Grading of Recommendations Assessment, Development and Evaluation (GRADE) for use of HPV vaccine in adults ages 27 through 45 years

Introduction

Three HPV vaccines are licensed for use in the United States: 9-valent and quadrivalent HPV vaccines (9vHPV and 4vHPV, Gardasil 9 and Gardasil, Merck & Co., Inc., Kenilworth, NJ) and bivalent HPV vaccine (2vHPV, Cervarix, GlaxoSmithKline, Rixensart, Belgium).123 Until October 2018, all were licensed for use in persons aged 9 through 25 or 26 years. Since late 2016, only 9vHPV has been available in the United States. In October 2018, FDA approved an expansion of the age indication through age 45 years for 9vHPV. HPV vaccination of adults in the United States was considered using Grading of Recommendations Assessment, Development and Evaluation (GRADE). The main policy question was, "Should catch-up HPV vaccination be recommended for primary prevention of HPV infection and HPV-related disease for all persons aged 27 through 45 years?"

Methods

The population of interest was adults aged 27 through 45 years at initiation of vaccination; intervention was catch-up vaccination with a complete 3-dose series of HPV vaccine (9vHPV, 4vHPV, or 2vHPV); comparison was persons through age 45 years with no catch-up HPV vaccination; and outcome was primary prevention of HPV infection and HPV-related disease.

Scientific literature was searched from January 1, 2006 through October 18, 2018 using five databases: Medline, Embase, CINAHL, Cochrane library, and ClinicalTrials.gov. Search terms used to identify clinical trials of efficacy for primary prevention of HPV-associated health outcomes and safety of 3 doses of any licensed HPV vaccine in adults (age 27–45 years at initiation of vaccination) are listed in the appendix. These searches identified 1,388 references.

Trials were excluded if they did not report original data on the relevant population, outcome, or intervention. Benefits were based on per-protocol analyses of HPV vaccine efficacy; immunogenicity data were also considered. Harms were any vaccine-related serious adverse events including deaths.

After reviewing titles and abstracts, we selected 100 references mentioning age 27 and older for detailed review. Of these, 16 publications were selected for inclusion, and 84 were excluded, 50 because they included duplicate data, and the others because they did not address the policy question: 15 did not report data on population of interest (not age-stratified), 11 did not report data on outcome of interest (not primary prevention), and 8 did not report data on intervention of interest (no HPV vaccination).

GRADE tables reference these 16 publications45678910111213141516171819; personal communication from the vaccine manufacturer providing additional age-limited analyses for previously published studies20; and a June 2018 ACIP presentation on clinical data submitted to FDA supporting 9vHPV use in adults21. In June 2019, GRADE tables were updated to include new results from a 9vHPV immunogenicity and safety trial in women age 16 through 45 years.22

Supplemental data may not directly address the policy question, and is not included in the formal GRADE scoring, but may be helpful for decision making. Supplemental GRADE tables include data from the 9vHPV trial22, an additional four studies reporting bridging immunogenicity and efficacy data from young adults23242526; two presentations on 9vHPV safety2324; and the updated FDA label for 9vHPV.1

Results

Outcomes of interest included individual benefits and harms. Benefits of interest were per-protocol analyses of HPV vaccine efficacy against persistent HPV infections; anogenital warts; HPV-related precancers including cervical or anal intraepithelial neoplasias (CIN or AIN); or HPV-related cancers including cervical, anal, penile, vaginal, vulvar, and/or oropharyngeal cancers. Immunogenicity to HPV vaccine types was also considered. Harms of interest were vaccine-related serious adverse events including deaths. (Table 1)

Included trials involved 9vHPV, 4vHPV, or 2vHPV (Table 2A), as well as supplemental immunobridging data (Table 2B). For 9vHPV, there was 1 included observational trial. For 4vHPV, there were 7 included trials: 3 randomized placebo-controlled trials, and 4 observational trials. For 2vHPV, there were 4 included trials: 2 randomized placebo-controlled trials, and 2 observational trials. Supplemental data included 4 trials providing bridging immunogenicity data, and 2 surveillance reports of post-licensure safety data for 9vHPV in the United States.

In per-protocol analyses, HPV vaccines showed significant efficacy against a combined endpoint of persistent vaccine-type HPV infections, anogenital warts, and/or cervical intraepithelial neoplasia (CIN) grade 1 or worse (Table 3A). For immunogenicity, post-vaccination seroconversion rates were high, and antibody geometric mean titers (GMTs) tended to be higher than those after natural infection. Seropositivity rates were still high at 3 to 10 years post-initial vaccination, although noticeably lower for HPV type 18 (Table 3B).

For serious adverse events, numbers were comparable among the vaccine group and the placebo group across 8 studies; there were no vaccine-related deaths (Table 3C).

GRADE was used to evaluate evidence for use of 9vHPV in adults. Initial evidence level was 1 for each outcome based on data from randomized controlled trials (Table 4). All were downgraded for indirectness since no randomized placebo-controlled trials were conducted on use of 9vHPV in this age range, and extrapolation from 4vHPV efficacy was based on immunobridging data. Outcomes for which the 95% confidence interval crossed 1 were further downgraded for imprecision. For men, evidence type for each outcome could be further downgraded for indirectness, since most trials enrolled women only.

In summary, evidence level for efficacy is 2 in women and 3 in men, and evidence level for immunogenicity is 2. Overall evidence type for benefits is level 2 (Table 5A). Overall evidence type for harms is also level 2 (Table 5B).

Supplemental immunobridging data showed non-inferior immunogenicity comparing HPV vaccination in mid-adults with HPV vaccination in young adults (Supplemental table 1), and comparing 9vHPV in young adults with 4vHPV in young adults (Supplemental table 2). Supplemental data on harms summarize U.S. postlicensure safety data for over 29 million doses of HPV vaccine (Supplemental table 3).

^{Abbreviations}

^{HPV, human papillomavirus}

^{2vHPV, bivalent HPV vaccine (Cervarix)}

^{4vHPV, quadrivalent HPV vaccine (Gardasil)}

^{9vHPV, 9-valent HPV vaccine (Gardasil 9)}

^{EGL, external genital lesions}

^{CIN, cervical intraepithelial neoplasia}

^{AIN, anal intraepithelial neoplasia}

^{CI, confidence interval}

^{M, months}

^{GMT, geometric mean titer}

^{mMU/mL, milliMerck Units per milliliter}

^{EU/mL, ELISA units per milliliter}

^{ED₅₀, effective dose producing 50% response}

^{RCT, randomized controlled trial}

^{Obs, observational trial}

Tables

Table 1: Important and critical outcomes related to HPV vaccination

Table 1 Footnotes

* No HPV-related cancers were reported in per-protocol analyses from any of the studies reviewed; data on these outcomes not necessarily expected in clinical trials of current duration/size

Table 2A: Characteristics of included studies

References in this table:456789101112131415161718192122

Table 2A Footnotes

* Age-restricted data obtained from Luxembourg, 201820

** Per-protocol results for benefits; intention-to-treat results for harms

Table 2B: Characteristics of included studies, supplemental

References in this table:21222324252627282930

Table 2B Footnotes

* Per-protocol results for benefits; intention-to-treat results for harms

Table 3A: Efficacy outcomes

References in this table:561421

Table 3A Footnotes

* Age-restricted data obtained from Luxembourg, 201820

** Per-protocol results

Table 3B: Immunogenicity outcomes

References in this table:4578910111315181922

Table 3B Footnotes

* Age-restricted data obtained from Luxembourg, 201820

** Per-protocol results

Table 3C: Harms

References in this table:568101214181922

Table 3C Footnotes

* Age-restricted data obtained from Luxembourg, 201820

** Intention-to-treat results

Table 4: Evidence for use of 9vHPV in adults ages 27 through 45 years

Table 4 Footnotes

RCT, randomized controlled trial; Obs, observational study

* Evidence levels: 1) High, 2) Moderate, 3) Low, 4) Very Low

** Strength of association, dose-response, plausible residual confounding, publication bias

¹ Downgraded for indirectness since no randomized placebo-controlled trials were conducted on use of 9vHPV in adults aged 27 through 45 years, and there are no 4vHPV efficacy trials in males aged 27 through 45 years; extrapolation of efficacy from 4vHPV across age and genders is based on supplemental bridging immunogenicity data

² Downgraded for imprecision since 95% confidence interval for efficacy includes 1

Table 5A: Summary of evidence for benefits

Table 5A Footnotes

* Evidence levels: 1) High, 2) Moderate, 3) Low, 4) Very Low

Table 5B: Summary of evidence for harms

Table 5B Footnotes

* Evidence levels: 1) High, 2) Moderate, 3) Low, 4) Very Low

Supplemental Data

Supplemental Table 1: Immunobridging comparing HPV vaccination in mid-adults with HPV vaccination in young adults

References in this table:410212223

Supplemental Table 2: Immunobridging comparing 9vHPV in young adults with 4vHPV in young adults

References in this table:262728

Supplemental Table 3: 9vHPV post-licensure safety data

References in this table:2324

Summary

After reviewing the available data including the GRADE analysis, in June 2019, ACIP recommended catch-up HPV vaccination for all adults through age 26 years. ACIP did not recommend catch-up vaccination of adults aged 27–45 years, but recognized that some adults who are not previously vaccinated may be at risk for new HPV infection and might benefit from vaccination in this age range; therefore, ACIP recommended shared clinical decision making regarding potential HPV vaccination for these individuals. See 2019 policy note Updated Recommendations of the Advisory Committee on Immunization Practices for Human Papillomavirus Vaccination of Adults.

Appendix: Search Methods