An Evaluation of the Safety and Immunogenicity of a Recombinant Protein-Based Pneumococcal Vaccine in ICR Mice and Cynomolgus Macaque Models

TL;DR

A new protein-based pneumococcal vaccine was tested in mice and monkeys and showed good safety and strong immune responses, suggesting it could offer broad protection against pneumococcal diseases.

Contribution

The study introduces a novel recombinant protein-based pneumococcal vaccine with broad coverage and demonstrates its safety and immunogenicity in animal models.

Findings

The vaccine showed no significant toxicity in ICR mice and cynomolgus monkeys.

Both low- and high-dose groups in monkeys had increased IgG titers and opsonophagocytic killing activity against pneumococcal strains.

Neutralizing antibody titers against pneumolysin were significantly elevated post-vaccination in monkeys.

Abstract

Background: Pneumococcal diseases remain a global threat due to the serotype-specific limitations of polysaccharide vaccines. This study evaluated a recombinant protein-based pneumococcal vaccine (PBPV) combining three PspA variants (PRX1/Family1Clade2, P3296/Family2/Clade3, P5668/Family2/Clade4) and detoxified pneumolysin (PlyLD). PspA targets conserved surface epitopes to block immune evasion and achieve broad coverage, while PlyLD neutralizes pore-forming toxins and enhances adaptive immunity. Methods: We evaluated the safety and immunogenicity of the PBPV in animal models. Acute toxicity studies were conducted by administering a single intramuscular injection to ICR mice, whereas chronic toxicity and immunogenicity studies were performed in cynomolgus monkeys via repeated intramuscular injections, with an equal number of male and female animals in both groups. Immune responses were assessed using ELISA, multiplexed opsonophagocytic killing assays (MOPAs), and neutralizing antibody assays. Results: Acute toxicity studies in ICR mice showed no signs of abnormal toxicity or irritation at one-dose levels. In the chronic toxicity study, cynomolgus monkeys received repeated intramuscular injections once every 3 weeks for a total of four administrations, at doses of one dose/monkey and five doses/monkey, followed by a 4-week recovery period. No significant systemic toxic reactions were observed, and the safe dose was determined to be five doses/monkey. In the immunogenicity study of monkey serum, both low-dose and high-dose groups demonstrated significant increases in antigen-specific IgG titers against each component; opsonophagocytic killing activity against pneumococcal strains from Clades 2, 3, and 4 from PspA Families 1 and 2; and neutralization antibody titers against pneumolysin post-vaccination. Conclusions: The recombinant protein-based pneumococcal vaccine exhibited a favorable safety profile and potent immunogenicity in animal models, indicating promise for broad protection against pneumococcal disease. These findings support the further development of PBPVs as a viable alternative to conventional polysaccharide-based vaccines.

Figures

Full text

1. Introduction

Streptococcus pneumoniae is one of the most globally important pathogenic bacteria [1,2,3]. In 2023, it caused approximately 634,000 deaths—accounting for 25.3% of global lower respiratory tract infection (LRTI) deaths and ranking first among pathogens in mortality [4]—and imposed a particularly heavy disease burden on children under 5 years of age and adults aged 65 years and older [5,6]. In China, the incidence of invasive pneumococcal disease (IPD) in children under 5 years of age ranks among the highest globally, primarily driven by serotypes 19F, 19A, and 3 [7]. Currently, vaccination with pneumococcal conjugate vaccines (PCVs) is an effective strategy for preventing IPD in children and adults [8,9].

Current vaccines primarily include the 23-valent polysaccharide vaccine (PPV23) and pneumococcal conjugate vaccines (PCV7, PCV10, PCV13, PCV15, PCV20, PCV21). PCVs induce T-cell-dependent immunity via carrier proteins, reducing vaccine-type invasive pneumococcal disease (IPD) and eliciting herd immunity [10,11]. However, they face several challenges: limited serotype coverage (with over 100 known serotypes [12]), serotype replacement [13], and an increase in adult cases caused by vaccine-type serotypes—suggesting insufficient indirect protection [14]. Additionally, non-vaccine serotypes and some persistent vaccine serotypes (some associated with antibiotic resistance) remain pathogenic [15].

To overcome the aforementioned serotype-dependent bottleneck, current research is focusing on next-generation strategies represented by conserved bacterial protein antigens [16], DNA/mRNA technology platforms [17], virus-like particles (VLPs) [18], and live attenuated vaccines [19]. Among these, protein vaccines target conserved antigens such as PspA, PhtD/PhtE, dPly, and PcpA [10,20], aiming to provide serotype-independent protection; some candidates, like PnuBioVax, have entered early-stage clinical trials [21]. Meanwhile, novel conjugation technology platforms—including bioconjugation [22], non-covalent conjugation (MAPS) [23], and outer membrane vesicles (OMVs) [24,25]—are aimed at maintaining immunogenicity while simplifying manufacturing and reducing costs. The core objective of these strategies is to transcend serotype limitations, achieve broader or even universal protection [26], and potentially block colonization and transmission at the source through inducing mucosal immunity. Although these technologies demonstrate potential for superior breadth and accessibility compared to conventional vaccines, their long-term safety, immunological durability, and feasibility of large-scale production require further validation.

Among numerous protein vaccine candidate antigens, Streptococcus pneumoniaesurface protein A (PspA) has attracted considerable attention due to its high conservation across pneumococcal strains and critical role in immune evasion by inhibiting complement deposition on the bacterial surface [27,28,29,30,31]. Based on sequence variations, PspA is classified into three families and six clades, with most invasive strains belonging to families 1 and 2, especially clades 2, 3, and 4 [31,32,33]. Significant cross-reactivity between clades within these families further supports the inclusion of PspA in polyvalent protein vaccine (PBPV) formulations to maximize coverage [28,34]. Additionally, pneumolysin (Ply)—another highly conserved protein—is frequently incorporated into vaccine designs to enhance immunogenicity and protective efficacy [29]. Combining these key conserved proteins into multivalent formulations holds promise for providing a more broad-spectrum and universal solution to address the pneumococcal disease burden [30,34].

In this study, we report a novel universal pneumococcal protein vaccine comprising three rationally designed, complementarily distributed PspA antigens (PRX1, P3296, P5668) and detoxified pneumolysin (dPly). This design not only overcomes the protective limitations of a single protein due to high sequence diversity via a multi-allelic PspA coverage strategy but also achieves synergistic dual mechanisms—adhesion blockade and virulence inhibition—through dPly’s toxin neutralization and natural adjuvant effects [30,35]. Compared to other technological platforms, this vaccine achieves a favorable balance between immunogenicity and safety, features a relatively simple manufacturing process, and offers superior cost-effectiveness, making it particularly suitable for vulnerable populations such as the elderly and immunocompromised individuals. Here, we systematically evaluated the preclinical safety and immunogenicity of this candidate vaccine in mouse and cynomolgus monkey models (acute toxicity tests in mice are in Supplementary Materials). The data aim to support IND application and subsequent clinical trial design, provide critical evidence for the design and initiation of human clinical trials, and explore feasible technical pathways for developing next-generation efficient and accessible pneumococcal vaccines.

2. Materials and Methods

2.1. Materials

2.1.1. Vaccine Preparation

The production strain for the recombinant pneumococcal surface protein is an E. coli strain (BL21) carrying the gene encoding the P3296 protein on a plasmid. The same applies to P5668, RX1, and Ply.

Cells from a frozen stock vial were spread onto an agar plate and incubated at 37 °C for 1820 h. A bacterial lawn was scraped from the plate to prepare a cell suspension, which was used to inoculate a shake flask containing 100 mL of culture medium. A 2 L fermenter was inoculated with a 2% (v/v) inoculum derived from this seed culture. When the culture OD600 reached 15.020.0, isopropyl β-D-1-thiogalactopyranoside (IPTG) (Sangon Biotech, Shanghai, China) was added to a final concentration of 1 mM for induction.

Bacterial cells from the fermentation culture were harvested by centrifugation and resuspended in PB buffer. The cells were then disrupted by high-pressure homogenization, and the soluble expressed antigen was collected. Impurity profiles, including host cell protein, host cell DNA, and endotoxin, were subsequently removed through sequential anion-exchange and cation-exchange (Cytiva, Marlborough, MA, USA) chromatography steps. Finally, the protein solution was subjected to buffer exchange ultrafiltration using a 10 kDa membrane cassette with 20 mM Tris (Affymetrix, Santa Clara, CA, USA) 150 mM NaCl buffer (pH 7.5). Subsequently, the ultrafiltered solution was subjected to sterile filtration through a 0.2 μm filter.

The quality control standards specify that (1) the purity of the purified protein shall be ≥90% as determined by SDS-PAGE; (2) the endotoxin content shall be <125 EU/mg; and (3) the residual host cell DNA and host cell protein shall be ≤50 ng/mg and ≤500 ng/mg, respectively.

The active ingredients of the PBPV were 50 µg each of PspA RX1, PspA3296, PspA 5668, and PlyLD. The recombinant proteins were fermented, purified, mixed, and prepared by adding an aluminum hydroxide adjuvant for adsorption. Each vaccine dose was formulated with 0.7 mg of aluminum hydroxide adjuvant.

2.1.2. Protein Analysis by SDS-PAGE

6× Protein Loading Buffer (Transgen, Beijing, China) was added to each Eppendorf tube containing 10 μg of protein sample. The tubes were vortexed and centrifuged. Then, 12% Protein Gels (1.5 mm, 10 wells) were used to analyze the samples, using a 1× Running Buffer (5× Running Buffer: 15.1 g of Tris, 94 g of glycine, and 5 g of SDS were weighed and dissolved with ultrapure water, and then diluted to 1000 mL. It was diluted with water to 1× running buffer before testing.) and a prestained protein ladder (Color Prestained Protein Standard, Broad Range (10–250 kDa)) (NEW England Biolabs, Ipswich, MA, USA). The gels were stained with the Coomassie Brilliant Blue Staining Solution for 60 min, and the stained gel was briefly rinsed with water to remove excess staining solution on the surface. Coomassie Stain Destaining Solution was added to submerge the gel. The destaining solution was replaced until the protein bands were clear and the background of the gel became transparent.

2.1.3. Animal Models

Four- to six-week-old ICR mice, weighing 1724 g, sourced from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China), were used in acute toxicity studies. Two- to five-year-old cynomolgus monkeys, weighing 2.55 kg, procured from Guangxi Guidong Primate Experimental Development Co., Ltd. (Hezhou, China), were employed in both chronic toxicity and immunogenicity studies.

2.1.4. Bacteria Strains, Cells, and Complements

Five Spn strains—TR6A, OP4, TR11A, ST14, and OP17F—were acquired from BEI Resources; the other strain, F38, was gifted by Tsinghua University. The HL60 cell line was procured from the American Type Culture Collection (ATCC), and complements were prepared by CanSino Biologics (Tianjin, China).

2.2. Methods

2.2.1. Safety Analysis

Repeated-Dose Toxicity Studies

In this experiment, also conducted by the JOINN (Beijing) New Drug Research Center Co., Ltd., forty cynomolgus monkeys were used in the study, randomly assigned to four groups (n = 10 per group) with equal numbers of males and females: negative control group, adjuvant control group, low-dose group, and high-dose group. The negative control group received sodium chloride injections at 2.5 mL/animal/dose; the adjuvant control group received aluminum hydroxide adjuvant (0.7 mg per dose) at 2.5 mL/animal/dose. The low- and high-dose groups were administered the recombinant pneumococcal protein vaccine (each dose containing 50 μg of each of the four antigens—PRx1, P3296, P5668, and PlyLD—plus 0.7 mg of aluminum hydroxide adjuvant) at 1 dose/animal and 5 doses/animal, respectively. Vaccines were given via intramuscular injection once every 3 weeks for a total of 4 administrations over 9 weeks, followed by a 4-week recovery period. The schematic of the experimental design is presented in Figure 1.

During the study, animals underwent clinical observations, body weight measurements, body temperature measurements, electrocardiograms (ECGs), hematology, coagulation function tests, blood biochemistry tests, urinalysis, ophthalmic examination, T lymphocyte subset distribution (CD3+, CD3 + CD4+, CD3 + CD8+, and CD3 + CD4+/CD3 + CD8+ ratio) tests, and serum cytokine (IL-2, IL-4, IL-5, IL-6, TNF-α, and IFN-γ) measurements. Three days after the last dose (Day 67), the first three animals of each sex in every group were euthanized; the remaining animals in each group were euthanized 4 weeks after dosing cessation (Day 92). All animals underwent gross necropsy, with major organs weighed to calculate relative organ weights, and histopathological examinations performed on over 40 types of organs and tissues.

2.2.2. Immunogenicity Analysis

Cynomolgus monkeys’ serum samples collected pre-immunization, as well as at 21 days post-first and post-second immunization in the chronic toxicity study, were used for immunization assessments and immune response evaluation.

Enzyme-Linked Immunosorbent Assay (ELISA)

The specific IgG antibody titers against P5668, P3296, PRX1, and PlyLD in cynomolgus monkeys’ serum samples were measured using an indirect enzyme-linked immunosorbent assay (ELISA). The protocol is exemplified below using P5668 as the antigen: purified P5668 protein was coated onto 96-well plates (Corning, NY, USA) at 5 µg/mL in coating buffer (100 µL/well) and incubated overnight at 4 °C. After discarding the coating antigen, plates were thoroughly washed. Each well was then blocked with 200 µL of 1% bovine serum albumin (BSA) at 37 °C for 1 h. Serially diluted test serum samples were added to the antigen-coated plates and incubated at 37 °C for 1 h. Following incubation, horseradish peroxidase (HRP)-conjugated secondary antibody (Nordic-MUbio, Susteren, The Netherlands) was added. Plates were thoroughly washed again before adding 3,3′,5,5′-tetramethylbenzidine (TMB) substrate for color development. The optical density (OD) values were measured at 450 nm using a microplate reader. The cutoff value for determining sample titers was set at 2.1 times the average OD value of the negative control. The detection protocols for specific IgG antibodies against P3296, PRX1, and PlyLD followed the same procedure as described for P5668.

Multiplexed Opsonophagocytic Killing Assay (MOPA)

To assess the PBPV-induced serum bactericidal activity, the Multiplexed Opsonophagocytic Killing Assay (MOPA) was performed as previously described with modifications [36,37,38].

For the MOPA, the bacteria were opsonized with immune sera (sera were heat-inactivated by incubation in a 56 °C water bath for 30 min) and complement (rabbit complement) for 30 min, followed by incubation with HL60 cells for 75 min at 37 °C with 5% CO_2_. A three-fold dilution series of serum samples was used to evaluate the bactericidal activity, and the reaction mixture was spotted onto Todd-Hewitt Yeast Agar (THYA) plates. After overnight incubation, the number of colonies formed was counted to assess bacterial killing. The opsonic index for each sample was calculated by determining the dilution of serum that killed 50% of the bacteria using a linear interpolation algorithm.

The PBPV antigens include PspA, which was designed based on the PspA types. PspA is classified into three families and six distinct clades. Consequently, the selection of target strains for the MOPA was based on the PspA types, with each clade represented by two different bacterial strains, each corresponding to a distinct serotype. The strains used in the assay included Streptococcus pneumoniae serotypes 6A and F38 (Family 1 Clade 2), 4 and 11 A (Family 2 Clade 3), and 14 and 17F (Family 2 Clade 4). This approach ensures that the assay evaluates bactericidal activity against a broad spectrum of pneumococcal strains, thus providing a comprehensive evaluation of the vaccine’s effectiveness across different PspA types.

Neutralizing Antibody Assays

A standardized operating procedure was developed for the neutralizing antibody assay against pneumococcal pneumolysin as described previously [39]. Briefly, 2% rabbit erythrocytes were utilized as indicators of hemolysis. Serum samples were pre-incubated with Ply prior to the addition of erythrocytes, enabling assessment of the serum’s ability to neutralize Ply-mediated hemolytic activity. The optical density at 450 nm (OD450) was measured, and the serum dilution yielding 50% inhibition of hemolysis was determined and designated as the neutralizing antibody titer. This titer was subsequently used to evaluate vaccine-induced immunity. All procedures were performed in strict accordance with the established SOP to ensure data accuracy and reproducibility.

2.3. Statistical Analysis

The analytical methods for the long-term toxicity experiment data are as follows: Levene’s test was used to assess the homogeneity of variance. If no statistically significant difference was observed (p > 0.05), one-way analysis of variance (ANOVA) was performed for statistical analysis. If ANOVA revealed a statistically significant difference (p ≤ 0.05), Dunnett’s test (a parametric method) was applied for comparative analysis. In cases of heterogeneity of variance (p ≤ 0.05), data were log-transformed (ln transformation), followed by a repeat Levene’s test. If the transformed data showed restored homogeneity of variance (p > 0.05), one-way ANOVA was conducted; if ANOVA still indicated significance (p ≤ 0.05), Dunnett’s multiple comparisons (parametric method) were performed. If the log-transformed data remained heterogeneous in variance (p ≤ 0.05), the original data were analyzed using the Kruskal–Wallis nonparametric test. A significant result (p ≤ 0.05) from the Kruskal–Wallis test prompted further pairwise comparisons via the Mann–Whitney U test. Log transformation was omitted if the data contained negative values. When zeros were present, they were replaced with one-tenth of the smallest positive value in the dataset before log transformation. For animal studies, no statistical analysis was carried out if the sample size (n) was less than 3.

Antibody titers are expressed as geometric mean titers (GMTs) with 95% confidence intervals (95% CIs). After logarithmic transformation using GraphPad software (version 8), statistical analyses and comparisons were conducted. When analyzing three or more groups, one-way ANOVAs (and nonparametric or mixed models) were used. Significant differences between groups were considered at p-values of 0.05, 0.01, and 0.001.

3. Results

3.1. Expression and Purification of Recombinant Pneumococcal Antigens

The expression of recombinant antigens P3296, P5668, PRX1, and PlyLD induced with 1 mM IPTG is shown in Figure 2. SDS-PAGE analysis indicated that the purity of each purified antigen exceeded 90%.

3.2. Safety Results

Repeated-Dose Toxicity Study Outcomes

During the study period, no mortality or moribundity was observed in any group. Clinical observations, body weight, body temperature, electrocardiography, ophthalmic examinations, urinalysis, and T lymphocyte subset analysis (Supplementary Table S1) revealed no treatment-related abnormalities across all groups.

Hematological analysis (key results are presented in Table 1; all results are in Supplementary Table S2) showed intergroup variations in mean values compared to concurrent same-sex controls, with statistically significant differences (p ≤ 0.05) limited to specific time points: a 57.7% increase in platelet count (PLT) in high-dose males at 3 days post-third dose; 35.3–47.4% decreases in reticulocyte count (Retic, both absolute and percentage) in adjuvant control, low-, and high-dose females at 3 days post-third dose; and 10.0% decreases in hemoglobin (HGB) alongside 114.3% increases in eosinophil count (Eos) in high-dose males and females at 3 days post-last dose. These changes were interpreted as non-toxicologically significant, as Eos elevation was immune-related but fully resolved during recovery, PLT elevation in high-dose males reflected individual variability without abnormal trends or coagulation issues, and Retic/HGB fluctuations fell within historical control ranges (HGB: 108.0–147.7 for females, 107.5–144.6 for males; Retic: 0.024–0.148 for females, 0.024–0.152 for males; Retic%: 0.36–2.86 for females, 0.33–2.99 for males) with no dose/time correlation.

Coagulation function assessment (Supplementary Table S3) identified significant differences (p ≤ 0.05) including 44.9% prolongation of activated partial thromboplastin time (APTT) in high-dose males at 3 days post-first dose (attributed to a single outlier with no subsequent abnormalities), 7.3% prolongation of prothrombin time (PT) in low-dose females at 3 days post-third dose, and 12.4% PT prolongation in low-dose females plus 25.8% APTT prolongation in high-dose females at 3 days post-last dose; all other changes were minimal, within historical controls (PT: 8.66–11.08 for females, 8.55–11.01 for males; APTT: 20.04–51.31 for females, 18.39–50.78 for males), and lacked dose/time correlation.

Blood biochemistry evaluation (key results are presented in Table 2; all results are in Supplementary Table S4) showed all parameters within historical ranges without consistent toxic patterns, with significant differences (p ≤ 0.05) limited to 2.7–4.6% increases in Na^+^/Cl^−^ in low-/high-dose females post-first dose, a 7.2% decrease in total protein (TP) in adjuvant control males post-third dose, and 2.9–3.2% Cl^−^ increases in low-/high-dose females post-last dose—all minor, non-dose/time-correlated, and consistent with normal hematocrit. Transient ALT/AST/CK elevations pre-dose or at other time points were attributed to handling/phlebotomy stress without hepatic/cardiac pathology.

Cytokine analysis (Table 3) revealed no treatment-related abnormalities; IL-2, IL-4, IL-5, and IFN-γ remained below quantification limits, while TNF-α changes (e.g., decreases in adjuvant control/low-dose males, increases in high-dose females post-first dose; decreases in low-dose females/males and high-dose females post-third dose) lacked dose/time correlation and clinical association, and a single pre-dose IL-6 elevation in negative controls was an outlier.

Pathological examination (Supplementary Table S5) showed no toxicologic trends in organ weights/ratios; significant differences (e.g., reduced heart weight in adjuvant control females, increased uterine weight/ratio in high-dose females) lacked dose correlation and histopathologic support. Grossly, no treatment-related lesions were noted at terminal sacrifice or recovery. Microscopically, mild-to-moderate granulomatous inflammation with macrophages was observed at injection sites in 6/6 adjuvant control, 5/6 low-dose, and 6/6 high-dose animals at 3 days post-last dose, with 1 adjuvant control and 4 high-dose animals showing myofiber necrosis/inflammation (consistent with aluminum adjuvant-induced local irritation), persisting partially at 4-week recovery; no treatment-related organ toxicity was found elsewhere. Overall, under study conditions, the vaccine demonstrated an acceptable subchronic safety profile with no significant toxicological concerns.

3.3. Immunogenicity Results

3.3.1. PBPV-Elicited Serum Antibody Responses

The anti-PRX1, anti-P3296, anti-5668, and anti-PlyLD IgG titers in the sera of cynomolgus macaques were measured with Endpoint-ELISA before vaccination (Day 1) and 3 weeks after each of the first two vaccinations (Days 22 and 43) in the low-dose and high-dose groups, and the results are summarized in Figure 3.

Following the first immunization, serum IgG antibody titers against P3296, P5668, PRX1, and PlyLD antigens increased significantly compared to pre-vaccination levels in both the low-dose and high-dose groups. For example, the geometric mean titer (GMT) of anti-P3296 rose from approximately 205 to 42,224 in the low-dose group and from 492 to 32,000 in the high-dose group, with both increases being statistically significant (p < 0.001). After the second immunization, antibody titers in both groups did not increase further but remained at high titers. The results indicate that the vaccine effectively elicited a specific IgG response upon primary immunization. Moreover, no significant difference in post-vaccination antibody titers was observed between the high- and low-dose groups, suggesting that the initial low-dose immunization sufficiently activated the immune system and induced a plateau in antibody production; therefore, administration of a higher booster dose did not significantly enhance antibody levels.

3.3.2. PBPV-Elicited Serum Opsonophagocytic Killing Activity

In both the low- and high-dose groups, serum opsonophagocytic activity (OPA) was measured in cynomolgus macaques before vaccination (Day 1) and 3 weeks after the first and second immunizations (Days 22 and 43), as shown in Figure 4. Except for serotype 38, which showed only a non-significant upward trend, OPA titers against all other tested strains increased significantly after two immunizations compared to pre-vaccination levels.

Specifically, for PspA Family 1 Clade 2, serotype 6A showed significantly elevated titers after both vaccinations in the low-dose group, and in the high-dose group, titers were also significantly higher than baseline after the second vaccination (p < 0.05); serotype 38 showed an upward trend but without statistical significance. In PspA Family 2 Clade 3, serotype 4 exhibited a marked immune memory effect in both groups, with titers increasing significantly after the first vaccination (p < 0.05) and remaining high after the second; serotype 11A showed a similar response pattern, though titers declined after the second vaccination in the high-dose group while remaining significantly above pre-vaccination levels (p < 0.05). In PspA Family 2 Clade 4, serotype 14 increased significantly in both groups, with a slight decrease after the second vaccination in the high-dose group though still significantly higher than baseline (p < 0.05); serotype 17F increased significantly only after the second vaccination in the low-dose group, whereas in the high-dose group, it was significantly elevated after both vaccinations compared to baseline (p < 0.05).

In summary, no clear dose-dependent difference in immunogenicity was observed between the low- and high-dose groups across serotypes, and the second immunization did not further enhance OPA titers, consistent with the binding antibody results. These findings indicate that the vaccine effectively induces opsonophagocytic activity against representative serotypes of PspA Families 1 and 2, suggesting its potential to elicit broad cross-protective immune responses.

3.3.3. Serum Neutralization Antibody Titers Against Pneumolysin

Pneumolysin, a major virulence factor of Streptococcus pneumoniae, is known to facilitate immune evasion. To evaluate the functional neutralizing capacity of the vaccine, serum neutralizing antibody (nAB) titers against pneumolysin were measured. The results are summarized in Figure 5.

In the low-dose group, cynomolgus macaques received 50 µg of each antigen (P3296, P5668, PRX1, and PlyLD) on Days 1 and 22. Baseline nAB titers were low, with a geometric mean titer (GMT) of 12.7. Following the first and second immunizations, GMTs increased to 132.3 and 155.8, respectively. Statistical analysis showed a significant rise in nAB titers after vaccination (p < 0.0001), indicating robust immune activation.

In the high-dose group, animals were administered 250 µg of each antigen on the same schedule. The pre-vaccination GMT was 8.0, which increased to 192.7 after the first dose and declined slightly to 146.9 after the second dose. Similar to the low-dose group, no further statistically significant increase was observed after the second immunization compared with the first.

Based on the results shown in the figure, no significant difference in immunogenicity was observed between the low-dose and high-dose groups across the serotypes tested, nor did the second vaccination further enhance immunogenicity within either dosage group. This observation is also consistent with the results obtained for binding antibodies.

4. Discussion

Globally, many countries have included pneumococcal conjugate vaccines (PCVs) in their national immunization programs (NIPs), effectively reducing the burden of invasive diseases caused by vaccine-type serotypes through herd immunity [40,41]. However, as vaccine-serotype transmission is suppressed, serotype replacement is emerging, while the risk of infection by non-encapsulated Streptococcus pneumoniaeis increasingly prominent—highlighting the importance of developing novel vaccines with broader protective efficacy [42]. Currently, although the development of multivalent conjugate vaccines continues, it still faces challenges such as carrier-mediated epitope suppression and variability in immune responses among different serotypes [20].

Among numerous protein-based vaccine strategies, Streptococcus pneumoniaesurface protein A (PspA) has emerged as one of the most studied targets for broad-spectrum pneumococcal vaccine research due to its high conservation and good immunogenicity [33,34,43,44,45]. However, significant variation in PspA across clades restricts the capacity of single-variant vaccines to provide comprehensive protection against strains spanning clades [31,33]. To this end, this study evaluated an innovative pneumococcal protein vaccine (PBPV) developed by CanSino Biologics, whose design incorporates two major PspA families covering three distinct clades to achieve broad protection against >98% of prevalent strains [32,46]. The vaccine also integrates pneumolysin (Ply), a key virulence factor, utilizing its neutralizing activity to further enhance cross-protection. Unlike traditional vaccines relying on capsular serotypes, this PBPV is constructed based on PspA clades, avoiding the coverage verification challenge associated with serotype complexity (known >100 types); with only six major PspA clades currently identified, it enables systematic evaluation of protection breadth. The vaccine has completed Phase Ia clinical trials, and its safety and immunogenicity data provide an important basis for further clinical development [44].

This study conducted a comprehensive preclinical assessment of the safety and immunogenicity of a recombinant pneumococcal protein vaccine. Results demonstrated that the vaccine exhibited acceptable safety and good tolerability, and induced robust immunogenicity, eliciting significant protective immune responses. The toxicology study design followed established principles, using a high-dose group to elicit detectable toxic reactions for assessing the safety margin [47,48]. For dose selection in the chronic toxicity study, in accordance with the recommendations of the General Principles for Technical Review of Preclinical Safety Evaluation of Biological Products for Prevention (China) [49] and the WHO Guidelines for Nonclinical Evaluation of Vaccines [50], toxicology studies preferably use the highest anticipated human absolute dose of the final adjuvanted vaccine formulation intended for clinical trials. Following this recommendation, a high-dose group (2.5 mL) was established in the repeated-dose study to establish an adequate safety margin, assess potential toxic effects, and determine a scientifically sound safe dose for the first-in-human clinical trial. The chronic toxicity study further confirmed the vaccine’s safety at the maximum tolerated dose (MTD) of 2.5 mL, with no systemic toxicity detected.

Notably, in the repeated-dose toxicity study, local pathological changes—including myofiber necrosis with inflammatory cell infiltration and/or macrophage granulomas of mild to moderate severity—were observed at the injection site 3 days after the last dose. These findings were seen in animals in the adjuvant control, low-dose, and high-dose vaccine groups, consistent with previously reported local reactions following intramuscular administration of aluminum-adjuvanted vaccines [51,52], suggesting that close monitoring of injection site reactions should be prioritized in subsequent clinical trials to further ensure vaccine safety.

In addition to routine safety evaluations, this study comprehensively analyzed the vaccine’s immunogenicity using serum samples—including antigen-specific antibody levels, opsonophagocytic activity, and toxin-neutralizing capacity—to provide a basis for subsequent clinical studies. Results showed that both low- and high-dose groups induced robust immune responses: specific IgG titers for each vaccine component significantly increased; potent opsonophagocytic killing was observed against multiple pneumococcal strains covering different PspA families and clades; and neutralizing antibody titers against pneumolysin (Ply) also markedly increased. Based on the physiological and immunological relevance of cynomolgus monkeys to humans, these results enhanced the feasibility of clinical translation and supported the development of this vaccine as a potential alternative to existing pneumococcal polysaccharide vaccines [53].

Building on the experience of Philippe Denoël et al., in their 2011 study using a rhesus monkey model of the PhtD-dPly vaccine—which showed that the survival rate post-challenge correlated with anti-PhtD/Ply antibody levels [54]—we next plan to conduct animal challenge studies to directly validate the protective efficacy of the vaccine against pneumococcal infection, and proceed with the systematic evaluation of its clinical safety and efficacy. Meanwhile, to further refine the vaccine evaluation framework, we will expand the strain range in immunogenicity assessments: the current study covers only clades 2, 3, and 4 of PspA families 1 and 2, followed by supplementation of representative strains from all clades (including family 3 and clade 6, which are clinically relatively rare); additionally, we selected only two serotypes per clade as indicator strains, and we are now exploring the introduction of more diverse serotypes (including PCV-covered and non-covered serotypes) per clade to systematically evaluate the vaccine’s protective efficacy against cross-serotypes and cross-clades.

5. Conclusions

In conclusion, this study strongly demonstrates that the recombinant protein-based pneumococcal vaccine exhibits favorable safety and induces robust immune responses in mice and cynomolgus monkeys. Compared with existing polysaccharide-based vaccines, a vaccine strategy combining multiple PspA protein clades with pneumolysin holds promise for providing broader protection against pneumococcal diseases.