Immunogenicity of an Escherichia coli-produced bivalent human papillomavirus vaccine under different vaccination intervals

ABSTRACT

A new Escherichia coli-produced human papillomavirus (HPV)-16/18 vaccine has been shown to be safe and highly efficacious and was recently licensed in China. As a post hoc analysis of the phase III trial, this study aimed to assess the impact of vaccination time deviations on the specific antibody response and guide the better usage of this vaccine in the real world. A total of 3689 healthy women aged 18–45 years old were randomly assigned to receive the bivalent HPV-16/18 vaccine according to a 0-, 1- and 6-month schedule with a wide vaccination interval. The first vaccination interval between the 1^st and 2^nd doses (the 1^st interval) was divided into three groups: 28–40 d, 41–50 d and 51–60 d. The second vaccination interval between the 2^nd and 3^rd doses (the 2^nd interval) was divided into three groups: 103–139 d, 140–160 d and 161–198 d. The reverse cumulative curves for the IgG of the three groups with different 1^st vaccination intervals or with different 2^nd vaccination intervals at month 7 almost overlapped for both HPV-16 and HPV-18. Compared with the standard vaccination schedule (a 1^st interval of 28–40 d and a 2^nd interval of 140–160 d) subgroup, all the subgroups had GMC ratios greater than 0.83, with the lower limit of 95% CIs higher than 0.64. In conclusion, a slight deviation in the vaccination time of the 2^nd and 3^rd doses has only a minor, insignificant impact on the immune response induced by the Escherichia coli-produced HPV-16/18 vaccine.

KEYWORDS:

• Human papillomavirus vaccine
• Escherichia coli
• bivalent
• vaccination interval
• immunogenicity
• antibody response

Introduction

Human papillomavirus (HPV) infection is the main cause of cervical cancer, and there is also evidence linking HPV infection with cancers of the anus, vulva, vagina, penis and oropharynx.^1,2 HPV-16 and HPV-18 contribute to approximately 70.8% of cervical cancer cases.³ Three current licensed eukaryotic expressed recombinant HPV vaccines, a bivalent vaccine (Cervarix®, HPV-16/18, by recombinant baculovirus-infected insect cells), a quadrivalent vaccine (Gardasil®, HPV-6/11/16/18, by Saccharomyces cerevisiae) and a nine-valent vaccine (Gardasil® 9, HPV-6/11/16/18/31/33/45/52/58, by S.cerevisiae) which is a successor to the quadrivalent vaccine (Gardasil®), are all administered in three doses (0/1/6 or 0/2/6 months).^4–6 They have been widely introduced into the national immunization program of 104 countries. However, most people in low-income countries, who have a higher risk of cervical cancer, have reduced access to the vaccine due to the limited production capacity and high cost of these vaccines.^7,8 Post-marketing data have suggested that these vaccines result in an effective reduction in the prevalence of HPV-16/18 and related precancerous lesions.^9–11 The Director-General of the World Health Organization made a global call for action toward the elimination of cervical cancer in May 2018; however, the aforementioned high cost and limited access to vaccines have considerably slowed down the pace of implementation.^12,13

A new Escherichia coli (E. coli)-produced HPV-16/18 human papillomavirus vaccine that is well tolerated and highly efficacious against high-grade genital lesions and persistent infection with HPV-16/18^14-16 has just been licensed in China for usage in women aged 9–45 years old, with a 2-dose schedule for girls aged 9–14 years old and a 3-dose schedule for women aged 15–45 years old. Although the standard 3-dose schedule for the new HPV vaccine is to be administered at 0, 1 and 6 months, it is difficult to strictly follow this schedule in the real world, and women may delay receiving a vaccination for various reasons. Due to the high efficacy observed in clinical trials, the minimum antibody level threshold that is correlated with protection is still unclear for HPV vaccines;¹⁷ however, specific antibody levels induced by HPV vaccines are a reasonable clinical indicator for the effect of vaccination. This study aimed to analyze the difference in the specific antibody levels induced by the vaccine administered at different intervals in the phase III clinical trial to assess the impact of vaccination time deviations and guide the better usage of this vaccine in the real world.

Methods

Study design and participants

This study was a post hoc analysis of a multicenter, randomized, double-blind and placebo-controlled phase III clinical trial, which was approved by independent ethics committees. The phase III clinical trial assessed the efficacy, immunogenicity and safety of the E. coli-produced bivalent human papillomavirus vaccine (HPV-16/18). It was conducted at five sites in China: Yangcheng County in Shanxi Province, Xinmi County in Henan Province, Fengning County in Hebei Province, Funing County in Jiangsu Province and Liuzhou City in Guangxi Province. A total of 7372 women aged 18–45 years old were enrolled and randomly assigned to receive three doses of the test vaccine (HPV-16/18) or control vaccine (hepatitis E vaccine, Hecolin®). Women who were healthy and nonpregnant, had an intact cervix, did not have acute cervicitis or acute lower genital tract infection, and had not been vaccinated with an HPV vaccine were eligible. Women who had a history of sexually transmitted diseases, condyloma and abnormal cervical cancer screening results, had previously suffered from cervical intraepithelial neoplasia (CIN); or had more than 4 sexual partners were excluded. More details about the inclusion and exclusion criteria have been described previously.¹⁴

Vaccine and vaccination schedules

The HPV-16/18 bivalent vaccine, with the brand name Cecolin®, is a mixture of two aluminum hydroxide adjuvant-absorbed recombinant L1 virus-like particles (VLPs) of HPV-16 and HPV-18 expressed in E. coli. It contains 40 μg of HPV-16 L1 VLPs, 20 μg of HPV-18 L1 VLPs, and 208 μg of aluminum adjuvant, which are suspended in 0.5 mL of phosphate-buffered saline (PBS).¹⁸ The vaccine was administered at day 0, month 1 (vaccination window of 28 to 60 d after the 1^st dose) and month 6 (vaccination window of 150–240 d after the 1^st dose).

Serum collection and antibody measurement

Serum samples were collected on day 0 before vaccination and month 7 (one month after the 3^rd dose). According to the protocol, the visit window for month 7 was 21 to 89 d after the 3^rd dose. Serum samples from day 0 and month 7 were tested by the National Institute for the Control of Pharmaceutical and Biological Products using the E. coli-expressed VLP-based enzyme-linked immunosorbent assay (ELISA) with the same assay kit that was used in previous papers for the test vaccine.^14,15 The levels of IgG against HPV-16 and HPV-18 were measured with a cutoff value of 3.0 IU/mL for HPV-16 and 2.1 IU/mL for HPV-18,^14,19 which were traceable to the World Health Organization international standards for antibodies against HPV-16 (NIBSC code 05/134) or HPV-18 (NIBSC code 10/140). Each well of the microtiter plate was coated with HPV-16 and HPV-18 VLPs, which were suspended in phosphate buffer at 4°C overnight. After blocking, diluted serum samples, diluted reference serum samples and controls were added to the plates separately and incubated for 45 minutes at 37°C. Then, after the plates were washed, horseradish peroxidase-conjugated goat anti-human IgG was added and incubated under the same conditions. After the second washing, tetramethylbenzidine was added, and the samples were incubated for 15 minutes at 37°C. Finally, 0.36 mol/L H₂SO₄ was added to stop the reaction, and the optical density (OD) was read at 450/620 nm. The antibody titers against HPV-16 and HPV-18 were calculated using the diluted sample with an OD that fell within the working range of the standard curve. The titers for negative serum samples were artificially set to half of the cutoff value.¹⁹

Statistical analysis

The immunogenicity of HPV-16 and HPV-18 was analyzed in the per-protocol set (PPS) for immunogenicity (PPS-I 16 and PPS-I 18, respectively). PPS-I 16 or PPS-I 18 included participants who followed the immunization schedules to receive three doses of the HPV vaccine according to the protocol; were negative for HPV-16 IgG or HPV-18 IgG at day 0; were negative for the HPV-16 or HPV-18 type of HPV DNA at day 0 and month 7; had serum samples collected at day 0 and month 7; and had serum samples for month 7 collected between 21 and 40 d after the third dose. Modified intention- to -treat (mITT) for immunogenicity 16 and 18 (mITT-I 16 and mITT-I 18) included participants who received three doses of the vaccine and had available serum samples for day 0 and month 7.

Considering the reasonable sample size in each group, we divided the first vaccination interval between the 1^st and 2^nd doses (the 1^st interval) into the following three groups: 28–40 d, 41–50 d and 51–60 d. In addition, the second vaccination interval between the 2^nd and 3^rd doses (the 2^nd interval) was divided into the following three groups: 103–139 d, 140–160 d and 161–198 d. The standard vaccination schedule (a 1^st interval of 28–40 d and a 2^nd interval of 140–160 d) was set as the reference. The geometric mean concentrations (GMCs) of HPV-16- and HPV-18-specific IgG induced by the deviated schedules were compared to those of the reference group. The 95% confidence intervals (CIs) of the GMC ratios of IgG at month 7 were calculated by analysis of covariance (ANCOVA). As age had an important impact on the antibody response, it was analyzed as a covariate.²⁰ All analyses were conducted with SAS software (version 9.4).

Results

Baseline characteristics of the participants

A total of 3689 women aged 18–45 years old were randomly assigned to receive the bivalent HPV-16/18 vaccine, and among them, 2026 and 2444 women were included in PPS-I 16 and PPS-I 18, respectively. The seroconversion rates for anti-HPV-16 and anti-HPV-18 at month 7 were 100% (2026/2026) and 99.9% (2442/2444), respectively. The two women who did not seroconvert were 21 and 41 years old. The GMC of the HPV-16 antibody was 798.0 IU/mL (95% CI 770.9 to 882.4 IU/mL), and the GMC of HPV-18 was 271.0 IU/mL (95% CI 263.0 to 281.8 IU/mL). Among the subgroups with different vaccination intervals for both HPV-16 and HPV-18, the average ages were similar; however, younger women were prone to delaying the second and third doses (Table 1).

Table 1. Baseline characteristics of the immunogenicity subgroups (per-protocol set).

	28-40d		41-50d		51-60d		Total
2^nd interval	N	average age $\left(\bar{X}\pm S\right)$	N	average age$\left(\bar{X}\pm S\right)$	N	average age $\left(\bar{X}\pm S\right)$	N	average age$\left(\bar{X}\pm S\right)$
103-139 d	353	30.9±7.34	498	32.7±7.71	279	29.2±7.01	1130	31.3±7.55
140-160 d	288	30.3±7.16	388	30.8±7.50	24	26.3±5.42	700	30.5±7.34
161-198 d	88	28.8±7.16	106	29.4±7.25	2	27±4.24	196	29.1±7.17
Total average age	729	30.4±7.26	992	31.6±7.66	305	29.0±6.94	2026	30.8±7.47
HPV-18
103-139 d	428	30.7±7.46	608	32.3±7.61	323	29.2±7.15	1359	31.1±7.56
140-160 d	338	30.8±7.23	480	30.7±7.42	32	26.8±5.92	850	30.6±7.33
161-198 d	114	28.8±6.99	119	29.9±7.53	2	23.5±0.71	235	29.3±7.26
Total average age	880	30.5±7.33	1207	31.5±7.58	357	28.9±7.06	2444	30.7±7.46

Antibody levels and GMC ratios compared with the standard vaccination schedule

Reverse cumulative curves for IgG in the three groups with different 1^st dose vaccination intervals or the three groups with different 2^nd dose vaccination intervals at month 7 almost overlapped, regardless of whether they were for HPV-16 or HPV-18 (Figure 1), indicating that deviations from the standard interval only had a minor, insignificant impact on the immune response.

Figure 1. Reverse cumulative curves for IgG antibody levels of the different vaccination interval groups (per-protocol set) at month 7. Panel 1A: HPV-16 IgG antibody levels of the three groups with different 1^st intervals; Panel 1B: HPV-16 IgG antibody levels of the three groups with different 2^nd intervals; Panel 1 C: HPV-18 IgG antibody levels of the three groups with different 1^st intervals; Panel 1D: HPV-18 IgG antibody levels of the three groups with different 2^nd intervals.

The GMCs from the different subgroups and their GMC ratios compared with the standard vaccination schedule (a 1^st interval of 28–40 d and a 2^nd interval of 140–160 d) are presented in Table 2. The data showed that compared with the values of the reference subgroup, the lowest GMC ratios of the other groups were 0.86 and 0.83 for HPV-16 and HPV-18, respectively. All the lower limits of the 95% CIs for the GMC ratios were greater than 0.64, indicating that receiving the second dose at 28 to 60 d after the first dose and receiving the third dose at 103 to 198 d after the second dose had minor, insignificant impacts on the immunogenicity of the vaccine for both HPV-16 and HPV-18. The subgroup with a 1^st interval of 51–60 d and a 2^nd interval of 161–198 d was not analyzed as it only included 2 people. Analyses of the mITT cohorts showed similar results (Supplementary Table 1).

Table 2. Analysis of antibody levels and GMC ratios in the subgroups (per-protocol set).

Vaccination intervals		Anti-HPV-16			Anti-HPV-18
1^st interval	2^nd interval	n	GMC (IU/mL)	GMC ratio* (95% CI)	n	GMC (IU/mL)	GMC ratio* (95% CI)
28-40d	140-160d	288	883.7	-	338	257.5	-
28-40d	103-139d	353	767.9	0.88 (0.79,0.99)	428	284.2	1.10 (0.98,1,23)
28-40d	161-198d	88	803.9	0.89 (0.75,1.05)	114	244.4	0.91 (0.77,1.08)
41-50d	140-160d	388	832.7	0.95 (0.86,1.06)	480	271.3	1.07 (0.96,1.18)
41-50d	103-139d	498	758.9	0.90 (0.81,1.00)	608	280.7	1.13 (1.02,1.25)
41-50d	161-198d	106	774.5	0.86 (0.73,1.00)	119	218.2	0.83 (0.71,0.97)
51-60d	140-160d	24	874.5	0.89 (0.64,1.23)	32	260.5	1.05 (0.78,1.41)
51-60d	103-139d	279	779.5	0.86 (0.77,0.97)	323	283.6	1.07 (0.96,1.21)
51-60d	161-198d	2	-	-	2	-	-

GMC: Geometric mean concentration.

* The standard vaccination schedule (a 1^st interval of 28–40 d and a 2^nd interval of 140–160 d) was set as the reference subgroup. GMC ratios were the GMC of each subgroup compared with that of the reference subgroup.

Supplementary information

Discussion

This study explored the immunogenicity of the new E. coli-produced bivalent human papillomavirus vaccine (Cecolin®) under different vaccination intervals. The data showed that for both anti-HPV-16 and anti-HPV-18, the GMC ratios for the subgroups with different vaccination intervals compared with the standard vaccination interval (a 1^st interval of 28–40 d and a 2^nd interval of 140–160 d) were higher than 0.83, and the lower limits of the 95% CIs were higher than 0.64, indicating that postponing the second dose to 60 d (2 months) after the first dose and receiving the third dose at 103 to 198 d after the second dose had little impact on the induced antibody levels. Although the protective antibody levels of HPV vaccines have not been determined, humoral immune response levels were used to extrapolate the protective effect of HPV vaccines when used in adolescents who cannot be enrolled in an efficacy clinical trial due to ethical constraints on gynecological examinations.^21,22 The noninferiority criterion, such as a lower 95% confidence bound for the GMC ratio greater than 0.5, is accepted by regulatory authorities worldwide for HPV vaccines, and this criterion was also used for the registration of this E. coli-produced HPV vaccine in China.¹⁶ Hence, the small differences in GMCs in this trial are considered to have no clinical significance, and flexible schedules for the second and third doses could maximize vaccination rates with little effect on the clinical performance of the vaccine.

There are three main assays widely used to measure HPV antibody levels: a pseudovirion-based neutralization assay (PBNA), a competitive Luminex immunoassay (cLIA) and an ELISA. Among them, the cLIA measures neutralizing antibodies that compete with one dominant neutralizing monoclonal antibody.²³ The PBNA measures the total functional neutralizing antibodies, which can block the entry of pseudovirions into cultured cells; therefore, the PBNA is regarded as the ‘gold standard’ for testing for the anti-HPV antibody.²⁴ However, it is difficult to use in large-scale studies because of the complicated, labor-intensive process involved. The ELISA detects IgG antibodies, including neutralizing and non-neutralizing antibodies, that bind to the HPV L1 VLP coating. Although the consistency of these two assays is poor when assessing natural infection-induced HPV antibodies due to nonspecific binding in the ELISA,¹⁹ they are highly correlated when measuring HPV vaccine-induced antibodies.^19,25–27 Therefore, IgG antibodies were used instead to assess the immunogenicity of the test vaccine in this study.

There are some factors that impact the immune response: the nature of the vaccine, the immunization schedule, the use of adjuvants and immune memory, all of which can be influenced by the dose, intervals between doses and route of immunization.^28,29 The antibody concentration and the number of memory B cells peaked one month after the first dose, and the secondary immune response was evoked when the memory B cells were reactivated after a second exposure to the same antigen. In the secondary response, the IgG class of antibodies appeared more quickly, and the concentrations were orders of magnitude higher than before. In addition, the binding affinity of the antibody for the antigen was significantly increased, and this result emerged when sufficient time had elapsed after the first immune response. Therefore, for most protein vaccines, the immunization schedules are 0, 1 or 2 and 6 months, which allow at least 4–6 months between the priming and the booster, with longer intervals generally associated with a greater response.^29,30 There are some studies exploring the immunogenicity of the bivalent HPV vaccine or the quadrivalent HPV vaccine administered in three doses with longer time intervals.^31–34 One of the studies showed that the antibody response was superior at one month after the third dose in the group with dose 2 taken on time and dose 3 delayed compared to the group with doses 2 and 3 both taken on time. Another study compared the immunogenicity of the HPV-16/18 vaccine when people were administered an alternative schedule (0, 1, and 12 months) or a standard schedule (0, 1, and 6 months), in which the seroconversion rates at one month after the third dose were 100% for HPV-16 in both groups and 99.7% (alternative schedule group) and 100% (standard schedule group) for HPV-18. The GMCs for HPV-16 and HPV-18 were comparable at one month after the third dose in the alternative and standard schedule groups (11884.7 EU/mL vs 10311.9 EU/mL for HPV-16; 4501.3 EU/mL vs 3963.6 EU/mL for HPV-18). The results of this study and the effects of longer vaccination intervals between the 3 doses, with longer than 60 d between the first 2 doses or 198 d between the last two doses in the vaccination schedule, deserve to be further studied in phase 4 large-sample and randomized clinical trials.

The major limitation of our study is that it is a post hoc analysis, and the participants in each subgroup were not well balanced. One subgroup only included 2 people which was insufficient for analysis. In addition, we lacked long-term data on the antibody level; therefore, whether the dose intervals influence the durability of antibody levels still needs further research.

In conclusion, slight deviations in the vaccination times of the 2^nd and 3^rd doses are acceptable, as such deviations has only a minor, insignificant impact on the immune response induced by the E. coli-produced HPV-16/18 vaccine.

Declaration of conflicts of interest

Z.-J.L., B.-Z.L., W.-D.H., H.-R.P. report being current employees of Xiamen Innovax. No other potential conflicts of interest relevant to this article were reported.

Supplemental Material

Supplemental data for this article can be accessed on the publisher's website.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (81673240); The Fujian Province Health Education Joint Research Project (2019-WJ-05): Health Economic Evaluation and Application Strategy of Human Papillomavirus Vaccine in Fujian Province; The Scientific Research Foundation of State Key Laboratory of Molecular Vaccinology and Molecular Diagnostics (2018ZY001).