Pneumococcal Vaccine for Adults Aged ≥19 Years …

Pneumococcal Vaccines and Administration

One PPSV and three PCVs are licensed and available for use in the United States (Table 1). These vaccines are PPSV23 (Pneumovax23), a 23-valent pneumococcal polysaccharide vaccine; PCV13 (Prevnar13), a 13-valent pneumococcal conjugate vaccine; PCV15 (Vaxneuvance), 15-valent pneumococcal conjugate vaccine; and PCV20 (Prevnar20), a 20-valent pneumococcal conjugate vaccine. The PPSV and PCV vaccines induce immune responses in different ways. Studies in mice have found that the polysaccharide vaccine induces a T-cell independent immune response and stimulates immediate B-cell responses (113). As a result, B-cells differentiate to plasma cells that produce antibodies. However, a T-cell independent immune response does not result in creation of serotype-specific memory B-cells (114). On the other hand, conjugate vaccines induce a T-cell dependent response. The polysaccharide antigen binds to the B-cells, and the peptides from the carrier protein are presented to carrier-peptide–specific helper T-cells. These helper T-cells enhance the immune response by the B-cells, and memory B-cells also are created (114,115). All PCVs use CRM197 (genetically detoxified diphtheria toxin) as a carrier protein (116–118).

PCV15 and PCV20 are administered intramuscularly at a dose of 0.5 mL. If PCV13 is used in situations where PCV15 or PCV20 is not easily accessible, it is administered intramuscularly at a dose of 0.5 mL. PPSV23 is administered either intramuscularly or subcutaneously at a dose of 0.5 mL. In a small RCT among adults aged ≥55 years, no significant differences were observed in the immunogenicity (measured as IgG geometric mean concentration [GMC] pre- and 1-month post PPSV23 vaccination for serotypes 3, 4, and 6B) by route of administration (119). Local adverse events were more common among those who received PPSV23 subcutaneously (18.9% [nine of 127] versus 7.1% [24 of 127]), whereas the frequency of systemic adverse events was the same in both groups (6.3% [eight of 127]) (119). All three pneumococcal vaccines (i.e., PCV15, PCV20, and PPSV23) are available in a single-dose, prefilled syringe. PPSV23 also is available in a single-dose vial. Additional information on pneumococcal vaccine administration is available in the package inserts (116,117,120).

Vaccine Efficacy and Effectiveness

Vaccine efficacy refers to the percentage by which the rate of the target disease among those who are vaccinated is reduced compared with the rate among unvaccinated persons in the context of a placebo controlled randomized trial. Vaccine effectiveness (VE) measures the same percent reduction in the rate of disease as efficacy, but in the context of routine, real-world use of the vaccine through observational studies. Vaccine efficacy and effectiveness data are currently not available for PCV15 or PCV20.

23-Valent Pneumococcal Polysaccharide Vaccine (PPSV23)

PPSV23 contains pneumococcal polysaccharide serotypes 1, 2, 3, 4, 5, 6B, 7F, 8, 9N, 9V, 10A, 11A, 12F, 14, 15B, 17F, 18C, 19F, 19A, 20, 22F, 23F, and 33F (Figure). Each 0.5-mL dose contains 25 μg of each polysaccharide type in isotonic saline solution containing 0.25% phenol as a preservative (120).

Clinical Efficacy and Effectiveness in Adults

Multiple RCTs and observational studies have reported that PPSV23 is protective against IPD. A meta-analysis of RCTs reported a pooled vaccine efficacy of 63% (95% CI = 10%–92%) against IPD due to any serotype (121); the RCTs included in this meta-analysis did not report PPSV23 efficacy against vaccine-type (VT) IPD (VT-IPD) (122–125). A recent meta-analysis of observational studies reported a pooled VE estimate against VT-IPD of 45% (95% CI = 37%–51%), which is lower than the estimate against all-cause IPD from RCTs, even when limiting to VE estimates ≤5 years since PPSV23 vaccination (126). The lower pooled VE estimate might be because of the presence of adults with immunocompromising conditions who were excluded upon enrollment from most of the RCTs. A pooled VE estimate against VT-IPD in adults without immunocompromising conditions was 60% (95% CI = 47%–69%) (126).

Estimates of the efficacy and effectiveness of PPSV23 against pneumococcal pneumonia have been more variable across studies. A pooled estimate of RCTs evaluating efficacy of PPSV23 against pneumococcal pneumonia was 35% (95% CI = −62% to 65%), but when results were limited to two RCTs deemed to have low risk for bias only, the pooled estimate was 64% (95% CI = 35%–80%) (121). These RCTs did not report efficacy against VT-pneumococcal pneumonia. Observational studies have reported VE estimates ranging from none or not significant (127,128) to moderate (40%–50%) (129,130) against pneumococcal pneumonia. The heterogeneity in the estimates reported in these observational studies is likely because of differences in study design, including how pneumococcal pneumonia cases were ascertained, study settings (community based versus hospital based), proportion of adults with underlying conditions, or age distribution. The pooled VE estimate against VT-pneumococcal pneumonia from five observational studies was not significant (VE = 18%; 95% CI = −4% to 35%) (126).

Clinical Efficacy and Effectiveness in Adults With Immunocompromising Conditions

One RCT among Ugandan adults living with HIV infection conducted during 1995–1998 observed no protection against IPD due to all serotypes (VE = −48%; 95% CI = −232% to 35%) or VT-IPD (VE = −114%; 95% CI = −431% to 14%). Of note, this study was conducted before highly active antiretroviral therapy was widely available, >10% of the vaccinated and placebo groups were lost to follow-up, and 28% of remaining participants in each group had died by the end of the study period (131). Observational studies have demonstrated mixed results regarding PPSV23 VE against IPD among adults with HIV infection (132–136). Three indirect cohort studies reported VT-IPD effectiveness by risk groups in the same study (137–139). Typically, adults with immunocompromising conditions had lower VE than adults without underlying medical conditions, and VE in older adults (aged ≥65 years) often was not significant when ≥2 years had passed since PPSV23 vaccination (137,138).

Duration of Protection

Data on duration of protection after PPSV23 vaccination are only available through observational studies. In a large, indirect cohort study from England and Wales, VE of PPSV23 against VT-IPD declined significantly with time since vaccination from 48% (95% CI = 32%–60%) among those vaccinated ≤2 years ago to 15% (95% CI = −3% to 30%) among those vaccinated ≥5 years ago (p<0.001) (137). In another study from England and Wales, VE of PPSV23 against VT-IPD declined from 41% (95% CI = 23%–54%) among those vaccinated ≤2 years ago to 34% (95% CI = 16%–48%) among those vaccinated 2–4 years ago and to 23% (95% CI = 12%–32%) among those vaccinated ≥5 years ago (138). Although this difference in VE by time since vaccination was not statistically significant (p = 0.13), a spline model indicated a decline in VE from approximately 50% to a plateau of 20%–25% after 5 years. Both studies found a declining trend in VE against VT-IPD regardless of age at vaccination. In the three studies on the duration of protection of PPSV23 against CAP, diagnostic criteria for determining pneumococcal CAP varied (129,140,141). Nevertheless, a declining trend in VE against CAP was observed with increasing time since vaccination in all identified studies.

Repeat PPSV23 Administration

Repeat administration of bacterial polysaccharides vaccines might induce immune tolerance or hyporesponsiveness to vaccine antigens (142). In the RCT of PPSV23 among Ugandan adults with HIV infection, researchers hypothesized that the increased risk for pneumococcal disease that was observed among persons who received PPSV23 compared with those who had received placebo was due to destruction of polysaccharide-responsive B-cell clones (131).

A systematic review of immunogenicity studies investigating repeat administration of PPSV23 found that revaccination with PPSV23 might induce a lower immune response than initial vaccination in the short-term (i.e., when measured ≤2 months after each vaccination); however, these differences were not statistically significant. When immune responses measured >2 years after vaccination were compared, antibody levels against certain serotypes were higher after revaccination with PPSV23 compared with initial vaccination (143). In a more recent systematic review of immunogenicity studies, immune responses were lower with repeat PPSV23 than with the initial vaccination for certain serotypes when the interval between PPSV23 doses was <5 years, whereas no evidence of hyporesponsiveness was noted in studies that used an interval of ≥5 years, an interval that has been recommended for adults with indications for ≥2 PPSV23 doses (144). Because no clinical trials are investigating multiple-dose PPSV23 regimens, the clinical relevance of hyporesponsiveness on pneumococcal disease is unclear.

Safety

A postlicensure review of PPSV23 safety was conducted using VAERS data during 1990–2013, which included 25,168 reports of adverse events after PPSV23 receipt. No safety signals or new or unexpected adverse events were identified in this study (145).

An updated search of VAERS data during January 1, 2014–August 31, 2022 found 24,926 reports; 23,763 (95.3%) were nonserious, 20,737 (83.2%) were in adults aged ≥19 years, and 15,550 (62.4%) were in females (P Moro, MD, CDC, personal communication, September 7, 2022). When PPSV23 was administered alone, injection site erythema (21.1%), injection site swelling (19.3%), and injection site pain (19.3%) were the most commonly reported adverse events (on the basis of MedDRA preferred terms) overall. These rates are similar to commonly reported adverse events across all vaccines administered to adults: injection site erythema (27.1%), erythema (24.7%), injection site swelling (23.6%), and injection site pain (23.3%). Of 35 deaths reported, 30 were adults, two were children, and three were of unknown age. Clinical review of these death reports did not reveal any pattern suggestive of a causal association with PPSV23. Half of the 26 reports of anaphylaxis after PPSV23 receipt were after receipt of PPSV23 alone. The VAERS reporting rate for 26 reports of anaphylaxis was 1 case per million PPSV23 doses distributed in the United States during the review period (January 1, 2014–August 31, 2022), similar to the incidence of vaccine-triggered anaphylaxis (0.8 per million doses administered) across all vaccines (146). Data mining analysis did not reveal disproportional reporting of any new or unexpected adverse events after PPSV23 receipt.

13-Valent Pneumococcal Conjugate Vaccine (PCV13)

PCV13 contains pneumococcal polysaccharide serotypes 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, and 23F, conjugated to CRM197, a nontoxic form of diphtheria toxin (Figure). Each 0.5-mL dose of vaccine contains approximately 2.2 μg of all serotypes except for 4.4 μg of 6B saccharides, 34 μg of CRM197 carrier protein, 100 μg of polysorbate 80, 295 μg of succinate buffer, and 125 μg of aluminum as aluminum phosphate adjuvant (118). In 2021, ACIP updated the PCV13 recommendations to replace PCV13 with PCV15 or PCV20 for adults who have not received a pneumococcal conjugate vaccine (16).

Clinical Efficacy and Effectiveness in Adults With and Without Immunocompromising Conditions

Studies evaluating PCV13 efficacy or effectiveness in adults are limited. A RCT among 84,496 Dutch pneumococcal vaccine–naïve, community-dwelling adults aged ≥65 years (CAPiTA trial) reported a 75% (95% CI = 41%–91%) vaccine efficacy against VT-IPD and 45% (95% CI = 14%–65%) vaccine efficacy against the first episode of VT-NBPP and noninvasive pneumococcal pneumonia (3). This trial excluded adults with immunocompromising conditions upon enrollment. VE estimates from three observational studies ranged from 47% to 68% against VT-IPD in adults aged ≥65 years (147–149).

Three test-negative, case-control observational studies evaluated PCV13 effectiveness against VT-NBPP with estimates ranging from 38% to 68%. Each study noted that most participants had ≥1 underlying or high-risk condition (128,150,151). A population-based cohort study of approximately 2 million adults aged ≥50 years in Spain did not demonstrate a protective effect of PCV13 against pneumococcal pneumonia or all-cause pneumonia, although the characteristics of those who received PCV13 were significantly different from those who did not receive PCV13, including older age and a higher proportion of adults with immunocompromising conditions among the PCV13 recipients compared with nonrecipients (152). Other observational studies have demonstrated the effectiveness of PCV13 against all-cause pneumonia (range = 5%–12%) (153–156).

Among adults without immunocompromising chronic medical conditions (i.e., adults with chronic heart, lung, or liver disease; asthma; or diabetes mellitus), post hoc analysis of the CAPiTA trial observed a 77% (95% CI = 44%–91%) vaccine efficacy against VT-IPD (157) and 40% (95% CI = 11%–60%) vaccine efficacy against a first episode of VT-pneumococcal pneumonia (158). Efficacy and effectiveness data regarding PCV13 in immunocompromised adults are limited. A test-negative study of Italian adults aged ≥65 years with immunocompromising or chronic medical conditions observed a 40% (95% CI = −128% to 89%) VE against VT-CAP (151).

Duration of Protection

Post hoc time-to-event analyses of the CAPiTA trial found a significant difference in disease-free survival among PCV13 recipients compared with placebo recipients for all outcomes throughout the duration of the trial (mean follow-up approximately 4 years). Vaccine efficacy against VT-NBPP and noninvasive pneumococcal pneumonia was 44% 1 year after vaccination and 45% ≤5 years after vaccination; vaccine efficacy against VT-IPD was 67% 1 year after vaccination and 75% ≤5 years after vaccination (159). An additional post hoc analysis of data from the CAPiTA trial among adults with chronic medical conditions (excluding those with immunocompromising conditions) also demonstrated significant and persistent efficacy of PCV13 against VT-CAP for an average duration of 4 years (158).

Safety

A postlicensure review of the safety of PCV13 was conducted using VAERS data during June 2012–December 2015 among adults aged ≥19 years (160). VAERS received 2,976 PCV13 reports; 2,800 (94.1%) were nonserious, 2,511 (86%) were in persons aged ≥65 years, and 465 (14%) were in persons aged 19–64 years. Injection site erythema (28%), injection site pain (24%), and fever (22%) were the most frequent adverse events (on the basis of MedDRA preferred terms) among persons aged 19–64 years. The most common adverse events in persons aged ≥65 years were injection site erythema (30%), erythema (20%), and injection site swelling (18%) among those who were administered PCV13 alone. The results of this study were consistent with safety data from prelicensure studies of PCV13.

A more recent review of the safety of PCV13 using VAERS data during January 2016–August 2022 found 17,033 reports (CDC, unpublished data, September 2022). The most common adverse events were local reactions similar to the previous publication (160). Data mining analysis did not reveal disproportionate reporting of any new or unexpected adverse events.

A cohort study at six sites from the VSD cohort during January 1, 2011–August 15, 2015, assessed the safety of PCV13 in adults aged ≥65 years compared with PPSV23 for increased risk for certain prespecified adverse events (i.e., cardiovascular events, Bell’s palsy, Guillain-Barré syndrome, syncope, erythema multiforme, thrombocytopenia, cellulitis and infection, allergic reaction, and anaphylaxis). A total of 313,136 doses administered of PCV13 and 232,591 doses of PPSV23 were included. No increased rate of adverse events after PCV13 administration in persons aged ≥65 years compared with PPSV23 was observed (161).

15-Valent Pneumococcal Conjugate Vaccine (PCV15)

PCV15 contains pneumococcal polysaccharide serotypes 22F and 33F in addition to the PCV13 serotypes, conjugated to CRM197 (Figure). Each 0.5-mL dose of vaccine contains 2.0 μg each of S. pneumoniae polysaccharide serotypes 1, 3, 4, 5, 6A, 7F, 9V, 14, 18C, 19A, 19F, 22F, 23F, and 33F and 4.0 μg of 6B, 30 μg of CRM197 carrier protein, 1.55 mg of L-histidine, 1 mg of polysorbate 20, 4.5 mg of sodium chloride, and 125 μg of aluminum as aluminum phosphate adjuvant (116).

Immunogenicity in Adults Without Previous Pneumococcal Vaccination

Phase II and III RCTs evaluated the immunogenicity and safety of 1 dose of PCV15 compared with PCV13 in multiple populations: healthy adults aged ≥50 years (162,163), adults aged 18–49 years who are American Indian or Alaska Native (a population with higher rates of IPD than the general U.S. population) (164), adults with ≥1 risk condition for pneumococcal disease (165), and adults aged ≥18 years with HIV infection (CD4 cell count of ≥50 cells/μL and plasma HIV RNA of <50,000 copies/mL) (166). Serotype-specific IgG GMCs and functional antibody responses using an opsonophagocytic activity (OPA) assay were measured 1 month after vaccination. Correlates of protection against pneumococcal diseases have not been established for adults. In one phase II RCT, adults aged ≥50 years were randomized to receive either PCV13, PCV15, or PPSV23. Primary and secondary endpoints were assessment of noninferiority^† of PCV15 compared with PPSV23, which were met for all 14 shared serotypes; IgG GMC and OPA geometric mean antibody titers (GMTs) were numerically higher for 13 of 14 serotypes (except for serotype 19F). Compared with PCV13, PCV15 induced numerically higher IgG GMCs and OPA GMTs for eight of 13 and seven of 13 shared serotypes, respectively; noninferiority was not assessed for these comparisons (162). In one phase III RCT among adults aged ≥50 years, noninferiority criteria^§ were met for PCV15 compared with PCV13 for the 13 shared serotypes, and OPA responses were statistically higher^¶ for shared serotype 3 and PCV15-unique serotypes 22F and 33F (163). In studies that evaluated the immunogenicity of PCV15 or PCV13 followed by PPSV23 2–12 months later (165–167), persons who received PCV15 had numerically similar or higher OPA GMTs for 9–13** shared PCV13 serotypes and a higher percentage of seroresponders (defined as subjects with a greater than fourfold rise in OPA GMT titer postvaccination compared with pre-PCV vaccination) for 5–11 shared serotypes compared with persons who received PCV13 when measured 1 month after receipt of PPSV23. Thus, immunogenicity of PCV13 followed by PPSV23 was considered to be generally comparable to immunogenicity of PCV15 followed by PPV23.

Immunogenicity in Adults With Previous Pneumococcal Vaccination

One phase II trial assessed the safety and immunogenicity of a dose of PCV15 compared with PCV13 among adults aged ≥65 years who previously received PPSV23 ≥1 year before PCV receipt (168). Compared with baseline, increases of IgG GMCs and OPA GMTs were observed for all PCV serotypes when measured 1 month after PCV administration. Compared with those who received PCV13, those who received PCV15 had numerically similar or higher IgG GMCs and OPA GMTs for the 13 serotypes contained in PCV13 and higher immune responses for PCV15-unique serotypes 22F and 33F when measured 1 month after PCV administration.

Safety

Safety of PCV15 was assessed in seven RCTs with 5,630 participants aged ≥18 years who received a single dose of PCV15. Most participants were immunocompetent; however, one study included 302 adults with HIV infection (166). Participants from these seven RCTs included persons who had been vaccinated with PPSV23 ≥1 year before receiving PCV15, those who received PCV15 followed by PPSV23, and those who received PCV15 concomitantly with a seasonal inactivated quadrivalent influenza vaccine (IIV4). The most frequently reported adverse reactions were injection site pain, fatigue, and myalgia. The rates of SAEs within 6 months of vaccination in a subset of studies (n = 4,751 for PCV15; n = 1,532 for PCV13) that assessed this outcome were 2.5% among PCV15 recipients and 2.4% among PCV13 recipients. No SAEs or deaths were considered to be related to the study vaccines (116,169). Postlicensure safety data are being collected and reviewed; however, limited data are available.

20-Valent Pneumococcal Conjugate Vaccine (PCV20)

PCV20 contains pneumococcal polysaccharide serotypes 8, 10A, 11A, 12F, 15B, 22F, and 33F in addition to PCV13 serotypes, conjugated to CRM197 (Figure). Each dose of vaccine contains approximately 2.2 μg of all serotypes except for 4.4 μg of 6B saccharides, 51 μg of CRM197 carrier protein, 100 μg of polysorbate 80, 295 μg of succinate buffer, 4.4 mg of sodium chloride, and 125 μg of aluminum as aluminum phosphate adjuvant (117).

Immunogenicity Among Adults Without Previous Pneumococcal Vaccination

A phase II study among adults aged 60–64 years and two phase III RCTs among adults aged ≥18 years evaluated immunogenicity and safety of PCV20 compared with PCV13 (170–172). Immunogenicity of PCV20 for the seven additional serotypes included in PCV20 but not in PCV13 was compared with PPSV23 (administered 1 month after PCV13) (170,172). These studies included adults with stable medical conditions (i.e., disease not requiring significant change in therapy or hospitalization for worsening disease during the weeks preceding receipt of the vaccine) but not adults with immunocompromising conditions. Compared with PCV13 recipients, PCV20 recipients elicited responses that met the noninferiority criterion (defined as the lower bound of the two-sided 95% CI of the ratio [PCV20/PCV13] of OPA GMT being >0.5) for all 13 serotypes in a phase III trial among adults aged ≥60 years (170); however, PCV20 recipients appeared to have lower GMTs and included a lower percentage of seroresponders to 12–13 of the 13 PCV13-shared serotypes (170,172). The clinical relevance of these findings is unknown. Compared with PPSV23 recipients, PCV20 recipients had numerically higher GMTs and a higher percentage of seroresponders to six of seven (excluding serotype 8) shared non-PCV13 serotypes (170,172); the noninferiority criterion was met for those six serotypes (170).

Immunogenicity Among Adults With Previous Pneumococcal Vaccination

Two phase III studies assessed the immunogenicity of PCV20 among adults aged ≥65 years who had previous pneumococcal vaccination (PCV13 only, PPSV23 only, or both PCV13 and PPSV23) (173,174). These studies included adults with stable medical conditions, but not adults with immunocompromising conditions. Immunogenicity to PCV20 was assessed 1 month after receipt of PCV20. In one study, participants either received PCV13 ≥6 months previously; PPSV23 1–5 years previously; or PCV13 followed by PPSV23, with PPSV23 ≥1 year previously (173). The other study was a post hoc analysis of a subset of participants who received pneumococcal vaccines (PCV13, PPSV23, or both) ≥6 months previously in a phase III clinical trial of the coadministration of PCV20 with a seasonal inactivated influenza vaccine (174,175). Neither study directly compared the immunogenicity of PCV20 with PPSV23 after receipt of both PCV13 and PPSV23. Compared with persons who previously received PPSV23 only, those who previously received both PCV13 and PPSV23 had numerically higher OPA GMTs of serotypes included in PCV20 in 19 of 20 PCV20 serotypes in one study (173) and for 15 of 20 PCV20 serotypes in another study (174). In addition, those who received PCV13 only had numerically higher OPA GMTs of serotypes included in PCV20 compared with adults who previously received PPSV23. Percent seroresponders (defined as a greater than fourfold rise in OPA titer from before to 1 month after PCV20 receipt) was numerically higher for 13–18 of 20 serotypes among adults who previously received PCV13 compared with adults who previously received PPSV23. Compared with adults who previously received PPSV23 only, percent seroresponders were numerically higher for four of 20 serotypes for adults who received both PCV13 and PPSV23 in one study (173) and for six of 20 serotypes in another study (174,175). Percent seroresponders in these studies typically were lower compared with those observed among pneumococcal vaccine–naïve adults (170). Regardless, these data demonstrate that PCV20 is immunogenic in those aged ≥65 years who previously received a pneumococcal vaccine.

Safety

Clinical Trials

Safety was assessed in seven trials among immunocompetent adults aged ≥18 years; the trials included a total of 6,343 participants who received PCV20 (117). Participants included persons who were naïve to pneumococcal vaccination and those who had previously received pneumococcal vaccination. In the pivotal phase III trial among pneumococcal vaccine–naïve adults, the most frequently reported adverse reactions were injection site pain, muscle pain, fatigue, headache, and joint pain (55%, 39%, 30%, 22%, and 13%, respectively, among adults aged ≥60 years; 50%–61%, 50%–67%, 40%–43%, 32%–39%, and 13%–15%, respectively, among adults aged 18–59 years) (117). Among participants across six of seven trials, SAEs reported within 6 months after vaccination occurred among 1.5% of PCV20 recipients (n = 4,552) and among 1.8% of controls (n = 2,496) (117). No SAEs or deaths were considered to be related to study vaccines across all seven trials (117,176).

Postlicensure Safety Monitoring

A postlicensure review of the safety of PCV20 was conducted using VAERS data during October 21, 2021–August 31, 2022. A total of 412 reports related to PCV20 were submitted to VAERS; 399 (96.8%) were nonserious, 389 (94.4%) were in adults aged ≥19 years, and 313 (75.9%) were in females. Thirteen SAEs were reported, including two deaths. Cause of death in a report of a female aged 51 years was acute brain herniation secondary to acute left temporal hemorrhage; cause of death in the second report was not specified. Among the 11 remaining SAEs reported, the most common adverse events were asthenia, nausea, fever, and gait instability. When PCV20 was administered alone (n = 274) the most common adverse events were injection site erythema (26.6%), injection site swelling (22.6%), and injection site pain (22.3%). These proportions are similar to commonly reported adverse events across all vaccines approved for use in adults: injection site erythema (22.6%), injection site pain (19.3%), and erythema (19%). Data mining analysis did not reveal disproportional reporting of any new or unexpected adverse events after PCV20 administration.

Immune Response After Repeat PCV Dosing

Adults Without Immunocompromising Conditions

Findings from six immunogenicity studies that evaluated responses after repeat PCV administration were reviewed. In two studies among pneumococcal vaccine–naïve adults aged ≥60 years, immune responses after a second PCV13 dose, 1 year after the initial PCV13 dose, were statistically significantly lower for the majority of serotypes compared with the initial PCV13 dose (177,178). In contrast, in two studies among pneumococcal vaccine–naïve adults aged ≥50 years, immune responses after a second PCV13 dose, delivered 3.5–5 years after the initial PCV13 dose, were comparable or statistically significantly greater for the majority of serotypes compared with the initial PCV13 dose (179,180). These findings suggest that longer intervals between PCV13 administrations might optimize immune responses to subsequent doses. In another study among adults aged ≥70 years who had received 1 dose of PPSV23 ≥5 years before enrollment, two administrations of PCV13, 1 year apart, resulted in a recovery of the responses to levels similar to those after the initial PCV13 dose, indicating that an initial dose of PCV13 does not have a negative impact on immune responses to subsequent PCV13 doses among adults previously vaccinated with PPSV23 adults (181). Finally, in post hoc analysis among adults who had received PCV13 ≥6 months before, immune responses 1 month after administration of PCV20 compared with baseline (i.e., day of PCV20 vaccination) were numerically higher for the 13 serotypes common to PCV13 and PCV20 as well as for the seven additional serotypes included in PCV20 (182). The percentage of subjects with a greater than fourfold rise in OPA titers was typically lower in this study compared with the percentage observed in the pivotal trial among pneumococcal vaccine–naïve adults (170).

Adults With Immunocompromising Conditions

In one study among adults aged ≥18 years with HIV infection who previously received PPSV23 ≥6 months before the first PCV13 dose, IgG GMCs to the PCV13 serotypes were measured 1 month after the second and third doses of PCV13 that were administered 6 months apart. IgG GMCs after the second and third doses of PCV13 were comparable to those after the first dose of PCV13 (183). In another study among pneumococcal vaccine–naïve persons aged ≥6 years with HIV infection who received 3 doses of PCV13 followed by a single dose of PPSV23 1 month apart, the largest increase in IgG GMCs or OPA GMTs for the PCV13 serotypes was after the first dose, and only modest increases were observed after the second and third doses (184).

Adults Who Received HSCT

HSCT recipients can have as high as 80 times the risk for IPD compared with the general population (86). HCST recipients have a poor response to PPSV23 when administered during the first year of transplantation or later, especially those with chronic graft versus host disease (GVHD) (185,186). Because of the prolonged process of lymphocyte recovery after HSCT, both children and adults who are HSCT recipients have been recommended to receive 3 doses of PCV13 starting at 3–6 months after HSCT (15,92). PPSV23 as the fourth dose of pneumococcal vaccine starting 12 months after HSCT has been found to induce immune responses to vaccine serotypes (187) and has been recommended to provide immunity to additional serotypes that are not included in PCVs, except for patients with chronic GVHD who have been recommended to receive PCV for their fourth dose because they are unlikely to respond to PPSV23 (92).

In a phase III RCT, children and adults who received allogenic HSCT were randomized to receive either 3 doses of PCV13 or PCV15 starting 3–6 months after HSCT, followed by a single dose of PPSV23 at 12 months (or a single dose of PCV13 or PCV15 for those who developed GVHD) (188). Immunogenicity measured 1 month after the third PCV dose indicated that IgG GMC and OPA GMT were typically comparable for the 13 shared PCV13 serotypes, and PCV15 had significantly higher immunogenicity for two additional serotypes not included in PCV13. In both intervention groups, PPV23 also was immunogenic (as assessed by geometric mean fold rises [GMFRs] in IgG GMC and OPA GMT) from prevaccination (month 12) to 30 days postvaccination with PPV23 (month 13) for the 13 shared serotypes in PCV13 and for the two additional serotypes unique to PCV15 (189,190). Serotype-specific IgG GMCs and OPA GMTs measured in month 13 were typically comparable between the two groups for the 13 shared serotypes contained in PCV13; compared with PCV13 recipients, PCV15 recipients had numerically higher IgG GMC and OPA GMT values for the two serotypes unique to PCV15. A fourth dose of PCV15 in patients with chronic GVHD also was immunogenic for all 15 serotypes contained in the vaccine (as assessed by GMFRs) from prevaccination (month 12) to 30 days postvaccination. The frequency of vaccine-related adverse events was higher in HSCT recipients aged ≥18 years who received PCV15 compared with those who received PCV13 after receiving 3 doses of PCV (injection-site adverse events: 93.1% versus 75.2%; systemic adverse events: 55% versus 43.4%) although the majority of adverse events were mild to moderate. The frequency of adverse events after receipt of PPSV23 at 12 months (injection-site adverse events: 71.4% versus 64.8%; systemic adverse events: 31.0% versus 29.6%) were comparable between the PCV15 and PCV13 groups. No clinical trials of PCV20 among HSCT recipients have been reported.

The optimal timing of vaccination among HSCT recipients has been discussed globally. The Dutch immunization schedule recommends starting vaccination 1 year after HSCT to allow time for immune reconstitution after allogenic HSCT (191). Alternatively, starting vaccination earlier gives the opportunity for HSCT recipients to have protection against pneumococcal disease earlier (186). In a trial that compared initiating 3 doses of PCV7 3 months or 9 months after HSCT followed by a dose of PPSV23 12 or 18 months after HSCT, the response rate (defined as the percentage of patients with antibody titers of ≥0.15 μg/mL to all the PCV7 serotypes 1 month after the third PCV7 dose) was not statistically significantly different between the two groups, although the GMC and response rates were lower in the early vaccination group at 24 months (192). For the serotypes unique to PPSV23, the early vaccination group had lower GMCs and percentage of responders at 24 months, although the differences between the two groups were not statistically significant (187).

To provide earlier and prolonged protection against IPD after HSCT, use of 4 doses of PCV13 (3 doses administered 1 month apart followed by a booster dose 6 months after the third dose of PCV) followed by a single dose of PPSV23 has been studied (193,194). One study assessed immunogenicity and safety of 4 doses of PCV13 doses followed by a single dose of PPSV23 1 month later among children and adults who received HSCT 3–6 months previous. Results indicated that the GMFR of IgG concentrations from baseline to 1 month after the third dose and from after the third dose to 1 month after the fourth dose increased significantly across all PCV13 serotypes in both children and adults; however, there was little change in the GMFR when comparing 1 month after the fourth dose with 1 month after the PPSV23 dose (194). Similar findings were reported in a separate study using a similar 5-dose vaccine schedule (4 doses of PCV13 followed by 1 dose of PPSV23 2 months after the fourth dose of PCV13) among adults who received allogenic HSCT 3–6 months previous (193). Of note, the interval between PCV and PPSV23 doses used in these trials was shorter than the current recommended schedule. Although these studies do not provide direct comparison of a schedule consisting of 4 doses of PCV with a schedule using 3 doses of PCV13 and 1 dose of PPSV23 at 12 months, findings from these studies suggest that use of 4 doses of PCV is immunogenic and the booster dose might provide additional protection against the serotypes covered in the vaccine. Regarding safety of using 4 doses of PCV among HSCT recipients, local and systemic reactions occurred more frequently after the fourth dose of PCV13, compared with after the first, second, and third doses of PCV13 (194). Vaccine-related SAEs occurred across vaccine doses in one study, including facial diplegia, injection site erythema and pyrexia, autoimmune hemolytic anemia, Guillain-Barré syndrome, and cellulitis (194).

Intervals Between PCV and PPSV23

Seven immunogenicity studies evaluating the immune response after a sequence of PCV13 or PCV15 followed by PPSV23 administered at intervals of 2, 6, or 12 months or 3–4 years were reviewed (165–167,177,179,195,196). Four were among adults aged ≥50 years without immunocompromising conditions (167,177,179,195). Three included younger adults (aged ≥18 years) who are considered to be at increased risk for pneumococcal disease: one among American Indian and non-American Indian adults with chronic medical conditions (165), one among adults with HIV infection (166), and one among adults with rheumatoid arthritis receiving biologic disease-modifying antirheumatic drugs (196). In addition to these seven studies, one PCV7 study that compared different intervals between PCV7 and PPSV23 among Alaska Native adults aged 55–70 years was included (197). Of eight studies, three that compared intervals ranging from 2 to 6 months between administration of PCV and PPSV23 in the same study found no significant difference in immunogenicity measured after PPSV23 receipt, although reactogenicity tended to be higher with shorter intervals (195–197). In a study conducted among pneumococcal vaccine–naïve adults aged 60–64 years that compared antibody responses to 1 dose of PCV13 with responses to PCV13 followed by PPSV23 1 year apart, the immune responses after PPSV23 were significantly lower compared with the responses after 1 dose of PCV13 for eight of 12 common serotypes (177). In another study among pneumococcal vaccine–naïve adults aged ≥50 years who received either PCV13 or PCV15 followed by PPSV23 1 year apart, immune responses (measured as GMFR in IgG concentrations and a greater than fourfold rise in OPA titers 1 month after vaccination compared with prevaccination) to PPSV23 were typically comparable to those measured 1 month after 1 dose of PCV for the serotypes contained in the PCVs (167). In another study that was conducted as an extension of a previous study in pneumococcal vaccine–naïve adults aged 50–64 years, antibody response to 1 dose of PCV13 was compared with responses to PCV13 followed by PPSV23 approximately 3.5 years apart (179). Immune responses after PPSV23 were significantly higher for seven of 12 common serotypes (179). These findings suggested that longer intervals between administration of PCV and PPSV23 might improve immunogenicity in immunocompetent adults, although a direct comparison between a 1- versus 4-year interval was not made.

Coadministration With Other Vaccines

Concomitant administration of PCV15, PCV20, or PPSV23 with seasonal inactivated quadrivalent influenza vaccine (specifically, PCV15 or PPSV23 and IIV4 [Fluarix] and PCV20 and adjuvanted IIV4 [Fluad]) has been demonstrated to be immunogenic and safe in adults (174,198,199). However, numerically lower pneumococcal serotype–specific OPA GMTs or GMCs were reported for certain serotypes when pneumococcal vaccines were coadministered with IIV4 compared with administration of pneumococcal vaccines alone.

Among adults aged ≥65 years who received 2 doses of BNT162b2 COVID-19 vaccine with the second BNT162b2 dose administered ≥6 months previous, slightly lower pneumococcal serotype–specific OPA GMTs were reported when PCV20 was coadministered with BNT162b2 compared with administration of PCV20 alone; however, this difference was not statistically significant. Pain at injection site within 10 days after vaccination and systemic events within 7 days after vaccination were reported more frequently among the group that received both PCV20 and BNT162b2 compared with the group that received PCV20 alone. The percentage of participants who reported any adverse events through 1 month after vaccination was similar across groups (PCV20 and BNT162b2 coadministration: 5.3% [95% CI = 2.6%–9.6%]; PCV20 only: 4.3% [95% CI = 1.9%–8.3%]; BNT162b2 only: 6.5% [95% CI = 3.4%–11.1%]) (200). No data are available on PCV20 coadministration with other COVID-19 vaccines. Evaluation of coadministration of PCV15 or PPSV23 with COVID-19 vaccines among adults aged ≥50 years is ongoing (201). In adults aged ≥50 years who received 2 doses of adjuvanted recombinant zoster vaccine (RZV), immune responses to coadministration of RZV and PPSV23 were noninferior to those in this age group who received these vaccines in sequence. Solicited adverse reactions were reported more frequently when the first doses of RZV and PPSV23 were coadministered (202). No data are available on coadministration with other vaccines (e.g., tetanus, diphtheria, acellular pertussis vaccine, or hepatitis B vaccine) in adults.