Preventive and therapeutic efficacy of mRNA/LNP vaccines encoding HPV16 E6 and E7 in an early-intervention HPV tumor mouse models Authors: Supichcha Saithong, Eakachai Prompetchara, Chutitorn Ketloy, Chirayus Khawsang, Kittipan Tharakhet, Papatsara Kaewpang, Nongnaphat Yostrerat, Pachara Wangsoontorn, Kunlanan Charsangbong, Kieu Lam, James Heyes, Drew Weissman, Kiat Ruxrungtham Published: 2026 ABSTRACT ABSTRACTPersistent infection with high-risk human papillomavirus (HPV) drives the development of cervical and other anogenital cancers, with limited therapeutic options for early-stage disease and precancerous lesions. There is a critical unmet need for effective therapeutic vaccines capable of inducing durable immune responses to eliminate early tumor cells and prevent disease progression. In this study, we developed lipid nanoparticle (LNP)-formulated mRNA vaccines encoding HPV16 E6 and E7 oncoproteins and evaluated their efficacy in a murine model using a low tumor inoculum (5,000–10,000 TC-1 cells) to represent a low tumor burden, early-intervention setting. E7 mRNA/LNP induced stronger antigen-specific T cell responses than E6 when administered intramuscularly at a low dose (1 µg). In therapeutic settings, initiated 3 d after tumor implantation, either as a single 10 µg dose or a two-dose regimen (10 µg followed by 5 µg, 6 d apart), E7 vaccination elicited robust T cell responses, significant tumor regression, and undetectable HPV16 E6/E7 DNA, outperforming E6 across regimens. In the preventive setting, administration of two 5 µg doses of HPV16 E7 prevented tumor establishment following TC-1 cell inoculation. Importantly, E7 vaccination induced durable immune memory, as demonstrated by sustained tumor protection and undetectable HPV E6/E7 DNA upon rechallenge at 5 months. These findings highlight the potential of HPV16 E7 mRNA/LNP as a therapeutic vaccine targeting low tumor burden HPV-associated disease and support further clinical development. A multivalent mRNA vaccine targeting multiple high-risk HPV genotypes is currently under development.KEYWORDS: HPV16, mRNA vaccine, therapeutic vaccine, E6, E7, early-intervention Introduction Human papillomavirus (HPV) infection is associated with a range of pre-malignant and malignant lesions, including cervical,1–3 anogenital, and head and neck cancers.4 HPV16 is a particularly high-risk genotype of HPV, accounting for approximately 70% of cases such cancers.5,6 During the progression of HPV-related malignancies, the viral oncoproteins E6 and E7 play critical roles by inactivating tumor suppressors such as p53 and retinoblastoma protein (pRb). These proteins also inhibit cyclin-dependent kinase (CDK) inhibitors, including p16INK4a, p21CIP1, and p27KIP1, leading to disrupted DNA repair, evasion of apoptosis, and uncontrolled cell proliferation.7 As a result, pathways involving DNA repair, cell growth, angiogenesis, and mitogenesis become upregulated, especially in cervical intraepithelial neoplasia (CIN), high-grade squamous intraepithelial lesions (HSILs), and invasive cancer.8 To date, five preventive HPV vaccines (Gardasil®4, Gardasil®9, Cervarix™, Cecolin®, and WalrinVax®) have shown strong efficacy in preventing HPV infection across different populations.9,10 However, these vaccines, based on virus-like particles (VLPs) derived from the L1 capsid protein, do not provide therapeutic benefit for individuals already infected.11 Natural HPV infection also induces only weak neutralizing antibody responses against L1.9 Currently, surgical excision is the standard treatment for HPV-related lesions, but it carries risks such as incomplete resection, infections, functional impairments, psychological distress, and prolonged recovery. Thus, therapeutic vaccines offer promise either as standalone interventions or adjuncts to standard treatments to improve outcomes.9 The oncoproteins E6 and E7 remain the most attractive targets for therapeutic HPV vaccine development due to their essential role in carcinogenesis and continued expression in malignant cells.12,13 Consequently, most therapeutic vaccine strategies focus on these antigens and are designed to elicit robust HPV-specific CD8+ cytotoxic T lymphocyte (CTL) responses.14–16 Among therapeutic platforms, DNA vaccines have progressed furthest in clinical trials, including INO-3112 (NCT02172911),17 GX-188E DNA (NCT01634503),18 and VGX-3100 (NCT01188850).19 However, DNA vaccines require electroporation devices to facilitate nuclear delivery, limiting ease of administration. The significant advancement of mRNA vaccines during the COVID-19 pandemic highlighted their considerable effectiveness.20,21 This led to the fast expansion of mRNA technology to address infectious diseases, cancer, and various other applications, including HPV-related cancers. Compared to DNA vaccines, mRNA vaccines encoding E7 elicit comparable or superior CTL responses at much lower doses, as shown in preclinical studies.14 These advantages prompted the development of an mRNA-LNP-based therapeutic HPV vaccine in this study. Despite significant progress in preclinical and early-phase clinical studies, no therapeutic HPV vaccine targeting CIN2/3 or advanced cervical cancer has yet received FDA approval.22 In mouse models, late-stage vaccination with large tumors improved survival but often failed to eliminate tumors.23,24 This discrepancy has contributed to disappointing clinical trial results: studies targeting CIN3 showed no significant lesion resolution,18,25 no significant reduction in lesion size in CIN2/326 and limited success in advanced cancer, with few or no complete clinical responses and no improvement in overall survival.27,28 Consequently, many of these vaccines are now evaluated in combination with immunotherapies such as pembrolizumab,29 imiquimod30 or nivolumab.31 These combination strategies are primarily limited to metastatic cervical and head and neck cancers and are not suitable for CIN3 due to unacceptable toxicity. Moreover, their use remains associated with significant concerns regarding treatment-related adverse events. Given these considerations, rather than continuing to test cancer immunotherapy strategies in end-stage patients with limited chances of success, the focus could shift to studying cancer vaccines in individuals with persistent infections or mild dysplasia who are at higher risk of developing HPV-related cancer. Addressing the problem at an earlier stage could increase the benefit derived from cancer vaccine treatment. In the era of mRNA vaccines, several strategies to enhance translation and expression have been established, including the use of cap analogues, capping enzymes, 5‘ UTR and 3’ UTR regions,32 poly(A) tails,33 and modified nucleosides.34 These modifications were incorporated into our mRNA constructs to aid in antigen expression, thereby improving the chances to elicit the cellular and humoral immune responses that suppress tumor growth. The de-oncogenic HPV16 E6 and E7 proteins were fused with an IgE signal peptide to direct them into the ER via the Sec translocon, where the signal peptide is cleaved. The proteins then traffic through the Golgi and are secreted into the extracellular space. Accordingly, the expressed antigens are expected to be present in the culture supernatant. Secretory targeting may facilitate uptake of extracellular antigen by professional APCs and subsequent cross-presentation to CD8+ T cells, a mechanism known to contribute to antitumor immunity.35–37 Each antigen was evaluated to determine if it could enhance adaptive immune responses to the target proteins prior to assessing their therapeutic and preventive efficacy in murine HPV16-positive TC-1 challenge models. We highlight the importance of modifying experimental models to better reflect early intervention by using low-dose TC-1 implantation and early vaccination before extensive tumor formation, thereby maximizing vaccine efficacy and informing future clinical translation. Results E6 and E7 proteins were highly expressed in mRNA-transfected VERO cells As designed (Figure 1(A)), our mRNA constructs were translated into the target proteins within the cytoplasm of transfected cells (Figure 1(B)) and secreted into cell culture supernatant, with the expected molecular weight of 16–20 kDa (Supplementary Figure 1). The translatability of each mRNA construct was assessed by comparing them to non-transfected conditions. Quantification of E6 and E7 expression was performed using specific antibody staining, with mean fluorescence intensities in transfected cells increased approximately 3-fold and 9-fold, respectively, compared to non-transfected cells (Figure 1(C)). Cellular HPV proteins expression analysis by immunofluorescence after mRNA transfection in VERO E6 cells. (A) Illustration of the mRNA structure of modified E6 and E7 (de-oncogenic proteins) fused with IgE signal peptide (SP). (B) Intracellular E6 and E7 proteins expression at 24 h post-transfection examined by immunofluorescence assay employing anti-E6 and anti‑E7 primary antibodies; the nuclei were counterstained with DAPI (blue). FITC-tagged 2nd antibodies (green) were used. (C) Green fluorescence intensity was measured randomly from 3 fields and compared between transfected and non-transfected cells. Immunization with HPV16 mRNA-LNP induced robust HPV‑specific cellular and humoral immune responses To evaluate the functionality of our vaccines encoding modified E6 and E7, mice were immunized intramuscularly (i.m.) with two doses of mRNA-LNP or normal saline solution (NSS) at a 14-day interval (Figure 2(A)). In C57BL/6 mice, HPV-specific IFN-γ-producing T cells determined by ELISpot assay were significantly increased in the E7 mRNA-LNP group, while a modest increase was observed in the E6 mRNA-LNP group (Figure 2(B), p = .0079 vs NSS control). Functional T cells from stimulated splenocytes were further characterized by flow cytometry analysis. The gating strategy for intracellular cytokine staining (ICCS) included sequential gating on live cells, CD3+, CD4+ or CD8+, and intracellular cytokine-positive populations (IFN-γ, granzyme B, and TNF-α). CD8+ T cells from the E7 mRNA-LNP group showed elevated expression of TH1 cytokines including IFN-γ, Granzyme B, and TNF-α. In contrast, E6 mRNA-LNP induced only weak cytokine responses (Figure 2(C)). CD4+ T cells demonstrated modest increases in cytokine expression, particularly IFN-γ and TNF-α, although these responses were not statistically significant (Supplementary Figure 2). Moreover, polyfunctional analysis revealed distinct functional profiles between CD4+ and CD8+ T cells following HPV16 E6 and E7 stimulation. CD4+ T cells were predominantly composed of single-positive IFN-γ-producing cells, whereas CD8+ T cells, particularly in the E7-stimulated group, showed increased Granzyme B-producing populations together with higher frequencies of double-positive and triple-positive functional subsets (Supplementary Figure 3). Additionally, significantly higher levels of HPV-specific IgG were detected in mice vaccinated with E7 mRNA-LNP compared to NSS controls. In contrast, specific IgG was undetectable in mice immunized with E6 mRNA-LNP (Figure 2(D)). Immunization of HPV16 mRNA-LNP elicits antigen-specific CD8+ T cell responses. (A) Schematic representation of the immunization regimens. C57BL/6 mice (N = 5) were immunized with 2 doses of mRNA-LNP encoding E6 and E7 (1 μg each) in 14 d interval. (B) IFN-γ-producing T cells from 5 × 105 cells of C57BL/6 mouse splenocytes restimulated with pooled peptides of HPV16 E6 or E7 for 40 h detected by ELISpot. (C) HPV16 E6 and E7-specific responses were determined by measuring the percentages of CD8+ T cells producing IFN-γ, Granzyme B, and TNF-α after 6-h poststimulation with E6 or E7 pooled peptides, analyzed by flow cytometry. (D) Total HPV16 E6 and E7‑specific IgG from immunized mice sera (1:10 dilution) at 2 weeks after the second dose. (E) Schematic representation of the experiment in outbred mice. JCl/ICR mice (N = 5) were immunized with 2 doses of 1 μg mRNA-LNP encoding HPV16 E6 with 14 d interval. (F) IFN-γ-producing T cells from 5 × 105 cells of JCl/ICR mouse splenocytes were restimulated with pooled peptides of HPV16 E6 or E6 for 40 h, as detected by ELISpot. In this study, mice that received NSS were used as a negative control. As in other reports,14,23,24 we observed a poor response against the E6 oncoprotein in C57BL/6 mice. To determine whether this was mouse strain-specific, we conducted an extended experiment using Jcl:ICR outbred mice. These mice were immunized with two doses of 1 µg E6 mRNA-LNP or NSS (Figure 2(E)). The results showed that Jcl:ICR mice mounted a much stronger HPV16-specific IFN-γ T cell response compared to C57BL/6 mice (816 SFU/106 vs. 162 SFU/106, Figures 2(B,F)). These findings indicate that E6 is immunogenic, but the magnitude of the immune response is influenced by host MHC background. Early prime-boost vaccination with E7 mRNA-LNP, either alone or in combination with E6 mRNA-LNP, successfully eliminates established tumors in mice To evaluate the anti-tumor efficacy of E6 and E7 mRNA-LNP vaccines in antigen-positive tumor models, we first assessed the kinetics of tumor growth by implanting varying numbers of TC-1 cells. Tumor formation occurred in a dose-dependent manner. Specifically, implantation of >2 × 104 TC-1 cells led to rapid tumor growth, with bioluminescence signal intensities reaching 109-1010 photons/sec within 12 d, resulting in euthanasia per humane endpoint criteria.14 Implantation with 1 × 104 cells extended mouse survival to approximately 2–3 weeks. A tumor inoculum of 5 × 103 cells was identified as the minimum dose that consistently established tumors without spontaneous regression (Supplementary Figure 4). To assess early therapeutic efficacy, we implanted 1 × 104 TC-1 cells to mimic low tumor burden approach. Mice were immunized with 10 µg of E6 or E7 mRNA-LNP on day 3 post-implantation, followed by a booster dose of 5 µg on day 9 (Figure 3(A)). Mice vaccinated with E7 mRNA-LNP showed marked tumor growth inhibition and complete remission by day 21 post-implantation. E6 mRNA-LNP conferred partial tumor suppression, achieving 87% inhibition by day 21 (average tumor volumes: E6 group ~309 mm3 vs. NSS control group ~2371 mm3) (Figure 3(B)). Co-administration of 10 µg each of E6 mRNA-LNP and E7 mRNA-LNP on day 3, followed by a boost with 5 µg each on day 9, conferred no additional benefit over E7 mRNA-LNP alone. NSS-immunized mice reached humane endpoints by day 11. Tumor regression study with early prime–boost immunizations in TC-1 Luc tumor mouse models. (A) Schematic representation of the immunization regimen. C57BL/6 mice (N = 3/group) were implanted with 1 × 104 TC-1 cells (s.c.) and subsequently treated with 10 and 5 μg of E6, E7, or E6+E7 mRNA-LNP on days 3 and 9 post‑tumor implantation. NSS-treated mice served as the negative control. (B) Kinetics of tumor volume measured by caliper. Tumor volume was recorded twice or thrice a week. (C) The percentage of survival was followed until day 21 post‑tumor implantation. Note: One mouse in the E6 mRNA-LNP group developed a wound rupture on day 21 and consequently met the humane endpoint criteria. (D) Representative bioluminescence images indicating TC-1 cells are displayed. (E) Representative pictures of immunofluorescent-stained sections from tumor mass or tumor-implanted area determined by HPV16 E6 (green), HPV16 E7 (red), and nuclear DNA (blue) of individual mouse from each group at day 21 post‑tumor implantation (magnification 630X). Tissue from a randomly selected mouse was designated as a representative sample for each group in this investigation. (F) Agarose gel (1%) electrophoresis was performed for HPV16 E6/E7 positive (target size ~209 bp) and negative samples. Each lane illustrated of individual mouse (N = 3/group). *Mice that did not undergo tumor implantation (N = 2) were used as a negative control, and a **reagent mixture without template DNA served as the control reagent. Note: As the tumor regression study with early prime–boost immunization and the single-immunization study were conducted in parallel under identical experimental conditions, the bioluminescence imaging data, tumor size, and survival rate statistics of the control (NSS) group, which received tumor implantation at 1 × 104 TC-1 cells, were shared between the two analyses. Tumor volume correlated strongly with bioluminescence signal intensity (Supplementary Figure 5), confirming the reliability of imaging as a sensitive method for detecting tumor progression. Notably, bioluminescence imaging detected tumor presence earlier than caliper measurements, particularly in the NSS group by day 7 post-implantation. Therapeutic vaccination significantly improved survival outcomes. Mice in the E7 and E6 + E7 groups exhibited 100% survival, while the E6 group showed a lower survival benefit (Figure 3(C-D)). To assess tumor regression in immunized mice, HPV16 E6/E7 protein expression and HPV16 E6/E7 DNA were evaluated at day 21 post-implantation using biopsies from tumor masses or the implantation sites. HPV16 E6/E7 protein and DNA were undetectable in the E7 and E6+E7 mRNA-LNP groups, and no signal was observed in the secondary antibody staining in NSS group (Supplementary Figure 6). In contrast, HPV16 E6/E7 DNA remained detectable in the E6 mRNA-LNP and NSS group (Figure 3(E–F)). These findings suggest that the E7 mRNA-LNP vaccine, in particular, can eliminate neoplastic cells and potentially eradicate HPV16 E6/E7 DNA, supporting its therapeutic potential in HPV-infected individuals. A single immunization of E7 mRNA-LNP completely eradicates tumor formation in an early therapeutic model A single dose of 10 μg E7 mRNA-LNP was administered 3 d after implantation of either 5 × 103 or 1 × 104 TC-1 cells (Figure 4(A)). Mice receiving E7 mRNA-LNP showed complete inhibition of tumor growth compared to the NSS group. Tumor volume and bioluminescence signal were undetectable by day 21, resulting in 100% survival in both vaccinated groups (Figure 4(B–D) and Supplementary Figure 7). Furthermore, HPV16 protein and DNA were undetectable in tumor biopsy samples, indicating that a single dose of 10 μg E7 mRNA-LNP was sufficient to eliminate both viral antigens and infected cells without requiring a booster dose (Figure 4(E–F)). These findings demonstrate that early single-dose therapeutic vaccination with E7 mRNA-LNP is capable of eradicating low tumor burden in animal model. Tumor regression study with a single immunization of E7 mRNA-LNP. (A) Schematic representation of the immunization regimen. C57BL/6 mice (N = 3/group) were implanted with 5 × 103 or 1 × 104 or TC-1 cells (s.c.) and subsequently treated with 10 μg of E7 mRNA-LNP on day 3 post‑tumor implantation. NSS-treated mice served as the negative control for tumor implantation. (B) Kinetics of tumor volume measured by using caliper. Tumor volume was recorded twice or thrice a week. (C) The percentage of surviving mice was monitored until day 21 post‑tumor implantation. Note: One mouse in the NSS implanted with 5 × 103 TC-1 cells group developed a wound rupture on day 21 and consequently met the humane endpoint criteria. (D) Representative bioluminescence images indicating TC-1 cells are displayed. (E) Representative pictures of immunofluorescent-stained sections for tumor mass or skin at tumor-implanted area were obtained, as determined by HPV16 E6 (green), HPV16 E7 (red), and nuclear DNA (blue) of individual mouse from each group at day 21 post‑tumor implantation (original magnification 630X). Tissue from a randomly selected mouse was designated as a representative sample for each group in this investigation. (F) Agarose gel (1%) electrophoresis was performed for HPV16 E6/E7 DNA positive and negative samples (target size ~209 bp). Each lane illustrated an individual mouse (N = 3/group). *DNA extraction from TC-1 cells (harboring HPV16 E6/E7 DNA) and **C3 cells (harboring whole genome of HPV16) were used as a positive control. ***Mice that did not undergo tumor implantation (N = 2) were used as a negative control, and a #reagent mixture without DNA template served as the reagent control. Note: As the tumor regression study with early prime–boost immunization and the single-immunization study were conducted in parallel under identical experimental conditions, the bioluminescence imaging data, tumor size, and survival rate statistics of the control (NSS) group, which received tumor implantation at 1 × 104 TC-1 cells , were shared between the two analyses. Determining the minimum effective dose for complete tumor clearance in the early therapeutic model To determine the lowest effective dose for tumor eradication, mice implanted with 1 × 104 TC-1 cells were treated with reduced doses (2.5 μg or 5 μg) of E7 mRNA-LNP (Figure 5(A)). A dose-dependent inhibition of tumor growth and a delayed increase in bioluminescence signal were observed. While both doses extended survival and delayed tumor progression, complete tumor clearance was not achieved in any mouse at these lower doses (Figure 5(B–D) and Supplementary Figure 8). These results suggest that, under our experimental conditions, a minimum of 10 μg E7 mRNA-LNP is required to achieve full tumor eradication in the early therapeutic model. Tumor regression study with dose optimization of E7 mRNA-LNP in TC-1 Luc tumor mouse models. (A) Schematic representation of the immunization regimen. C57BL/6 mice (N = 3/group) were implanted with 1 × 104 TC-1 cells (s.c.) and subsequently treated with different doses of E7 mRNA-LNP (2.5 and 5 μg of E7 mRNA-LNP) on day 3 and day 10 post‑tumor implantation. NSS-treated mice served as the negative control. (B) Kinetics of tumor volume which measured by using caliper. Tumor volume was recorded twice or thrice. (C) The percentage of surviving mice was monitored until day 23 post‑tumor implantation. (D) Representative bioluminescence images indicating TC-1 cells are displayed. Preventive immunization protected mice from tumor development The preventive efficacy of E6, E7, and combined E6+E7 mRNA-LNP vaccines was evaluated in a tumor challenge model. Mice were immunized twice with 5 μg of each individual mRNA-LNP formulation before being challenged with 1 × 104 TC-1 cells (Figure 6(A)). All NSS-treated control mice developed tumors rapidly and reached humane endpoints by day 13. In contrast, mice immunized with E7 and E6+E7 mRNA-LNP were completely protected from tumor development, showing no detectable tumor masses or bioluminescence signals. These groups achieved a 100% survival rate throughout the 54-d observation period post-challenge. In comparison, mice receiving E6 mRNA-LNP alone showed only partial protection, with delayed tumor onset relative to NSS-treated controls (Figures 6(B–D) and Supplementary Figure 9(A)). Preventive efficacy of HPV16 mRNA vaccines mRNA in TC-1 Luc challenged mice model. (A) Schematic representation of the immunization regimen in C57BL/6 mice (N = 3/group). At 14 d after two immunizations with 5 μg of E6, E7, or E6 + E7 mRNA-LNP, mice were subsequently implanted with 1 × 104 TC-1 cells (s.c.) for the first TC-1 Luc challenge, and tumor growth was routinely monitored. Mice were rechallenged with 5 × 103 TC-1 cells at day 141 (s.c). NSS-treated mice with age-matched served as the negative control at day 141. (B) Representative bioluminescence images indicating TC-1 cells were displayed for the first challenge study. (C) Kinetics of tumor volume which measured by using caliper for the first challenge study. Tumor volume was recorded twice or thrice a week. (D) The percentage of surviving mice was monitored until day 140 post‑tumor implantation. Note: One mouse in the E6 mRNA-LNP group developed wound rupture on day 30 and consequently met the humane endpoint criteria. (E) Kinetics of tumor volume measured by using caliper for the second challenging. Tumor size volume recorded twice or thrice a week. (F) The percentage of surviving mice was monitored until day 182 post‑tumor implantation for the second challenge study. (G) Representative bioluminescence images indicating TC-1 cells are displayed for the second challenge study. (H) IFN-γ-producing T cells from 5 × 105 cells of mouse splenocytes at day 182 restimulated with specific pooled peptides of HPV16 E6 or HPV16 E7 for 40 h were detected by ELISpot. (I) Agarose gel (1%) electrophoresis was performed for HPV16 E6/E7 DNA positive and negative samples (target size ~209 bp). Each lane illustrated an individual mouse sample (N = 2-3/group). #Reagent mixture without DNA template served as the reagent control. *DNA extraction from TC-1 cells were used as a positive control. Note:One mouse in the E6+E7 mRNA-LNP (rechallenge study) group died on day 130 due to a cause unrelated to vaccination, prior to the rechallenge To assess the development of long-term protective immunity, tumor-free vaccinated mice were re-challenged with TC-1 cells at day 141 after the initial tumor challenge. Age-matched NSS-treated mice served as controls (Figure 6(A)). Complete protection against tumor reestablishment was observed in both E7 and E6+E7 mRNA-LNP groups, as evidenced by the absence of tumor formation and no detectable bioluminescence signals, resulting in 100% survival (Figures 6(E–G) and Supplementary Figure 9(B)). Further evaluation at day 182 revealed that splenocytes from E7 or E6+E7 immunized mice maintained high levels of HPV16 E7-specific IFN-γ-producing T cells (Figure 6(H)). Additionally, neither HPV16 DNA nor HPV proteins were detectable in these groups (Figure 6(I)), confirming complete tumor cells clearance. These findings indicate that preventive immunization induced durable immune memory, which enabled rapid and effective responses to tumor antigens, thereby preventing tumor establishment upon re-exposure. Discussion Several therapeutic HPV vaccines are currently in clinical trials, targeting high-grade squamous intraepithelial lesions (HSILs), including precancerous cervical intraepithelial neoplasia (CIN II/III), as well as advanced cancer stages. Despite extensive research and development, major challenges remain. Tumors often develop immune escape mechanisms that limit the efficacy of therapeutic vaccines targeting tumor-specific antigens, thereby reducing clinical success rates. To address these limitations, redirecting vaccine strategies toward persistent high-risk HPV infections or early-stage dysplasia could provide multiple advantages and represents a promising direction for vaccine-based cancer treatment. Potential benefits include the availability of standard-of-care therapies for comparison, a large at-risk population particularly in regions with limited HPV DNA screening and the feasibility of integrating therapeutic vaccines into existing immunization programs. This approach could reduce the dependency on adult HPV screening, enable combination with a single-dose preventive HPV vaccine, and potentially lower the risk of immune evasion by implantation of early intervention approach. In this study, we developed mRNA vaccines encoding HPV16 E6 and E7 and evaluated their efficacy in a mouse model designed to mimic low tumor burden. Due to the species specificity of papillomaviruses, infection-based animal models of HPV-induced carcinogenesis are limited.38,39 Previous attempts to establish in vivo infection models using Mouse papillomavirus type 1 (MmuPV1) have been successful only in immunodeficient mice (Hsd:Athymic Nude-Foxn1nu) or genetically immunocompetent FVB/N mice, which required rigorous UVB and 4NQO exposure to induce infection.40 To overcome these challenges, we used a TC-1 tumor cell-based model with low-dose inoculation and early vaccination in C57BL/6 mice, providing a practical and robust model to assess HPV therapeutic strategies prior to irreversible tumor progression. Notably, implantation of 1 × 104 TC-1 cells led to consistent tumor growth without spontaneous regression, and tumor volumes were comparable to models using ten-fold higher inoculum.23 Our mRNA vaccine design was the inclusion of a human IgE signal sequence to enhance translation efficiency and promote trafficking of E6 and E7 antigens to the endoplasmic reticulum. This strategy successfully yielded intracellular and secreted HPV16 E6/E7 protein expression. Among vaccinated mice, those receiving the E7 mRNA-LNP vaccine developed robust antigen-specific immune responses. Notably, early administration of E7 mRNA-LNP 3 d after tumor cell implantation either as a single dose or in a prime-boost regimen resulted in complete tumor regression and significantly prolonged survival. We further demonstrated that E7 mRNA-LNP vaccination achieved complete clearance of HPV16 E6/E7 proteins and DNA at tumor implantation sites. Importantly, vaccinated mice were fully protected upon tumor rechallenge approximately 5 months later, demonstrating the induction of long-term immunological memory. The E7 mRNA-LNP vaccine thus exhibited dual therapeutic and preventive potential. Interestingly, lower mRNA doses were sufficient for preventive protection compared to therapeutic clearance, underscoring the value of early intervention. In comparison to a previous preclinical study,24 our mRNA vaccine offers several advantages over existing HPV16 vaccines based on an mRNA platform. We observed that two doses of 1 µg E7 mRNA elicited stronger immune responses by stimulating IFN-γ-producing T cells from mouse splenocytes. In the therapeutic setting, a single 10 μg dose led to complete tumor clearance without requiring self-amplifying RNA or additional adjuvants or drug combinations.14,41 Notably, no previous HPV vaccine studies have reported complete clearance of HPV DNA following vaccination. Although in vivo imaging systems (IVIS) offer high sensitivity for tumor tracking, PCR remains the gold standard for HPV DNA detection due to its capacity for single-copy sensitivity.42–44 Our study confirmed complete tumor elimination at the genomic level, further validating the efficacy of the E7 mRNA-LNP vaccine. Despite its lower efficacy, E6 mRNA-LNP still conferred measurable therapeutic benefit. The diminished performance is likely due to limited E6 epitope recognition in C57BL/6 mice. This finding aligns with previous reports showing that E6-targeted vaccines often require co-administration with E7 for optimal efficacy in this strain.14,23,24 To explore the impact of host genetic background, we evaluated E6 mRNA immunogenicity in outbred Jcl:ICR mice. These mice mounted stronger E6-specific IFN-γ responses compared to C57BL/6, supporting the role of host MHC restriction. Even in the C57BL/6 model, E6 mRNA-LNP delayed tumor onset in both therapeutic and preventive contexts, suggesting that E6 remains a valuable vaccine target, warranting further investigation in more genetically diverse settings. This study has several limitations, including the relatively small number of animals per experimental group and the absence of independent repeat experiments, which may limit statistical power and robustness. Although consistent trends were observed, the findings require validation in larger, independently replicated studies. In addition, the use of the TC-1 tumor model, although widely employed for evaluating HPV-targeted immunotherapies, does not fully recapitulate the stepwise progression and immune microenvironment of human HPV-associated precancerous lesions. While the use of a lower tumor inoculum may partially approximate earlier disease stages, it does not reflect key features of tumor evolution, including stromal remodeling, angiogenesis, and immune suppression. In particular, characterization of the tumor microenvironment such as vascularization, stromal activation, and myeloid cell infiltration was not performed and would provide important mechanistic insight. Therefore, the TC-1 model should be interpreted as a proof-of-concept system for assessing vaccine-induced immune responses and antitumor activity, rather than a direct surrogate for early HPV lesion biology. Future studies using more physiologically relevant models will be important to further validate these findings. In addition to E6 and E7, other viral antigens such as HPV E2 may represent important targets for therapeutic vaccination in early-stage disease. E2 is expressed during persistent infection and in premalignant lesions but is often lost upon viral genome integration during malignant progression.45 Immunological targeting of E2 has been shown to elicit cellular immune responses and may contribute to viral clearance in early lesions. Therefore, inclusion of E2 in multivalent vaccine strategies could broaden immune targeting and improve efficacy across different stages of HPV-associated disease.46–48 This study establishes the therapeutic potential of mRNA-LNP vaccines targeting HPV16 E6 and E7, enhanced by the inclusion of an IgE signal sequence. Our vaccine demonstrated significant benefits in early intervention models, including tumor suppression, undetectable HPV16 E6/E7 DNA, and durable memory responses. E7 mRNA-LNP, in particular, offers distinct advantages in both therapeutic and preventive applications. The ability to completely eliminate HPV genomic material following early treatment highlights the clinical potential of this approach in eradicating high-risk HPV infections before malignant progression. Conclusions To support the early eradication of HPV-associated tumors and viral genomes, we established a reliable low tumor burden model for proof-of-concept evaluation. Using this model, we demonstrated that mRNA-LNP vaccines encoding HPV16 E6 and E7 are effective in both preventive and therapeutic contexts. E7 mRNA vaccination elicited robust T cell responses, significant tumor regression, and effective HPV16 E6/E7 DNA clearance, outperforming the E6 mRNA vaccine in both models. In a rechallenge study conducted 5 months post-vaccination, mice previously vaccinated with E7 mRNA demonstrated durable T cell memory, tumor protection, and sustained HPV-DNA clearance. These findings strongly support the further clinical development of HPV16 E7 mRNA-LNP vaccine. A multivalent version targeting at least five high-risk HPV genotypes-covering approximately 90% of global HPV-associated cancers is currently in development. Method details Ethics statement The researchers strictly followed the principles and guidelines set by the Institute of Animals for Scientific Purposes Development, National Research Council of Thailand. All studies were conducted in accordance with protocols approved by the Committees on Care of Laboratory Animals at the Faculty of Medicine, Chulalongkorn University (IACUC approval no. 024/2566). Antigen design To generate 2 constructs from HPV16 E6 and E7, each ten gene consensuses collected from different countries were retrieved from GenBank randomly then performing multiple alignments (GenBank number: KP874716.1, FJ610152.1, AY686584.1, MN961217.1, AB889494.1, KU684317.1, AF472508.1, AF472509.1, and NC_001526.4). The de-oncogenic forms of the E6 and E7 proteins were designed by introducing several mutations to disrupt their interactions with p53 and pRb, thereby ensuring safety. Mutations at C70G and C113G were introduced into E6, while mutations at C24G, E26G, and C91G were introduced into E7, respectively.49,50 AfeI and SpeI restriction sites were added before the first codon of IgE leader sequences (MDWTWILFLVAAATRVHS)24,51 and after the last codon of E protein, respectively. The humanized‑codon of the designed gene fragment was synthesized by GenScript (Piscataway, NJ, USA) and cloned in pUC57 plasmid. Plasmid cloning, transformation, and in vitro transcription (IVT) Plasmid pUC57 carrying the antigen-encoding HPV16 E6 and E7 was subcloned into pUC-ccTEV vector for in vitro transcription (IVT). The plasmids were propagated in E. coli (Stbl3TM, Invitrogen, Carlsbad, CA, USA) and extracted by EndoFree® Giga Kit (Qiagen, Hilden, Germany). The plasmids were linearized with restriction enzymes AflII and NotI (New England BioLabs, Ipswich, MA). Nucleoside-modified mRNA was synthesized through IVT by replacing uridine triphosphate (UTP) with N1-methylpseudouridine (m1Ψ) triphosphate (TriLink Biotechnologies, San Diego, CA, USA), as previously described.34 The transcription reaction was carried out by T7 RNA polymerase (Invitrogen, Carlsbad, CA, USA) on a linearized plasmid. The transcribed mRNA included a 101-nucleotide poly(A) tail. The mRNA capping was achieved using the trinucleotide cap1 analog, CleanCap (TriLink Biotechnologies, San Diego, CA, USA). The capped mRNA was purified using cellulose column to remove dsRNA by centrifuge tube filters (Corning® Costar® Spin-X®, Merck Millipore, Darmstadt, Germany) and cellulose fibers (C6288, Sigma-Aldrich, USA).52 IVT mRNA was analyzed on an agarose gel to assess its integrity. Additional quality control measures were performed to check for the presence of double-stranded RNA (dsRNA) by dot blot analysis and endotoxin contamination prior to encapsulation into lipid nanoparticles (LNPs). The mRNAs were stored frozen at −80°C until use. mRNA encapsulation mRNA was encapsulated by lipid nanoparticles (LNP) proprietary to Genevant Sciences Corporation: Patent Application WO2020097540A1 (Vancouver, British Columbia, Canada). The LNPs were composed of 3D-P-DMA, PEG2000-C-DMA, cholesterol, and DSPC and were formulated by mixing nucleic acid and lipid solutions at a ratio of 3:1 (aqueous:lipid) using a NanoAssemblr® Ignite™ (Precision NanoSystems), in accordance with the manufacturer’s instructions. The resulting mRNA-LNPs were subsequently preserved in a sucrose-based cryoprotectant solution. The final products were measured the diameter which have average hydrodynamic diameter of 60–80 nm with a polydispersity index (PDI) <0.2 as measured by using a Zetasizer Advance (Malvern Panalytical, UK) and an encapsulation efficiency ~95% by using Quant-it™ RiboGreen RNA Assay Kit (Invitrogen, Carlsbad, CA, USA). Cells VERO cells, the kidney epithelial cell line from African green monkey (ATCC, Old Town Manassas, VA, USA), were cultured in Dulbecco’s Modified Eagle’s Medium (DMEM) supplemented with 10% (V/V) heat-inactivated fetal bovine serum (HIFBS), 1X L-glutamine, 1X penicillin–streptomycin at 37°C with 5% CO2. TC-1 Luc cells, kindly provided by T.-C. Wu (Johns Hopkins University, USA), were used for the establishment of a tumor mouse model. Originally, TC-1 cells were derived from a tumor induced by lentiviral transduction of E6, E7, and H-ras of C57BL/6 lung keratinocytes, then passaged in vivo and later incorporated with a firefly luciferase gene for tracing purpose.53 TC-1 cells were cultured in RPMI 1640 medium supplemented with 10% heat-inactivated fetal bovine serum (HIFBS), 1X L-glutamine, 1X penicillin–streptomycin at 37°C with 5% CO2. The C3 tumor cell line was kindly provided by S.H. van der Burg (Leiden University Medical Center, The Netherlands). This cell was made by transfecting B6 mouse embryonic cells with a plasmid harboring the full-length HPV16 genome (E+L).54 C3 cells were cultured in Iscove’s modified Dulbecco’s medium (IMDM) supplemented with 10% HIFBS, 1X penicillin–streptomycin, 100 ug/ml Kanamycin Sulfate and 2 × 10−5 Molar 2-Mercaptoethanol at 37°C with 5% CO2. All reagents were from Gibco™, Carlsbad, CA, USA. The fifth to tenth passages were used for tumor experiments. mRNA transfection and in vitro protein expression analysis by immunofluorescence staining For cellular protein expression analysis, a total 5 × 106 of VERO cells (Negative HPV genome) were seeded into six-well plates (Thermo Fisher Scientific, MA, USA) for 24 h. After that, 2.5 μg of IVT mRNA of HPV16 E6 or E7 in the presence of Lipofectamine™ MessengerMAX™ Transfection Reagent (Invitrogen, Carlsbad, CA, USA) was transfected to the cells according to the manufacturer’s instructions. Transfected VERO cells were fixed and permeabilized, then stained with mouse anti-HPV16 E7 (Santa Cruz, STCSC-6981, Texas, USA) or mouse anti-HPV16 E6 + HPV18 E6 Abs (Abcam, ab70, Cambridge, UK), followed by rat anti-mouse IgG FITC as a secondary antibody (Invitrogen, Carlsbad, CA, USA). Nuclear DNA was stained by using 4, 6-diamino-2-phenylindolehydrochloride (DAPI) (A4099, Applichem, Germany). The images were analyzed with ZEISS LSM 800 (Carl Zeiss, Germany) at 630X magnificent and measured mean fluorescence intensity (MFI) by ZEN imaging software. ELISpot assay Mouse spleens were collected, and splenocytes were isolated. Splenocytes (5 × 105 cells) were cultured for 40 h by using ELISpot Plus: Mouse IFN-γ (ALP) (MABTECH) pre-coated IFN-γ 96-well plates in the presence 2 μg/ml of a pooled of wild-type HPV16 E6 or HPV16 E7 peptide (Pepmix, JPT Peptide Technologies, Berlin). ELISpots were performed according to the manufacturer’s instructions by ELISpot plate reader (ImmunoSpot, USA). Results were reported as spot-forming cells (SFCs) per 106 splenocytes. Intracellular cytokine staining (ICS) by flow cytometry Flow cytometry staining was conducted on peptide-stimulated splenocytes. Monoclonal antibodies for surface staining included CD3, CD4, and CD8. For intracellular staining, antibodies against IFN-γ, Granzyme B, and TNF-α were used (all were from BioLegend, CA, USA). After stimulation of 106 splenocytes for 1 h at 37°C and 5% CO2 in the presence of 2 μg/ml with pooled of HPV16 E6 or HPV16 E7 peptides (Pepmix, JTP peptide technologies, Berlin), Protein Transport Inhibitor Cocktail (Invitrogen, Carlsbad, CA, USA) was added and incubated for 5 h. Live cells were stained with viability dyes (Invitrogen, Carlsbad, CA, USA) according to manufacturer’s instructions. Cells were washed once then blocked with 1 μg of TruStain FcX™ PLUS (anti-mouse CD16/32) (BioLegend, CA, USA) for 15 min at 4°C. Surface staining antibodies were mixed in cell staining buffer (BioLegend, CA, USA) then incubated for 30 min at 4°C. For the staining of IFN-γ, TNF-α, and Granzyme B, samples were fixed and permeabilized with Cyto-Fast Fix/Perm (BioLegend, CA, USA) following manufacturer’s instructions. Intracellular targets were stained for 30 min at 4°C in the dark. Data were acquired on an Attune N × T Flow Cytometer (Thermo Fisher Scientific, MA, USA) and analyzed with FlowJo 10 software (Tree Star, USA). Tumor mouse model and treatment In therapeutic study, female C57BL/6 mice at 4–5 weeks old, 3 mice/group were engrafted subcutaneously (s.c.) with of TC-1 cells in 100 μl of FBS-free RMPI 1640 medium. After implantation, tumor size was daily measured to ensure that the tumor was successfully engrafted by bioluminescence imaging (described below) in parallel with Vernier Calipers measurement method (tumor volume was calculated with the formula: Lenght2 × Width)/2 mm3. After tumor implantation, mice were immunized with mRNA-LNP vaccine following the schematic regimen of animal plan. Tumor growth was monitored 2–3 times a week. In preventive study, female C57BL/6 mice at 4–5 weeks 3 mice/group were immunized with two doses of 5 μg mRNA-LNP with 14 d interval. Two weeks after the second immunization, mice were engrafted subcutaneously for the first challenging with TC-1 cells of 1 × 104 cells and rechallenge with 5 × 103 cells. According to the predefined survival endpoint criteria, mice reaching a total bioluminescence flux of 109 p/s were considered to have reached the experimental endpoint, consistent with previous bioluminescence-based tumor monitoring studies.14 Humane endpoints were applied when mice exhibited >10% body weight loss, tumor volume >2,000 mm3, or tumor ulceration/rupture in cases where the bioluminescence cutoff had not been reached, in accordance with institutional animal care guidelines. Bioluminescence imaging TC-1 cells were evaluated by in vivo bioluminescence imaging using the IVIS Lumina III Spectrum imaging system (Revvity, MA, USA). An aqueous solution of D-Luciferin Potassium Salt (Revvity, MA, USA) at 150 mg/kg in Ca2+ Mg2+-free Dulbecco’s Phosphate-Buffered Saline (DPBS) was injected intraperitoneally (i.p.). After mice were anesthetized in a ventilated anesthesia chamber with 2% isoflurane in oxygen, emitted photons from live animals were quantified 20 min after luciferin administration. Regions of interest (ROI) were quantified to obtain the total flux values expressed in photons per second (p/s) using IVIS Living Image 4.0 software. HPV16 E6/E7 DNA detection from tissue by PCR DNA was extracted from 10 μg tissue sections from tumor mass or area of tumor implantation by using a kit (DNeasy® Blood & Tissue, Qiagen) according to the manufacturer’s instructions. HPV16 E6/E7 DNA was amplified by using GoTag Flexi DNA polymerase (Promega, Wisconsin, USA) with forward 5’-GTGGACCGGTCGATGTATGTCT-3’ and reverse primers 5’-TCCGGTTCTGCTTGTCCAGC-3.’55 PCR products were determined by running at 1% gel electrophoresis. HPV16 E6/E7 proteins detection from tissue by immunofluorescent analysis Tissue from tumor mass or skin at implantation site was prepared in Tissue-Tek O.C.T Compound (Sakura Finetek, CA, USA), and the sections were subsequently stained. HPV oncoproteins were detected with primary antibodies: mouse anti-HPV16 E7 (Santa Cruz, STCSC-6981, Texas, USA) and rabbit anti-HPV16 E6 (Bioss, MA, USA). The secondary antibodies were goat anti-rabbit IgG with Alexa Fluor® 488 (Abcam, Cambridge, UK) and goat anti-mouse IgG labeled with Alexa Flour 647 (BioLegend, CA, USA). Nuclear DNA was counterstained by using DAPI (Invitrogen, Carlsbad, CA, USA). The slides were mounted with ProLong™ Gold Antifade mountant (Invitrogen, Carlsbad, CA, USA). The images were analyzed with ZEISS LSM 800 (Carl Zeiss, Germany) at 630X magnification. Total IgG by ELISA HPV16 E6 or E7 IgG measurement was performed employing indirect ELISA. In brief, 50 ng of recombinant HPV16 E6 (Abcam, ab226447, Cambridge, UK) or HPV16 E7 (Abcam, ab237790, Cambridge, UK) were coated to 96 half well plates. The 10-fold serially diluted mice sera were added to well. After incubating for 1 h at 37°C, the plates were thoroughly washed with washing buffer (PBS +0.5% Tween 20, PBST). Then, goat anti-mouse IgG (H+L) labeled with horseradish peroxidase (SeraCare, Milford, USA) was added for an additional 1 h. After washing, the signals were detected by adding tetramethylbenzidine (TMB) substrate (BioLegend, CA, USA). The reactions were then stopped with 25 µL of 0.16 N sulfuric acid. The absorbance was measured at a wavelength of 450 nm using a Varioskan microplate reader (ThermoFisher Scientific, Vantaa, Finland). Statistical analysis Statistical analyses were performed using GraphPad Prism software. Comparisons between two independent groups were conducted using the nonparametric Mann–Whitney test. Survival curves were analyzed using the Kaplan–Meier method. Statistical significance was reported as follows: *p < .05; **p < .01; ***p < .001; and ****p < .0001. Supplementary Material Supplementary figure_.docx Acknowledgments The authors acknowledge all the members of the Chula VRC for their input and support. TC-1 tumor cells kindly provided by Tzyy-Choou Wu (Johns Hopkins University, United States). C3 tumor cells kindly provided by S.H. van der Burg (Leiden University Medical Center, Netherlands). We also thank Dr.Mohamad-Gabriel Alameh (Department of Pathology and Laboratory Medicine, Children’s Hospital of Pennsylvania) for his inputs in the initial step of this project. S.S., E.P., C.KL., K.R., J.H., K.L., and D.W.: study conception and design, S.S., E.P., K.T., and P.K.: data collection, S.S., C.KL., E.P., and K.R.: analysis and interpretation of results, S.S., C.KS., K.T., P.K., N.Y., P.W., and K.C.: reagent preparation and analysis, S.S., C.KS., and E.P.: draft manuscript preparation. S.S., E.P., C.KL., and K.R.: grant funding acquisition. All authors reviewed the results and approved the final version of the manuscript. Biography Eakachai Prompetchara, Ph.D., is a Principal Investigator at the Chula Vaccine Research Center, Faculty of Medicine, Chulalongkorn University, Bangkok, Thailand. He earned his Ph.D. in Biomedical Sciences and subsequently undertook postdoctoral training in immunology and vaccine research. His primary research focus is the development of nucleic acid-based vaccines, particularly mRNA vaccine platforms, including antigen design, delivery systems, and immunogenicity optimization. Dr. Prompetchara has been actively involved in national mRNA vaccine programs targeting emerging infectious diseases, including SARS-CoV-2 and zoonotic viruses, with work spanning antigen engineering, preclinical evaluation, and translational development. He has served as a principal or co-investigator on multiple competitively funded projects supported by the National Vaccine Institute and international funding agencies. His work aims to advance locally sustainable mRNA vaccine manufacturing and rapid-response vaccine strategies for outbreak preparedness. Funding Statement This research was funded by the National Vaccine Institute (NVI), Thailand, under Grant Nos. 2568.1/36 and 2569.1/23. This work was also partially supported by the UK–Southeast Asia Vaccine Manufacturing Research Hub. The Hub is funded by the UK Department of Health and Social Care through UK International Development funding and is managed by the Engineering and Physical Sciences Research Council (EPSRC). The views expressed in this publication are those of the authors and do not necessarily reflect those of the Department of Health and Social Care. S.S. was additionally supported by a postdoctoral fellowship from the Second Century Fund (C2F), Chulalongkorn University. E.P. was supported by the Thailand Research Fund under Grant No. MRG6180163 and by the Faculty of Medicine, Chulalongkorn University, under Grant No. RA-MF-09/68. Disclosure statement DW is named on patents of mRNA and the use of modified mRNA in lipid nanoparticles. KL and JH are employees of Genevant Sciences Corporation and are named on patent describing lipid nanoparticles. The remaining authors declare no competing interests. Supplementary material Supplemental data for this article can be accessed online at https://doi.org/10.1080/21645515.2026.2686510 References Associated Data Supplementary Materials Supplementary figure_.docx