Cancer Vaccine Immune
Chapter Five
General Discussions, Conclusions and Future Directions
5.1 Discussion
The overall objective response rate to current cancer vaccine formulations is only 3.3% [1]. Among the various reasons for this failure in immune response, two important issues stand out: 1. Current cancer vaccines can only induce “weak” qualitative and quantitative T cell responses. 2. The immunosuppressive tumor microenvironment inhibits anti-tumor T cell activity at the effector phase [2, 3]. The major goal for improvement of response to cancer vaccines is to develop immunotherapy strategies that can activate robust and lasting immune responses against cancer antigens and, at the same time, be able to reverse the ‘immunosuppressive milieu' of the tumor microenvironment. The goal of this research was to evaluate PLGA nanoparticles (NP) as delivery vehicles for targeting vaccine components to dendritic cells (DCs) for meeting those challenges. Formulating antigens in PLGA-NP offers distinct advantages over soluble formulation. Some of these advantages include 1. protecting the antigen from proteolytic degradation. 2. delivering the antigen to the phagocytic cells (mainly DCs) in a targeted and prolonged manner. 3. avoiding antigen entry to systemic circulation. 4. particulate antigens are more efficiently cross-presented (than soluble antigens). Furthermore, PLGA-NP facilitate co-delivery of various immunomodulators (such as TLR ligands) to DCs. Co-delivery of antigens along with TLR ligands to the same DC result in concomitant antigen processing and presentation in addition to triggering of TLR signaling leading to the generation of mature DCs capable of activating cellular immune responses.
Lipid A anchor of bacterial LPS is a well characterized TLR4 ligand that is responsible for both immunostimulatory properties and toxicity of LPS. Previous studies have shown that the toxicity of LPS can be largely reduced by selective hydrolysis of the anomeric phosphate group of lipid A molecule. In fact, monophosphoryl lipid A (MPLA_ is now extensively studied in various vaccine formulations in clinic, with an excellent record of safety. However, being extracted and purified from bacterial culture, the process of MPLA manufacture is laborious, inconsistent and often leads to batch to batch difference in terms of activity and structure of MPLA. With this in mind, several lipid A analogues have been designed and chemically synthesized. Synthetic lipid A analogues are very promising adjuvants for future cancer vaccine formulations. In addition to their potent immunostimulatory activity, lower toxicity (compared to LPS), the process of their manufacture is consistent and often lead to highly pure and stable molecules of single structure.
We believed that the initial step towards the development of lipid A-based vaccine is the establishment of a satisfactory analytical method for their quantification. In fact, previous attempts to analyze lipid A have been hampered by its extremely low sensitivity to UV detection. This problem had been circumvented by pre-column derivatization of lipid A (in a 3 hours-chemical reaction). In the present studies (chapter 2), we have developed and validated a LC-MS-based method for quantitative analysis of lipid A analogues inside PLGA-NP. This novel method holds several advantages over the previously reported methods for lipid A quantification. Some of these advantages include; 1.Ease. 2. High sensitivity. 3. Quick analysis time. 4. High specificity 5. High accuracy and reproducibility. 6. Suitable for simultaneous quantification of several lipid A analogues. 7. Avoid the need for any pre-column derivatization or radiolabeling. As a preliminary application, this method has been used in the quantification of two synthetic lipid A analogues in PLGA-NP (7-acyl lipid A and PET lipid A). Both analogues (in soluble form) exert similar immunostimulatory effect on BMDCs, as measured by upregulation of co-stimulatory molecules (CD40 and CD86) and secretion of cytokines (IL-6, IL12 and TNF-a). However, when both analogues were formulated inside PLGA-NP, the encapsulation efficiency of 7-acyl lipid A was almost 3 fold higher than that of PET lipid A. The presence of an extra lipid chain in 7-acyl lipid A may account for its higher lipophilicity and hence better encapsulation in PLGA-NP. Different formulation parameters could be further optimized to refine PET lipid A-NP and to increase the encapsulation efficiency of PET lipid A in PLGA-NP. However, these optimization studies were not performed in the current series of investigations. In fact, 7-acyl lipid A was the adjuvant of choice in all our subsequent studies.
Our next goal was to formulate a model antigen (ovalbumin: (OVA)) along with 7-acyl lipid A in PLGA-NP and to evaluate the potential of this formulation to stimulate OVA specific T cell responses (chapter 3). PLGA-based vaccines are designed to target DCs both in vitro and in vivo. DCs exhibit several features which are required for the generation of efficient T cell mediated immune responses. These features include: wide distribution allover the body surfaces, high efficiency in taking up antigens and transferring them to secondary lymphoid organs, intrinsic migratory capacity, ability to cross-present exogenous antigens to CD8+ T cells through multiple pathways (described in chapter 1), constitutive expression of MHC class I, class II molecules as well as co-stimulatory molecules, ability to secrete high amount of IL-12 that derives Th1-biased immune response and unique ability to stimulate immunological naive T cells. Our results showed that PLGA-NP could deliver the encapsulated protein (OVA) to DCs both in vitro and in vivo. Released OVA could be further processed to give CD4 (OVA323-339) and CD8 epitope (SIINFEKL), resulting in simultaneous activation of CD4+ and CD8+ T cells. The presence of 7-acyl lipid A along with OVA in the same nanoparticle formulation further increased the extent of T cell activation, as measured by proliferative response, up-regulation of activation markers and ability to secrete IFN-γ. These results are in agreement with recent studies that reported enhanced and prolonged antigen presentation when antigen is co-delivered with TLR9 ligand in PLGA-NP.
Particulate antigens are more efficiently taken up and cross-presented by DCs (compared to soluble antigens). Similarly, particulate delivery of TLR ligands offers several advantages over their administration in a soluble form. Interestingly, persistent signaling through TLR was required for overcoming tumor-induced immunosuppression mediated by Treg cells. Particulate delivery systems could facilitate a sustained TLR signaling in DCs. Consequently, avoid the need of repeated administration or high dosages of TLR ligands. In fact, most of TLR ligands may show serious side effects when administered in high dose. For example, repeated daily injections of 60 µg of CpG in mice caused drastic damages to lymphoid tissues and hepatic toxicity after 14 days of treatment. Delivery of TLR ligands in PLGA-NP would permit the use of very small doses and limit the non-specific immune activation and/or toxicity that may result upon systemic administration of those compounds. A previous study in our lab had shown that the effective dose of TLR9 ligands (CpG) needed for in vivo priming of antigen-specific T cell response could be reduced by 10-100 fold when it was delivered in PLGA-NP. This dose sparing effect is probably not limited to CpG only, but can also be extended to other TLR ligands and/or immunomodulatory molecules. Particulate delivery can also facilitate simultaneous delivery of more than one TLR ligand in the same formulations. Warger et al have recently shown a synergistic effect of combined-TLR triggering (TLR3 and TLR7) on BMDCs, as evidenced by a faster and more sustained secretion of pro-inflammatory cytokines (IL-6 and IL-12). More importantly, CD4+ and CD8+ T cells activated by DCs treated with a combination of both TLR3 and TLR7 ligands were completely resistant to Treg-mediated immune suppression. Untreated DCs or DCs treated with a single TLR ligand were unable to reverse Treg mediated suppression of CD4+ and CD8+ T cell responses. These results imply that properly activated DCs could overcome and/or reverse Treg mediated immune suppression.
Given the demonstrated effectiveness of 7-acyl lipid A as a vaccine adjuvant, and the ability of DCs pulsed with OVA along with 7- acyl lipid A in PLGA-NPs to dramatically enhance in vitro and in vivo OVA specific CD4+ and CD8+ primary T cell responses (chapter 3), we hypothesized that particulate delivery of cancer antigen (TRP2) and 7-acyl lipid A would be very effective at eliciting in vivo potent cellular responses that could mediate therapeutic anti-tumor response in the poorly immunogenic murine B16 melanoma model. The capability of TRP2/7-acyl lipid A-NP vaccine strategy to break self-tolerance and to induce superior anti-tumor effect (chapter 4) could be explained through numerous mechanisms. Co-delivery of 7-acyl lipid A along with TRP2180-188 to the same DC poulation provides the three signals required for optimum CTL activation. DC stimulated with TLR ligand increase the expression of peptide/MHC I complex on the cell surface (signal 1), upregulate costimulatory molecules e.g CD40, CD80 and CD86 (signal 2), and secrete various cytokines e.g. IL-12 (signal 3). The three signals combined lead to enhanced activation and proliferation of TRP2 specific CD8+ T cell. Another avenue for breaking self-tolerance is through the ability of TLR activated DCs to reverse the T regulatory (Treg) suppressive effects. In fact, it has been recently shown that IL-6 secreted by TLR4-activated DCs renders antigen specific T cells refractory to the suppressive activity of Treg. Other studies have shown that stimulation of DCs with TLR ligands enhances the proliferation of antigen specific T cells, making it harder for Treg cells to inhibit them. Recent studies from our lab have shown that particulate delivery of 7-acyl lipid A leads to 1000-fold increase in the amount of IL-6 secreted by DCs (relative to soluble form). Furthermore, co-delivery of OVA and 7-acyl lipid A in PLGA-NP to DCs have dramatically enhanced the extent of in vitro primary CD4+ T cell activation by 1000 fold (compared to soluble formulation). Vaccine delivery system capable of inducing IL-6 production by DCs, and activation of primary T cell responses to this extent, may assure overcoming Treg-mediated immunosuppression through involvement of activated T cells.
5.2 Conclusions
1. A rapid, easy and reliable LC/MS-based analytical method for lipid A quantification was developed and validated.
2. Co-delivery of antigen and 7-acyl lipid A in PLGA-NP results in clonal expansion and activation of primary CD4+ T cell responses in vivo.
3. DCs targeted in vitro with antigen and 7-acyl lipid A co-encapsulated in PLGA-NP induce robust activation of primary CD8+ T cell responses in vitro
4. PLGA-NP can efficiently deliver encapsulated antigens to both MHC class I and MHC class II processing pathways, resulting in simultaneous activation of CD8+ and CD4+ T cell responses, respectively. Presence of 7-acyl lipid A in the same nanoparticle formulation dramatically enhances IFN-γ secretion by activated CD4+ and CD8+ T cells.
5. PLGA-NP co-encapsulating the poorly immunogenic melanoma antigen, TRP2 along with 7-acyl lipid A (TRP2/7-acyl lipid A-NP) are able to break self-tolerance against TRP2 and generate robust CD8+ T cell immune responses in both normal and tumor bearing mice.
6. Vaccinating B16 melanoma bearing mice with TRP2/7-acyl lipid A-NP resulted in the induction of therapeutic anti-tumor immunity, activation of IFN-γ secretion by CD8+ T cells as well as innate immune cells and finally, reversal of the immunosuppressive network in the tumor microenvironment.
5.3 Future directions
5.3.1 Evaluating the efficacy of PLGA-based cancer vaccine in a mouse spontaneous tumor model
The promising results that we obtained from B16 melanoma tumor model (chapter 4) is a real driving force for us to rigorously evaluate the efficacy of our vaccination strategy in a more challenging and realistic tumor model. Unfortunately, transplanted tumor models harbor several drawbacks that limit their applicability to human disease and make them poor predictors of vaccine efficacy in patients. First, transplanted tumors don't grow in the anatomically appropriate site; as they are usually inoculated subcutaneously (s.c) or intravenously (i.v). As a result, they don't mimic the organ-specific physiology characteristic of the naturally occurring tumor. More over, transplantable tumors progress very rapidly following inoculation, and the immune system is suddenly exposed to them, whereas spontaneous tumors develop much more slowly. This slow development provides sufficient time for the battle between host immunity and cancer, with two possible outcomes; tumor elimination or most probably, tumor escape. Finally, most transplantable mouse tumors are not spontaneously metastatic, so immunotherapeutic studies using these models are not particularly relevant for most types of human cancer, where the metastatic spread of the disease is the major cause of death [11]. Assessing whether PLGA-based cancer vaccine is capable of eliminating metastatic tumors will be a great challenge and will truly demonstrate the effectiveness of our vaccine formulation.
Previous studies in our lab has shown that bone marrow derived DCs treated with MUC1/7-acyl lipid A PLGA-NP stimulated the proliferation of T cells derived from human MUC1 transgenic mice, suggesting that particulate delivery of antigen and 7-acyl lipid A could break tolerance to self antigen (MUC1). These results are in agreement with our current results obtained from B16 melanoma tumor model (chapter 4). One of our future goals is to test the efficacy of PLGA-based vaccine against MUC1 expressing tumors that have comparable onset, progression, staging, and pathology to the human cancer for which the vaccine is designed.
Mukherjee et al [14] have developed and characterized a mouse line that expresses human MUC1 as a self molecule (with a tissue distribution similar to that in human) and spontaneously develops MUC1-expressing tumors of the mammary gland. These double transgenic mice are called MUC1-expressing mammary tumor (MMT) mice, and were developed by mating mice that carry the human MUC1 transgene driven by its own promoter (MUC1.transgenic mice) with mice that carry the polyomavirus middle T oncogene under the transcriptional control of the mouse mammary tumor virus promoter [14]. Middle T antigen specifically associate with and activates the tyrosine kinase activity of a number of signal transduction proteins such as the c-Src family, phosphatidylinositol 3'-kinase, Ras, and c-Myc. Activation of these proteins leads to the promotion of cell growth and/or survival and result in widespread transformation of the mammary epithelia resulting in rapid production of multifocal mammary adenocarcinomas [15]. MMT transgenic mouse model harbor several advantages that make it an excellent model for evaluating vaccine-induced anti-tumor immune response. Some of these advantages are listed below.
1. The expression of MUC1 in MTT mice provides a useful target for immunotherapy as well as a marker for detecting ongoing immune responses: Similar to MUC1.Tg mice, MMT mice show immune tolerance against MUC1 and therefore suitable for evaluation of vaccination strategies to break self-tolerance against MUC1 antigens and evaluation of any potential autoimmunity. In addition to the MMT model, MET [17] and MUC1/MIN [18] double transgenic mouse models also develop spontaneous MUC1 expressing tumors in the pancreas and intestine, respectively. However, mammary tumors in particular, are superior to the other tumors, as it can be followed by palpation and are very useful for the assessment of therapeutic vaccines, as tumor location alleviates the need to sacrifice the animal to determine clinical outcome. Mammary tumors are also ideal for prophylactic vaccine studies, as the tumors develop after birth, allowing enough time for the immunizations to be performed prior to tumor development [19]. This is in contrast to MET model, where the mice exhibit acinar dysplasia “pre-cancerous lesion” at birth, which very rapidly progressesto invasive cancer by week 10, resulting in losing 90% of the mice by week 16 [20].
2. In MMT model, spontaneous tumors develop in 100 % of the female mice: Whereas the incidence of spontaneous tumors in many of genetically modified mouse models can be very low (less than 10%), and tumor development can take up to several months [13], 100% of the female MMT mice get tumors within reasonable time frame (10-16 weeks). Thus it is very useful in experimental settings.
3. MMT model closely mimics human breast cancer situation: Immunohistochemistry and western blot studies have shown that spontaneous mammary gland tumors in MTT mice expressed MUC1 protein in a similar pattern to what is observed in human [15]. In addition, Tumors pathology in MTT mice closely resembles scirrhous carcinomas of the human breast (highly fibrotic with dense connective tissue separating individual nests of tumor).
4. MMT tumors are spontaneously metastatic: Lung and bone metastasis were also detected in the MTT mice by the age of 17-24 weeks. Thus it is an excellent model to investigate the efficacy of prophylactic and therapeutic immunotherapies against metastatic spread of the disease. It is worth mentioning that the mortality among breast cancer patients is directly associated with lung and bone metastasis. MMT model is then more useful than MUC/MIN model, in which no visible metastasis was observed [18].
For all the aforementioned reasons, we believe that MMT mouse is very reliable and appropriate model that closely mimics human cancer and will be an excellent tool for future investigations on the effects of PLGA-based vaccine on self-tolerance, immunity and autoimmunity to MUC1 antigens.
5.3.2 Assess the relative contribution of vaccine-induced CD4+ and CD8+ T cell responses in mediating anti-tumor activity
The relative contribution of CD4+ and CD8+ T cells to the efficacy and success of cancer vaccines differ from one vaccination strategy to another [4-7]. It also depends on whether the vaccine is employed in prophylactic or therapeutic setting [4]. One of our future goals is to investigate the ability of PLGA-based cancer vaccine formulations to stimulate CD4+ and CD8+ T cell responses in MTT spontaneous tumor model, and to assess the relative contribution of vaccine-induced CD4+ and CD8+ T cell responses in mediating anti-tumor immunity.
To perform these studies, MUC1 derived MHC class I and class II restricted epitopes will be co-encapsulated (along with 7-acyl lipid A) in PLGA-NP. The ability of this formulation to break self-tolerance against MUC1 antigen and to induce Th and CTL responses against MUC1 expressing tumor cells will be examined in both prophylactic and therapeutic vaccine setting. The relative role of CD4+ and CD8+ T cells in mediating prophylactic and/or therapeutic anti-tumor immunity will be investigated by in vivo depletion with specific monoclonal antibodies (mAb). Comparing vaccine efficacy and long-term survival between different groups will demonstrate whether PLGA-mediated tumor regression is dependant on CD4+ or CD8+ T cells, or both combined.
On the light of these results, we can work on modification of physical properties of PLGA to shift the delivery of encapsulated antigens to either cytoplasm (for MHC I presentation and CD8+ T cell activation), to the endosome (for MHC II presentation and CD4+ T cell activation), or to both MHC I and II pathways. Previous studies in our lab [9] have shown that, cytoplasmic delivery of PLGA content is affected by difference in molecular weight of PLGA. Small molecular weight PLGA (6000 g/mol) delivered its content faster than larger molecular weight polymer (60,000 g/mol). Further studies may be needed to define additional physical properties that can be adjusted to tailor-made cytoplasmic delivery of PLGA (such as; hydrophobicity, lactic/glycolic ratio, particle size, surface charge), and the effect of these parameters on the ability of PLGA-NPs to induce either CD4+ and/or CD8+ T cell responses in the MTT model. These proposed investigations will clearly define the way to modify PLGA formulations for optimum cellular immunity that can mediate anti-tumor immune responses.
5.3.3 Combination studies between PLGA-based vaccination and adoptive cell therapy
The magnitude of the anti-tumor immune responses has been demonstrated to be a significant factor in tumor eradication 38. Augmentation of anti-tumor immune responses could be achieved by adoptive transfer of ex vivo expanded effector T cells 13. Previous animal studies have shown that a frequency of tumor specific T cells of at least 1-10% of CD8+ T cells is needed for successful eradication of established tumors 13 . In human, this corresponds to an approximate dose of 2 to 20 × 109 CD8+ T cells. Many methods have been employed to expand effector T cells in vitro (such as treatment with cytokines or non-specific triggering of TCR/ co-stimulatory molecules). These methods could successfully expand effector cells to more than 1010 cells in vitro within 14-28 days14,15. However, such methods are laborious, expensive and time consuming. The robust expansion of T cells by DCs loaded with particulate antigen (chapter 3) can be employed in adoptive transfer therapy. Interestingly, recent studies have demonstrated that a combination of in vivo vaccination and adoptive T-cell transferresults inmore robust anti-tumor activity than the use of eachstrategyindividually 39. Particulate delivery of antigen plus 7-acyl lipid A to DCs in vitro could be an optimum strategy to generate a population of T cells of desired magnitude with defined phenotypic and functional properties in a very short time. Those expanded T cells could be optimized to be used in adoptive transfer studies along with in vivo vaccination protocol.
5.3.4 Particulate delivery ofa-Galactosyl-Ceramide; A Strategy for NKT Cell Activation:
Natural killer T (NKT) cells are a specialized subset of T cells that share some properties of NK cells and conventional T cells (Mercer, Ragin, & August, 2005). Whereas conventional CD8+ and CD4+ T cells recognize peptide antigens in association with MHC class I and MHC class II, respectively, NKT cells respond to glycolipids in the context of CD1d molecule (Brigl & Brenner, 2004). a-galactosyl-ceramide (a-GalCer) is a synthetic ligand that is efficiently presented on CD1d molecules to NKT cells (Kawano et al., 1997). a-GalCer-stimulatedNKT cells don't kill tumor directly, but instead they stimulate down stream innate and adoptive anti-tumor immune responses through secretion of large amount of cytokines needed for recruitment and activation of various immune cells (such as DCs, NK cells and CTLs).
Previous studies have shown that i.v injection of ex vivo a-GalCer-loaded DCs resulted in to 100- fold expansion of several subsets of NKT cells in a group of patients with advanced cancer. Expanded NKT cells could be detected for up to 6 months after vaccination. NKT activation was associated with an elevated serum level of IL-12 and IFN-γ inducible protein -10 (IP-10)(Chang et al., 2005). These observations can open a new avenue for targeted delivery of a-GalCer to DCs in vivo using PLGA-NP. Particulate delivery of a-GalCer to DCs will lead to sustained expansion of NKT cells in vivo resulting indownstream activation of both innate and adoptive anti-tumor immune responses.
5.3.5 Direct assessment of NK cell activation (induced by PLGA-based vaccination)
The involvement of NK cells in mediating anti-tumor responses has beenwell described in different experimental systems. Furthermore, a cross-talk between NK cells and DCs has been found to orchestrate both innate and adoptive anti-tumor immune responses. NK cells can facilitate adaptive anti-tumor immunity by producing IFN-γ and other cytokines that lead to the recruitment and activation/maturation of DCs. Lysis of tumor targets by NK cells could further provide DCs with increased access to tumor antigens (to be cross-presented to CTL). On the other hand, activated DCs constitute a source of numerous cytokines that in turn induce NK cell activation. In particular, DCs-derived IL-15, IL-12/IL-18 and IFNα/β could induce NK cell proliferation, IFN-γ secretion and cytotoxic function, respectively. In the light of these findings, systemic administration of IL-15, IL-12/IL-18 and IFNα/β was employed for harnessing NK cells to reject established tumors. However, the systemic administration of these cytokines (particularly IL-2 and IFN-α) often lead to non-specifically activation of a broad range of immune cells, resulting in serious undesirable side effects. Co-delivery of tumor antigen and 7-acyl lipid A in PLGA-NP can selectively stimulate DCs to produce NK cell-activating cytokines. This strategy might provide a precise route for triggering NK-cell activation and circumvent the need systemic cytokine administration. One of our next research goals is to directly assess the relative contribution of NK cells to the anti-tumor effects of our vaccination strategy. NK cell activation induced by PLGA-based vaccination will be assessed (with respect to increase in number, cytokine and functional activity).
These proposed investigations will clearly define the way to modify PLGA formulations for optimum cellular immunity that can mediate anti-tumor immune responses. These investigations will help up to formulate a single vaccine strategy capable of activating both CTL and NK cells. Such vaccine can act as a double-edged sword, capable of targeting and killing both MHC class I positive and negative tumor cells.
In summary, we believe that PLGA nanoparticles are as competent carriers for future cancer vaccine formulations. The aforementioned proposed investigations (5.3.1-5.3.5) will clearly and define the way to refine PLGA formulations for optimum cellular immunity. Results obtained could help in designing a single vaccination strategy capable of activating robust CD4+, CD8+ T cells, NK cells and NKT cells that could mediate full-scale anti-tumor immune responses.
We provide a professional essay writing service that thousands of our customers use as an effective way of improving their grades, improving their research and saving them lots of time.
