Cancer vaccines: limited success but the research should remain viable

Therapeutic cancer vaccines are designed to induce robust T-cell effector and memory responses. From the clinical point of view, benefits of cancer vaccines administered to treat established tumors have shown some efficacy; however, challenges remain. The choice of target antigen is of paramount importance, and mutated tumor antigens represent attractive targets. Combining vaccines with immunomodulating agents in a properly designed clinical setting are important issues to optimize the benefits offered by cancer vaccines.

Cancer immunotherapy has emerged as a treatment modality upon recent advances in our understanding of immune pathways regulating endogenous antitumor immunological responses. Numerous clinical studies, consisting of immune checkpoint blockade and adoptive transfer of genetically engineered T-lymphocytes, yielded impressive clinical results in various types of cancer. These studies have demonstrated a link between the outcome of cancer immunotherapies and potentiation of previously induced immune responses. Although the development of cancer immunotherapy has reached an important level among cancer therapies, the contribution of therapeutic vaccines to cancer treatment remains debatable. The initial success with the US FDA approval for Provenge^®, and a plethora of phase II trials showing a large variability of vaccine efficacy, was followed by the failure of a large number of phase III trials to reach their primary endpoints. These failures raised the question why vaccines have failed and how vaccines can induce clinical efficacy in a subset of patients and in which treatment combination. For instance, in the MAGRIT phase III trial which was terminated due to lack of improvement in progression-free survival or overall survival, a Th1-associated immune gene signature of the tumors before immunization was identified that predicted high responsiveness to the vaccine with a significant improvement in disease-free interval hazard ratio [1,2]. Similar findings were also reported in several other trials where immunological responses to the vaccine served as surrogates for clinical efficacy [3,4]. Thus, it seems to be quite possible that the success or failure of cancer vaccines depends on the magnitude of the antitumor responses mediated by patients’ immune system. This suggests that therapies aiming at positively modulating patients’ antitumor immunity represent promising treatment modalities.

Compromised immunological responses to the vaccine may reflect a broader immune deficiency. For example, in studies with melanoma patients, vaccines have failed whereas immune checkpoint inhibitors have demonstrated durable clinical benefits [5,6]. The reason for this discrepancy may be explained by the different mechanisms of action utilized by these two immunotherapeutic approaches. Immune checkpoint inhibitors act by counteracting tumor-mediated suppression, thus re-inforcing the exhausted antitumor CD8⁺ T cells and re-instating endogenous antitumor immunity [7]. The mechanism of action of vaccines requires functionally intact dendritic cells that can effectively present the vaccine antigen, and CD8⁺ T cells capable of responding to the vaccine, both of which are functionally inactive in melanoma and other types of cancer [8,9]. Restoring the deficient immune system before vaccinations may result in the generation of more efficient immunological and clinical responses. In this case, an important issue will be to choose the best approach that would most successfully complement vaccines. T-cell exhaustion by the persistent exposure to tumor antigens is one major mechanism that blocks effective antitumor immunity. Programmed cell death-1 (PD-1) has been identified as a marker expressed by exhausted CD8⁺ T cells in cancer, and reversing exhaustion of this type of cells by anti-PD-1 antibodies has been shown to induce disease control and prolong overall survival in some types of cancer [10]. Given that vaccines are designed to promote functionally active tumor-specific CD8⁺ T cells, PD-1 expression on these cells will significantly limit vaccine efficacy. Consequently, the field of therapeutic vaccination in cancer will move forward through the design of protocols reversing exhaustion and increasing vaccine efficacy.

The field of therapeutic vaccination will be also advanced through our better understanding of tumor evolution and tumor escape from immune surveillance. Given the crucial role of oncoantigens in cancer development and progression, targeting such antigens by vaccines should largely reduce the appearance of tumor-escape variants and those which will emerge may have impaired tumorigenic potential. A crucial component of cancer immunotherapy will be to control malignancies by robustly targeting mutations regardless of their oncogenic activities. Recent reports have suggested that patients who responded to checkpoint inhibitors harbor cancers with a high frequency of somatic mutations which generate tumor-specific epitopes [11]. This class of tumor-specific ‘neoantigens’ may represent an ideal class of target antigens for active immunotherapy mainly for two reasons: first, they are highly tumor specific; and second, they can be immunogenic because, in contrast to normal self epitopes, they are seen as ‘foreign’ (i.e. not subjected to negative thymic selection). Studies have shown that enhancing neoantigen-specific CD8⁺ T-cell responses could improve antitumor responses [12]. Furthermore, there is increasing evidence to suggest that neoantigens can also be presented to T cells in the context of major histocompatibility complex class II molecules [13]. The participation of CD4⁺ T cells recognizing neoantigens would be required to optimize the lytic activity of CD8 T⁺ cells targeting neoantigens. Thus, the optimal vaccine platform used to elicit robust antitumor responses may have to include both CD8⁺ and CD4⁺ T cells specifically recognizing neoantigens. Tumor-specific CD4⁺ T cells may also be needed for inducing antitumor humoral immunity. Antibodies may be useful therapeutics against tumors in general by activating complement-mediated cytotoxic mechanisms, mediating antibody-dependent cytotoxicity, and giving rise to immune complexes triggering additional antitumor responses and determinant spreading. A striking example in this respect was provided by the IMPACT phase III study, where overall survival benefit in vaccinated patients was correlated with increased immunoglobulin G production against prostatic acid phosphatase but also other tumor antigens (B-cell determinant spreading) [3].

Another issue which needs to be addressed in order to improve the clinical benefits offered by therapeutic cancer vaccines relies in the proper clinical setting. Although we understand how immunotherapy works, the subject surrounding the optimal clinical design for the application of cancer vaccines is rather complex because it has been difficult to demonstrate clinical benefits in the various clinical settings. Therapeutic vaccination in the metastatic setting has yet to demonstrate clinical significance. This may be due to the heavy tumor burden which generates a hostile tumor microenvironment with a plethora of suppressor elements dampening antitumor responses. To circumvent this issue, recent clinical trials were performed with patients at the adjuvant setting with minimal residual disease and a high risk of relapse [14]. The rationale behind vaccination in this clinical setting is that patients with minimal tumor burden still have a fully competent immune system capable of developing robust antitumor responses. Moreover, vaccinating in the adjuvant setting has the advantage to avoid accumulation of T cells at large hostile tumor burdens where they might become dysfunctional. Recent reports from clinical trials support the application of therapeutic vaccination as an adjuvant therapy in patients with low tumor load post surgery or in patients with more indolent disease [15,16]. In these settings, therapeutic vaccines have significantly reduced the frequency of recurrences. Importantly, booster inoculations are essential to maintain any immunity with peptide cancer vaccines. In the NeuVax Phase II trial with disease-free patients at high risk for recurrence, immunity was noted to wane with time and this corresponded with increased recurrences noted in the vaccine arm. Booster inoculations were able to maintain immunity, and those who received scheduled booster inoculations were less likely to recur [17].These findings are in line with the emerging body of evidence supporting immunotherapy in patients with low tumor burden. Proving efficacy of cancer vaccines alone in this setting will allow the use of novel adjuvants and combination therapy to expand the indications to more aggressive and advanced disease.

Results from various clinical trials have demonstrated that monotherapies are unlikely to confer significant clinical benefits to patients because of the serious obstacles provided by the tumors which diminish antitumor immunity [18]. Moreover, tumor cells generate adaptive immune resistance, a process which enables them to evade immune attacks [19]. Therefore, interventions are needed to re-instate anticancer immune responses by actively counteracting the immune inhibitory mechanisms of tumor cells. In this respect, clinically effective antitumor responses are dependent on the modulation of more than one immune pathway which will enable the induction of robust T-cell responses against the tumor. Under this frame, it is reasonable to suggest that cancer vaccines are more likely to generate clinical effects when combined with other modalities. Despite the failures from vaccination studies, we should recognize that cancer vaccines may generate meaningful antitumor responses under the appropriate conditions provided that patients have a functioning immune system that can respond to the vaccine. Indeed, even in studies where vaccines have been successful, the immunological responses generated were suboptimal and could be improved under the combined treatment with immune checkpoint inhibitors [20]. Besides immunomodulatory antibodies, several other immune-modulating molecules targeting oncogenic pathways have been approved for treatment [18]. In addition, several other molecules targeting inhibitory or stimulatory pathways are being developed. The combined use of these medicines with cancer vaccines is the subject of intense investigation and holds promise for the improvement of cancer vaccines.

Financial and competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.