicated groups and pneumococcal capsular polysaccharide (CPS) levels among HIV-infected patients receiving the 23-valent pneumococcal polysaccharide vaccine (PV). Patients were stratified on the basis of CD4 cell count (!200 vs. ⭓200 cells/mm3). Shaded and white columns indicate pre- and postvaccination values, respectively. HAART 1st PV, HAART recipients who underwent their first vaccination with PV; HAART re-PV, HAART recipients who underwent revaccination with PV; no-HAART 1st PV, patients who did not receive HAART who underwent their first vaccination with PV. *P ! .05 , compared with prevaccination levels; **P ! .01 , compared with prevaccination levels.
sponse, the makeup of the response also needs to be examined. Among HIV-uninfected subjects, the IgG produced in response to pneumococcal polysaccharides belongs predominantly to the immunoglobulin variable region gene family 3 (VH3). HIVinfected patients have reduced expression of VH3-positive antibodies to pneumococcal CPSs [27, 28]. This effect seems to revert with receipt of HAART and is independent from the quantitative IgG response, when compared with patients who are not receiving HAART [34]. In addition, among HIV-
infected subjects, the opsonic capacity of specific anticapsular IgG seems to be impaired [29, 30]. Conjugate vaccine seems to improve this function [18], compared with antibodies induced by polysaccharide vaccine. However, whether HAART has any effect on the opsonic activity of antibodies induced by vaccination with PV has not been specifically evaluated. In the present study, patients with CD4 cell counts of ⭓200 cells/mm3 did better than did patients with CD4 cells counts of !200 cells/mm3. This finding supports the recommendation HIV/AIDS • CID 2003:37 (1 August) • 445
of vaccination for all patients with CD4 cell counts of ⭓200 cells/mm3 and for those in whom the CD4 cell count has increased to ⭓200 cells/mm3 with treatment. Guerrero et al. [23] found that, among vaccinated patients with a CD4 cell count of !100 cells/mm3, pneumococcal vaccination reduced the risk of subsequent pneumonia, although the degree of reduction was lower than that achieved by vaccinated patients with higher CD4 cell counts (⭓200 cells/mm3). With the criteria we examined, patients with a CD4 cell count of !200 cells/mm3 had modest but statistically significant responses to vaccination. These findings suggest that the evaluation of immunogenicity of pneumococcal vaccine based on IgG levels may have clinical correlation among HIV-infected patients. On the basis of our results and those of others, the best approach appears to be to vaccinate HIV-infected patients early in the course of their HIV disease [7, 35]. The current CD4 cell count appears to be related to antibody response (there were better overall responses among patients with a CD4 cell count of ⭓200 cells/mm3), and the impact of CD4 cell count at initial vaccination continues to be apparent at the time of revaccination. After initial vaccination and after revaccination, the immunogenicity of pneumococcal PV among HIV-infected patients receiving long-term HAART is modest, especially among patients with a CD4 cell count !200 cells/mm3. Whether quantitative or qualitative changes in antibody response induced by HAART have an impact on the efficacy of pneumococcal immunization still needs to be determined.
7.
8.
9.
10.
11.
12.
13.
14.
15.
Acknowledgments
We thank Hoang M. Phang and Elizabeth Goldsmith for technical assistance.
References 1. Simberkoff ME, El Sadr W, Schiffman G, Rahal JJ Jr. Streptococcus pneumonia infections and bacteremia in patients with acquired immunodeficiency syndrome with report of a pneumococcal vaccine failure. Am Rev Respir Dis 1984; 130:1174–6. 2. Redd SC, Rutherford GW III, Sande MA, et al. The role of human immunodeficiency virus infection in pneumococcal bacteremia in San Francisco residents. J Infect Dis 1990; 162:1012–7. 3. Masur H, Kaplan JE, Holmes KK. Guidelines for preventing opportunistic infections among HIV-infected persons—2002: recommendations of the US Public Health Service and the Infectious Diseases Society of America. US Public Health Service/Infectious Diseases Society of America. Ann Intern Med 2002; 137:435–78. 4. Huang KL, Ruben FL, Rinaldo CR Jr, Kingsley L, Lyter DW, Ho M. Antibody responses after influenza and pneumococcal immunization in HIV-infected homosexual men. JAMA 1987; 257:2047–50. 5. Janoff EN, Douglas JM Jr, Gabriel M, et al. Class-specific antibody response to pneumococcal capsular polysaccharide in men infected with human immunodeficiency virus type 1. J Infect Dis 1988; 158: 983–90. 6. Glasser JB, Volpe S, Aguirre A, Simpkins H, Schiffman G. Zidovudine
446 • CID 2003:37 (1 August) • HIV/AIDS
16.
17.
18.
19.
20.
21.
22.
23.
24.
improves response to pneumococcal vaccine among persons with AIDS and AIDS-related complex. J Infect Dis 1991; 164:761–4. Rodriguez-Barradas MC, Musher DM, Lahart C, et al. Antibody to capsular polysaccharides of Streptococcus pneumoniae after vaccination of human immunodeficiency virus–infected subjects with 23-valent pneumococcal vaccine. J Infect Dis 1992; 165:553–6. Mascart-Lemone F, Gerard M, Libin M, et al. Differential effect of human immunodeficiency virus infection on the IgA and IgG antibody responses to pneumococcal vaccine. J Infect Dis 1995; 172:1253–60. Nielsen H, Kvinesdal B, Benfield TL, Lundgren JD, Konradsen HB. Rapid loss of specific antibodies after pneumococcal vaccination in patients with human immunodeficiency virus–1 infection. Scand J Infect Dis 1998; 30:597–601. Ahmed F, Steinhoff MC, Rodriguez-Barradas MC, Hamilton RG, Musher DM, Nelson KE. Effect of human immunodeficiency virus type 1 infection on the antibody response to a glycoprotein conjugate pneumococcal vaccine: results from a randomized trial. J Infect Dis 1996; 173:83–90. Rodriguez-Barradas MC, Groover JE, Lacke CE, et al. IgG antibody to pneumococcal capsular polysaccharide in human immunodeficiency virus infected subjects: persistence of antibody in responders, revaccination in non-responders, and relationship of immunoglobulin allotype to response. J Infect Dis 1996; 173:1347–53. Kroon FP, van Dissel JT, Revensberen E, Nibbering PH, van Furth R. Enhanced antibody response to pneumococcal polysaccharide vaccine after prior immunization with conjugate pneumococcal vaccine in HIV-infected adults. Vaccine 2000; 19:886–94. Kaplan JE, Hanson D, Dworkin MS, et al. Epidemiology of human immunodeficiency virus–associated opportunistic infections in the United States in the era of highly active antiretroviral therapy. Clin Infect Dis 2000; 30(Suppl):S5–14. Blazevic V, Sahgal N, Kessler HA, Landay AL, Shearer GM. T cell responses to recall antigens, alloantigen, and mitogen of HIV-infected patients receiving long-term combined antiretroviral therapy. AIDS Res Hum Retroviruses 2000; 16:1887–93. Kroon FP, Rimmelzwaan GF, Roos MTL, et al. Restored humoral immune response to influenza vaccination in HIV-infected adults treated with highly active antiretroviral therapy. AIDS 1998; 12:F217–23. Paul S, Gilbert HM, Ziecheck W, et al. The impact of potent antiretroviral therapy on the characteristics of hospitalized patients with HIV infection. AIDS 1999; 13:415–8. Taconelli E. Tumbarello M, de Gaetan K, et al. Highly active antiretroviral therapy decreases the incidence of bacteremia in human immunodeficiency virus–infected individuals. Clin Infect Dis 1998; 27: 901–2. Feikin DR, Elie CM, Goetz MB, et al. Randomized trial of the quantitative and functional antibody responses to a 7-valent pneumococcal conjugate vaccine and/or 23-valent polysaccharide vaccine among HIVinfected adults. Vaccine 2001; 20:545–53. Tasker SA, Wallace MR, Rubins JB, Paxton WB, O’Brien J, Janoff EN. Reimmunization with 23-valent pneumococcal vaccine for patients infected with human immunodeficiency virus type 1: clinical, immunologic, and virologic responses. Clin Infect Dis 2002; 34:813–21. Amendola A, Tanzi E, Zappa A, et al. Safety and immunogenicity of 23-valent pneumococcal polysaccharide vaccine in HIV-infected former drug users. Vaccine 2002; 20:3720–4. Musher DM, Phan HM, Watson DA, Baughn RE. Antibody to capsular polysaccharide of Streptococcus pneumoniae at the time of hospital admission for pneumococcal pneumonia. J Infect Dis 2000; 182:158–67. Gebo KA, Moore RD, Keruly JC, Chaisson RE. Risk factors for pneumococcal disease in human immunodeficiency virus–infected patients. J Infect Dis 1996; 173:857–62. Guerrero M, Kruger S, Saitoh A. et al. Pneumonia in HIV-infected patients: a case control study of factors involved in risk and prevention. AIDS 1999; 13:1971–5. Breiman RF, Keller DW, Phelan MA, et al. Evaluation of effectiveness
25.
26.
27.
28.
29.
30.
of the 23-valent pneumococcal capsular polysaccharide vaccine for HIV-infected patients. Arch Intern Med 2000; 160:2633–8. Dworkin MS, Ward JW, Hanson DL, Jones, JL, Kaplan JE. Pneumococcal disease among human immunodeficiency virus–infected persons: incidence, risk factors, and impact of vaccination. Adult and Adolescent Spectrum of HIV Disease Project. Clin Infect Dis 2001; 32: 794–800. French N, Nakiyingi J, Carpenter LM, et al. 23-valent pneumococcal polysaccharide vaccine in HIV-1 infected Ugandan adults: doubleblind, randomized and placebo controlled trial. Lancet 2000; 355: 2106–11. Abadi J, Friedman J, Mageed RA, Jeffferis R, Rodriguez-Barradas MC, Pirofski L. Human antibodies elicited by a pneumococcal vaccine express idiotypic determinants indicative of Vh3 gene segment usage. J Infect Dis 1998; 178:707–16. Chang Q, Alpert P, Abadi J, Pirofski L. A pneumococcal capsular polysaccharide vaccine induces a repertoire shift with increased VH3 expression in peripheral B cells from HIV-uninfected, but not HIVinfected individuals. J Infect Dis 2000; 181:1313–21. Janoff EN, O’Brien J, Thompson P, et al. Streptococcus pneumoniae colonization, bacteremia, and immune response among persons with human immunodeficiency virus infection. J Infect Dis 1993; 167:49–56. Janoff EN, Fasching C, Ojoo JC, O’Brien J, Gilks CF. Responsiveness
31.
32.
33.
34.
35.
of human immunodeficiency virus type 1–infected Kenyan women with or without prior pneumococcal disease to pneumococcal vaccine. J Infect Dis 1997; 175:975–8. Vernacchio L, Romero-Steiner S, Martinez JE, et al. Comparison of an opsonophagocytic assay and IgG ELISA to assess responses to pneumococcal polysaccharide and pneumococcal conjugate vaccines in children and young adults with sickle cell disease. J Infect Dis 2000; 181: 1162–6. Musher DM, Phan HM, Watson D, Baughn RE. Antibody to capsular polysaccharide of Streptococcus pneumoniae at the time of hospital admission for pneumococcal pneumonia. J Infect Dis 2000; 182:158–67. Kroon FP, Van Dissel JT, Ravensbergen E, et al. Antibodies against pneumococcal polysaccharides after vaccination in HIV-infected individuals: 5-year follow-up of antibody concentrations. Vaccine 1999; 18:524–30. Subramaniam K, Segal R, Lyles R, Rodriguez-Barradas M, Pirofski L. Qualitative change in antibody responses of human immunodeficiency virus–infected individuals to pneumococcal capsular polysaccharide vaccination associated with highly active antiretroviral therapy. J Infect Dis 2003; 187:758–68. Janoff EN, Tasker SA, Stevenson M, et al. Immune activation and virologic response to immunization in recent HIV type 1 seroconverters. AIDS Res Hum Retroviruses 1999; 15:837–45.
HIV/AIDS • CID 2003:37 (1 August) • 447
