Introduction

Immunotherapy in oncology can be defined as the manipulation of the immune system to recognize and destroy cancer cells [1]. In this context, the application of cancer vaccines represents the logical evolution and extension of their use in infectious diseases. However, attempts to reproduce the same results registered in the latter have been rather disappointing. There are a few exceptions, which are the generation of prophylactic vaccines against hepatitis B virus (HBV) and human papillomavirus (HPV), which subsequently impact the incidence and mortality of liver and cervical cancer, respectively [2]. While these are hailed as successes, their oncologic efficacy is indirect, through the prevention of viral infections. The choice of the right target antigen is essential while designing a vaccine [3]. Tumor-associated antigens (TAAs) are self-antigens abnormally expressed by tumor cells. Since high-affinity T cells recognizing self-antigens are eliminated during development by our immune system’s central and peripheral tolerance mechanisms, TAA-directed cancer vaccines face the challenge of activating any remaining low affinity T cells. Despite these challenges, the TAA vaccines are the most studied cancer vaccines thus far. On the other hand, tumor-specific antigens (TSAs) are often patient specific, coming from nonsynonymous mutations or genetic alterations, or even virally introduced genetic information in cancer cells. In this situation, the TSAs recognized by high-affinity T cells are less likely to be subject to central tolerance and induce autoimmunity [4].

Cancer vaccines can be categorized as cellular, viral vector, or molecular (peptide, DNA, or RNA) (Fig. 1) [3]. In this review, we aim to highlight the advances in cancer vaccines and the remaining hurdles to overcome.

Fig. 1
figure 1

Different categories of cancer vaccines on the basis of the vector used (Credit: created with BioRender.com)

Full size image

TAA Vaccines

As mentioned above, most cancer vaccines have targeted TAAs, which include cancer/germline antigens normally expressed only in immune privileged germline cells such as MAGE-A1, MAGE-A3, and NY-ESO-1 [5,6,7], cell lineage differentiation antigens, normally not expressed in adult tissues, such as tyrosinase, gp100, and MART-1 (PSA and prostatic acid phosphatase (PAP)) [8,9,10], and antigens that are overexpressed in cancer cells such as hTERT, HER2, mesothelin, and MUC-1 [11,12,13]. Developing such vaccines presents several challenges. TAAs, as self-antigens, B cells, and T cells that strongly recognize these antigens may have been removed from the immune repertoire by central and peripheral tolerance. Given this problem, any cancer vaccine TAA should be able to break the tolerance through stimulation of the low affinity or even rare TAA-reactive T cells remaining [14]. A mechanism to stimulate and increase T cell affinity is the use of strong adjuvants, co-stimulators, and repeated vaccination [15]. Despite this, though, in many cases the immune responses registered have been low and the clinical benefit marginal.

The most relevant and reliable measure of T cell activation is the quantity and quality of tumor-infiltrating T cells (TILs). Such analyses have become common in cancer vaccine development. A further challenge is that targeting TAAs, even ones overexpressed by the cancer itself, might result in increased toxicity. On-target, off-tumor toxicity has been observed in clinical studies. Chimeric antigen receptor–engineered T cell therapy (CAR-T) targeting colorectal carcinoembryonic antigen (CEA) causes severe colitis in a high percentage of patients, as this antigen is also expressed in normal intestinal tissue [16].

Oncogenic Viral Antigens

The potential impact and importance of developing cancer-preventing vaccines targeting viral antigens are quite simple to understand if we consider that approximately 10% of human cancers worldwide are caused by viral infections [17]. Those antigens are potentially highly immunogenic, and sometimes molecular drivers of oncogenesis. An example of a successful vaccine is the one comprising HBV surface antigens, highly effective in preventing infection and reducing the incidence of hepatocellular carcinoma (HCC). Similarly, a vaccine comprising HPV-like particles has provided protection against HPV infections and pre-cancerous lesions [18,19,20,21]. These vaccines are effective in preventing cancer, but have a main limitation: they lack or have only a very modest activity on established cancers. This is likely due to the fact that humoral immunity cannot efficiently eradicate large numbers of virus-infected cancer cells, which instead require cell-mediated immune responses. An alternative strategy, for instance in HPV-induced cancer, has been the development of distinct HPV vaccines targeting T cell epitopes of the viral E6 and E7 oncoproteins. These oncoproteins are expressed within infected cells and then processed and presented to stimulate cytotoxic T cells. Several different E6 and E7 vaccines are being tested in patients with cervical intraepithelial neoplasia (CIN), cervical cancer, and head and neck cancer [22,23,24] (Table 1).

Table 1 Cellular cancer vaccines under active clinical development
Full size table

For established cancer, in 2018, the FDA approved the first oncolytic virus for cancer treatment, talimogene laherparepvec (T-VEC) [25]. It relies on direct intratumoral injections to overcome dilution and neutralization in blood. It induces cell lysis and promotes antitumor immune responses locally and in distant lesions [26••, 27, 28•]. In a randomized phase II trial, T-VEC was combined with an anti-CTLA4, ipilimumab, in first or second line, and the combination showed a significantly higher objective response rate (ORR) compared to ipilimumab alone in patients with metastatic melanoma [27]. In the same patient population, the phase III OPTiM study demonstrated improved progression-free survival (PFS), ORR, and overall survival (OS) of T-VEC alone compared to GM-CSF [28•].

It is essential to stress that data coming from relatively small phase II studies should be confirmed, whenever possible, in larger phase III trials. It is not rare for preclinical and clinical data on small patient samples not to be confirmed in larger studies. A good example is the PROSTVAC-VF/Tricom vaccine that used recombinant poxviruses expressing prostate-specific antigen (PSA) for priming, followed by subsequent booster doses of a fowlpox virus encoding PSA. The study showed OS benefit in prostate cancer [29], but a more recent phase III trial of PROSTVAC in castration-resistant prostate cancer was discontinued due to futility at interim analysis [30].

To disrupt the tumor microenvironment (TME), viruses have been engineered to express targeted antigens and immunomodulatory molecules. Examples are the vaccine TG4010 that contains the modified vaccinia virus (MVA)–expressing tumor antigen, MUC-1, the immunostimulatory cytokine, IL2 [31], the TroVax which is an MVA-expressing oncofetal antigen 5T4 (MVA-5T4) [32], and the MG1 that is a version of the oncolytic Maraba virus engineered with added transgene capacity for targeted expression of TAAs and immunomodulatory agents [33]. The latter has been assessed in non-small cell lung cancer and human papilloma virus (HPV)–associated tumors [34, 35]. Furthermore, the MEDI5395, an attenuated Newcastle disease virus (NDV) engineered to express GM-CSF, entered phase I clinical trials in 2019 and results are awaited in 2021. Lastly, the B cell/monocyte-based vaccine, BVAC-C, transfected with recombinant viruses, such as HPV 16/18 E6/E7, has shown some activity in activating virus-specific T cells in a phase I study of patients with recurrent cervical cancer. In the trial, 10 patients who had experienced recurrence after at least one prior platinum-based combination chemotherapy received three intravenous infusion of BVAC-C. It was well tolerated, and of the 8 patients evaluable, one partial response (12.5%) and four stable diseases (50%) were seen. Immunologic response analysis showed that BVAC-C induced activation of natural killer T cells, natural killer cells, and HPV E6/E7 specific CD4 and CD8 T cells upon vaccinations in all patients evaluated. A phase II study is underway [36], as it is a phase I study of BVAC-B, transfected with recombinant HER2/neu, in patients with gastric cancer [37].

Peptide-Based Vaccines

Many peptide vaccine clinical trials have been conducted with demonstration of immune responses, yet significant clinical benefit has been elusive. Often, only single antigen–based short peptides are used. These may not be able to overcome antigen heterogeneity or loss of antigen expression within the tumor or stimulate robust immune responses [38, 39]. In contrast to short peptides, the use of multivalent synthetic long peptides (SLPs), containing both MHC class I and class II epitopes, can elicit a balanced induction of both CD8 and CD4 T cells [40].

SLP immune-therapeutics have been developed. They consist of highly immunogenic long peptides engineered to avoid central tolerance mechanisms by efficiently delivering antigens to dendritic cells (DCs), inducing CD4 + and CD8 + T cell responses [41]. Early clinical trials have shown a good safety profile and promising activity. For instance, in a phase II trial, the SLP vaccine ISA101 combined with the anti-PD-1 immune checkpoint antibody nivolumab was well tolerated in 24 patients with HPV-16–positive cancer. The efficacy appeared superior to that of nivolumab monotherapy [42•]. Furthermore, a phase I/II study of ISA101 combined with standard platinum-based chemotherapy in 77 patients with metastatic HPV-16–positive cervical cancer showed a strong correlation between strong vaccine-induced HPV-16–specific T cell response and OS [43].

The SVN53-67/M57-KLH (SurVaxM) is another peptide-based vaccine consisting of an SLP mimic engineered to trigger an immune response by targeting survivin, which is highly expressed in many cancers [44, 45]. The vaccine SurVaxM is under investigation in a phase I study in patients with survivin-positive neuroendocrine tumors (NCT03879694) [46].

A novel technology platform, T-win, was developed to allow identification, design, and validation of immune modulatory peptide-based vaccine candidates targeting the TME [47]. T-win vaccination has led to an antitumor response in vitro and vivo and synergizes with anti-PD-1 antibody treatment [48]. It is likely that T-win vaccination may lead to the expansion of T cells counteracting and modulating the immune suppressive environment within the TME.

The major T-win technology challenge is to activate the most potent anti-Treg immune response, while minimizing autoimmunity and subsequent toxicity.

DNA Vaccines

Similar to peptide vaccines, DNA and RNA vaccines have the advantage of relatively simple and inexpensive production. They can also trip nucleic acid sensors that activate DCs, including certain TLRs, STING, AIM2, and DAI pathways; hence, adjuvant co-stimulators are often less important.

DNA vaccination holds great promise in cancer. They use plasmids to ensure the delivery of tumor antigen–encoding genes. DNA vaccines allow the encoded antigen to be presented by MHC classes I and II, with subsequent activation of both CD4 and CD8 T cells and, indirectly, humoral immunity [49]. Furthermore, the intrinsic elements of plasmid DNA can also activate the innate immune response due to the recognition of the double-stranded DNA structure by cytosolic sensors [50]. Despite encouraging preclinical data and the improvement in the delivery techniques, DNA vaccines have not revealed high immunogenicity in human trials so far [51].

RNA Vaccines

RNA cancer vaccines offer advantages over DNA vaccines. In fact, RNA is more susceptible to degradation by ubiquitous RNases and this could be undermined by chemical modifications and incorporation of modified nucleosides such as pseudouridine [52, 53]. RNA, unlike DNA, cannot be integrated into the genome; therefore, it has no oncogenic potential. Furthermore, RNA only needs to enter the cytoplasm, whereas DNA needs to enter the nucleus, thus facing an additional barrier, the nuclear membrane. Many mRNA vaccine platforms have been developed recently and validated [54,55,56]. The possibility of engineering the RNA sequence has made synthetic mRNA more manageable than before. Furthermore, efficient and non-toxic RNA carriers have been developed that allow prolonged antigen expression in vivo [57].

RNA vaccines have traditionally been based on mRNA in trials to date. This approach is being challenged by the use of RNA replicons [58]. As the latter are self-replicating, they are thought to be longer lasting than mRNA vaccines and may require fewer vaccinations to elicit the desired response. Transfection efficiency and the duration of RNA replicons before degradation could be further improved with novel vaccine delivery approaches. Two possibilities consist of condensing RNA with protamine and encapsulating it into liposomal particles.

Recently, a phase I trial in metastatic melanoma patients assessed mRNA expressing a variety of TAAs grouped together in a liposome [59]. The antigens triggered T cell responses which were accompanied by disease control or tumor response.

Conclusion and Future Perspectives

Cancer immunotherapy has experienced tremendous progress in the last decade, with improvement of our understanding of cancer biology and immune escape mechanisms. It is therefore an exciting time in the field of immune therapies, including cancer vaccines, with myriad promising developments.

Considering the increasing number of approved monoclonal antibodies for cancer treatment, the development of antibody inducing vaccines represents an important opportunity to improve the armamentarium of therapeutic strategies against many tumor types. Of note, the effectivity of a polyclonal antibody response is expected to exceed the one of monoclonal antibodies, as reported in both preclinical studies that demonstrate pronounced antitumor responses and in early clinical trials showing benefit in patients with advanced cancer [60].

Furthermore, the exponential expansion over the past decade in the ability to sequence the genetic profile of an individual cancer patient has opened the door to a deeper understanding of cancer’s underlying biology through the checkpoint blockade therapy response, as well as to finding better antigens to target, though computational assessment of the mutations that have the most potential in stimulating the immune response of each patient.

With the concept of personalized cancer therapy and immunotherapy, panels of genomic and proteomic biomarkers predictive for response following molecular profiling of tumor and host cells using next-generation sequencing are expected to further help to shape the treatment and improve outcomes for patients with cancer. Moreover, vaccination strategies are expected to reduce hospital visits, resulting in enhanced quality of life, and most of these strategies are extremely cost-effective, keeping affordable costs for anticancer treatments, and offering socio-economic benefits, especially when compared to the prohibitively high drug costs of most recently developed anticancer agents.

As a future perspective, it is likely that some cancer vaccines could become the next preferred combination partner for long-term cancer treatments, serving as a platform that is easily combinable with existing therapies, such as immune checkpoint inhibitors, which have already dramatically risen therapeutic expectations in numerous cancers. This would provide innovative treatment options in which either combination therapy can be given as multi-target vaccines or vaccination is combined with conventional therapy or immunotherapy.

There are still hurdles to overcome in order to maximize success.

Future efforts should focus on optimization of target identification, streamlining of most appropriate vaccination strategies, and adjuvant development.

Another major concern is the current lack of validated biomarkers predictive of vaccine efficacy. The concept that vaccine-induced TILs increase is a plausible possibility, but the quantity and quality of TILs required for clinical efficacy are still unknown and probably vary for different vaccines and cancer settings. Furthermore, understanding which subtypes of T cells are more relevant for an effective cancer vaccine, and how to more selectively stimulate them, remains a little-understood challenge.

New strategies to improve outcomes are essential. These may include combinations of cancer vaccines with agents that increase MHC expression. Each novel approach will be accompanied by potential toxicities and unexpected challenges. Cautious optimism is warranted in the face of early successes seen in recent clinical trials for oncolytic vaccines. If an approach were to prove successful, it could revolutionize cancer therapy this decade the way checkpoint inhibition did in the previous decade.