Turning the corner on therapeutic cancer vaccines

Abstract

Recent advances in several areas are rekindling interest and enabling progress in the development of therapeutic cancer vaccines. These advances have been made in target selection, vaccine technology, and methods for reversing the immunosuppressive mechanisms exploited by cancers. Studies testing different tumor antigens have revealed target properties that yield high tumor versus normal cell specificity and adequate immunogenicity to affect clinical efficacy. A few tumor-associated antigens, normal host proteins that are abnormally expressed in cancer cells, have been demonstrated to serve as good targets for immunotherapies, although many do not possess the needed specificity or immunogenicity. Neoantigens, which arise from mutated proteins in cancer cells, are truly cancer-specific and can be highly immunogenic, though the vast majority are unique to each patient's cancer and thus require development of personalized therapies. Lessons from previous cancer vaccine expeditions are teaching us the type and magnitude of immune responses needed, as well as vaccine technologies that can achieve these responses. For example, we are learning which vaccine approaches elicit the potent, balanced, and durable CD4 plus CD8 T cell expansion necessary for clinical efficacy. Exploration of interactions between the immune system and cancer has elucidated the adaptations that enable cancer cells to suppress and evade immune attack. This has led to breakthroughs in the development of new drugs, and, subsequently, to opportunities to combine these with cancer vaccines and dramatically increase patient responses. Here we review this recent progress, highlighting key steps that are bringing the promise of therapeutic cancer vaccines within reach.

Fig. 1

Therapeutic cancer vaccine target types. Targets for tumor vaccines fall into two general classes: tumor-associated antigens (TAAs) and tumor-specific antigens (TSAs). TAAs are self-antigens that are either preferentially or abnormally expressed in tumor cells, but may be expressed at some level in normal cells as well. As self-antigens, T cells that bind with high affinity to TAAs are typically deleted from the immune repertoire by central and peripheral tolerance mechanisms, and thus a cancer vaccine using these antigens must be potent enough to “break tolerance.” TSAs, comprised of antigens expressed by oncoviruses and neoantigens encoded by cancer mutations, are truly tumor-specific and as such high-affinity T cells may be present and strongly activated by these antigens. Although individual oncoviral antigens are expressed in specific tumor types (e.g., the HPV E6 and E7 antigens in cervical cancer), this occurs in many patients. Similarly, neoantigens encoded by oncogenic driver mutations may be prevalent across patients and tumor types, and hence are referred to as shared neoantigens. The majority of neoantigens are unique to individual patients’ tumors (private neoantigens), and thus require generation of a personalized therapy

Fig. 2

Mechanism of T cell activation and cancer cell killing. Activation of cytotoxic T cells (CTLs) depends on three signals: T cell receptor (TCR) engagement (signal 1), co-stimulation (signal 2), and an inflammatory stimulus (signal 3) via cytokines. T cell priming is initiated in tumor draining lymph nodes by specific binding of a TCR to its cognate peptide-major histocompatibility (MHC) complex displayed on an antigen-presenting cell (APC), particularly dendritic cells (DCs). This triggers a signaling cascade from the TCR complex that ultimately can regulate nuclear gene expression programs that transforms the T cell from a resting state to a state of activation and proliferation. Signal 1 alone, however, is insufficient for full activation, and the T cell must receive co-stimulation from the APC. Important co-stimulatory signals events include binding of the CD80 and CD86 ligands on APCs to the T cell CD28 receptor, and the binding of the OX40 and 4-1BB ligands to their receptors. TCR activation in the absence of co-stimulation can lead to T-cell anergy. In addition, co-stimulator ligands are depressed in tumors, and this can be overcome both by adjuvants that activate pattern recognition receptors on APCs, which upregulate expression of co-stimulatory ligands, and by antibodies that agonize the co-stimulatory receptors on T cells. Tumor cells also overexpress co-inhibitors, including CTLA-4 and PD-1, which normally function as T cell checkpoints to deactivate T cell activation after an infection is cleared. Antagonist antibodies have been developed to overcome this suppressive mechanism, and have demonstrated good clinical efficacy in some cancer types. Cytokines, including type I interferons and interleukin (IL)-12, provide the third necessary activation signal, and support the stimulation the expansion and differentiation of CD8 T cells into effector and memory CTLs. In addition, CD4 T_H1 helper T cells can significantly amplify and sustain CTLs, primarily by supplying IL-2, whereas various other cells, including CD4 T_reg cells, myeloid-derived suppressor cells (MDSCs), and M2-type tumor-associated macrophages (TAMs) can significantly dampen CTL activation and function. T cells must also migrate to and infiltrate the tumor, and tumors employ numerous countermeasures to block this. These include physical barriers created by abnormal vasculature and stromal cell build up, as well as disruption of chemokines that guide T cells to the tumor