Therapeutic Vaccination against Human Papillomavirus Type 16 for the Treatment of High-Grade Anal Intraepithelial Neoplasia in HIV⁺ Men

This article is featured in Selected Articles from This Issue, p. 4019

Several malignancies, in particular anal carcinoma, are observed in excess among HIV⁺ patients (1). While anal cancer is an uncommon disease in the general population, in HIV⁺ men who have sex with men (MSM) the incidence increased from 22 per 100,000 person-years in the pre-combination antiretroviral therapy (cART) era to 137 in the late cART era (2, 3). Like cervical cancer, anal cancer is causally linked to infections with high-risk human papillomaviruses (HPV), predominantly HPV type 16 (HPV16), and is preceded by premalignant high-grade anal intraepithelial neoplasia (HGAIN or AIN 2/3), also referred to as anal high-grade squamous intraepithelial lesions (aHSIL; refs. 4–6). In a recent meta-analysis, the pooled prevalence of aHSIL was 22.4% in HIV⁺ MSM (7). Prophylactic vaccination with the quadrivalent HPV vaccine reduced the rates of both low- and high-grade AIN among HIV-negative MSM (8), which underscores the role of HPV-specific immunity in controlling the development of dysplasia. As in cervical intraepithelial neoplasia (CIN), early diagnosis and treatment of AIN has been shown to prevent malignancy (9).

Several strategies have been developed to treat AIN, including electrocautery, infrared coagulation, photodynamic therapy, topical imiquimod, or topical 5-fluorouracil (5FU), but success rates are low. Persistence or recurrence of lesions occurs frequently after treatment (10–12). Thus, there is a need for additional strategies in AIN management.

An alternative approach might be therapeutic vaccination with a synthetic long peptide (SLP) vaccine comprising a mix of long overlapping peptides from the HPV16 oncoproteins E6 and E7 (HPV16 SLP). This strategy was tested in women with vulvar high-grade squamous intraepithelial lesions (vHSIL), a condition with a comparable pathogenesis, including failure to raise effective HPV-specific immune responses that control established HPV-induced lesions (13). In two independent trials (13, 14), therapeutic vaccination with HPV16 SLP in women with HPV16-positive vHSIL was well tolerated, and proved to be clinically effective in > 50% of these patients, while the reported spontaneous regression is < 2% (15). Clinical reactivity was significantly associated with the induction of a strong and broad HPV16-specific immune response (13, 14, 16). The obvious similarities between the immunocompetent women and the HIV⁺ men on cART in the failure of the immune system to combat HPV suggest that vaccine-mediated activation of HPV16-specific immunity could have an effect in the latter group of patients. In this phase I to II dose-finding study, we assessed the safety and efficacy of intradermal administration of the HPV16 SLP vaccine in HIV⁺ men with therapy-resistant or recurrent HPV16-positive HGAIN.

We performed an investigator-initiated, government-granted, phase I to II dose-finding study (VACCAIN-T study) to assess the safety and efficacy of intradermal administration of the SLP vaccine SLP-HPV-01, with or without the administration of pegylated interferon α (IFNα; PegIntron), in HIV⁺ men with therapy-resistant or recurrent HPV16-positive HGAIN. The SLP-HPV-01 vaccine consists of two drug products, HPV-DP-7P and HPV-DP-6P. HPV-DP-7P contains seven E6 peptides. HPV-DP-6P contains four E7 peptides and two E6 peptides (14).

The study was a dose-response study, with four different dosage schedules (1-5-10; 5-10-20; 10-20-40; and 40-40-40-40 μg of SLP-HPV-01 administered intradermally with a 3-week interval, each drug product in a different limb), tested in 10 patients per dose level (DL). In each dosage group, 5 patients also received, one hour after each vaccination, a subcutaneous injection of 0.5 μg/kg body weight of pegylated IFNα-2b (PegIntron, Schering-Plough, the Netherlands) within a 10-cm radius of each vaccination site, yielding a total of 1 μg/kg body weight. The rationale for intradermal administration of the vaccine, the starting dose, and the use of PegIntron as adjuvant is provided in Supplementary Material 1. Each of the eight vaccination schedules was given to cohorts (C1-C8) of 5 patients, starting with the lowest DL1 without IFNα, followed by DL1 with IFNα (Table 1). The vaccination schedule in HIV⁺ men that would induce an HPV16-specific response by IFNγ–enzyme-linked immunospot (ELISPOT) comparable with that of previously vaccinated vHSIL patients with a clinical response (13) would be considered the optimal schedule.

The study was conducted in accordance with the Declaration of Helsinki. The study protocol was approved by the ethics review board at the Dutch Central Committee on Research Involving Human Subjects (CCMO), the Hague in the Netherlands, and overseen by an independent data and safety monitoring board. The trial is registered at ClinicalTrials.gov, NCT01923116.

Male patients over 18 years of age attending the HIV and Dermatology outpatient clinics of the Academic Medical Center in Amsterdam, the Netherlands, were eligible for inclusion if they were HIV⁺ MSM; had a CD4 count >350/μL (maximum 3 months before screening visit); had biopsy-proven intra-anal HGAIN caused by HPV16, resistant to, or recurring after previous treatment with cauterization (or other ablative treatment), 5FU, or imiquimod; had a Karnofsky performance score of ≥60; and had normal pretreatment laboratory blood values: white blood cell counts (WBC) >3×10⁹/L, lymphocytes >1×10⁹/L, thrombocytes >100×10⁹/L, and hematocrit >30%. In consenting patients with HGAIN resistant to, or recurring after previous treatment, the causative HPV type was determined in stored biopsies of these patients, using in situ PCR and laser capture microdissection (LCM; ref. 17). Only patients with HPV16-induced lesions were eligible for the current study. Exclusion criteria were immunosuppressive medication or other diseases than HIV associated with immunodeficiency; a life expectancy <1 year; a history of anal carcinoma; contraindications to receive IFNα, i.e., severe cardiac, thyroid, hepatic, or central nervous system disease including severe depression in the past; previous HPV vaccination; and current hepatitis C treatment with IFNα. Written informed consent was obtained from all participants.

According to the clinical development paradigm for cancer vaccines by the Cancer Vaccine Trial Clinical Working Group (CVCTWG; ref. 18) a group size of 10 patients per DL was chosen. Thereby, the trial satisfied the CVCTWG criteria for a trial with the objectives to cumulate a safety database, determination of dose and schedule, and demonstration of biologic activity (i.e., any effect of the vaccine on the target disease or the host immune system, e.g., immune response), as the minimal requirements would be at least 20 patients and at least 6 patients per dose group.

Patients stayed under clinical observation for 4 hours after each vaccination, with 1-hour checks of side effects, blood pressure, heart rate, and injection sites. Monitoring for spontaneously reported adverse events (AE) and injection-site reactions (redness, swelling, pain, induration) was done weekly for 3 weeks after each vaccination. Clinical assessments and laboratory tests (routine hematology and chemistry) were performed before the second, third, and if applicable the fourth vaccination and thereafter every 3 months for a total of 18 months of follow-up. At each follow-up visit, the former injection site was evaluated for any visible persisting skin reaction. AEs were graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 4.0 (19).

Screening for AIN was performed by high-resolution anoscopy (HRA), as described previously (20). HRA by the same, experienced anoscopist was performed at inclusion, and repeated at 3, 6, 12, and 18 months. At each HRA session, detailed (annotated) photos of lesion sites were obtained. To assign the correct biopsy site and to distinguish between persisting and new lesions, previous photos were compared with the current HRA view, and the exact lesion description was taken into account. If necessary, a second anoscopist reviewed the photos to achieve consensus. Biopsies of (former) lesion sites were obtained at each HRA.

For histologic analysis, formalin-fixed, paraffin-embedded tissue sections were stained with hematoxylin and eosin and, in case of any doubt of diagnosis, immunohistochemically stained for P16 [Immunologic (ILM0632-C1), clone MX007 on Ventana Benchmark Ultra]. Grading of the lesions was performed by two experienced pathologists according to the criteria of the Armed Forces Institute of Pathology, with consensus review by the two pathologists. HPV detection and genotyping on paraffin-embedded sections of biopsy specimens was performed as described in Supplementary Material 2. All histologic and virology results were evaluated in a blinded fashion with respect to clinical outcome.

Venous blood samples were obtained before the first vaccination (baseline, BL), 3 weeks after the first vaccination (visit 1, v1), 3 weeks after the second vaccination (visit 2, v2), 3 weeks after the third vaccination (visit 3, v3), and if applicable 3 weeks after the fourth vaccination (visit 4, v4).

The blood samples were analyzed to determine the vaccine-induced HPV16-specific immune response by the lymphocyte stimulation test (LST) for ex vivo detection of HPV16-specific proliferation and the validated 4-day IFNγ-ELISPOT assay for the quantification of the numbers of HPV16-specific IFNγ producing T cells. Details on the analyses are provided in Supplementary Material 1. These assays have shown a strong correlation with clinical efficacy in previous trials (13, 16, 21).

The primary clinical endpoints were toxicity/safety and the regression of the HPV16-positive lesions at 3, 6, and 12 months, as assessed by HRA, with biopsies of lesion sites. Complete response (CR) was defined as disappearance of lesions as assessed by HRA, confirmed by biopsy of the former lesion site or, if lesions persisted, histologic resolution of AIN. Partial response (PR) was defined as regression from high- to low-grade AIN. No response (NR; stable or progressive disease) was defined as histologic persistence of HGAIN (18).

Secondary endpoints were regression of lesions at 18 months, and HPV16-specific immunity in blood as assessed by several complimentary tests, including the LST and the IFNγ-ELISPOT assay.

AEs were reported using descriptive statistics. Clinical responses at 12 months, as assessed by HRA, were reported as the percentage of patients with a given response (CR, PR, NR). The clinical responses were evaluated for patients who received at least one vaccination (intention-to-treat).

Clinical responses at 18 months were assessed by HRA. Prespecified analyses were used for the measurement and reporting of vaccine-induced HPV16-specific T-cell responses (13, 16, 21).

To assess if the dose of intradermally administered HPV16-SLP would generate a similar immune response as the clinical responders in our previous study with HPV16-positive vHSIL patients (13), responses were compared. The sum of spots, measured per 100,000 peripheral blood mononuclear cells (PBMC) to all six different HPV16 peptide pools used for immunomonitoring, was compared at each time point of vaccination in the HGAIN patients to the time point with the highest vaccine-induced response of the vHSIL patients, using the nonparametric Mann–Whitney test. A significantly lower response was considered to be inferior.

In post hoc exploratory analyses, the effect of IFNα on the vaccine-induced HPV16-specific T-cell response was analyzed by comparing the sum of spots per 100,000 PBMC between patients with or without IFNα at each DL using the nonparametric Mann–Whitney test. Because there were no significant differences between the two groups of patients at each DL, all patients in one DL were grouped to study the correlation between the DL and the detection of HPV16-specific T-cell responses by 0, 1, or 2 different assays, tested by χ² test for trend, grouping DLs 1 and 2 as well as 3 and 4 together.

The correlation between immunologic responses and the highest clinical outcome at 18 months was analyzed by comparing the sum of spots measured per 100,000 PBMC to all the different HPV16 peptide pools at each of the time points of the patients with NR to the same time points of the patients with PR or CR using the nonparametric Mann–Whitney test. In addition, this test was used to compare the strength of the vaccine-induced HPV16-specific T-cell response, defined as the mean of the maximal response to each peptide pool over all time points, between NR and PR/CR at 18 months.

The data generated in this study are available upon request from the corresponding author.

Overall, 272 HGAIN lesions of 134 patients were screened. In 81 patients (60.4%) all lesions were HPV16-negative, leaving 53 eligible patients. For various reasons, 13 patients could not be enrolled, and the remaining 40 HIV⁺ men received the vaccinations (Supplementary Fig. S1). The median age of the patient group was 53.5 [interquartile range (IQR), 46.3–58.5] years, with a median nadir CD4⁺ T-cell count of 164×10⁶/L (IQR, 70–275) and a median current CD4⁺ T-cell count of 770×10⁶ cells/L (IQR, 565–960). All but one participant were on antiretroviral therapy; 38 patients had a suppressed HIV plasma viral load (HIV-RNA < 40 copies/mL). In 15 patients the HPV16-positive HGAIN lesions were refractory to earlier treatment, in 22 patients lesions recurred after previous initial successful treatment, and 3 patients had both refractory and new lesions.

After completion of the vaccination series in the DL Cohorts C1 and C2, C3 and C4, and finally C5 and C6, the blood samples of each patient were analyzed in a single experiment, to determine the time course of the HPV16-specific immune response and to assess whether the vaccine-induced HPV16-specific T-cell responses were comparable in strength with that of the vHSIL patients who showed complete histologic regression of their lesions in the previous study (13). As this was not the case, a further dose escalation was performed to complete cohorts C7 and C8. At this DL, no significant difference in the strength of the response was found when compared with the vHSIL patients (Fig. 1).

In general, the vaccine was well tolerated. One patient, in the highest dose group (C8), developed a generalized rash, itching, sweating, nausea, and vomiting within 30 minutes after the third administration of the SLP-HPV-01 vaccine, but before IFNα was administered. He remained hemodynamically stable and clemastine was administered. Within 2 hours all signs and symptoms had disappeared. For this reason, the fourth vaccination was not given to this patient. Other grade 3–4 events were found to be unrelated to the vaccine: one tendon rupture; one traumatic fracture; two episodes of severe transaminase elevations, which turned out to be caused by an acute episode of hepatitis E (hepatitis E-RNA positive) and concomitant use of stanozolol, respectively; and 2 patients experienced vasovagal collapses, respectively 9 and 11 days, and 11 months after the vaccinations.

Almost all patients experienced transient grade 1–2 AEs (Tables 2 and 3). The majority of patients receiving the vaccine experienced tiredness during the first 3 weeks after vaccination. The majority of patients also receiving IFNα experienced its well-known side effects during the week after administration: tiredness, malaise, fever, chills, flu-like symptoms, muscle pain, and headache. These symptoms abated after the first week. In both groups the main side effects observed after week 3 were tiredness, flu-like symptoms, and elevated transaminases. As these long-term side effects often started months after vaccination, the causal link with the vaccine was often unclear.

Common injection-site reactions were mostly mild (grade 1), with redness and swelling, especially at the higher DLs (Table 3). In addition, 16 patients (40%) experienced mild itching at the vaccination site during the first 3 weeks after vaccination; in only 2 patients (5%) this occurred after 3 weeks. Eight patients (20%) experienced mild pain at the vaccination site, mainly during the first week.

Results of HRA during follow-up are shown in Table 1. Two patients were excluded, both from DL1-C2, because in retrospect the lesions were not caused by HPV16. One patient (C3) had no immunologic response at 3 months follow-up and withdrew from the study at his request, and in 1 patient (C5) at 6 months an anal carcinoma was suspected. This was not confirmed after additional biopsies, but he was excluded from the study and received additional treatment. Therefore, 36 patients were evaluated at 12 months of follow-up (Supplementary Fig. S1).

At 12 months follow-up, 5 of 38 intention-to-treat patients had a complete clinical and histologic response (CR) and 1 patient had a PR of the HPV16-induced lesions. The response rate for this group at 12 months was 15.8% (6/38). After this visit 16 additional patients were excluded by the investigators, because they received additional treatment, in general electrocautery, upon failure to respond clinically or because of withdrawal of informed consent (Supplementary Fig. S1). Therefore, 20 patients could be evaluated at 18 months of follow-up. Seven of them had a CR; one had a PR. In total, 8 patients showed a CR and one patient a PR at 12 and/or 18 months of follow-up, indicating an overall clinical response rate of 23.7% (9/38). The highest DLs (DL3/4) comprised the majority (n = 6) of clinical responders.

A vaccine-induced HPV16-specific T-cell response to at least one of the 6 tested peptide pools at one or more time points after the first vaccination was found in 19 of the evaluable 36 patients (52.8%) by proliferation in the LST and in 27 of the 36 (75%) as measured by IFNγ-ELISPOT (Supplementary Tables S1 and S2). There was no indication for more or stronger reactivity when IFNα was injected near the vaccination site, as the strength of the responses of individual patients in each group showed overlap (Supplementary Fig. S2). A positive association was observed between the DL and the detection of vaccine-induced responses to none, one, or both assays (P < 0.0001, χ² test). Notably, all but 1 patient in the highest DL4 displayed reactivity in both assays (Fig. 2A). A subset of 8 patients, of whom the PBMC showed ELISPOT reactivity against 2–6 different peptide pools and of whom almost half showed a clinical response, were used to analyze the type of T cells responding to the vaccine. In all patients a strong CD4⁺ T-cell–mediated response was detected. The CD8⁺ T-cell response was weaker and detected in half of the patients (Supplementary Fig. S3).

Exploratory analyses assessing the relationship between the vaccine-induced HPV16-specific T-cell response and clinical outcome indicated that the clinical responders tended to show more often reactivity in two different assays (P < 0.0001; χ² test; Fig. 2B). They also displayed more vaccine-induced HPV16-specific T cells already after the first vaccine administration (P = 0.008; Mann–Whitney test; Fig. 3A). Moreover, the strength of the ELISPOT response tended to be higher in the CR/PR group than in nonresponders (P = 0.08; Mann–Whitney test; Fig. 3B). Notably, we failed to detect a vaccine-induced HPV16-specific response in one of the patients with a CR.

The intradermal injection of the HPV16-SLP vaccine was well tolerated in our cohort of HIV⁺ men. Adverse effects attributable to the HPV16-SLP vaccine were mainly tiredness and mild local skin reactions. The skin reactions lasted for more than 3 months in the majority of patients in the highest DL groups, but no ulcerations were seen, which was a problem for use in patients with precancerous cervical lesions who were treated with the vaccine in combination with the adjuvant Montanide (13, 22). The vaccine was immunogenic at the highest dosage level (four vaccinations of 40 μg/peptide), with the strength of the vaccine-induced response being close to the levels observed previously in the patients with HPV16-induced vHSIL, who received a higher subcutaneous dosage of the vaccine in incomplete Freund's adjuvant (Montanide ISA-51), and of whom 50% showed a clinical response (13, 16). At this highest dosage level, 4 of 10 patients had a clinical response, and the immune responses were significantly earlier and tended to be higher in patients with a clinical response. However, one of the clinical responders did not show a detectable vaccine-induced response, which may be explained by the spontaneous regression rate of about 20% in HGAIN (23, 24). Predictors of regression were, amongst others, age below 45 years and no persistent HPV16 (24). As we only included HPV16-positive HGAIN resistant to, or recurring after previous treatment, we would expect a lower spontaneous regression rate in our patients. Therefore, the majority of clinical responses in the highest DL4 are considered to be related to the administration of the SLP-HPV-01 vaccine. Despite the fact that some patients displayed a strong T-cell response to the vaccine, this was not always met with a clinical response. We had observed the same in our vaccine trials in patients with high-grade vulvar lesions. More in-depth studies in that group of patients revealed that the lack of a preexisting coordinated immune cell infiltration in the lesion precluded the induction of a clinical response even when patients did mount strong T-cell reactivity to the vaccine (25).

The clinical response rate of 40% in the highest DL of the HPV16-SLP vaccine is lower than reported response rates for HGAIN treatment by conventional ablative therapies like infrared coagulation (62%; ref. 26), cryotherapy (60%; ref. 27) or electrocautery (69%; ref. 11). However, in the current trial only patients with HGAIN resistant to, or recurring after—often multiple rounds of—conventional treatment were included. Moreover, no restrictions in lesion size or number were applied. Both could contribute to the lower response rate for vaccination in the current study population.

Notably, in only 40% of the screened patients the HGAIN lesion was caused by HPV16. According to a recent systematic review, anal cancer in HIV⁺ men is caused by HPV16 in 67% of cases (28), and therefore also HGAIN caused by other HPV types requires treatment. The observed percentage HGAIN caused by HPV16 (39.6%) was lower than the 51% reported in the meta-analysis (28). However, anal swabs and even biopsies in HIV⁺ men often contain more than one HPV type (17, 28), and we showed earlier that analysis of anal swabs and whole tissue sections is not sufficient to determine the causative HPV type when multiple HPV types are present (17). We therefore used microdissection (LCM) and in situ PCR to determine the single causative HPV type of the lesion, explaining the lower HPV16 positivity rate.

In our trial, we used intradermal injections as the administration route for the vaccine. In the vHSIL study (13), the vaccine was given subcutaneously and the adjuvant Montanide ISA-51 served as a depot function, next to an adjuvant function. However, given local tolerability issues, in particular the development of long-lasting skin ulcerations, an alternative route was considered. Intradermal dosing is known to have about a 10-fold dose sparing effect compared with subcutaneous administration. Using the intradermal route, the dermis can serve as a depot for prolonged release of the vaccine. Several studies support the idea that intradermal injection allows dose sparing if effector T-cell but also antibody responses are intended (29, 30). Intradermal administration of the SLP vaccine has been tested clinically in healthy volunteers and cancer patients as part of a skin test (31). Women with cervical neoplasia (n = 11) and healthy individuals (n = 19) were intradermally challenged with a single dose of 8 different pools of HPV16 E2, E6, and E7 peptides (at 10 μg per peptide pool; ref. 31). The injections were perceived as mildly painful and no AEs were observed. The majority of skin reactions appeared significantly earlier in HPV16-positive patients (<8 days) than in healthy subjects (8–25 days). The development of late skin reactions in healthy subjects was associated with the appearance of circulating HPV16-specific T cells and the infiltration of both HPV16-specific CD4⁺ Th1/Th2 and CD8⁺ T cells into the skin (31). These data showed that the intradermal injection of pools of HPV16 SLPs is safe and results in the migration of HPV16-specific T cells into the skin as well as in an increase in the number of circulating HPV16-specific T cells (31). Given this response, 1-5-10 μg/peptide pool was the starting dose for our current trial, but consistent strong vaccine-induced T-cell responses were only observed at the highest DL in this study.

We investigated earlier whether combination of IFNα with a SLP (p53-SLP) vaccine was both safe and able to improve the induced p53-specific IFNγ response (21). When compared with an earlier performed trial with p53-SLP vaccination alone, the combination with IFNα was found to induce significantly more IFNγ producing p53-specific T cells. In the current trial, where the effect of IFNα was tested head-to-head, no consistent effect of IFNα on the vaccine-induced HPV16-specific IFNγ response was observed. We have also tested the potential of IFNα to improve the vaccine-specific T-cell responses in another trial, where higher doses of vaccine were used, but in that study no effect of IFNα was found either (32).

Finally, a potential concern was the immunogenicity of vaccines in HIV⁺ patients. For that reason, a current CD4 level > 350 × 10E6/L was an inclusion criterion in our study, and the median CD4 level was 770 × 10E6/L CD4⁺ cells. The currently available prophylactic HPV vaccines show very good humoral immunogenicity in HIV-infected persons (33, 34).

There are a few earlier studies reporting successful use of therapeutic vaccination for HPV-induced precursor lesions. VGX-3100 consists of two DNA plasmids encoding optimized synthetic consensus E6 and E7 genes of HPV16 and HPV18. In women, use of this vaccine reported a 49.5% histopathologic regression rate in CIN2/3 as compared with a 30.6% response rate in placebo recipients (35). There are two earlier studies on therapeutic vaccines for HGAIN in HIV⁺ men. The vaccine ZYC101 was tested in 12 HIV⁺ men with anal dysplasia. ZYC101 is composed of plasmid DNA encapsulated in biodegradable polymer microparticles. The plasmid DNA encodes multiple HLA-A2–restricted epitopes derived from the HPV16 E7 protein. The investigational agent was well tolerated. Three subjects experienced partial histologic responses. Using a direct ELISPOT assay, 10 of 12 subjects demonstrated increased immune responses to the peptide epitopes encoded within ZYC101; each continued to show elevated immune responses 6 months after the initiation of therapy (36). In the second study, a therapeutic vaccine consisting of a fusion of the HPV16 E7 protein and the Mycobacterium bovis heat shock protein 65 (SGN-00101) was tested in 15 HIV⁺ subjects with HGAIN. The vaccine was well tolerated, and 5/15 showed evidence of histologic regression (37). However, no further studies have been reported for these vaccines.

In conclusion, in HIV⁺ men with HPV16-positive HGAIN the intradermally administered HPV16-SLP vaccine with or without IFNα (PegIntron) had satisfactory immunogenicity at the highest DL associated with clinical efficacy in 4 of 10 individuals. Since only 39.6% of the screened patients had an HGAIN lesion caused by HPV16, and 67% of anal cancers are reportedly caused by HPV16 (28), further studies should aim at inclusion of antigens from other subdominant HPV types (18, 33, 52) associated with anal cancer. In addition, the immunogenicity and efficacy of the current vaccine can conceivably be enhanced by addition of a defined adjuvant such as a TLR ligand or by combination with checkpoint blockade (38, 39).

S.H. van der Burg reports personal fees from ISA Pharmaceuticals during the conduct of the study; S.H. van der Burg also reports personal fees and other support from Mendus AB and PCI Biotech, as well as other support from Frame Pharmaceuticals outside the submitted work. In addition, S.H. van der Burg has a patent for WO 02/07006 issued and licensed to ISA Pharmaceuticals, a patent for WO 2006/115413 issued and licensed to ISA Pharmaceuticals, a patent for WO 2998/147187 issued and licensed to ISA Pharmaceuticals, and a patent for WO 2009/002159 issued and licensed to ISA Pharmaceuticals. W.J.T.A. Krebber reports personal fees from ISA Pharmaceuticals BV outside the submitted work, as well as beneficiary of a management participation plan of ISA Pharmaceuticals BV. C.J.M. Melief reports personal fees from ISA Pharmaceuticals during the conduct of the study, as well as personal fees from ISA Pharmaceuticals outside the submitted work; in addition, C.J.M. Melief is an inventor on numerous patents and patent applications for use of SLPs as therapeutic vaccines against lesions caused by high-risk HPV16 pending and issued. H.J.C. de Vries reports grants from the Netherlands Organization for Health Research and Development during the conduct of the study. J.M. Prins reports grants from the Netherlands Organization for Health Research and Development and Stichting Pathologie Onderzoek en Ontwikkeling during the conduct of the study. No disclosures were reported by the other authors.

The funder of the study had no role in study design, data collection, data analysis, data interpretation, writing of the report, or in the decision to submit the paper for publication. The corresponding author had full access to all the data in the study and had final responsibility for the decision to submit for publication.

K.C.M. Gosens: Resources, data curation, formal analysis, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. S.H. van der Burg: Conceptualization, resources, data curation, formal analysis, supervision, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. M.J.P. Welters: Resources, data curation, formal analysis, supervision, validation, investigation, writing–original draft, project administration, writing–review and editing. S. Boekestijn: Resources, investigation, project administration. N.M. Loof: Resources, investigation, project administration. W.G.V. Quint: Formal analysis, supervision, validation, investigation, methodology, writing–original draft. C.J.M. van Noesel: Resources, validation, investigation, methodology, writing–original draft. A.C. van der Wal: Resources, validation, investigation, methodology. O. Richel: Conceptualization, supervision, funding acquisition, visualization, methodology. W.J.T.A. Krebber: Conceptualization, resources, funding acquisition, methodology. C.J.M. Melief: Conceptualization, supervision, validation, visualization, methodology, writing–original draft. H.J.C. de Vries: Conceptualization, data curation, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft. J.M. Prins: Conceptualization, data curation, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing.

This study was funded by the Netherlands Organization for Health Research and Development (ZonMw), grant 95103003, and was supported by the (non-commercial) Stichting Pathologie Onderzoek en Ontwikkeling.

We thank all participants in the trial. We thank Ilina Ehsan, Vanessa van Ham, and Lien van der Minne for their help with isolating PBMC from the blood samples (LUMC, Leiden); Hans-Erik Nobel for his help with trial management (Amsterdam UMC, Amsterdam); Henk van den Munckhof for his help with HPV-genotyping (DDL, Rijswijk); Sonja Visscher for her contribution in the early phase of the trial (ISA Pharmaceuticals, Oegstgeest); and Ciska van Doesum-Wolters for critical reading of the manuscript and valuable suggestions (ISA Pharmaceuticals, Oegstgeest). We thank the members of the DSMB: prof. R.J.M. ten Berge (Amsterdam UMC, Amsterdam), dr. F.P. Kroon (LUMC, Leiden), and prof. P.M.M. Bossuyt (Amsterdam UMC, Amsterdam).

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.