It may not be necessary to give three doses of the human papillomavirus (HPV) vaccine targeting viral types 16 and 18 to achieve its full protection against cervical infection, according to a combined analysis of data from a pair of randomized controlled trials among women aged 15–25 from several world regions.¹ Four years after vaccination, the vaccine’s efficacy relative to a control vaccine was 76–86% against new infection with carcinogenic HPV types 16 and 18 among women who received all three doses, and was comparable among women who received only one or two of the scheduled doses. Substantial albeit lower vaccine efficacy was observed with three doses for the closely related carcinogenic HPV types 31, 33 and 45 that are not included in the vaccine formulation. Such cross protection was not evident among women who received one dose; however, among women who received two doses, a delay in receipt of the second dose from one month to six months was associated with some cross-protective efficacy.

Investigators performed a post hoc analysis that combined data from the National Cancer Institute’s Costa Rica Vaccine Trial and the PATRICIA (Papilloma Trial Against Cancer in Young Adults) trial, which was conducted in Europe, Asia, Latin America and North America. In both trials, young women were randomized to receive three doses of the bivalent HPV-16/18 vaccine (Cervarix) or a control vaccine (the hepatitis A vaccine) at baseline, one month and six months; some women, however, failed to receive all scheduled doses.

At baseline, the women provided information about risk factors and gave a blood sample for HPV antibody testing, and, if sexually active, underwent collection of cervical cells for cytology and HPV DNA testing. Incident infections were ascertained by HPV DNA testing of cervical samples among sexually active women annually, or more often if needed. The investigators calculated differences in detection of an incident HPV infection at any one time during the four-year follow-up period between the HPV-vaccinated and control- vaccinated women by number of doses received, after excluding women who had less than a year of follow-up. They also did similar analyses for incident infections that persisted for at least six months and at least 12 months. (Persistent infection with a carcinogenic HPV type is the cause of cervical cancer.)

Vaccine efficacy against incident HPV 16 and 18 infection, the study’s primary endpoint, was assessed among 24,055 women aged 15–25 women who were DNA-negative for either or both of those types at enrollment. In this subset of women, 93% received three doses of their assigned vaccine, 5% received two doses and 2% received one dose. The most common reasons for missing doses were pregnancy at the time of a scheduled dose and referral for colposcopy for further evaluation of cervical changes.

Analyses showed that efficacy of the vaccine for incident HPV type 16 or 18 infection was 77% for women receiving all three doses, 76% for those receiving two and 86% for those receiving one. None of the differences between groups were statistically significant. Efficacy against infections that persisted for at least six months and for at least 12 months were also statistically indistinguishable according to doses received. Findings were essentially the same in a more restricted subset of women that excluded those who, at the time of enrollment, tested positive for any of 14 high-risk HPV types, were HPV 16 or 18 seropositive or had cervical cytology abnormalities.

Among women who received three doses, vaccine efficacy was 60% for HPV types 31, 33 and 45. This cross protection extended to incident infections that persisted for at least six months or for at least 12 months.

In an additional analysis restricted to women in the Costa Rica Vaccine Trial who had received two doses, the efficacy of the vaccine against incident infection with HPV types 16 and 18 was similarly high whether women received their second dose one month after the first dose (i.e., on schedule) or six months after the first dose (75% and 83%, respectively). The vaccine had 68% efficacy against incident infection with HPV types 31, 33 and 45 among those receiving the second dose at six months (a value similar to that seen with all three doses), but it was not efficacious against these types when the second dose was received one month after the first.

The analysis offers new evidence that one or two doses of HPV vaccine provides protection similar to that of the full three-dose course, according to the investigators. Women were not randomized by dose, few received just one dose and regulators typically do not accept incident HPV infection as a surrogate for cervical cancer, they acknowledge; therefore, the findings are unlikely to lead to policy change. Nonetheless, "these data strongly argue for a direct assessment of one-dose efficacy of the HPV-16/18 vaccine…. If one-dose HPV vaccine administration provides strong protection against HPV-16/18 for the long term, this approach might be what is necessary to overcome the barriers prohibiting vaccine uptake in many world regions," the investigators note.

The author of an accompanying comment contends that the findings appear to be valid given that within each dose group, women who received the HPV vaccine and control vaccine were well matched with respect to HPV risk and other potential confounders.² Furthermore, one-time HPV detection is actually "a high bar for vaccine efficacy estimates," as it may simply result from recent intercourse and not true infection, making it more challenging to assess the vaccine’s true impact. "If this finding [of adequate one-dose efficacy] is confirmed, it opens up a great opportunity to extend the reach of protection using HPV vaccines to more people than we would have previously thought possible," she concludes.—S. London

REFERENCES

1. Kreimer AR et al., Efficacy of fewer than three doses of an HPV-16/18 AS04-adjuvanted vaccine: combined analysis of data from the Costa Rica Vaccine and PATRICIA trials, Lancet Oncology, 2015, 16(7):775–786.

2. Brotherton JML, Could one dose of bivalent HPV vaccine prevent cervical cancer? Lancet Oncology, 2015, 16(7):739–740.