Ipilimumab in prostate cancer

Abstract

Introduction:

Immune checkpoint inhibitors, such as ipilimumab, are a new class of immunotherapeutic agents that have shown significant efficacy in melanoma. A number of ongoing clinical trials are investigating the role of ipilimumab in prostate cancer, either alone or in combination with immunomodulating agents such as radiation and chemotherapy, and in combination with cancer vaccines.

Areas covered:

This article reviews the molecular basis, preclinical and clinical evidence on the safety and efficacy of ipilimumab in prostate cancer. Medical literature search using MEDLINE and online abstracts database of national meetings form the basis of this article.

A number of preliminary clinical studies suggest the potential therapeutic utility of immune checkpoint inhibitors such as ipilimumab in prostate cancer. Pending the results of large-scale studies, the rationale of combining ipilimumab with standard anticancer therapeutics such as radiation, cytotoxic chemotherapy and other immunotherapeutic agents can be of great value in reducing mortality and morbidity in prostate cancer.

1. Introduction

The scientific community’s painstaking efforts to break the immune system’s tolerance to tumors have finally borne fruit. The first breakthrough came in the form of sipuleucel-T (PROVENGE®, Dendreon Corp.), which in 2010 became the first therapeutic vaccine approved by US Food and Drug Administration (FDA) for the treatment of prostate cancer. To create the vaccine, a patient’s autologous peripheral blood mononuclear cells are isolated and co-cultured ex vivo for 36 to 44 h with a fusion protein of prostatic acid phosphatase and granulocyte-macrophage colony-stimulating factor (GM-CSF), then reinfused back into the patient. The efficacy of this innovative approach was established in the Phase III IMPACT trial, which showed a 22% relative reduction in risk of death (HR [hazard ratio] = 0.78; 95% confidence interval [CI], 0.61 – 0.98; p = 0.03) in patients with asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer (mCRPC) [1]. An alternative strategy currently being studied in a Phase III trial involves a prime-boost regimen using the prostate-specific antigen (PSA)-expressing PROSTVAC-VF vaccine [2]. PROSTVAC-VF (PSA-TRICOM, BN ImmunoTherapeutics) is a poxvirus-based vaccine engineered to contain PSA and three immune costimulatory molecules (B7.1, ICAM-1 and LFA-3) within a vaccinia virus prime and fowlpox virus vector boost. The vaccine has been tested in several Phase II trials. In one such trial involving 125 patients with asymptomatic or minimally symptomatic mCRPC, PROSTVAC-VF showed an improved 3-year survival rate of 30 vs 17% and a median survival of 25.1 vs 16.6 months compared with an empty vector [3,4].

Overall, prostate cancer is the second most common cause of death from cancer in men in the United States (28,170 estimated deaths in 2012) [5,6]. A number of anticancer immunotherapy strategies have shown preclinical and clinical efficacy either as monotherapies or, more recently, in combination strategies (Table 1). However, once prostate cancer has metastasized, it can be only transiently controlled by androgen-deprivation therapy (ADT), and the 5-year relative survival rate is 27.8%, fueling an urgent search for more effective, durable treatments for mCRPC. With the proven success of ipilimumab in melanoma, a number of completed and ongoing clinical studies are investigating the potential role of immune checkpoint inhibitors in prostate cancer therapies.

Table 1.

Immunotherapeutic antitumor strategies used alone or in combination.

Strategies	Example	Refs.
Immunostimulants	IL-2, IL-15, IL-7, GM-CSF, IFN	[71–76]
Therapeutic vaccines	Peptides/proteins, recombinant vectors, tumor cells, APCs	[77]
Immune checkpoint inhibitors	Anti-CTLA-4 antibodies, anti-PD-1 and anti-PD-L-1 antibodies, anti-LAG3 antibodies, anti-TIM3 antibody, anti-B7-H3 and anti-B7-H4 antibodies	[8–11,77]
Miscellaneous immunomodulators and drug-immune conjugates	Costimulatory receptor agonists (CD134, glucocorticoid-induced TNF receptor-related protein), monoclonal anti-CD25-diptheria toxin, monoclonal anti-TGF-β, IL-2 and diphtheria toxin conjugate, NHS-IL-12, hu14.18-IL-2, radiation, cytotoxic chemotherapy, small molecule inhibitors	[43,78–82]
Adoptive cellular therapy	Autologous TILs, allogeneic and autologous peripheral T cells, gene-modified T cells (chimeric TCR), NK cells	[83–85]

APC: Antigen-presenting cell; CTLA-4: Cytotoxic T lymphocyte-associated antibody 4; IL: Interleukin; PD-1: Programmed death 1; TCR: T-cell receptor; TIL: Tumor-infiltrating lymphocyte.

Ipilimumab (Yervoy®; Bristol-Myers Squibb), a fully human IgG1κ monoclonal antibody (mAb) that targets cytotoxic T lymphocyte-associated antigen-4 (CTLA-4), see Drug summary Box 1, is the first in a class of therapies targeting T-cell activation and regulation to be licensed in the broad category of agents known as immune checkpoint blockers/inhibitors. It has an approximate molecular weight of 148 kDa and is produced in mammalian (Chinese hamster ovary) cell culture [7]. Tremelimumab (Pfizer) is another fully human IgG2 mAb that targets CTLA-4 and blocks its functionality. As a group, the immune checkpoint inhibitors block diverse pathways involved in moderating an evolving antitumor immune response, thereby abrogating tolerance. CTLA-4 is one of the best studied molecules that acts as a check on specific effector and regulatory immune cells [8]. Programmed cell death protein-1 (PD-1) and its ligand PD-ligand-1 (PD-L-1), a different set of immune checkpoint molecules, have also been studied extensively. Anti-PD-1 and anti-PD-L-1 treatments have been shown to have antitumor activity in solid tumors [9,10]. Novel agents that target other checkpoint molecules, such as lymphocyte activation gene 3 (LAG3) and B7-H3, are in clinical trials, while B7-H4 and T-cell membrane protein 3 (TIM3) inhibitors are in preclinical development [8,11].

Box 1. Drug summary.

Drug name	Ipilimumab
Phase	Phase III clinical trials
Indication	Prostate cancer
Pharmacology description	CTLA-4 inhibitor
Route of administration	Injectable, in solution
Chemical structure	Fully human mAb
	Fully human immunoglobulin (IgG1κ) consisting of four polypeptide chains; two identical heavy chains primarily consisting of 447 amino acids each with two identical kappa light chains consisting of 215 amino acids each linked through interchain disulfide bonds
Pivotal trial(s)	[60,97]
Pharmaprojects – copyright to Citeline Drug Intelligence (an Informa business). Readers are referred to Pipeline (http://informa-pipeline.citeline.com) and Citeline (http://informa.citeline.com).

2. Molecular function of CTLA-4

In order to initiate an adaptive response, immune T cells require two separate signals from an antigen-presenting cell (APC). The first signal involves the interaction of a T-cell receptor (TCR) with an antigenic peptide–MHC ligand complex. The best characterized costimulatory signal involves the interaction of B7 with CD28. The combination of the two signals determines subsequent effector cell activation or anergy. There are also naturally occurring CD4⁺CD25⁺ regulatory T cells (Tregs) that specifically express the transcription factor forkhead box P3 (FOXP3) and are essential for maintaining immunological self-tolerance and homeostasis [12], the latter of which provides checks and balances between tolerance and autoimmunity.

CTLA-4 is a CD28 protein homologue that participates in peripheral tolerance, along with FOXP3 and transforming growth factor-β (TGF-β) [13–16]. Despite their structural similarity and shared genes on human chromosome 2q33, there are a number of differences between CTLA-4 and CD28. CTLA-4 and CD28 share the same B7 ligands (CD80 and CD86), but CTLA-4 can bind the ligands with higher affinity and avidity and is thought to act as a competitive antagonist for CD28 [17]. CD28 is constitutively expressed on the plasma membrane of both resting and activated naïve T cells, whereas CTLA-4 is not expressed by resting naïve T cells, but becomes detectable 2 days after activation [18]. TCR stimulation leads to the trafficking of CTLA-4 from the Golgi apparatus to the plasma membrane [19]. However, ongoing constitutive endocytosis keeps its expression level very low compared to CD28 [20].

The role of CTLA-4 as a regulator of immune tolerance and autoimmunity was described in preclinical studies for the first time in the mid-1990s. Investigators noted that CTLA-4 knockout mice died just 3 weeks after birth from massive post-thymic lymphoproliferation and autoimmunity [16,21,22]. Subsequently, it was shown that CTLA-4 blockade with antagonistic mAbs broke immune tolerance, resulting in effective antitumor immunity and autoimmunity [23]. Several pivotal observations followed the discovery of CTLA-4; however, the exact role of the CTLA-4 receptor in negative immune regulation is still the subject of considerable debate. The finding that CTLA-4 expression is controlled by both intracellular trafficking and transcription, the discovery of three different isoforms and the observation that the binding of CTLA-4 to its ligands on effector T cells initiates extrinsic cell signals have given rise to a multitude of models describing the role of CTLA-4 in the immune response [24]. CTLA-4 exerts negative regulation on CD28 costimulation through the upregulation and cell-autonomous engagement of CTLA-4 on activated effector T cells that leads to downregulation of IL-2 production, thereby terminating the immune response [25]. On the other hand, CTLA-4 can inhibit T cells by involving others cells (the extrinsic model). The extrinsic path involves release of immune-inhibitory cytokines by the CTLA-4-bearing cell and several pathways of B7 ligand sequestration (Figure 1).

Figure 1. Hypothesized pathways of CTLA-4-related immunosuppression.

CTLA-4 can act intrinsically on the T cell that expresses it. Proposed intrinsic pathways may involve downstream inhibitory pathways or competitive inhibition of the positive pathway mediated by CD28. Alternatively, CTLA-4 may act via involvement of other cells, such as APCs, or processes outside of the T cell that expresses it. Extrinsic pathways include inhibitory cytokine release by immune cells. There are multiple theories on how CTLA-4 aids in extrinsic sequestration of the B7 ligand, thereby preventing its positive function. Data presented at the AACR Tumor Immunology: Multidisciplinary Science Driving Basic and Clinical Advances meeting in Miami last week. Recent data (unpublished) from Allison’s lab suggests that anti-CTLA4 antibodies may also deplete intratumoral T-regs by antibody dependant cellular cytotoxicity.

The presence of CTLA-4 has also been described on CD4⁺CD25⁺FOXP3⁺ Tregs [24]. The CTLA-4 molecule seems to be essential to Tregs’ suppressive function, where it is constitutively expressed [26]. These observations are not definitive proof of the role of CTLA-4 in Treg suppressive function; in fact, contradictory evidence exists in both murine and human models [27–30].

Anti-CTLA-4 antibody may also enhance the avidity of cytotoxic T lymphocytes (CTLs), thus making them more effective [31,32]. To test this hypothesis, investigators determined an optimum dosing schedule for combining an anti-CTLA-4 mAb and with a viral vector consisting of live recombinant vaccinia (rV) [33]. This poxviral-based vaccine was designed to express the transgene for carcinoembryonic antigen (CEA) plus TRICOM (rV-CEA/TRICOM). Tetramer dissociation and cytotoxicity assays measured T-cell avidity. There was a profound difference in tetramer dissociation and a 10-fold difference in T-cell functional avidity in mice receiving both rV-CEA/TRICOM and anti-CTLA-4 compared to mice treated with vaccine alone. These findings raised the possibility that CTLA-4 blockade could have a potentiating effect on T-cell avidity against tumors.

3. Ipilimumab in prostate cancer: preclinical studies

The earliest evidence that ipilimumab showed promise in antitumor immunotherapy was published in 1994. Using murine fibrosarcoma Sal N and murine colon adenocarcinoma 51BLimlO cells in mice, Leach et al. demonstrated for the first time the permissive role of CTLA-4 antibodies in the rejection of established tumors [23]. Tumor rejection was also associated with immunity to tumor rechallenge.

In preclinical studies with a transplantable prostate cancer cell line (pTC1) derived from TRAMP (transgenic adenocarcinoma of the mouse prostate) mice, Kwon et al. modified pTC1 to express the murine B7.1 costimulatory ligand [34]. Their results showed that this led to tumor rejection in syngeneic mice, proving that the rejection was T-cell-mediated. They showed that in vivo antibody-mediated blockade of CTLA-4 led to wild-type (unmodified) tumor rejection.

Following radiation, chemotherapy, cryoablation or immunotherapy, dead or dying tumor cells can release an ‘antigen cascade’. Through cross-presentation, these peripherally released antigens may gain entry into the MHC class I pathway, where APCs can present them to CD8⁺ T cells as a priming step in the immune response (Figure 2) [35–37]. Cross-presentation may act synergistically with immune-based therapies, augmenting their antitumor effect. Using poorly immunogenic metastatic mouse mammary carcinoma 4T1, Demaria et al. showed that combination therapy with anti-CTLA-4 antibody and radiation was associated with a significant survival advantage compared to control mice [37]. The improved survival correlated with inhibition of lung metastases and required CD8⁺ but not CD4⁺ T cells.

Figure 2. Simplified model of cross-presentation and CTL stimulation by multiple modalities.

By killing and/or modifying tumor cells, ionizing radiation and cytotoxic chemotherapy may induce an antigen cascade that releases multiple MHC-1-restricted tumor-associated antigens. APCs such as dendritic cells, when exposed to these antigens and appropriate maturation signals, cross-present the antigens to CD4 T cells (not shown) and CD8 CTLs. The resulting cytotoxic immune response targets modulated and nonmodulated tumor cells not killed by the primary treatment modality, setting off a dynamic immune response late into therapy.

The combined effect of CTLA-4 blockade and tumor cryoablation demonstrated synergy in a transplantable TRAMP C2 prostate cancer mouse model [38]. TRAMP C2 tumor cells were injected on the left flank of male B6 mice. The mice were treated with tumor cryoablation and/or anti-CTLA-4 antibody when tumors reached 5 to 8 mm in diameter, about 28 days after tumor cells were injected. As evidence of synergy, at tumor rechallenge, there was a rejection rate of 44% in mice treated with cryoablation and anti-CTLA-4 antibody alone, compared to no rejection in controls. Preclinical studies in mice bearing non-prostatic, poorly immunogenic tumors also support the use of ipilimumab in novel combinatorial approaches in humans [39,40].

Hurwitz et al. examined the combined effect of whole tumor cell vaccine and anti-CTLA-4 antibody in 14- to 16-week-old male TRAMP mice that spontaneously develop prostatic adenocarcinoma as a consequence of expression of the SV40 TAg oncogene [41]. They found that only the combination therapy resulted in a significant reduction in tumors.

There is emerging evidence that ADT and cytotoxic chemotherapy can enhance antitumor immunity by acting directly on the immune system or by immunogenic modulation [42–44]. The biological mechanism may involve alteration of subpopulations of CD4⁺ and CD8⁺ cells, inhibition of Tregs, increased thymic output of T cells, increased intratumoral trafficking of effector cells and decreased tolerance of tumor cells [45–50]. A chemotherapeutic agent such as docetaxel can upregulate one or more surface molecules (Fas, ICAM-1, MUC-1, CEA and MHC class I) in both sensitive and resistant human carcinoma cell lines [42,43]. Radiation can have a similar effect on the phenotype of tumor cells, as demonstrated by clinical trials in prostate cancer that showed clinical evidence of epitope expansion and antitumor immunity [51,52].

4. Anti-CTLA-4 antibody in prostate cancer: clinical evidence

The success of ipilimumab in advanced melanoma is highlighted by its approval by the FDA in 2010. A unique set of toxicities referred to as immune-related adverse events (irAEs) has been seen with the use of anti-CTLA-4 antibodies. These irAEs include skin and mucosal rash (47 – 68%), diarrhea/colitis (all grades, 44%), hepatotoxicity (3 – 9%), hypophysitis (1 – 6%), episcleritis/uveitis (< 1%), immune-related pancreatitis (< 1.5%), lymphadenopathy/sarcoid-like syndrome (anecdotal) and transient peripheral neuropathies (< 1%) [53]. The events are typically dose-dependent and dermatological toxicities tend to occur before the visceral manifestations. It is mechanism-based as the CTLA-4 blockade leads to a compromised immune tolerance to selected normal tissue antigens [54]. Discontinuation of immune checkpoint inhibitor and immunosuppressive medications is standard treatment for irAEs [53]. Neither our experience nor published reports show a consistent correlation between irAEs and dose or tumor response. Risk evaluation and mitigation strategy (REMS) for ipilimumab, prepared by Bristol-Myers Squibb in conjunction with FDA, should be referred to in managing patients with moderate to severe irAEs. REMS consists of a communication plan to inform healthcare professionals of the serious risks, early identification and an overview of recommended management.

A pilot study with safety as the primary endpoint used a single 3 mg/kg i.v. dose of ipilimumab as monotherapy in 14 patients with mCRPC [55]. Overall, ipilimumab was well tolerated, with attributable irAEs noted in one patient who developed grade 2 pruritus and grade 3 rash that responded to steroids. Of the 14 patients, 2 had a decline in PSA of > 50% that lasted 135 and 60 days, respectively; 8 of 14 patients had a < 50% decline in PSA. Two patients with bidimensionally measurable disease who had baseline and follow-up imaging showed no radiographic objective response. Since this pilot trial was completed, more than 10 clinical trials exploring the role of ipilimumab in prostate cancer, including two Phase III trials, have been completed or are ongoing (Tables 2 and 3). The two ongoing Phase III studies will address overall survival (OS) as the primary endpoints. One Phase III, randomized, placebo-controlled trial (clinical trials.gov identifier: NCT00861614) that is open to recruitment aims to enroll 800 patients with mCRPC who have been exposed to docetaxel [56]. Ipilimumab at 10 mg/kg will be used as induction–maintenance strategy in the treatment group following radiation therapy. The second randomized, placebo-controlled clinical trial (clinical trials.gov identifier: NCT01057810) will look into the survival impact of ipilimumab in patients with asymptomatic or minimally symptomatic chemo-naïve mCRPC [57]. The projected enrollment for the trial is 600 and a 10 mg/kg dose of ipilimumab will be employed in an induction–maintenance strategy.

Table 2.

Summary of published trials employing ipilimumab in prostate cancer.

Study agent	Phase	Disease	(n)	Endpoints	Results	irAEs (≥ grade 3)	Refs.
Ipilimumab	I	mCRPC	14	Safety, PSA response	≥ 50% PSA declines in 2/14 pts	1	[55]
Concurrent ipilimumab + GVAX	I	Chemo-naïve mCRPC	28	Primary: MTD and safety Secondary: PSA response, TTP, immune response, survival	50% greater decline in PSA in 7 (25%) pts	9	[64,86]
Concurrent ipilimumab + PSA-TRICOM + GM-CSF	I	mCRPC	30 (80% chemo-naïve)	Primary: safety Secondary: PSA response, OS, immune response	PSA decline in 14/24 (58%) chemo-naïve pts; 6 with > 50% PSA decline. Median Halabi-predicted OS: 18.5 months Actual median OS: 31.8 months with 74% survival probability at 24 months	12	[63,87]
Concurrent Ipilimumab + GM-CSF	I	mCRPC	36	Primary: safety Secondary: PSA response, radiographic response	≥ 50% PSA declines in 3/6 pts at 3 mg/kg and 1/6 pts at 10 mg/kg	10 (all grade 3)	[61,87]
Ipilimumab ± radiation	I/II	mCRPC	50	Safety, PSA response, radiographic response	Decline in baseline PSA of ≥ 50% in 8/50 evaluable pts in the ipilimumab 10 mg/kg ± radiation group	Colitis (16%), diarrhea (8%), hepatitis (10%)	[58]
Ipilimumab vs ipilimumab + docetaxel	II	mCRPC	43 chemo-naïve	Safety, PSA response	PSA declines of ≥ 50% in total of 3 pts in both arms	5/52 SAEs in 18 pts were irAEs (grade not reported)	[59,88]
Ipilimumab + ADT vs ipilimumab	II	Advanced CaP	108	Safety, PSA and clinical response	Ipilimumab + ADT group showed more undetectable PSA by 3 months (55 vs 38%)	Ipilimumab + ADT: colitis 4.5%; diarrhea 4.5%	[61,90]

ADT: Androgen-deprivation therapy; MTD: Maximum tolerated dose; OS: Overall survival; PSA: Prostate-specific antigen; pts: Patients; SAE: Severe adverse event; TTP: Time to progression.

Table 3.

Summary of unpublished clinical trials employing ipilimumab in prostate cancer.

Study drugs	Phase	Eligibility	Estimated enrollment	Primary outcome	Refs.
Ipilimumab + ADT	II	Metastatic castration-sensitive prostate cancer	48	PSA response	[91]
Neoadjuvant ipilimumab + ADT	II	High-risk prostate cancer; undergoing prostatectomy	20	Local and systemic immune variables	[92]
Ipilimumab alone vs ipilimumab + GM-CSF	II	mCRPC	54	PSA response	[93]
Ipilimumab + ADT	II	mCRPC	30	PSA response	[94]
Ipilimumab	I/II	mCRPC	66	Safety	[95]
Ipilimumab + abiraterone acetate	I/II	mCRPC	25	Safety	[96]
Ipilimumab vs placebo	III	Chemo-naïve mCRPC	600	OS	[57]
Ipilimumab vs placebo following radiation	III	Chemo-treated mCRPC	800	OS	[56]

ADT: Androgen-deprivation therapy; mCRPC: Metastatic castration-resistant prostate cancer; PSA: Prostate-specific antigen.

Numerous clinical trials have suggested the effectiveness of combining multiple modalities to obtain synergy with anti-CTLA-4 mAbs. The results of a Phase I/II clinical trial of ipilimumab and radiation in mCRPC (in chemo naïve subjects and those with prior docetaxel) have been reported [58]. Patients (n = 33) received dose-escalated ipilimumab with or without a single dose of radiation (8 Gy/lesion, up to three lesions per patient). Later, the cohort receiving 10 mg/kg of ipilimumab was expanded to 50 patients, 34 of whom received radiation. Of the 50 PSA-evaluable patients in the 10 mg/kg ± radiation group, 8 patients had a PSA decline of ≥ 50%.

A randomized Phase II trial of 3 mg/kg ipilimumab with or without concurrent docetaxel in mCRPC showed a confirmed PSA decline of ≥ 50% in 3 of 43 patients, with no objective responses [59]. Of 52 severe adverse events, 5 (reported in 3 patients) were irAEs. In this study, the 6% rate of irAEs was less than the 10 to 15% rate of grade ¾ irAEs noted in a Phase III study of ipilimumab (3 mg/kg) in melanoma [60].

The combination of ipilimumab and ADT has also shown clinical activity. A Phase II trial randomized patients newly diagnosed with locally advanced prostate cancer to receive ADT with or without a single dose of ipilimumab (3 mg/kg). At 3 months, 55% of patients in the combination group had undetectable PSA as against 38% in patients receiving ADT alone [61].

A Phase I dose-escalation trial of another anti-CTLA-4 antibody, tremelimumab, in combination with androgen receptor blockade (high-dose bicalutamide) in patients (n = 11) with biochemically recurrent prostate cancer was recently reported [62]. With safety as the endpoint, the trial reported three grade 3 irAEs: two patients receiving 6 mg/kg tremelimumab had diarrhea and rash; one patient receiving 3 mg/kg tremelimumab had colitis. In a follow-up period of 12 to 18 months, two patients had asymptomatic disease progression, and three reported improvements in PSA doubling time of 23.4 to 41.8 months (baseline for the group was 6.3 months; range: 2.7 – 11.3).

Ipilimumab has also been clinically tested in combination with immune adjuvants such as GM-CSF. A Phase I trial combined GM-CSF with escalating doses (0.5 – 10 mg/kg) of i.v. ipilimumab on day 1 of each cycle × 6 cycles in 36 patients with mCRPC [63]. The combination was well tolerated, with 10 reporting irAEs (rash, panhypopituitarism, temporal arteritis and diarrhea). Notably, 3 of 6 patients treated with 3 mg/kg had PSA declines of ≥ 50%.

CTLA-4 blockade increases T-cell avidity, raising the possibility of synergy with a therapeutic cancer vaccine [32]. This strategy was tested in a Phase I study that combined GVAX with escalating doses of ipilimumab (0.3 – 5 mg/kg) in patients with mCRPC [64]. A total of 12 patients were enrolled in the dose-escalation cohort. At the 3 mg/kg dose level, three patients had grade 2 hypophysitis; two had grade 3 hypophysitis; three had colitis (one grade 1 and two grade 2) and one had grade 3 hepatitis. At the 5 mg/kg dose level, two patients had grade 3 hypophysitis and one patient had grade 4 sarcoid alveolitis (a dose-limiting toxicity). Seven patients (25%) who received either 3 or 5 mg/kg ipilimumab had PSA declines of ≥ 50%. A strong association between irAEs and PSA response was noted in the dose-escalation cohort, but not in the expansion cohort. Two patients in the dose-escalation cohort showed a clear regression of bone metastases. Complete regression of abdominal lymphadenopathy was seen in one of four patients with measurable disease at baseline. Stable bone metastases lasting 3 to 27 months were seen in 15 patients in the escalation and expansion cohorts.

In a trial involving patients with mCRPC, PSA-TRICOM was tested in combination with an initial six planned cycles of ipilimumab in escalating doses of 1 to 10 mg/kg and expansion cohorts of 3 and 5 mg/kg [65]. Of the 30 patients enrolled (80% chemotherapy naïve), 14 patients discontinued ipilimumab because of disease progression, while no dose-limiting toxicities were reported. Grade 3 or grade 4 irAEs were identified at all doses above 1 mg/kg, including diarrhea or colitis (n = 4), grade 3 rash (n = 2), grade 3 raised aminotransferases (n = 2), grade 3 endocrine irAEs (n = 2) and grade 4 neutropenia (n = 1). Of the 24 chemotherapy-naïve patients, 14 (58%) had PSA declines, 6 (25%) of which were > 50% (2 of these 6 declines were > 90%). Of 12 patients, 3 with measurable disease had unconfirmed partial responses on CT imaging. The actual median OS for all patients was 34.4 months (95% CI: 29.6 – > 41), with a 2-year OS of 73% (95% CI: 55.6 – 85.8). In the two combination trials involving ipilimumab, the median OS of 29.2 months in combination with GVAX (95% CI: 9.6 – 48.8) and 34.4 months in combination with PSA-TRICOM (95% CI: 29.6 – > 41) suggests the promise of improved outcomes over vaccine alone in mCRPC [64,65]. Together, these trials suggest that, when used in combination, vaccine does not enhance the toxicity of anti-CTLA-4 antibody. Preliminary evidence of efficacy for this combination strategy needs to be confirmed in large randomized studies.

Ongoing experience, in general with immunotherapeutic agents, has revealed a unique characteristic of clinical response as opposed to the standard cytoreductive therapies [66–68]. Vaccines such as sipuleucel-T and PROSTVAC-VF generate an immune response that grows significantly over time. Due to delayed response, the resultant antitumor effect is not amenable to prediction with standard PSA kinetics model and PFS is relatively unaffected compared to the OS as randomized data show [60,69]. Ipilimumab, in metastatic melanoma patients, demonstrated an OS advantage without a change in disease progression. Decreased reliance on PSA for response (as suggested by most recent Prostate Cancer Clinical Trials Working Group criteria) [70], development of new immune biomarkers and their validation are thus paramount in future trials involving immunotherapeutic agents.

5. Expert opinion

The traditional view that prostate cancer is poorly immunogenic has been challenged by the approval of sipuleucel-T and by preliminary studies of vaccines and immune checkpoint inhibitors. Recent studies have shown acceptable safety for the combination of vaccine and ipilimumab, as well as encouraging clinical outcomes (25% of patients showed PSA declines of ≥ 50% in vaccine combination studies) [64,65]. Larger combination studies will be required to confirm this observed impact on OS.

Given the encouraging safety data on novel immune checkpoint inhibitors, such as anti-PD-1, combination studies of anti-PD-1 with vaccines such as PSA-TRICOM or sipuleucel-T may be the next logical step [3]. The rationale for combination studies with immune checkpoint inhibitors could also be extended to include combinations with standard anticancer agents, such as chemotherapy, radiation and small molecular inhibitors. Optimum sequencing of the various treatment modalities would need to be determined. For patients with bone metastases, once radiation has exposed endogenous cancer-specific antigens by inducing tumor necrosis and expansion of the T-cell repertoire through crosstalk, an immune checkpoint inhibitor may help to sustain or even enhance the antitumor immune response. The efficacy of ipilimumab alone or in combinatorial testing remains to be tested in large-scale setting and the results of the two ongoing Phase III trials [56,57] are eagerly awaited.