As discussed above, adaptive immune responses are primarily mediated by dendritic cells (DC), and a cancer vaccine, like any other vaccine, would be expected to depend on the presence of a population of host DC that are numerically and functionally intact. Unfortunately, this is often not the case in cancer patients, in which DC are often dysfunctional [16]. One approach to overcoming this dysfunction, then, might be to generate new autologous DC outside of the patient’s tolerogenic environment. This is indeed the approach involved in the generation of Sipuleucel-T, which was the first cancer vaccine approved by the FDA for the treatment (rather than prevention) of cancer. The product is individually manufactured for each patient with PC, in a process which comprises multiple steps (Fig. 9.1a). Briefly, patients undergo a standard leukopheresis procedure and peripheral blood mononuclear cells (PBMC) are separated and then incubated with PAP2024, a fusion protein that links the antigen PAP to the granulocyte-macrophage colony-stimulating factor (GM-CSF). After approximately 36 h of incubation, cells are washed and suspended in Lactate Ringer Injection, for infusion back into the patient. In this approach, the GM-CSF serves as the adjuvant that helps to activate dendritic cells and other cells in the infusion product. The process is repeated three times at 2-week intervals [17].

After a series of phase I and II trials, two relatively small randomized, double-blind, placebo-controlled phase III trials (D9901 and D9902A) showed evidence for the clinical activity of Sipuleucel-T [18]. Both studies enrolled men with asymptomatic or minimally symptomatic mCRPC, and both suggested clinical activity via an increased survival. Thus, the multicenter, randomized, double-blind, placebo-controlled IMPACT trial was designed with primary endpoint of OS. trial trail enrolled 512 patients with asymptomatic or minimally symptomatic mCRPC randomized 2:1 to receive Sipuleucel-T or placebo. Resulting data showed that treatment with Sipuleucel-T was increased median OS by 4.1 months (25.8 versus 21.7 months; p = 0.032, HR = 0.78) [19]. As in the prior phase III studies, median time to progression was not different between groups, and radiographic responses were not generally observed. Sipuleucel-T was generally well tolerated, and 92 % of patients received all three infusions. The most common adverse events (AEs) in the Sipuleucel-T group included chills (in 51.2 %), fever (22.5 %), fatigue (16.0 %), nausea (14.2 %), and headache (10.7 %); all these were graded as mild or moderate. Adverse events of grade 3 or more within 1 day after infusion were rare, and were reported in 23 of 338 patients (6.8 %) in the Sipuleucel-T group and 3 of 168 patients (1.8 %) in the placebo group.

The final Sipuleucel-T product is heterogeneous, comprising mature antigen-presenting cells (APC) and other cell types, including T cells, B cells, and natural killer cells [17]. Once infused, the autologous ex vivo-activated APC are thought to prime PAP-specific CD4+ and CD8+ T cells in a manner similar to a classical vaccine-mediated prime-boost regimen, where the first infusion primes the immune system and subsequent infusions boost the response [20]. Recently, a combined analysis of immunological data from the previously cited phase III trials (D9901, D9902A, and IMPACT) was completed [21]. The authors demonstrated that APC activation in the infused product occurred with the initial dose, and increased with subsequent doses. The median cumulative APC activation with Sipuleucel-T across the three dose preparations was 26.7 [95 % CI 21.5–33.6]. Furthermore, these data demonstrated antigen-specific T-cell proliferation and IFN-γ ELISPOT activity in pre-culture cells obtained at weeks 2 and 4 (but not week 0). Finally, T-cell activation-associated cytokines were noted in the second and third doses of Sipuleucel-T, supporting the idea that the first infusion of activated, antigen-loaded APCs primes T cells in vivo. Thus, the second and third doses are of Sipuleucel-T are biologically different from the first, and each dose contains progressively more activated APCs and possibly a greater proportion of antigen-specific T-cells with the capacity to recognize and kill PC cells. Based on these studies, antigen-specific immune responses, in the form of either PAP-specific antibodies, or a T cell response to PAP were observed in 78.8 % of monitored subjects. Finally, the authors demonstrated a positive correlation between OS and cumulative APC activation and antigen-specific immune responses. Taken together these important analyses show that Sipuleucel-T induces an antigen-specific immune response, and that response appears to be associated with a survival benefit in treated patients.

ProstVac VF is a poxvirus-based vaccine directed against PSA (Bavarian Nordic, USA); this technology was developed over several years by a group at the NIH [22], and its current iteration includes a number of critical modifications to optimize immunogenicity. First, the vaccine involves a heterologous prime boost in which the initial vaccine is based on a modified vaccinia ankara (MVA) backbone, followed by a series of booster vaccines with a fowlpox backbone. This is because the immune response to the MVA backbone is quite robust, so boosting with an identical vaccine is limited by the host’s immune response to the viral backbone. To further increase immunogenicity, the vaccine was engineered to incorporate a triad of costimulatory molecules, including leukocyte function-associated antigen 3 (LFA-3), the T cell costimulatory molecule B7.1, and the adhesion molecule intercellular adhesion molecule 1 (ICAM-1) [23] (Fig. 9.1b). Additionally, administration of GM-CSF at the vaccination site is used to help recruit local dendritic cells (APC) and enhance antigen presentation.

GVAX prostate (Aduro Biotech, Berkeley, CA) is a cell-based immunotherapy comprised of a combination of irradiated allogeneic tumor cell lines modified with granulocyte-macrophage colony-stimulating factor (GM-CSF) gene [26] (Fig. 9.1c). The product includes two prostate carcinoma cell lines: the androgen-sensitive LNCaP, as well as the castration-resistant line PC3 [27]. The concept underlying whole-cell vaccines is that these cells will serve as a diverse source of multiple tumor and tissue specific antigens, at least some of which will correspond to antigens in the patient’s tumor. These vaccines are thus considered polyvalent. The primary advantage to such an approach is that it’s possible that the inclusion of multiple antigens might prevent tumors from escaping immune pressure by down-regulating the expression of a single tumor-associated antigen. The major disadvantage of GVAX vaccines is that they’re relatively difficult to monitor immunologically, since the key target antigen/antigens are not known for any particular patient.

Although the focus of this chapter is the management of metastatic castration-resistant prostate cancer (mCRPC), and this is the setting where the majority of clinical trials of cancer immunotherapy have been performed, from an immunological perspective, it is important to highlight that biochemical recurrence provides a unique opportunity for immunological intervention in patients with prostate cancer. In this setting the several of the immunosuppressive mechanisms (such as Treg cells, myeloid-derived suppressor cells (MDSCs) and transforming growth factor-β (TGFβ)) associated with an advanced tumor burden are expected to be at a minimum, and it is likely that immunotherapy approaches would be more beneficial [20]. But it must be noted that the single prostate cancer vaccine that is FDA approved (Sipuleucel-T) is approved only in the metastatic state, and that clinical development of prostate cancer drugs is somewhat hampered by the absence of clear markers of response in the non-metastatic setting.

Among men with mCRPC, then, which patients are most appropriate for immunotherapy with Sipuleucel-T, and perhaps more importantly, when in the treatment sequence should this agent be given? In terms of patient status, the enrollment criteria for all 3 phase III trials of this agent stipulated that men be either asymptomatic or minimally symptomatic, so it’s clear that men with advanced, symptomatic disease are most likely not appropriate candidates. In terms of sequencing, it is not well appreciated that approximately 20 % of the men treated in the phase III trial of Sipuleucel-T (IMPACT) had been treated with prior chemotherapy [19]. Subgroup analysis suggested a benefit in both chemotherapy-experienced and chemotherapy-naïve patients, so prior chemotherapy in and of itself is not an absolute contraindication to vaccine treatment, as long as patients are either asymptomatic or minimally symptomatic. Perhaps more relevant at the current time are the second-generation hormonal treatments such as abiraterone-acetate [32] and enzaluatmide [33]. These are well-tolerated agents, often administered in the pre-chemotherapy setting. Their sequencing with immunotherapy has not yet been well studied, and in the case of abiraterone acetate a particular concern arises from the notion that this agent is usually administered with a low dose of prednisone (5 mg BID). Recent data from a randomized phase II trial of Sipuleucel-T administered either concurrently (N = 30) or sequentially (N = 32) with Abiraterone acetate + prednisone (AA + P) suggested that the administration of AA + P did not appear to influence the parameters of Sipuleucel-T associated with potency (CD54 count), and that evidence for a prime-boost effect was still documented in the product in the presence of prednisone [34]. While clinical follow-up of these patients continues, the data provide important evidence that co-administering AA + P is not obviously contraindicated. Still, though, it’s not clear whether the first therapy for men with progressing mCRPC should be Sipuleucel-T or a second-generation hormonal agent. Clearly, large randomized clinical trials, following men for overall survival would be necessary to answer that question definitively. Those trials are not currently underway, and are in fact unlikely to be initiated. A second set of data, albeit retrospective, add a bit of additional insight. In a retrospective analysis of the large randomized phase III trial of Sipuleucel-T, patients were subgrouped based on initial PSA, and those data correlated with potential survival benefit [35]. In this study, patients with the lowest quartile of initial PSA (<22.2 ng/mL) appeared to enjoy the largest survival benefit from Sipuleucel-T (41.3 versus 28.3 months for placebo), while patients with the highest quartile of initial PSA (>134.1) appeared to achieve a less obvious survival benefit (18.4 months with Sipuleucel-T versus 15.6 months for placebo). So, taken together these data would suggest that the most appropriate patients for Sipuleucel-T treatment would be those with early stage disease, a lower PSA, and fewer symptoms.

As discussed above, T cells have a crucial role in anti-tumor immunity. Recent developments in basic immunology show that T cell function is physiologically modulated by the interaction of a series of cell surface proteins (called immune checkpoints) with their individual ligands [3, 41, 42]. In normal physiology, these molecules likely serve to prevent an over-exuberant immune response to infection from causing autoimmunity, but in the case of cancer these molecules likely attenuate and/or prevent an anti-tumor immune response. Perhaps the best studied of the checkpoint molecules is Cytotoxic T Lymphocyte Antigen-4 (CTLA-4), originally cloned as an analog of the costimulatory molecule CD28. While the function of CTLA-4 was controversial for some time [43], definitive studies using knockout mice were performed by a number of groups, with concordant results. Mice with germline deletion of CTLA-4 develop widespread lymphadenopathy and autoimmunity, and die within 21–28 days of age, attesting to a critical role for this molecule in attenuating a T-cell driven immune response [44, 45]. Subsequent studies by the Allison group showed that blocking CTLA-4 with a monoclonal antibody could potentiate anti-tumor immunity in an animal model [46]. A monoclonal anti-CTLA-4 antibody was developed clinically, and is now FDA-approved for patients with metastatic melanoma [47].