Abstract
In clinical settings, cancer vaccines as monotherapies have displayed limited success compared to other cancer immunotherapeutic treatments. Nanoscale formulations have the ability to increase the efficacy of cancer vaccines by combatting the immunosuppressive nature of the tumor microenvironment. Here, we have synthesized a previously unexplored cationic polymeric nanoparticle formulation using polyamidoamine dendrimers and poly(D,L-lactic-co-glycolic acid) that demonstrate adjuvant properties in vivo. Tumor-challenged mice vaccinated with an adenovirus-based cancer vaccine [encoding tumor-associated antigen (TAA)] and subsequently treated with this nanoparticulate formulation showed significant increases in TAA-specific T cells in the peripheral blood, reduced tumor burden, protection against tumor rechallenge, and a significant increase in median survival. An investigation into cell-based pathways suggests that administration of the nanoformulation at the site of the developing tumor may have created an inflammatory environment that attracted activated TAA-specific CD8+ T cells to the vicinity of the tumor, thus enhancing the efficacy of the vaccine.
INTRODUCTION
Cancer immunotherapy represents an important treatment strategy for patients with inoperable cancers such as late-stage melanoma (1). A primary goal of cancer immunotherapy is to break immune tolerance in the immunosuppressive tumor microenvironment (TME) and mount an efficient immune effector response against tumor cells (2). Several U.S. Food and Drug Administration (FDA)–approved cancer immunotherapy strategies exist for melanoma treatment including checkpoint blockade [ipilimumab (anti–CTLA-4) and pembrolizumab (anti–PD-1)], oncolytic viral therapy (talimogene laherparepvec), and combinational checkpoint blockade [nivolumab (anti–PD-1) + ipilimumab] (3). One approach that has shown great promise in preclinical therapeutic settings is the use of cancer vaccines where adjuvants in combination with tumor-specific antigens (TSAs) or tumor-associated antigens (TAAs) are administered to generate tumor-specific cytotoxic T cell responses (2, 4). Currently, Sipuleucel-T is the only FDA-approved therapeutic cancer vaccine; this involves the reinfusion of the host’s blood cells [containing dendritic cells (DCs)] pretreated and activated with a recombinant fusion antigen [granulocyte-monocyte colony-stimulating factor (GM-CSF)–prostatic acid phosphatase]. However, its marginal increase in median survival time (4.1 months) (5) and the high cost (≈$100,000 per patient) prohibit its widespread adoption. Currently, a personalized DC-based melanoma cancer vaccine is in a phase 2b clinical trial (NCT02301611) but runs the risk of following the path of Sipuleucel-T by having a high cost associated with it, thus making it inaccessible to most patients not adequately insured.
A viable and cost-effective alternative to cell-based vaccines that can be delivered without the need for harvesting tissues from the patient includes using viral-based vaccines; these are an efficient means of delivering DNA encoding TAAs or TSAs and eliciting effective cytotoxic T lymphocyte (CTL) responses (6). While each viral vector has its advantages and disadvantages, the replication-deficient adenovirus has been proven to have many advantages, and its few disadvantages can be readily surmounted. The replication-deficient serotype 5 adenovirus (Ad5) has well-documented production techniques, can produce high viral titer stocks, and can encode relatively large DNA inserts, which can result in multiple whole-TAA/TSA expression (7). Along with high-efficiency gene transduction, the Ad5 has been shown to have a tropism for DCs, the most potent professional antigen-presenting cell (APC) population (8, 9). Given that DCs are the basis for all current cell-based cancer vaccines with FDA approval for those in clinical trials, the Ad5 cancer vaccine may be a viable and less expensive alternative (5, 7, 10, 11). Ad5 has also been proven to be well tolerated while being highly immunogenic in humans (12, 13). An unfortunate downside to using Ad5 is the reduction in efficacy due to neutralizing antibodies resulting from prior exposure to the wild-type virus; however, this can be readily circumvented using a gelatin matrix such as Gelfoam to deliver the Ad5-based vaccine (14–16).
Ideally, TSAs would be preferred over TAAs as the immunogen of choice for a cancer vaccine because of their inherently greater immunogenicity (2, 17, 18). However, TSAs lack the possibility of being prepared on a large standardized scale as they are often patient specific (18). Also, TAA-based vaccines can be produced ahead of diagnosis since it is already known that a patient diagnosed with a certain type of cancer will likely express well-defined TAAs. For example, patients with melanoma will have tumors expressing a suite of defined TAAs [including tyrosinase-related protein 2 (TRP2)]. However, regardless of whether TAA-based or TSA-based cancer vaccines are implemented, they both must overcome the myriad of immunosuppressive properties of the TME (19). Numerous strategies have been tested in preclinical studies in an attempt to improve the potency of TAA-based and model TSA-based cancer vaccines (2, 20, 21). It has been previously demonstrated that combining the Ad5 cancer vaccine (encoding a model TSA) with intratumoral administration of the adjuvant cytosine guanine oligonucleotide (CpG ODN) significantly reduces tumor growth and increases survival in mice, along with increasing the proportion of antigen-specific CD8+ T cells in the TME and peripheral blood (15, 22). Adjuvants formulated into nanoparticles (NPs) may be effective at modulating the immunosuppressive nature of the TME. Studies have demonstrated the increased efficacy of adjuvants when formulated into NPs (4, 23), and NPs of less than 500 nm have been shown to effectively accumulate in DCs (24).
Thus, formulating an NP-based adjuvant system may provide a possible means to further enhance the immunogenicity of Ad5-based cancer vaccines and cancer vaccines in general. Nanoscopic compounds such as dendrimers have created new avenues toward the development of novel delivery systems (25). Since being introduced in 1984, polyamidoamine dendrimers (PAMAMs) have gained the attention of many researchers as a tool for gene delivery (26–28) and drug delivery (29–31). Despite their great potential and biomedical applications, PAMAMs are known to be toxic; however, this can be overcome through chemical modifications that shield the highly cationic surface (32, 33). This, of course, limits the very attributes that differentiated PAMAM to begin with. In an effort to yield the beneficial attributes of PAMAM without reducing their biomedical application through chemical modification, we have developed a previously unexplored PAMAM-based NP formulation using a combination of poly(D,L-lactic-co-glycolic acid) (PLGA) and PAMAM to form NPs that limit the toxicities associated with unmodified PAMAM. This NP differs from previously reported formulations where PAMAM was physically adsorbed to the surface of PLGA NPs (34), or the PAMAM was chemically modified to yield beneficial attributes. This PLGA/PAMAM NP formulation will be termed PM and represents the first time that PLGA and PAMAM were used to make bilayer NPs from a single polymer solution that contains both polymers. To investigate the potential adjuvant properties of PM, we have used them in combination with an antitumor vaccine, Ad5 encoding the melanoma TAA, tyrosinase-related protein 2 (Ad5-TRP2). Melanoma represents an appropriate model to explore the effectiveness of this formulation for several reasons. The National Institutes of Health Surveillance and Epidemiology and End Results Program places malignant melanoma as the fifth most commonly diagnosed cancer in the United States, with an estimated 96,480 new cases in 2019 (35). Melanoma, if detected early (melanoma in situ), can be treated with surgery or targeted therapeutic agents based on a patient’s mutation status; however, for advanced-stage melanoma (stage III or IV), few options are available to patients (36). Thus, immunotherapy represents a viable alternative for these patients. Here, we have demonstrated that therapeutically combining PM with Ad5-TRP2 resulted in inhibition of tumor growth and increased survival in melanoma-challenged mice, which was, in part, due to greatly increased systemic CD8+ T cell responses and to the enhanced accumulation of CD8+ T cells to the tumor and tumor periphery. Mice that survived tumor-free (TF) long-term after receiving this treatment were also protected from tumor rechallenge.
RESULTS
Particle synthesis, characterization, and uptake
PAMAM can be classified according to a generation number, referring to the degree of branching from the core and being related to the number of surface amines present on the PAMAM. PMs were synthesized using the nanoprecipitation method summarized in Fig. 1A from a mixture of PLGA and PAMAM (ethylenediamine core, generation 5), and the resultant NPs will be referred to as PMG5. Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) images showed that PMG5 were spherical with smooth surfaces (Fig. 1, B to D). Fabricating PMG5 before each use is not practical for translation into the clinic. Thus, to evaluate their stability under appropriate storage conditions, PMG5 were lyophilized (freeze-dried) with and without sucrose, a cryoprotectant. The lyophilization of PMG5 without sucrose resulted in significant aggregation as indicated by an increase in the size and polydispersity index (PDI) of the PMG5 (Fig. 1, E and F). This effect was circumvented by adding sucrose to the formulation, which resulted in a significant decrease in the size and PDI of PMG5 when compared to PMG5 lyophilized without sucrose. It has been previously established that PAMAMs exhibit intrinsic fluorescence in the blue region with a slight shift in lambda max with increasing generations (37–39). Taking advantage of this feature, the intrinsic fluorescence of PAMAM was used to evaluate the uptake of PMG5 by bone marrow–derived DCs (BMDCs), and fluorescence (an indicator of uptake) was assessed by flow cytometry. Results shown in Fig. 2D demonstrate clearly that BMDCs incubated with PMG5 (163.9 ± 0.61 nm) had significantly greater fluorescence (due to the PMG5 being taken up by BMDCs) than BMDCs incubated with larger submicrometer-sized (523.9 ± 15 nm) and micrometer-sized particles (1278.3 ± 27 nm) (also made from PLGA and PAMAM) after 48 hours of incubation.
