|Home | About | Journals | Submit | Contact Us | Français|
HIV-infected men are at increased risk for anal cancer. Human papillomavirus (HPV) vaccination may prevent anal cancer caused by vaccine types.
AIDS Malignancy Consortium Protocol 052 is a single-arm, open-label, multi-center clinical trial to assess the safety and immunogenicity of the quadrivalent HPV (types 6, 11, 16, 18) vaccine in HIV-infected men. Men with high-grade anal intraepithelial neoplasia or anal cancer by history or by screening cytology or histology were excluded. Men received 0.5 mL intramuscularly at entry, week 8, and 24. The primary endpoints were seroconversion to vaccine types at week 28, in men who were seronegative and without anal infection with the relevant HPV type at entry, and grade 3 or higher adverse events related to vaccination.
There were no Grade 3 or greater adverse events attributable to vaccination among the 109 men who received at least one vaccine dose. Seroconversion was observed for all 4 types: type 6 (59/60, 98%), type 11 (67/68, 99%), type 16 (62/62, 100%), type 18 (74/78, 95%). No adverse effects on CD4 counts and plasma HIV-1 RNA were observed.
The quadrivalent HPV vaccine appears safe and highly immunogenic in HIV-infected men. Efficacy studies in HIV-infected men are warranted.
Anal cancer is caused by persistent infection with high-risk types of human papillomavirus (HPV) . Approximately 66% of anal cancers are caused by HPV type 16 and 5% by HPV type 18 . HIV-1-infected men who have sex with men (MSM) are at increased risk for anal cancer , and the incidence may have increased further in the era of antiretroviral therapy (ART) [4–8]. Anal cancer prevention strategies have emerged that screen for high-grade anal intraepithelial neoplasia (HGAIN) through an algorithm of cytology and high resolution anoscopy (HRA)-directed biopsies, prompting removal or destruction of HGAIN lesions . Options for primary prevention are currently limited.
The use of the quadrivalent HPV vaccine has been shown to prevent persistent cervical HPV infection caused by vaccine types (6, 11, 16 and 18) in women, and to prevent nearly 100% of high-grade cervical, vaginal and vulvar intraepithelial neoplasia associated with vaccine types, the precursor lesions to invasive cancers at these sites, in women who were not infected at baseline [10, 11]. This vaccine prevented 93% of persistent anal infections with vaccine types in young HIV-uninfected men who have sex with men .
The quadrivalent HPV vaccine is a recombinant protein virus-like particle vaccine with a proprietary adjuvant, Amorphous Aluminum Hydroxyphosphate Sulfate . HIV-1-infected individuals have lower rates of antibody conversion to similar vaccine constructs such as the hepatitis A vaccine and hepatitis B vaccine than in HIV-uninfected individuals [14–16]. This study was designed to test the hypothesis that the quadrivalent HPV vaccine is safe and immunogenic in HIV-1-infected men.
AIDS Malignancy Consortium (AMC) Protocol 052 was a multi-center single arm open-label, pilot trial conducted at eight clinical trial sites (Boston University Medical Center, Boston, MA; Denver Public Health, Denver, CO; Laser Surgery Care, New York, NY; Montefiore Medical Center, Bronx, NY; University of California, Los Angeles, CA; University of California, San Francisco, CA; Virginia Mason Medical Center, Seattle, WA; Weill-Cornell Medical College, New York, NY). Data collection and follow-up for 18 months after entry are ongoing. Institutional review boards of the participating institutions approved the study, and each patient gave written informed consent.
Inclusion criteria were as follows: men aged 18 years or older; laboratory documentation of HIV-1 infection; if receiving ART, receipt of ART for at least 6 months prior to entry and no change in ART within 30 days of entry, CD4 count ≥200 cells/µL and plasma HIV-1 RNA level <200 copies/mL; if not receiving ART, CD4 count >350 cells/µL and no plans to start ART within 28 weeks of entry; anal cytology result that was normal, atypical squamous cells of uncertain significance (ASCUS) or low grade squamous intraepithelial lesions (LSIL); having an absolute neutrophil count >750 cells/µL, hemoglobin ≥ 9.0 g/dL, platelet count ≥100,000/µL, aspartate and alanine aminotransferase ≤ 3 times the upper limit of normal; total or conjugated bilirubin <2.5 times the upper limit of normal; calculated creatinine clearance by Cockcroft-Gault equation ≥60 mL/min; and a Karnofsky performance score ≥ 70.
Participants were excluded if they had anal or perianal carcinoma, high-grade squamous intraepithelial lesions (HSIL), atypical squamous cells suggestive of HSIL or invasive carcinoma on cytology, or HGAIN on high resolution anoscopy (HRA)-guided biopsy, at any point prior to entry. Participants were excluded if polymerase chain reaction (PCR) testing of anal swabs detected DNA for both HPV 16 and 18. Participants with recent or expected use of systemic antineoplastic agents, immunomodulatory treatments, or investigational vaccines were excluded. Participants with hemophilia, prior splenectomy or prior receipt of HPV vaccines were excluded.
At entry, week 8 and week 24, participants received the quadrivalent HPV (types 6, 11, 16 and 18) recombinant vaccine 0.5 mL intramuscularly. The primary immunogenicity endpoint was determined 4 weeks after the third vaccination.
Clinical assessments were made at each vaccination, and week 4, week 12 and week 28. Plasma HIV-1 RNA, CD4 cell counts, and routine safety laboratory monitoring were assessed at entry, week 4, week 12, and week 28. Participants were provided with a diary to report targeted symptoms occurring within 5 days of vaccination. Participants were contacted 24–48 hours after each vaccination by study personnel to assess for vaccine-related symptoms. All laboratory abnormalities, signs and symptoms were graded using the National Cancer Institute Common Terminology Criteria for Adverse Events Version 3.0 . Anal swabs for cytology and HRA with directed anal biopsies of lesions consistent with anal intraepithelial neoplasia (AIN) were performed to assess eligibility and at week 28. Briefly, HRA is visualization of the anal canal and perianal skin using a colposcope after application of 3% acetic acid . Criteria analogous to those used in cervical colposcopy are used to identify areas suspicious for AIN . Biopsies of these areas are obtained using standard biopsy forceps. A questionnaire to assess tobacco use and sexual behavior was administered at entry and week 28. Oropharyngeal samples for HPV DNA PCR and anti-HPV IgA were collected but are not yet available for analysis.
Anal swabs for HPV DNA PCR testing were obtained to assess eligibility, and at entry and week 28 as described previously . This testing was performed at a central laboratory. Specimens negative for beta-globin gene amplification were excluded from analysis. The results of PCR were recorded on a scale from 0 to 5 based on the intensity of the signal on the dot-blots, as described previously . Anal swab specimens to assess eligibility were assayed for HPV types 16 and 18 only. However, four participants were missing anal swab specimens from entry or the specimen was unsatisfactory. Stored anal swab specimens obtained within 45 days of entry to assess eligibility were re-assayed for the full panel of HPV infections.
Serum was collected for assessment of antibody responses at entry and week 28. We analyzed vaccine type epitope-specific neutralizing anti-HPV antibodies with a competitive Luminex-based immunoassay (Merck Research Laboratories) as described previously [22, 23]. These assays were performed under the direction of Merck Research Laboratories at PPD Vaccines and Biologics Laboratory (Wayne, PA).
Separate analyses were conducted for HPV types 6, 11, 16 and 18. The primary endpoint was seroconversion at week 28. This was defined as being seronegative at entry (i.e. having no detectable type-specific anti-HPV antibodies or detectable antibodies below the pre-specified cut-off for the assay) and seropositive at week 28 (i.e. antibody concentrations above the assay cut-off). For inclusion in the primary per-protocol analysis for a given HPV type, participants were required to be seronegative for that type, and to have no HPV DNA detected for that type by PCR of the anal swab at entry and screening. Participants were also required to have received 3 vaccine doses. The proportion of participants experiencing seroconversion and the accompanying exact binomial lower one-sided 95% confidence interval were reported for types 6, 11, 16 and 18.
The study was designed so that using a one-sided alternative type 1 error rate of 5%, a sample size of 40 eligible participants in the per-protocol analysis for a given type would provide approximately 80% power to exclude a seroconversion rate of 45% or less assuming that the seroconversion rate was 65%. Because this was a single-group pilot trial, the type-specific Ab responses were reported with one-sided 95% confidence intervals with precision for the lower bound of efficacy. Very few data existed on the rate of seropositivity in this population at the time of AMC052 protocol design [24, 25]. We assumed that 50% of participants would be excluded from the primary analysis because of seropositivity prior to vaccination, and 10% would have detectable HPV DNA for a given type at entry. The sample size was increased an additional 10% to account for attrition for a planned sample size of 110 participants.
All 109 participants who received at least one vaccine dose were included in the safety analyses. The primary safety endpoint was occurrence of grade 3, 4, or 5 adverse events that were possibly, probably or definitely related to vaccination as determined by the site investigator. The proportion of participants experiencing such adverse events and corresponding one-sided 95% confidence interval were reported. Pre-planned secondary safety analyses were to report the proportion of participants experiencing a grade ≥2 AE related to vaccination, and to report changes in plasma HIV-1 RNA and CD4 cell counts stratified by ART use.
The geometric mean concentration (GMC) of anti-HPV antibodies and the accompanying 95% confidence interval were calculated for the primary per-protocol population for each type as a pre-specified secondary endpoint. Participants with undetectable antibody concentrations were nominally assigned a value of 50% of the detection limit for the assay. The plan in the protocol was to assess predictors of seroconversion using logistic regression. Because the number of participants without seroconversion was smaller than hypothesized by the protocol, the logistic regression analysis was not performed. In a post-hoc analysis, we analyzed the relationship of CD4 cell count at entry, nadir CD4 count, ART use, age and antibody concentration at entry, and HPV DNA detection to log-transformed antibody concentrations at week 28 was assessed using multivariable linear regression.
One hundred twelve participants were enrolled between January 9, 2008 and November 24, 2008. The flow of study participants who gave informed consent for study participation is shown in the Figure. Of note, of 123 participants excluded, 50% had HGAIN by histology and 5% had HSIL based on cytology. Baseline characteristics of the 112 participants who entered the study are shown in Table 1. Anal cytology results were normal in 48%, ASCUS in 35%, and LSIL in 17%. No AIN was found on HRA in 71% of participants and 29% had low-grade AIN.
Twelve participants were not included in the per-protocol immunogenicity analyses: 2 participants did not receive any vaccine, 2 were subsequently found to be ineligible (one did not receive any vaccine doses and one received all 3 vaccine doses), 4 received one vaccine dose but did not complete study follow-up, and one participant did not receive the third vaccine dose per protocol and remained in study follow-up. Three additional participants were missing either the baseline or week 28 serum samples.
Immunogenicity endpoints are shown in Table 2. The observed proportion of participants with seroconversion at week 28 (and the lower one-sided 95% confidence interval) in the primary per-protocol analysis were for type 6: 98% (92%), for type 11: 99% (93%), for type 16:100% (95%), and for type 18: 95% (89%). All of these confidence intervals excluded the 45% threshold supporting the primary hypothesis. When including all 100 participants in the per-protocol analysis without regard to baseline seropositivity or HPV DNA detection, 97%, 98%, 99%, and 96% had detectable antibodies at week 28 to type 6, 11, 16 and 18, respectively.
In a post-hoc analysis, we analyzed the baseline predictors of antibody concentrations at week 28. For all types, higher baseline concentrations were significantly associated with higher concentrations at week 28. Current ART use at baseline was associated with higher anti-HPV 16 concentrations (.38 log10 mMU/mL; 95% CI .05–0.72) and anti-HPV 18 concentrations (.36 log10 mMU/mL; 95% CI .03–.69). Type-specific DNA detection on anal swab at entry was associated with lower anti-HPV 11 concentrations (−.48 log10 mMU/mL; 95% CI −.86 to −.10), lower anti-HPV 16 concentrations (−.39 log10 mMU/mL; 95% CI −.68 to −.09), and marginally with lower anti-HPV 6 concentrations (−.46 log10 mMU/mL; 95% CI −.95 to .02). CD4 cell counts, nadir CD4 count, and age were not associated with antibody concentrations after adjusting for the other variables in the model.
Anal cytology and histology results were available for 105 participants at week 28. Cytology results were normal in 50 (48%) participants, unevaluable in 2 (2%), ASCUS in 33 (32%), LSIL in 16 (15%), and HSIL in 3 (3%). Twelve participants (11%) were found to have HGAIN on histology. HGAIN or HSIL was found at week 28 in 3 of 21 participants with HPV 16 detected at entry and in 1 of 8 participants with HPV 18 detected at entry. Table 3 summarizes incident and persistent HPV infection and progression to HGAIN from baseline to week 28.
There were no grade 3, 4 or 5 events that were related to vaccination among the 109 participants who received at least one vaccine dose (0%; 1-sided 95% CI 2.7%). One subject died 22 weeks after the third vaccination due to a ruptured hepatocellular carcinoma caused by hepatitis C infection. This was reported to be unrelated to vaccination. Four other participants experienced Grade 3 events not related to vaccination: cholecystitis (with concomitant grade 4 serum glucose and lipase elevations), non-septic arthritis, headache, and leukopenia. Grade 2 injection site reactions were observed in 9 (8%). Other grade 2 reactions that were related to vaccination were observed in 5 (5%) including one participant with recurrent tinnitus possibly related to vaccination leading to withholding the third vaccine dose. Grade 1 or 2 injection site reactions were observed after the first, second and third vaccinations in 18%, 17%, and 12% respectively.
For those on ART at baseline, the median CD4 counts at entry, week 4, week 12 and week 28 were 514, 540, 558 and 591 cells/µL, respectively. Plasma HIV-1 RNA levels in this subgroup were <200 copies/mL in 96%, 97%, 97% and 93% respectively. For those not on ART, the CD4 cell counts were 544, 528, 517 and 513 cells/µL and the geometric mean plasma HIV-1 RNA levels in this subgroup were 3.6, 3.4, 3.4 and 3.4 log10 copies/mL respectively.
In this pilot study, the data suggest that the quadrivalent HPV vaccine is safe and elicits anti-HPV antibodies in a high proportion of HIV-1-infected men. The proportion of men exhibiting seroconversion was 95% or greater for each of the HPV types included in the vaccine. For those with pre-existing anti-HPV antibodies, the vaccine induced a marked increase in antibody concentrations (Table 2). No serious adverse events attributable to vaccination were observed. The injection site reactions were mild or moderate and were similar to other commonly administered vaccines for HIV-1-infected adults. There were no obvious effects on CD4 cell counts or HIV-1 levels. The CD4 cell counts increased slightly for those on ART and decreased slightly for those not on ART.
The proportion of participants exhibiting seroconversion was higher than hypothesized during the study design. Approximately 65% of HIV-1-infected subjects experience antibody conversion after hepatitis A vaccination  and 18–71% after hepatitis B vaccination [15, 16]. The markedly higher antibody conversion to the quadrivalent HPV vaccine may be due to differences in the populations studied; AMC052 participants had a median CD4 count of 517 cells/L, and 92% had a plasma HIV-1 RNA level <10,000 copies/mL. We also excluded patients with current or past HGAIN at study entry. Alternatively, these apparent differences may be due to differences in assay sensitivity or the inherent immunogenicity of the vaccines.
The antibody concentrations induced by this vaccine in the primary per-protocol population were lower than those reported in other studies. For example, the anti-HPV 16 antibody concentrations are approximately 50% those of HIV-1-uninfected women 34–45 years of age  and 40% of those reported for HIV-1-uninfected men and women 16–26 years of age . The clinical significance of lower antibody concentrations is unknown as there is no established threshold correlating with efficacy in any population studied to date. Interestingly, the concentrations were similar to those of HIV-uninfected men who have sex with men of 18 to 26 years of age , suggesting that the lower antibody concentrations were not a direct effect of HIV infection. We did not observe a relationship between age, CD4 cell counts or nadir CD4 cell counts and antibody concentrations at week 28, but did observe higher anti-HPV 16 and anti-HPV 18 antibody concentrations in those receiving ART. Further study is clearly needed to determine the role of HIV status, sexual behavior and other potentially unmeasured factors that may determine immune response to the quadrivalent HPV vaccine. These regression results should be interpreted with caution since this was a post-hoc analysis and these results were not controlled for multiple comparisons.
The quadrivalent HPV vaccine is known to be a preventive vaccine and does not treat or prevent disease from prevalent infection with vaccine types [10, 11]. For a given vaccine type, between 60% and 78% of the per-protocol population were both seronegative and HPV DNA negative in the anal canal. This implies that a high proportion of men may potentially benefit from an HPV preventive vaccine despite being older than previously studied populations and having significant prior exposure to HPV infections through receptive anal sex and multiple sexual partners [12, 27]. However, efficacy would need to be confirmed by clinical trials.
This study has several limitations. The clinical significance of these findings is unknown. This study was not designed to establish the efficacy of this vaccine for preventing HPV infection or HGAIN in this population. This study had entry criteria that limit the generalizability of these results. For example, this study did not enroll patients with low CD4 counts or many participants with high plasma HIV-1 RNA levels. Nearly 30% of men entering screening were excluded because of HGAIN or HSIL found during screening, and many others were likely not approached for participation because of prior HGAIN. Further studies are needed to establish the safety and immunogenicity of the vaccine in these groups.
Anal cancer and its precursor, HGAIN, are relatively common problems in HIV-1-infected men. In this multi-center, single-arm pilot study, the quadrivalent HPV vaccine was safe and highly immunogenic in HIV-1-infected men. These data support further study of HPV vaccination in HIV-1-infected men in conjunction with anal cancer screening as measures to reduce the incidence of anal cancer.
The authors would like to thank Merck Research Laboratories for supplies of the quadrivalent HPV vaccine and Pharmaceutical Product Development, Inc. (Mark Esser, PhD) for performing the competitive Luminex Immunoassay. We thank the people who have made significant contributions including Mary Crowe, B.A., Maria Da Costa, M.S., Fred Fishman, B.S., Donna E. Neumark, RN, Ph.D., Valery Hughes, N.P., Marya Krogstad RN, BSN, MFA, Cecelia Marquez, M.D., Christina Megill P.A.-C, Frances Moran, R.N., Lacey Siekas, ARNP, Alen Voskanian, MD, Danielle C. Varney, P.A.-C, Tim Wright, B.A.
Funding/Support: K23 AI 55038 to Dr. Wilkin, UL1 RR024996 (Weill-Cornell Clinical and Translational Science Center (CTSC)), U01 CA121947-01 to Dr. Mitsuyasu, M01-RR00865 (UCLA CTSC), and UL1 RR024131 (UCSF Clinical and Translational Science Institute). Merck Laboratories funded the HPV antibody assays, and provided the vaccine doses for this study.
Potential Conflicts of Interest:
Drs. Lee, Aboulafia, Berry, Cohn, Mitsuyasu, Ms. Lensing, and Ms. Jay have no financial disclosures to report. Dr. Wilkin’s employer, Weill-Cornell Medical College, has received grant funding from Merck for research-related costs of clinical trials for which he was the Cornell PI. Dr. Stier was on the speaker’s bureau for Merck until 2008. Dr. Einstein advised or participated in educational speaking activities, but did not receive an honorarium, from Merck. In specific cases, his hospital, Montefiore Medical Center has received payment for his time spent for these activities. Montefiore has received grant funding from Merck for research-related costs of clinical trials for which he was Montefiore PI. Dr. Goldstone has received grant funding from Merck for research-related costs of clinical trials conducted Laser Surgery Care. He has served as a consultant and a paid speaker for Merck. He has received grant support from Qiagen. Dr. Saah is an employee of Merck Laboratories and owns stock in Merck Laboratories. Dr. Palefsky has served as a consultant for Merck. His employer, University of California San Francisco, has received grant funding from Merck for advisory board activities and research-related costs of clinical trials for which he was the USCF PI.
This manuscript was presented in part the 17th Conference on Retroviruses and Opportunistic Infections, February 19th, 2010, San Francsico, CA.
Timothy Wilkin, Division of Infectious Diseases, Weill Cornell Medical College, New York, NY.
Jeannette Y. Lee, University of Arkansas for Medical Sciences.
Shelly Y. Lensing, University of Arkansas for Medical Sciences.
Elizabeth A. Stier, Boston University Medical Center.
Stephen E. Goldstone, Laser Surgery Care; New York, NY.
J. Michael Berry, University of California San Francisco.
Naomi Jay, University of California San Francisco.
David Aboulafia, Virginia Mason Medical Center and University of Washington.
David L. Cohn, Denver Public Health, University of Colorado Denver Health Sciences Center.
Mark H. Einstein, Albert Einstein College of Medicine and Albert Einstein Cancer Center Montefiore Medical Center.
Alfred Saah, Merck Labortories.
Ronald T. Mitsuyasu, University of California Los Angeles.
Joel M. Palefsky, University of California San Francisco.