Characteristics of included studies

We identified 227 records by running the pre-defined search strategies on the three selected databases. After removing duplicates, 148 papers were assessed for eligibility by title and abstract. Papers were screened and selected as illustrated in Figure 1. Two papers were published on 9vHPV clinical trials’ protocols, describing the RCTs’ design and rationale but not reporting original data; they were excluded from the review [Reference Luxembourg32, Reference Chen, Gesser and Luxembourg33]. Ten papers met the inclusion criteria and were included in the review.

Fig. 1. PRISMA flowchart of included published papers.

Included papers’ characteristics are reported in Table 1.

Table 1. Characteristics of included published studies, reporting findings from randomized, controlled trial on 9vHPV

9vHPV, human papillomavirus 9-valent vaccine; 8vHPV, HPV 8-valent vaccine; 5vHPV, HPV 5-valent vaccine; qHPV, HPV quadrivalent vaccine; MCV4, meningococcal A/C/Y/W-135 vaccine; Tdap, tetanus, diphtheria, acellular pertussis vaccine; Tdap-IPV, diphtheria, tetanus, pertussis, and poliomyelitis vaccine; Pap test, Papanicolaou test; Yrs, years; Dose form., dose formulations.

* Subjects who underwent randomization.

Included papers reported findings from eight different clinical trials on 9vHPV's efficacy, immunogenicity, and/or safety. Of them, one paper reported data on 9vHPV clinical efficacy [Reference Joura34], 10 papers reported data on 9vHPV immunogenicity [Reference Joura34–Reference Vesikari43], and nine papers reported data on 9vHPV safety [Reference Joura34–Reference Luxembourg38, Reference Schilling40–Reference Vesikari43].

In particular:

• • one paper reported the findings of the three Phase II RCTs that were conducted to select the best vaccine formulation to undergo Phase III evaluation [Reference Luxembourg38];

• • one paper reported preliminary data on 9vHPV clinical efficacy in females aged 16–26 years, as well as on 9vHPV immunogenicity and safety in this population, compared with qHPV [Reference Joura34];

• • one paper reported data on 9vHPV immunogenicity and safety in males aged 16–26 years, compared with qHPV [Reference Van Damme41];

• • one paper reported data on immunogenicity and safety of 9vHPV in 9–15-year-old females, compared with qHPV [Reference Vesikari43];

• • two papers reported on 9vHPV bridging efficacy to 9–15-year-old populations [Reference Van Damme42], 16–26-year-old heterosexual males and men having sex with men (MSM) [Reference Castellsagué35], respectively;

• • one paper reported the findings of the Phase III trial conducted to evaluate vaccines’ lot consistency [Reference Luxembourg39];

• • two papers reported data on concomitant administration of 9vHPV with other vaccines, compared with non-concomitant administration [Reference Kosalaraksa37, Reference Schilling40];

• • one paper reported data on 9vHPV immunogenicity and safety in young females previously immunized with qHPV vaccine [Reference Garland36].

The sections below, together with Tables 2–4, report the synthesis of included RCTs’ findings, by outcome. Meta-analysis of RCTs’ estimates could not be performed due to studies’ heterogeneity in terms of: study design, tested interventions, and comparisons, as well as targeted study populations.

Table 2. HPV 9-valent vaccine clinical efficacy and bridging efficacy findings of included studies

MSM, men-having-sex with men; 9vHPV, human papillomavirus 9-valent vaccine; qHPV, HPV quadrivalent vaccine; GMT, geometric mean titer.

* Intention-to-treat populations (all other estimates are on per protocol populations).

Table 3. HPV 9-valent vaccine immunogenicity findings of included studies

qHPV, human papillomavirus quadrivalent vaccine; GMT, geometric mean titer.

* The P value for non-inferiority was <0·001 for all comparisons HPV 9-valent vaccine vs. qHPV vaccine.

Table 4. The 9vHPV safety findings of included studies; adverse events (AEs, %)

9vHPV, human papillomavirus 9-valent vaccine; qHPV, HPV quadrivalent vaccine.

* The difference in AE incidence between vaccines was statistically significant only for injection-site swelling (P = 0·003).

Dose formulation data

Data from the Phase II dose formulation study were published in June 2015 [Reference Luxembourg38]. The paper reports the findings of three RCTs conducted to compare immunogenicity and safety of seven vaccine candidates with licensed qHPV vaccine. The vaccine candidates were: three dose formulations (low-, mid-, and high-dose formulations) of an 8-valent HPV6/11/16/18/31/45/52/58 vaccine (Study 1), three dose formulations of a 9-valent HPV6/11/16/18/31/33/45/52/58 vaccine (Study 2), and qHPV vaccine concomitantly administered with a 5-valent HPV31/33/45/52/58 (Study 3). All vaccines were administered on a 3-dose schedule at months 0, 2, and 6. Six-hundreds and eighty, 1242 and 623 girls aged 16–26 years were, respectively, randomized in the three studies. Within each study, the primary immunogenicity objective was to demonstrate that month 7 anti four original vaccine HPV types (anti-HPV6/11/16/18) geometric mean titers (GMTs) were non-inferior in subjects who received experimental vaccines compared with subjects in the control group. Non-inferiority was defined as the lower bound of the two-sided 95% CI of the GMT ratio (experimental arm/control arm) being >0·5 for each of anti-HPV types 6/11/16/18. Among all vaccine candidates, the 9vHPV mid-dose formulation containing 30/40/60/40/20/20/20/20/20 mg of HPV6/11/16/18/31/33/45/52/58 virus such as particles, and 500 mg of amorphous aluminum hydroxyl-phosphate sulfate adjuvant was selected for all subsequent Phase III efficacy, immunogenicity, and safety evaluations.

Successful 9vHPV vaccine candidate was reported to: (i) provide non-inferior antibody responses compared with qHPV vaccine with respect to the four HPV types covered by both vaccines (the lower bound 95% CI GMT ratio exceeded 0·5 for all HPV types), (ii) be strongly immunogenic against five additional oncogenic HPV types (>95% subjects seroconverted at month 3), and (iii) be generally well tolerated. Injection-site adverse events (AEs) were slightly higher in the selected 9vHPV vaccine as compared with qHPV (92·4% vs. 90·3%).

Efficacy

The 9vHPV efficacy findings of included studies are summarized in Table 2. One paper has been published on 9vHPV clinical efficacy so far [Reference Joura34]. It reports preliminary data of a Phase III multinational multicenter efficacy study conducted on 14 215 women aged 16–26 years comparing 9vHPV to qHPV. The primary efficacy outcome was the combined incidence of several conditions related to 9vHPV additional HPV types (HPV-31, 33, 45, 52, 58) – with follow-up over a 54-month period. Analysis in the modified intention-to-treat population (which included subjects both not HPV-infected and HPV-infected at the time of vaccination, who received at least one dose of vaccine and for whom there was at least one measurement of efficacy) showed high-grade cervical, vulva, and vaginal disease incidence to be the same in the 9vHPV and qHPV arms, irrespective of HPV testing results (14·0 per 1000 persons-years, risk reduction = 0·7, 95% CI −15·7 to 14·8), while when restricting the analysis to not HPV-infected participants at the time of vaccination, 9vHPV risk reduction was 42·5%; (95% CI 7·9–65·9), this reaching 100% (95% CI 70·4–100) when only considering disease related to 9vHPV HPV types. Analysis conducted on the per-protocol population (subjects who received all three doses of vaccine within 1 year, who were seronegative on day 1 and PCR-negative from day 1 through month 7 and had no protocol violations) reported 9vHPv vaccine efficacy to be 96% or above for all considered clinical outcomes related to HPV types 31, 33, 45, 52, or 58 as well as persistent infection (⩾6 months’ duration, Table 2) [Reference Joura34].

Two published immunobridging efficacy studies inferred 9vHPV efficacy in males and females aged 9–15 years [Reference Van Damme42], in heterosexual males and in MSM aged 16–26 years [Reference Castellsagué35], comparing immunogenicity data between individuals in the intervention arms and 16–26-year-old female controls. By definition, bridging studies generate immunogenicity data to support the extrapolation of data on efficacy or safety obtained under specific circumstances of use (in our case 9vHPV efficacy in 16–26-year-old females) to other situations (e.g. different age groups or different populations) [44]. Both studies reported month 7 GMTs in the intervention arms to be non-inferior (lower bound of the two-sided 95% CI of the GMT ratio >0·67) to the control arm for all 9vHPV vaccine types, as well as seroconversion non-inferiority in more than 99% of study participants (defined as the lower bound of the two-sided 95% CI for the differences in seroconversion rates being >−5 percentage points for each 9vHPV HPV type) (Table 2). Data on 9vHPV immunogenicity MSM aged 16–26 years showed lower antibody response compared with heterosexual males for all tested HPV types (GMT ratios MSM/women for all 9vHPV vaccine types are reported in Table 2) [Reference Castellsagué35].

Immunogenicity

All included articles assessed 9vHPV immunogenicity. The majority of studies adopted a non-inferiority design. Immunogenicity outcomes included intervention–control comparisons of month 7 GMTs for anti-HPV types and proportion of studies’ participants with seroconversion to the 9vHPV vaccine types at 4 weeks after the administration of the third vaccine dose (month 7).

The immunogenicity analyses findings of the dose-formulation Phase II study has been described in the previous section [Reference Luxembourg38] and reference was previously made to the published lot-consistency study – conducted to meet regulatory requirements – that reported three different lots of 9vHPV to elicit equivalent antibody response and seroconversion rates [Reference Luxembourg39]. In addition, data are available on: (i) 9vHPV vs. qHPV immunogenicity [Reference Joura34, Reference Van Damme41, Reference Vesikari43]; (ii) 9vHPV immunogenicity in females aged 16–26-years old vs. other age and gender populations to infer bridging efficacy [Reference Castellsagué35, Reference Van Damme42]; (iii) 9vHPV immunogenicity when co-administrated with other vaccines (including meningococcal, tetanus, diphtheria, pertussis, polio vaccines) [Reference Kosalaraksa37, Reference Schilling40]; and (iv) 9vHPV immunogenicity in subjects previously vaccinated with qHPV [Reference Garland36].

Three studies reported immunogenicity data comparing 9vHPV to qHPV in females aged 16–26 years [Reference Joura34], in females aged 9–15 years [Reference Vesikari43] and in males aged 16–26 years [Reference Van Damme41]. These are summarized in Table 3. They showed: (i) 9vHPV non-inferiority as compared with qHPV with regard to HPV types 6, 11, 16, and 18 in females and males aged 16–26 years (GMT ratios ranging from 0·80 and 1·19 and from 0·89 and 1·23, respectively, Table 3) [Reference Joura34, Reference Van Damme41]; (ii) 9vHPV non-inferiority as compared with qHPV with regard to HPV types 16 and 18 in females aged 9–15 years [Reference Vesikari43]; and (iii) virtually all studies’ participants seroconversion for HPV 31/33/45/52/58 at month 7 (in females aged 9–15 years and 16–26 years and in males aged 16–26 years [Reference Joura34, Reference Van Damme41, Reference Vesikari43]; Table 3).

Two studies have been published on concomitant administration of 9vHPV with other vaccines, such as meningococcal (MCV4), tetanus, diphtheria, pertussis (Tdap) and polio vaccines, reporting non-inferior immune response in subjects receiving 9vHPV concomitantly with other vaccines as compared with non-concomitant administration [Reference Kosalaraksa37, Reference Schilling40]. In particular, authors reported concomitant administration of 9vHPV together with MCV4 and Tdap to elicit non-inferior GMTs and seroconversion rates for all 9vHPV antigens as well as non-inferior immune response and seroconversion to MCV4 and Tdap vaccines’ components as compared with non-concomitant vaccines’ administration in subjects aged 11–15-years old [Reference Schilling40]. Similarly, findings from another RCT showed all 9vHPV HPV types 7 month GMTs and seroconversion non-inferiority in the concomitant group compared with the non-concomitant group. Non-inferiority of immune response was established for diphtheria, tetanus, all pertussis, and polio antigens for both groups [Reference Kosalaraksa37].

One study has been published reporting findings from an RCT that assessed 9vHPV immunogenicity vs. placebo in females aged 12–26, previously vaccinated with qHPV [Reference Garland36]. Seroconversion at month 7 was reported to be >98% for all 9vHPV HPV types, with marked elevations in GMTs. Data from cross-study analysis showed anti-HPV 31/33/45/52/58 GMTs to be lower than in study subjects administered 9vHPV vaccine who had not previously received qHPV vaccine [Reference Garland36].

Safety

Nine papers reported data on 9vHPV vaccine safety; of them, one is the Phase II RCTs conducted to select the right 9vHPV formulation [Reference Luxembourg38], three made comparisons between 9vHPV and qHPV vaccine [Reference Joura34, Reference Van Damme41, Reference Vesikari43], two made comparisons between 9vHPV administered in different age and gender populations (immunobridging studies) [Reference Castellsagué35, Reference Van Damme42], two assessed 9vHPV safety comparing concomitant and non-concomitant administration with other vaccines [Reference Kosalaraksa37, Reference Schilling40] and one assessed 9vHPV safety against placebo in girls previously vaccinated with qHPV vaccine [Reference Garland36]. In five studies, participants recorded injection-site events (AE, within 5 days after vaccination) and systemic events (within 15 days after vaccination) on VRCs (vaccination report cards) [Reference Joura34, Reference Castellsagué35, Reference Luxembourg38, Reference Van Damme41, Reference Van Damme42]. Injection-site AEs were rated by study subjects to be mild/moderate/severe. AEs causality was assessed by investigators and classified as possibly, probably, or definitely vaccine related. Serious AEs were pre-defined as any AE resulting in deaths or in which discontinuation due to AEs was reported. AEs were summarized as frequencies and percentages by study arm and AE type. The 9vHPV safety findings are reported in Table 4. Overall, 9vHPV vaccine recipients were more likely than qHPV vaccine recipients to have AEs related to the injection site (90·7% vs. 84·9% in females aged 16–26) [Reference Joura34], while systemic AEs distribution was similar between intervention and control groups (55·8% vs. 54·9% in female aged 16–26) [Reference Joura34]. Discontinuation rates because of vaccine-related AE were rare and all but one study [Reference Joura34] reported no vaccine-related serious AEs [Reference Joura34, Reference Luxembourg38, Reference Van Damme41, Reference Vesikari43]. In immunobridging studies injection-site or systemic AEs were lower in heterosexual men and MSM aged 16–26 and in males and females aged 9–15 years as compared with controls (females aged 16–26 years) [Reference Castellsagué35, Reference Van Damme42]. In studies assessing concomitant vaccines administration, injection-site AEs of swelling after 9vHPV and Tdap-IPV were more frequent in concomitant administration arms as compared with non-concomitant ones (after 9vHPV: 14·4% vs. 9·4% [Reference Schilling40], after Tdap-IPV 21·7% vs. 31·3% [Reference Kosalaraksa37]); the risk difference between the groups being statistically significant. No other statistically significant differences were reported in terms of AEs after all vaccines’ administration between study arms [Reference Kosalaraksa37, Reference Schilling40]. In both studies, few subjects discontinued because of an AE and no deaths were reported [Reference Kosalaraksa37, Reference Schilling40]. In the placebo-controlled trial conducted in females who previously received qHPV, injection-site AEs were more frequent in the intervention arm (91·1% vs. 43·9) and increased with subsequent 9vHPV vaccine doses, the incidence of serious vaccine-related AEs and discontinuation was reported to be low and no subjects died during the study [Reference Garland36].

Ongoing clinical studies

One hundred and forty records were initially retrieved searching the selected clinical trials’ registries and platforms (Fig. 2). After removing duplicates, 87 RCTs were excluded as they did not include 9vHPV administration in the intervention or control study arms. Twenty-four relevant clinical trial protocols met the inclusion criteria, referring to 16 clinical trials conducted on the 9vHPV vaccine (some RCTs were registered more than once in different RCT registries; Table 5).

Fig. 2. PRISMA flowchart of included registered trials (update August 25, 2016).

Table 5. Existing randomized controlled trials on human papillomavirus 9-valent vaccine

* As reported by ClinicalTrials.gov.

† As reported by European Union Clinical Trials Register.

As of August 25, 2016, 16 Phase II or Phase III multicenter multinational RCTs have been conducted on 9vHPV, the first of which was started in February 2007 [45] while the last was started in August 2016 [46]. Thirteen (81·2%) RCTs are directly sponsored by pharmaceutical companies, two are sponsored by universities [46, 47], one by the US National Cancer Institute [48] and one by the Danish Department of Infectious Diseases [49].

Of the 16 identified RCTs, eight (50%) have been completed, three (18·8%) are active but not recruiting [50–52], four (25%) are ongoing or recruiting [47–49, 53], and one (6·3%) is not recruiting participants yet [46].

For eight RCTs (50%) results or preliminary results have been published in scientific peer-reviewed journals (Tables 1–4, results described in previous sections); one completed RCTs does not have published studies [54].

Of the active RCTs, three are assessing immunogenicity and tolerability of less than three-dose 9vHPV vaccine schedules [46, 47, 51]. The first one – started in December 2013 and no longer anymore – is a 37-month safety and immunogenicity study conducted with the aim of assessing whether investigational two-dose regimens (0, 6 months and 0, 12 months) administered in males and females aged 9–14 years elicit non-inferior immunogenicity and safety profiles as compared with three-dose regimens administered to females aged 16–26 [51]. The second one – started in September 2015, and still recruiting – aims to assess immunogenicity of 9vHPV and bHPV vaccines administered to males and females aged 9–10 years, according to a 0–6 month schedule to infer possible interchangeable use of the two vaccines [47]. The third one – about to start recruiting females aged 9–14 years in Tanzania – will compare immunogenicity between one-dose and two-dose 9vHPV and bHPV vaccines and three-dose regimens of both vaccines [46].

One active RCT is a long-term follow-up study of the first 9vHPV vaccine RCT (NCT00543543), currently being conducted in countries with centralized cervical cancer screening infrastructures (Denmark, Norway, and Sweden) that aims to evaluate the long-term effectiveness, immunogenicity, and safety of 9vHPV vaccine in females aged 16–26 years [52].

Within ongoing European RCTs, one efficacy immunobridging study started in July 2016 and currently recruiting subjects in four European countries aims at assessing immunogenicity and tolerability of the 9vHPV vaccine in females aged 27–45 years compared with females aged 16–26 years [53]. Another RCT – currently recruiting subjects in Denmark – aims at inferring 9vHPV vaccine efficacy in HIV-infected sexually active MSM, measuring changes in the prevalence of HPV types in either anus, oral cavity, or penis as well as well as antibody response [49].

One Phase IIA RCT started in March 2016 aims at assessing immunogenicity of a prime and deferred-booster dosing 9vHPV vaccine schedule among females aged 9–11 years [48]. Primary and secondary objectives are to determine persistence and stability of the immune response to HPV types 16/18 and HPV types 6/11/31/33/45/52/58, respectively, between 6, 12, 18, and 24 months after the prime dose and prior to the administration of the second dose [48].