Direct vaccination with mRNA encoding tumor antigens is a novel and promising approach in cancer immunotherapy. CureVac's mRNA vaccines contain free and protamine-complexed mRNA. Such two-component mRNA vaccines support both antigen expression and immune stimulation. These self-adjuvanting RNA vaccines, administered intradermally without any additional adjuvant, induce a comprehensive balanced immune response, comprising antigen specific CD4+ T cells, CD8+ T cells and B cells. The balanced immune response results in a strong anti-tumor effect and complete protection against antigen positive tumor cells. This tumor inhibition elicited by mRNA vaccines is a result of the concerted action of different players. After just two intradermal vaccinations, we observe multiple changes at the tumor site, including the up-regulation of many genes connected to T and natural killer cell activation, as well as genes responsible for improved infiltration of immune cells into the tumor via chemotaxis. The two-component mRNA vaccines induce a very fast and boostable immune response. Therefore, the vaccination schedules can be adjusted to suit the clinical situation. Moreover, by combining the mRNA vaccines with therapies in clinical use (chemotherapy or anti-CTLA-4 antibody therapy), an even more effective anti-tumor response can be elicited. The first clinical data obtained from two separate Phase I/IIa trials conducted in PCA (prostate cancer) and NSCLC (non-small cell lung carcinoma) patients have shown that the two-component mRNA vaccines are safe, well tolerated and highly immunogenic in humans. Copyright © 2012 John Wiley & Sons, Ltd.
Recent clinical data provide encouraging evidence that immunotherapy can successfully contribute to the treatment of cancerous diseases [1-4]. The field of immunotherapy comprises a broad number of different approaches, such as vaccination with recombinant protein  or with major histocompatibility complex (MHC) class I-derived peptides [4, 5], vaccination with nucleic acid-based vaccines [6, 7], vaccination with autologous modified dendritic cells , adoptive T cell transfer  and treatment with monoclonal antibodies such as anti-CTLA-4 [2, 9], anti-PD1 , CD40  and OX40 . Despite the varied strategies, the goal of all these approaches is the same: to stimulate the immune system and to mobilize it to use its cellular and molecular tools in the fight against cancer. There are different ways to achieve this and the success of the immunotherapy will depend on how efficient it is at stimulating the immune system to fight on the right side. Based on the existing comprehensive knowledge of how the ‘healthy’ immune system works, we can define a set of different requirements that should be met by the ideal cancer vaccine. To induce a broad, potent and long lasting immune response, the ideal cancer vaccine should result in a high level of stable antigen expression, should activate the adaptive and innate immune system, and should induce T helper (TH)1 and TH2 responses. The induction of B cell responses, multifunctional and lytically active T cells, as well as memory cells, will be required to provide a sustained anti-tumor effect. The immune response should be fast and boostable by repeated vaccination and the vaccine should avoid the induction of negative regulators. Finally, because the chance of successfully treating cancer with a monotherapy is still very low, the vaccine should be able to be combined with other therapies (such as chemotherapy, radiation, monoclonal antibodies), providing an enhanced therapeutic effect.
We have recently published a novel and more effective mRNA vaccine design: the two-component mRNA-based cancer vaccines with self-adjuvanting activity . The two-component vaccines contain free and protamine-complexed mRNA and supported both antigen expression and immune stimulation, mediated by Toll like receptor 7 (TLR7). The two-component vaccines activated the adaptive and innate immune system to induce balanced humoral, as well as T cell mediated immunity. This balanced immune response was based on the induction of antigen specific CD4+ T helper cells and cytotoxic CD8+ T cells. Immunization with the two-component vaccines induced sustained memory responses, mediated by antigen specific memory T cells. Treatment of mice with the two-component mRNA vaccine mediated a strong anti-tumor response against ovalbumin (OVA)-expressing tumor cells not only in a prophylactic, but also in a therapeutic setting. These and recently obtained clinical data provide evidence that the two-component mRNA vaccine is a highly interesting candidate for a powerful cancer vaccine and fulfills many of the requirements for an ‘ideal’ cancer vaccine.
In the present study, we demonstrate further characteristics of the two-component mRNA vaccines that make them very potent cancer vaccines and show how mRNA vaccines can be combined with other therapies to further improve their therapeutic effect.
Materials and methods
Natural killer (NK)1.1-depleting antibody (clone PK136), CTLA-4-blocking antibody (clone 9H10) and isotype control antibodies [mouse immunoglobulin (Ig)G2a,k and polyclonal hamster IgG, respectively] were purchased from BioXCell (West Lebanon, NH, USA). Fluorescein isothiocyanate-labeled anti-mouse CD3e antibody (clone 145–2C11) and PerCP-Cy5.5-labeled anti-NK1.1 antibody (clone PK136) were from eBioscience (Frankfurt/Main, Germany). SIINFEKL and control peptides were obtained from Bachem AG (Weil am Rhein, Germany). Docetaxel (Taxotere) was from Sanofi-Aventis, and cisplatin was from Pfizer Inc (Frankfurt, Germany).
CureVac GmbH proprietary technology generates mRNA molecules with increased stability and translatability (claimed in the patents: EP 1392341 and EP 1604688). All mRNA vaccines used in the present study were produced in accordance with this technology.
Two-component mRNA vaccine
RNActive-based mRNA was protamine-formulated as previously described . Briefly, RNA complexation consist of a mixture of 50% free RNActive-based mRNA (component 1) and 50% mRNA complexed with protamine at a weight ratio of 2:1 (component 2). First, mRNA was complexed by the addition of protamine-Ringer lactate solution and, after stable complexation, free mRNA was added.
Mice and cell lines
C57BL/6 mice were obtained from Janvier Laboratories (Le Genest-Saint-Isle, France) and were 7–9 weeks of age at the beginning of the experiment. E.G7-OVA, a mouse T cell lymphoma cell line stably expressing Gallus gallus OVA (GgOVA) and parental EL4 cells were purchased from LGC Promochem GmbH (Wesel, Germany).
Luciferase expression in vitro
6 × 105 HeLa cells in 200 µl opti-MEM (Invitrogen, Carlsbad, CA, USA) were electroporated with 6 µg of different mRNAs encoding Photinus pyralis luciferase (PpLuc) in triplicate. Cells from a single electroporation were split into three wells on 24-well plates in 1 ml of RPMI-1640 medium to determine the luciferase level at three different time points. Some 6, 24, or 48 h after transfection, medium was aspirated and cells were lysed in 200 µl of lysis buffer [25 m m Tris, pH 7.5 (HCl), 2 m m ethylenediaminetetraacetic acid, 10% glycerol, 1% Triton X-100, 2 m m dithiothreitol, 1 m m phenylmethanesulfonylfluoride]. PpLuc activity was measured as relative light units in a BioTek SynergyHT plate reader at 5-s measuring time using 50 µl of lysate and 200 µl of luciferin buffer [25 m m glycylglycin, pH 7.8 (NaOH), 15 m m MgSO4, 2 m m ATP, 75 µ m luciferin].
Detection of OVA specific antibodies
Blood samples were taken retro-orbitally and analyzed for the expression of GgOVA-specific antibodies using an enzyme-linked immunosorbent assay (ELISA). The antibody titer was represented as a reciprocal dilution limit, defined as specific signal above the limit of detection (mean of optical density in serum of control mice, plus the three-fold SD of this group).
Enzyme-linked immunosorbent spot (ELISpot)
Splenocytes were isolated 6 days after the last vaccination. Splenocytes from OVA-vaccinated and control mice were stimulated with 1 µg/ml of either relevant (SIINFEKL) or control MHC class I restricted peptides. Secreted interferon (IFN)γ was detected using a standard ELISpot protocol and measured using a plate reader (Immunospot Analyser; CTL Analysers LLC, Shaker Heights, OH, USA).
Quantitative real-time polymerase chain reaction (qRT-PCR)
Total RNA was extracted by homogenization with RNeasy isolation kit (Qiagen, Hilden, Germany) in accordance with the manufacturer's instructions. Expression levels of GgOVA were quantified via qRT-PCR by MFTServices (Tübingen, Germany). All reactions were performed in triplicate.
Ten days after tumor challenge, mice were euthanized and complete tumors were excised. Punch biopsies (6 mm) from the center and the periphery of the tumor were removed and stored in RNAlater RNA Stabilization Reagent (Qiagen). Total RNA was extracted by homogenization with an RNeasy isolation kit (Qiagen) in accordance with the manufacturer's instructions. Some 10 µg of purified RNA from the center and the periphery of the tumor were combined and used for microarray analysis.
RNA integrity and quantity was evaluated on Bioanalyser-2100 (Agilent Technologies Inc., Santa Clara, CA, USA). Gene expression analyses were performed by MFTServices via whole-genome RNA microarray (Affymetrix UK Ltd, High Wycombe, UK) and analyzed with Ingenuity IPA Software (Ingenuity Systems, Redwood City, CA, USA).
C57BL/6 mice were injected intraperitoneally either with 200 µg of NK1.1-specific antibody (clone PK136) to deplete NK cells or 200 µg of mouse IgG2a,κ as a control. NK-cell depletion was monitored by cell surface staining of CD3 and NK1.1 on splenocytes and whole blood cells with flow cytometry. The efficiency of NK-cell depletion was always between 90% and 96%.
Statistical analysis was performed using GraphPad Prism, version 5.01 (GraphPad Software Inc., San Diego, CA, USA). The statistical tests used to analyse the differences between the test groups with a significance level of 5% were: nonparametric Mann–Whitney U-test (antibody ELISA, IFNγ ELISpot) and two-way analysis of variance with Bonferroni post-test for grouped data (tumor growth kinetics).
Results and Discussion
mRNA two-component vaccines lead to high antigen expression, are self-adjuvanting and induce a balanced immune response. The core of the two-component vaccines consists of a sequence-modified messenger RNA, encoding the desired tumor antigen. Starting with mRNA generation 1, we modified the molecule stepwise in its coding and noncoding regions improving both the level and duration of protein expression as detailed in several patents (EP 1392341 and EP 1604688). Protein expression from four mRNA generations is compared in Figure 1A using mRNAs coding for P. pyralis luciferase. First-generation mRNA includes an in vitro transcribed m7G Cap, an A70 poly(A) tail, and Promega's ‘luc+’ coding sequence (Promega, Madison, WI, USA). Both the level and the duration of PpLuc expression are dramatically improved in CureVac's fourth-generation mRNA.
As shown previously, the two-component vaccines activate the innate immune system via TLR-7 . This self-adjuvanting property makes the addition of any foreign adjuvant obsolete. To show that self-adjuvanticity is a unique characteristic of mRNA vaccines, mice were vaccinated either with the two-component OVA encoding mRNA vaccine or with recombinant OVA or with OVA-derived MHC class I epitope SIINFEKL. No adjuvant was added. The analysis of IFNγ secretion in splenocytes from vaccinated mice revealed that only the mRNA vaccine was able to induce cytotoxic T cell responses. In the absence of the adjuvant, both recombinant protein- or peptide-based vaccines completely failed to induce T cell responses (Figure 1B). Interestingly, a humoral response was induced by both the mRNA and protein vaccines; however, the levels of IgG2a antibodies (indicator of a shift towards a TH1 response) were significantly higher for the mRNA approach (Figure 1C). The imbalanced immune response, completely lacking cytotoxic T cells, resulted in the failure of protein and peptide-based vaccines to mediate tumor protection (Figure 1D). By contrast, the two-component vaccines, in the absence of foreign adjuvant, were sufficient to induce a balanced immune response and mediate complete protection against OVA expressing tumor cells. Tumors that escaped the control of the immune system after vaccination with the two-component mRNA vaccine were ovalbumin negative as tested by RT-PCR (Figure 1E).
mRNA two-component vaccines induce very fast and boostable immune responses. For any therapeutic cancer vaccine, it is highly desirable to induce a fast and strong immune response. mRNA vaccines are able to induce high T cell responses within 2 weeks, after just two vaccinations (Figure 2A). Interestingly, the number of antigen specific T cells could be doubled by a more frequent vaccination schedule comprising four vaccinations administered within 10 days (Figure 2A). This demonstrates that T cell responses, induced by mRNA two-component vaccines, can be boosted very efficiently within a very short time period. This is an important property that has not been seen with a lot of other vaccines. With these vaccines, after priming, the immune system remains insensitive to further vaccinations and only after the T cell contraction phase immune responses can be boosted . On the other hand, the active tumor immunotherapy for these vaccines spans several months when multiple vaccinations are administered. Therefore, we checked whether immune responses induced by mRNA vaccines can be also boosted when the time interval between vaccinations is longer. Indeed, having two versus four vaccinations administered with a time interval of 2 weeks between single vaccinations, the immune responses could be boosted very efficiently. We observed a doubling of antigen specific T cells and a dramatic increase of humoral responses, yielding higher IgG1 (20-fold) and IgG2a (100-fold) antibody titers. Overall, these results demonstrate that optimization of the vaccination schedule for the two-component mRNA vaccines is possible and, at the same time point, critical for maximizing immune responses.
The two-component mRNA vaccines induce the expression of a broad panel of genes at the tumor site. These genes are connected with chemotaxis of immune cells to the tumor site, as well as T cell and NK cell activation. When testing the therapeutic effect of the mRNA vaccines in aggressively growing tumor models, it is necessary to induce a strong immune response as fast as possible. Satisfyingly, frequent vaccination with the two-component mRNA vaccine (two vaccinations per week) resulted in strong anti-tumor immunity in vaccinated mice (Figure 3A). This raised the question of which processes take place at the tumor site. Given the fact that tumor inhibition is visible macroscopically between the second and third vaccination, we checked whether any changes in mRNA expression levels can be detected after just two intradermal vaccinations. Tumor tissue was isolated from mice vaccinated twice with the OVA mRNA vaccine or treated with buffer to extract RNA from all samples and perform gene expression analysis. Clear differences in expression pattern were seen between vaccinated and control mice (data not shown). Compared to buffer-treated mice, intradermal vaccination with OVA vaccine induced a substantial change in the expression of 67 different genes in the tumor (Figure 3B). Remarkably, the most heavily regulated genes were either NK cell-related genes or genes important for inducing a T cell response (Figure 3C), further confirming that two-component mRNA vaccines trigger the activation of innate and adaptive anti-tumor immune responses. The fact that several up-regulated genes were identified for specific pathways provides strong evidence that real functional changes are induced at the tumor site after vaccination. Intriguingly, similar immune gene profiles in human tumors, analyzed in recent years, were found to correspond to prolonged disease-free survival and improved outcome in cancer patients (Table 1) [15-18]. Interestingly, our results correlate very well with the literature, supporting, amongst others, an important role of chemokines CXCL9 and CXCL10 as prognostic markers. The fact that, after vaccination with the two-component mRNA vaccine, many genes associated with so-called ‘good’ inflammation were up-regulated suggests that the induced changes may directly correlate with a better prognosis. In addition, we found that mRNA transcripts of several members of IFNγ-inducible GTPases (guanylate-binding protein and p47-family) are also up-regulated after vaccination (data not shown). The functions of these proteins in cancer are poorly understood but it has been described that they can inhibit mammary tumor growth in mice  and can be correlated to improved survival in patients with colorectal cancer . Taken together, two-component mRNA vaccination increases the intra-tumoral expression of a panel of genes described to be important for a beneficial anti-tumor immune response.
|NK cell-related genes||Increased numbers of tumor-infiltrating NK cells correlates with a positive prognosis for patients with different tumor types (e.g. colorectal cancer, pulmonary adenocarcinoma, gastric carcinoma, squamous cell lung carcinoma)||[53-58]|
|T cell-related genes||Infiltration of T lymphocytes in different tumor types is correlated with a favourable prognosis for cancer patients||[59-66]|
|CXCL9 (MIG)||Presence of CXCL9 and CXCL10 correlates with increased numbers of tumor-infiltrating T cells and prolonged disease-free survival in colorectal cancer|||
|CCL5 (RANTES)||Expression of immune-related genes (e.g. CCL5) in tumor microenvironment correlates with prolonged disease-free survival in CRC patients|||
|IFNγ||Co-modulated expression of genes of the TH1 adaptive immunity (such as IFNγ) has a beneficial effect on clinical outcome in colorectal cancer||[53, 59, 67]|
|IFNγ-inducible GTPases||Increased expression of GBP-1 in tumor lesions is correlated with improved survival in colorectal cancer patients|||
The two-component mRNA vaccines activate NK cells which contribute to the anti-tumor effect during therapeutic vaccination. The gene expression analysis of tumor tissue suggested that, after vaccination, NK cells may be recruited to the tumor site. This result prompted us to clarify the role of these cells in mRNA vaccine induced anti-tumor activity. After depletion of NK1.1-positive cells (Figure 4A), the anti-tumor effect of the OVA vaccine was partially reduced, indicating a functional contribution of NK cells to the anti-tumor effect induced by the mRNA two-component vaccine (Figure 4B). The effect of NK cells on tumor control can involve different mechanisms. It is well known that NK cells perform a multifunctional role in anti-cancer immunity. On the one hand, NK cells are able to kill tumor cells directly via several well described mechanisms, such as the granule exocytosis pathway, the death-receptor pathway and through antibody-dependent cellular cytotoxicity [21, 22]. On the other hand, NK cells can also influence the shape of immune responses indirectly. It is well known that a subpopulation of NK cells, CD56hiCD16–CCR7+ in humans [23, 24] and NK1.1+CD27hiCD11bhi in mice , produce large amounts of chemokines and cytokines such as IFNγ, tumor necrosis factor-α and interleukin (IL)-15, thus modifying the activation of dendritic cells (DC) and T cells [26-28]. Moreover, it has been shown that NK cells can interact with DCs directly via a NK cell receptor and thereby eliminate immature DCs . Whether the NK-cell effect, during mRNA two-component vaccination, is mediated directly or indirectly cannot be answered yet. Because E.G7-OVA is a tumor cell line, expressing high levels of cell surface MHC class I molecules (the natural ligand of several inhibitory NK cell receptors such as Ly49C) [29-31], an indirect NK cell-mediated effect is likely. Nevertheless, we have previously shown that the mRNA two-component vaccine induces the production of IL-12  in vivo, a cytokine important for direct NK cell activation . Although the exact molecular mechanism of NK cell contribution can be not explained in detail, we could show that, after local intradermal injection, the mRNA two-component vaccines have the capacity to activate both the innate and the adaptive immune responses and that the NK cells play a considerable role in the anti-tumor activity of the vaccine. We expect that these NK cell-mediated effects shown in preclinical animal models are also of importance in humans as a result of the similar administration route and mode of action.
The two-component mRNA vaccines can be combined with chemotherapeutics yielding improved therapeutic effect. Our data so far demonstrate that the two-component mRNA vaccines activate the innate and adaptive immune system in a comprehensive and multifunctional manner. Although this property supports the exceptional position of mRNA vaccines among immunotherapeutics, we have no doubt that future clinical developments will be based on combination therapies (i.e. combination of established standard therapies with novel approaches to increase the therapeutic effect). Among the standard therapies, chemotherapy is still the first-line treatment for many malignancies. In the past, the observed negative effect of many chemotherapeutics on lymphocytes [32-34] raised doubts about the suitability of chemotherapy for combination with immunotherapy. However, recent clinical research addressing this point demonstrates that a group of chemotherapeutics can induce immunogenic cell death and act synergistically with immunotherapy [35-37].
The great challenge for such combination therapies is to find an optimal treatment dosage and schedule for both therapies to induce a positive effect (and avoid negative effects). With this in mind, we tested the combination of mRNA vaccines with the two standard chemotherapeutics: docetaxel and cisplatin.
Docetaxel is used for the treatment of a variety of human malignancies including hormone-refractory prostate carcinoma and non-small cell lung carcinoma [38-40]. Although docetaxel was approved for the treatment of prostate carcinoma several years ago, there is no ‘gold standard’ regarding treatment schedules and doses. In our combination experiments, docetaxel was used at a dose of 25 mg/kg corresponding to the Food and Drug Administration (FDA) approved dose in humans of 75–100 mg/m2. Docetaxel was applied intraperitoneally as a single dose after three preceding vaccinations with OVA mRNA vaccine. Docetaxel administration was then followed by OVA mRNA vaccinations 3 days later (Figure 5A). Combining the OVA vaccine with docetaxel led to a significant further delay in tumor growth compared to the OVA mRNA vaccination or docetaxel alone. For combination therapies, the treatment schedule plays a pivotal role for their therapeutic efficacy. Thus, we also tested whether treatment of tumor-bearing mice might start with a single dose of docetaxel 2 days before vaccination with OVA mRNA vaccine (data not shown). With this treatment regimen, the delay of tumor growth was not increased compared to the single therapies. These data are in line with previous reports demonstrating enhanced anti-tumoral responses of recombinant viral vector vaccines in combination with docetaxel only when docetaxel was given after prime-boost vaccination . Moreover, docetaxel given before or at the same time point as viral vaccination led to decreased responses, reflecting the negative influence of docetaxel during the induction phase of the immune response. We made exactly the same negative observation when administering chemotherapy during the T cell induction phase (data not shown). Taken together, our data and previously published data advocate the combination of docetaxel with active vaccination after an immune response has been established.
Because we observed positive effects in the docetaxel combination therapy, we used an analogous treatment schedule for combination therapy with cisplatin at a single intravenous dose of 1.2 mg/kg (Figure 5B). The mRNA OVA vaccine significantly increased the therapeutic efficacy of cisplatin. Mice treated with combination therapy showed a slightly improved tumor delay compared to mice treated with immunotherapy alone, in line with our expectation that cisplatin administered after the induction phase has no negative but rather a positive effect on tumor inhibition. Cisplatin is known for the treatment of several cancers, including non-small cell lung cancer, testicular cancer, ovarian cancer and cervical cancer [42-47], and is also known for its immunomodulatory properties. It has been reported that cisplatin induces MHC class I expression, increased susceptibility of tumor cells to CTL killing, a decrease in myeloid suppressor cells in the spleen and peripheral blood of tumor-bearing mice and activation of NK cells [48, 49]. There is a strong possibility that the effects of the two-component RNA vaccine, such as strong induction of cytotoxic CD8+ T cells, reduction of myeloid-derived suppressor cells  and increased NK cell functionality (Figure 4B), might be able to strongly enhance the immunomodulatory effects of cisplatin resulting in the observed anti-tumoral responses.
In summary, our data demonstrate that the two-component mRNA vaccines can be combined with standard chemotherapeutics such as docetaxel or cisplatin when chemotherapy is applied after the induction of specific immune responses. Such combination therapy is feasible in the clinical setting and has the potential to provide an improved therapeutic effect.
Another possibility for increasing the therapeutic efficacy of mRNA vaccines is to combine them with other non-antigen specific immunotherapies. One example for such approach is the recently FDA approved ipilimumab, a human anti-CTLA-4 antibody for the treatment of melanoma. This antibody increases the activation of T cells by blocking the CTLA-4 receptor, which is responsible for the attenuation of the signal cascade [50, 51]. This blockade is rarely associated with severe side-effects as a result of the non-selective activation of autoimmune cells . However, in most cases, a positive clinical outcome is associated with the appearance of autoimmune symptoms. We investigated the effect of combining anti-CTLA-4 antibody with the OVA mRNA vaccine in a therapeutic tumor model. We administered anti-CTLA-4 at a lower concentration (100 µg/mouse) than previously reported in most studies (> = 200 µg/mouse) to avoid side-effects and to check whether this concentration may be suitable for combination therapy. Interestingly, although this concentration of anti-CTLA-4 antibody alone did not show any anti-tumor effect when combined with the OVA vaccine, a strong synergistic effect was obtained (Figure 6A) causing complete tumor rejection in some mice. To clarify whether complete tumor rejection was connected to induced epitope spreading, we challenged the responders with parental OVA-negative EL4 tumor cells. By contrast to the control mice, all three responders were protected against tumor challenge with EL4 cells, providing evidence for induced epitope spreading (Figure 6B).
These data indicate that a combination of anti-CTLA-4 with mRNA vaccines may allow exploitation of the therapeutic potential of ipilimumab at the same time as reducing the dose required and avoiding its side-effects.
The two-component mRNA vaccines represent a novel and highly potent vaccine approach. The first clinical data demonstrate their safety, tolerability and strong immunogenicity in humans [68, 69]. The data reported in the present study demonstrate how this innovative technology can be effectively transferred into the clinic by optimizing the vaccination schedule, as well as how it can provide a platform for different combination therapies. We are convinced that combination approaches will play central role in future clinical developments, opening the possibility for attacking tumors via complementary, synergistically-acting mechanisms.
The authors declare that there are no conflicts of interest.