Summary Basis of Decision for Cervarix ™

3.3.1 Human Pharmacology

The pharmacodynamics of Cervarix™ were assessed through the analysis of immunogenicity and are described in section 3.3.2 Clinical Efficacy. Pharmacokinetic studies were not submitted as they are generally not required for injectable vaccines.

3.3.2 Clinical Efficacy

Six studies were conducted to investigate the efficacy and immunogenicity of Cervarix™, in which a total of 23, 281 subjects were vaccinated. Of these, 12,363 received Cervarix™, and 10,918 received a control.

• Study HPV-001
• Study HPV-007 (an extension of Study HPV-001)
• Study HPV-008
• Study HPV-012
• Study HPV-013
• Study HPV-014

Two double-blind, randomized, controlled clinical studies (Study HPV-001/HPV-007 and Study HPV-008) included 19, 778 females 15-25 years of age at enrolment. In Study HPV-001/HPV-007, the follow-up lasted for 6.4 years, and in Study HPV-008, the follow-up was approximately 39 months. Efficacy endpoints included surrogate markers of cervical cancer: cervical intraepithelial neoplasia (CIN) grade 2 and 3 lesions or cervical adenocarcinoma in situ (AIS), precursors of squamous cell carcinoma and adenocarcinoma of the cervix, respectively. Other endpoints included an assessment of efficacy in the prevention of 6-month persistent infection and 12-month persistent infection.

Of all the clinical studies, Study HPV-008 was regarded as the pivotal study for this drug submission. The primary objective of the study was to demonstrate the efficacy of Cervarix™ compared with control in the prevention of histopathologically-confirmed CIN2+ (defined as CIN 2, CIN 3, AIS, and invasive cancer) associated with HPV-16 or HPV-18 cervical infection detected within the lesional component of the cervical tissue specimen [by polymerase chain reaction (PCR)], overall and stratified according to initial (Month 0) HPV-16 or HPV-18 serostatus [by enzyme-linked immunosorbent assay (ELISA)].

Study HPV-008 enrolled females who were vaccinated regardless of baseline HPV DNA status, serostatus or cytology. These females reflect a general population inclusive of females naïve (without current infection and without prior exposure) or non-naïve (with current infection and/or with prior exposure) to HPV. Before vaccination, cervical samples were assessed for oncogenic HPV DNA (HPV-16, -18, -31, -33, -35, -39, -45, -51, -52, -56, -58, -59, -66 and -68) and serostatus of HPV-16 and HPV-18 antibodies.

There were two study cohorts:

According to Protocol (ATP)

The According to Protocol (ATP) cohort for efficacy analysis included:

• all females who received 3 doses of vaccine for whom efficacy endpoint measures were available
• all females who were HPV DNA(-) and sero(-) at baseline for the HPV type considered in the analysis
• all females who were HPV DNA(-) at month 6 for the HPV type considered in the analysis
• normal or low-grade cytology (ASC-US or LSIL) at baseline (females with high-grade cytology were excluded)
• all females who met all eligibility criteria
• all females who complied with procedures defined in the protocol, and with no elimination criteria during the study

Total Vaccinated Cohort (TVC)

The total vaccinated cohort (TVC) included:

• all females who received at least 1 dose of the vaccine for whom efficacy endpoint measures were available
• all females were included irrespective of the HPV DNA status and serostatus at baseline

The primary endpoint was protection against CIN2+ lesions associated with HPV-16 and/or HPV-18 cervical infection among females who are negative for HPV DNA (by PCR) at baseline for the corresponding HPV type.

Secondary endpoints also included:

• Protection against persistent infection (6-month and 12-month definition) with HPV-16 or HPV-18 among females who are negative for HPV DNA at baseline for the corresponding HPV type.
• Protection against persistent infection (6-month definition) with the following oncogenic HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68.
• Protection against CIN2+ and CIN1+ associated with HPV-16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, and 68.

Cervarix™ significantly reduced the incidence of CIN2+ lesions due to HPV-16 and HPV-18, compared to HAV, in individuals who did not have evidence of HPV-16 or HPV-18 infection at the time of vaccination. The efficacy in the ATP cohort was 98.1% [confidence interval (CI) 88.4, 100]. In the TVC, irrespective of HPV status at time of vaccination the efficacy was 54.7% (CI 39.5, 66.3).

Sub analyses by initial HPV status in the TVC cohort shows efficacy against those who were DNA and serologically negative for HPV-16/18 at baseline of 97.7% (CI 91.1, 99.8). In a much smaller cohort of individuals negative for HPV-16/18 by DNA at baseline, but with serological evidence of prior infection, the efficacy was 88.5% (CI 10.8, 99.8), the wide CI in part due to the smaller numbers in this cohort (approximately 1700 compared to just over 8000 in the DNA negative, seronegative cohorts). In individuals with DNA evidence of HPV-16/18 infection at baseline, the numbers were small (around 300 per cohort), and there was no evidence of efficacy with a reduction of 32.8% but a negative lower bound of the confidence interval (CI -27.4, 65.3) in DNA positive, seronegative individuals, and those with active infection and who were seropositive there was a negative result, -13.8% (CI -77.6, 26.7).

There was supportive evidence for efficacy against HPV-16/18 with reductions in persistent infections with HPV-16/18 as measured over 12 month intervals and over consecutive samples at 6 month intervals.

Irrespective of HPV type and without consideration of HPV status at time of vaccination, Cervarix™ still showed efficacy with a reduction of 30.4% (CI 16.4, 42.1) and in a naïve cohort the efficacy was, as expected better at 70.2% (CI 54.7, 80.9).

In study HPV-008, post-hoc analyses for vaccine efficacy, adjusted for multiplicity, were conducted in the ATP and TVC cohorts to assess the impact of Cervarix on CIN2/3 or AIS due to 12 non-vaccine oncogenic types (HPV-31, -33, -35, -39, -45, -51, -52, 56, -58, -59, -66, -68). Vaccine efficacy in prevention of CIN2/3 or AIS associated with HPV-31 was 91.3% (99.7% CI: 43.7, 99.8) in the ATP cohort and 38.9% (99.7% CI: -25.2, 71.4) in the TVC cohort. Vaccine efficacy in prevention of CIN2/3 or AIS associated with HPV-45 was 100.0% (99.7% CI: -298.4, 100.0) in the ATP cohort 100.0% (99.7%CI: 29.0, 100.0) in the TVC cohort.

3.3.3 Clinical Safety

Study results from the pivotal Study HPV-008, showed an increased rate of local and immediate post-vaccination symptoms in patients that received Cervarix™ compared to those that received a licensed HAV vaccine. The local injection site reaction included pain, redness, and swelling. General adverse events included fatigue, fever, gastrointestinal symptoms, headache, rash, arthralgia, myalgia, and urticaria.

Compared to the HAV vaccine control there were more local and systemic reactions to Cervarix™ and there were more grade 3 reactions, although most were of short duration.

Grade 3 myalgia was more common in the Cervarix™-treated group (1.8% vs. 0.6%) and the CIs did not overlap.

One subject experienced anaphylactic shock in the Cervarix™ group, compared to one report of anaphylactic reaction in the HAV group.

The duration of adverse events was short, and there was no significant difference in rates after different doses. Seropositive subjects did not experience more adverse reactions than those who were HPV-16/18 positive at enrolment, although there is a suggestion that the DNA negative group had lower rates of reactions.

Overall, vaccination with Cervarix™ was associated with more local and general symptoms in the short-term 7 and 30 day periods after vaccination compared to those receiving an HAV vaccine. Adverse events resulting in school or work absenteeism were limited.

Serious adverse events show several differences, notably for spontaneous abortions. It should be noted that as this safety data has been prepared near the end of the study, most subjects will have been followed for four years with vaccinations at 0, 6 and 12 months, thus the temporal relationship of many of these events to vaccination is low.

In both the pivotal study and in the analyses of combined safety data, there were more reports of spontaneous abortions in the Cervarix™-treated groups compared to control groups. The increase was marginal, but when the analysis only considered pregnancy outcomes in subjects with an LMP close to the date of vaccination, the increase in the rate of spontaneous abortions in the Cervarix™ group compared to control groups was more marked, although not statistically significant. In 761 women who had their last menstrual period within 30 days prior to, or 45 days after a vaccine dose and for whom pregnancy outcome was known, spontaneous abortion occurred in a higher proportion of women who received Cervarix™ (13.6%) compared to those receiving a control substance (9.6%). It is not known if this is due to a vaccine related effect.

While a numerical increase in the number of spontaneous abortions in those receiving Cervarix™ compared to control groups was noted, the difference was not statistically significant, and varied between studies and cohorts within studies, such as regional cohorts, and was not apparent when pregnancy outcomes from North American participants were analyzed. Appropriate warnings and clear information on this issue is included in the Product Monograph. Other SAEs that were more common in the HPV group included pyelonephritis and pelvic inflammatory disease.

Apart from a very small number of rare disorders with an immune basis, there was no increase in autoimmune disorders or diseases classified as chronic disorders which were first diagnosed during the study period. Numbers of these new onset autoimmune and chronic disorders were too small to draw any firm conclusions. The sponsor intends to further monitor these events post licensure.

Post-marketing Data

The post-marketing data do not add significantly to the safety information for Cervarix. The rates of spontaneous reporting do not appear to be unusual and the range of reports is mainly as expected for a vaccine.

There are some adverse events that were not noted in the pivotal study, such as angioedema, but this is not unexpected in a vaccine.

Pregnancy registers are being set up in Europe and the United States and pregnancy outcomes after vaccination will continue to be monitored from ongoing studies.