Currently Approved Prophylactic HPV Vaccines

Diane M Harper


Expert Rev Vaccines. 2009;8(12):1663-1679. 

In This Article

Vaccine Efficacy

Both Cervarix and Gardasil have been found to be highly efficacious in preventing HPV-16/18-associated cervical intraepithelial neoplasias (CINs) (Table 2) in women 15–16 years of age through to 25–26 years of age who were seronegative and DNA negative at study entry for the relevant HPV types associated with each vaccine..[15,16] Table 2 delineates the differences in efficacy when the study population changes either in terms of baseline characteristics, the number of injections received or when the cases of CIN 2+ begin to be counted (CIN 2+ means CIN 2, CIN 3, adenocarcinoma in situ, cervical squamous cell carcinoma and/or cervical adenocarcinoma).

The efficacy of Gardasil changes only slightly when the per-protocol (PP) population includes those who missed their injection at the specified protocol timing (protocol violations), as seen in the unrestricted susceptible population. Over 97% of the subjects received all three doses within 1 year in FUTURE II,[17] and over 95% of the subjects received all three doses within 1 year in FUTURE I.[18] The 95% efficacy seen in this population of which the majority received all three doses of vaccine within 1 year indicates that the implementation regimen of the three doses can be quite flexible within a 1-year span without affecting the vaccine efficacy.

When women are included regardless of baseline serologic and HPV status who have received one or more injections (intent-to-treat [ITT] population), the efficacy of Gardasil over an average of 36 months falls to 44% against CIN 2+ caused by HPV-16/18. By counting cases after day 1, the data show that more than one dose of vaccine is necessary to induce a high efficacy, and that the cases counted prior to a full series of vaccination reflect infection and lesion development occurring during the nonprotected state. This is important for public-health tabulations of vaccine coverage – tabulating at least one dose will not reflect any population efficacy; population coverage can only be estimated from those individuals receiving three doses within 1 year. An additional decrease in efficacy is attributed to women already infected or who had already developed a CIN lesion from one of the vaccine-relevant types at the time of first vaccination.

Gardasil's efficacy further falls to 18% for an average of 44 months in the ITT population when the cause of CIN 2+ can be any HPV infection. This low efficacy from simulated real-life populations is expected when evaluating efficacy against CIN 2+ caused by any HPV infection, as 50% of CIN 2+ lesions are caused by high-risk types of HPV other than 16 and 18. Similarly, an additional decrease in efficacy is also expected in the ITT population reflecting conditions of an already actively infected population: 20% of the baseline population were infected with one of the vaccine-relevant HPV types at baseline, had serologic evidence of prior exposure or entered the study with abnormal cytology.[17] Uncharacteristically, however, the cross-protection benefits seen in later Gardasil studies[19,20] are not reflected by any increase in efficacy in this ITT calculation, questioning the clinically measurable benefit of cross-protection by Gardasil.

In comparison to the 18% efficacy seen by Gardasil, the efficacy of Cervarix in the naive total vaccinated cohort was 70% against all CIN 2+ regardless of HPV causation, and 30% in the total vaccinated cohort, which included those with current and past HPV-16/18 infections.

At this time, Cervarix and Gardasil show durations of efficacy lasting 6.4 and 5 years, respectively, from the Phase II trial data (Table 3), with Cervarix trials still ongoing. Gardasil Phase II trials were stopped at 5 years. There are ongoing trials in Costa Rica, the Netherlands, Iceland, Denmark, Sweden and Finland for both Cervarix and Gardasil planned for at least 10 years.[21–23]

Therapeutic Efficacy

It is important to establish with firm data, not just indirect evidence, that neither Cervarix nor Gardasil offers any therapeutic efficacy, nor any acceleration in viral clearance, nor any accelerated progression to cervical cancer. The data available to date are presented in Table 4, showing a lack of therapeutic efficacy in women positive for a vaccine-relevant HPV infection at baseline. Women who had a CIN lesion of any grade attributed to HPV-6, -11, -16 and/or -18 at study entry did not clear their lesion within 3 years after three Gardasil injections,[18] nor did they develop an increased number of CIN 2+ lesions over the placebo group. Likewise, in women who already had an active HPV-16/18 infection at the time of vaccination, there was no clearance of the HPV-16/18 infection or acceleration to CIN 2+ within 12 months with Cervarix vaccination.[24]

Baseline Seropositivity

The highest vaccine efficacies for Gardasil and Cervarix are reported in women 15–26 years of age who are both seronegative and PCR negative for the relevant vaccine types prior to vaccination. The limited but statistically significant data in Table 5 show that despite prior exposure to HPV-16/18 (seropositivity), women without active HPV-16/18 infection at the time of first vaccination continued to have 100% vaccine efficacy to prevent CIN 2+ caused by HPV-16/18. High efficacy in seropositive women without active infection is important to vaccinating young girls, approximately 10% of whom are already seropositive by 10 years of age for at least HPV-16 from unknown causes,[8–10] but without active infection. Having confidence that the vaccines perform at high efficacy rates despite prior infection alleviates concerns that vaccine performance might be compromised when targeting a supposedly naive adolescent population that is not actually completely naive. Serendipitously, the continued high efficacy of HPV vaccination in women 15–26 years of age who are already seropositive but PCR negative supports targeting vaccination to the newly sexually active population, a population far more accepting of a sexually transmitted infection vaccination than the parents of young adolescent girls.[101] This population of women vaccinated within the first year of sexual activity has a greater prevention of HPV-16/18 persistent infections and CIN 2+ disease within 3 years of vaccination in absolute rate reductions (57/1000 women) than does vaccinating virgins prior to the onset of their sexual activity (17/1000 women).[25]


Table 6 details the known data for the protection provided against persistent infection of 6 months duration or longer, and for CIN 2+ lesions caused by high-risk types other than HPV-16 and -18 for Cervarix and Gardasil. Cervarix provides individual cross-protection for HPV-31, -33 and -45 persistent infections.[26] Gardasil provides protection against HPV-31. All of the cross-protection provided by Gardasil in grouped classifications is due to the solo strength of HPV-31 coverage.[18] Ongoing trials are evaluating the efficacy of a pentavalent supplemental vaccine to Gardasil explicitly containing HPV-31, -33, -45, -52 and -58 to broaden Gardasil's coverage with eventual replacement of Gardasil with the nonovalent vaccine.[102] In the cross-protection studies, persistent HPV infection outcomes are more appropriate than CIN 2+ end points, as they unequivocally establish that infection has been present for some time, in contrast to the incidental presence of an HPV type in a lesion in which multiple types are also detected, but not necessarily causative,[27,28] which occurs in the majority of CIN 2+ lesions.[29] Extrapolating potential reduction in cervical cancers from direct and cross-protection of the HPV vaccines under assumptions of complete vaccine coverage, lifetime duration of efficacy and clinically relevant efficacies, indicates that Cervarix may reduce the incidence of both squamous and adenocarcinoma of the cervix by 81% and Gardasil by 73% (Table 7).

Reductions in Use of Secondary Screening

Both Cervarix and Gardasil have reduced the number of women referred to colposcopy for an abnormal Pap test by 26 and 20%, respectively, within 3 years of vaccine use (Table 8).[26,30] Because of the additional cross-protection, Cervarix has prevented in 3 years 69% of the excisional therapies associated with reproductive morbidity compared with 42% for Gardasil.[26,30] The rate of loop electrosurgical excision procedures in the USA is currently 500 per 100,000 women screened;[31] this could be reduced to as low as 155 per 100,000 women with universal vaccination coverage (Figure 1).

Figure 1.

Estimated maximal potential reduction in reproductive morbidity from excisional therapies for CIN 2/3 disease by vaccination, assuming all excisional therapies are associated with reproductive morbidity. Reduction in LEEPs for CIN 2/3 per 100,000 women screened per year in the USA for the current management scenario compared with the potential for reduction if there is 100% vaccination coverage, lifetime vaccine efficacy and 100% continued screening in women who receive either Cervarix® or Gardasil®.
CIN: Cervical intraepithelial neoplasia; LEEP: Loop electrosurgical excision procedure.

Protection beyond the Cervix for Gardasil

Table 9 details the efficacies against diseases not related to cervical cancer prevention. HPV-6 and -11 are the assumed cause of 90% of genital warts.[32] Vaccine efficacy for protection against the development of condyloma caused by HPV-6, -11, -16 or -18 in women 16–26 years of age lasting on average 44 months was reported at 80% for the ITT population and 99% for the PP population for Gardasil – both very high levels of efficacy.[103,104] Inexplicably, however, one would expect the efficacy of Gardasil against all condyloma regardless of HPV type to be very similar to the efficacy against condyloma caused by HPV-6 and -11, the purported cause of 90% of genital warts. The data show significantly less efficacy for condyloma caused by all HPV types at 63%. Potentially the attributable proportion of HPV-6 and -11 for genital warts is not 90% as previously thought, but actually much lower.

Protection against condyloma caused by HPV-6, -11, -16 or -18 in boys and men is 89% lasting for 29 months.[33] Unfortunately, there are no reports of efficacy against all condyloma regardless of HPV causation for males to date. We do not know whether there is a similar significant decrease in overall condyloma prevention in men as there is in women.

Gardasil's efficacy against vulvar intraepithelial neoplasia (VIN) 1 caused by HPV-6, -11, -16 or -18 in an ITT population was not significant after 36 months, trialed in a population of almost 18,000 women.[104] Vaginal intraepithelial neoplasia (VaIN) 1 caused by HPV-6, -11, -16 or -18, on the other hand, showed nonreassuring significant efficacy both in the PP and the ITT populations as both had quite broad confidence intervals.[104] Overall protection from VIN 1, VaIN 1 or condyloma caused by any HPV type was reported at 41% for the ITT population. This overall efficacy is dominated by the protection from condyloma as efficacies for VIN 1 and VaIN 1 contribute little to the overall protection of VIN 1/VaIN 1/condyloma from any HPV type. Gardasil does not have national regulatory approval for the prevention of VIN 1 or VaIN 1.

Gardasil's efficacy against VIN 2/3 and VaIN 2/3 caused by HPV-16/18 after an average of 44 months of follow up is 100% in the PP populations, with small overlapping decreases to 69 and 85%, respectively, in the ITT populations.[103] Both vaginal and vulvar cancer precursors have limited association with HPV infection,[34,35] explaining the extremely wide efficacy confidence intervals for VIN 2/3 from any cause and no significant efficacy for VaIN 2/3 from any cause.[36] While Gardasil is approved for and has the potential to reduce vaginal and vulvar cancers caused by HPV-16/18 by two-thirds of their current incidence, it is unlikely that Gardasil will have a measurable clinical effect on the prevention of vaginal cancers from any cause overall based on these nonsignificant data.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: