Enhancing immune responses to tumor-associated antigens

Abstract

The goal of vaccine-based cancer immunotherapy is to induce a tumor-specific immune response that ultimately reduces tumor burden. However, the immune system is often tolerant to antigens presented by the tumor, as the cancer originates from within a patient and is therefore recognized as self. This article reviews selected clinical strategies for overcoming this immune tolerance, and approaches to enhance generation of immunity to tumor-associated antigens by activating innate immunity, potentiating adaptive immunity, reducing immunosuppression, and enhancing tumor immunogenicity. Success in the field of cancer vaccines has yet to be fully realized, but intelligent choice of immunomodulators, tumor antigens and patient populations will likely lead to clinically relevant uses for cancer vaccines.

Introduction

Following the vision of Dr. William Coley, tumor immunologists have sought to develop cancer therapies by directing a patient’s own immune system to destroy tumor cells.¹ Advances in our understanding of genomics, proteomics and immunology have led to the clinical development of numerous cancer immunotherapies. Identification of tumor-associated antigens (TAAs) has facilitated the development of various cancer vaccination strategies to elicit tumor-specific immunity potentially capable of reducing tumor burden. However, strategies to induce tumor-specific immunity in patients must overcome several obstacles: (1) insufficient and poorly functional populations of antigen-presenting cells (APCs) and lymphocytes; (2) inducing immunity potent enough to overcome potential tolerance mechanisms without inducing unacceptable autoimmune toxicities; (3) poor immunogenicity of antigens expressed by tumor cells; and (4) immunoregulatory pathways that dampen the tumor-specific immune response. In numerous clinical trials to date, various strategies have been used in combination with cancer vaccine modalities in an effort to enhance the immune response to TAAs, including cytokine support, toll-like receptor (TLR) agonists, costimulation, chemotherapy, radiotherapy, ablative therapy, immunoregulatory modulation, adjuvants and vector-driven immunity.

Tumor-Associated Antigens

Many potential targets for cancer immunotherapy have been identified to date by a variety of methods (Table 1). Methods of identifying immunologically relevant tumor antigens include patient antibody responses, genomics, biochemical strategies, SEREX and reverse immunology.^2–6 Most TAAs that are over-expressed in malignant tissue are also expressed in normal tissue or originate as oncofetal antigens, and thus peripheral or thymic tolerance to TAAs often exists. It is important to note that when an antigen is targeted for cancer immunotherapy, the activated T-cells induced by vaccination recognize only the peptide-major histocompatibility complex (MHC) complexes of the tumor antigen on the cell surface, and not the surface protein. Thus, targets of vaccine therapy need not be cell-surface proteins. Additionally, a secondary, costimulatory signal is needed to fully activate adaptive T-cell responses. Identification of lipid antigens for natural killer T-cells and tumor stroma-associated antigens will further expand the repertoire of available tumor targets.^7–9 A current challenge is to determine which of these TAAs warrant further development.

Table 1.

Partial list of current and potential targets for cancer vaccine therapy

Carcinoma-associated antigens	Endogenous Retroviral gene products	Leukemia/lymphoma antigens	Melanoma antigens	Viral antigens
Brachyury	HERV-H	Aberrant class II	MAGE	HPV
CA125	HERV-K	Anti-idiotype	MART
CEA		B1	gp100
EGFR		CD19	Tyrosinase
HER-2/neu		CD20	CD2
KSA		CD22	CD3
Mesothelin		CD25	GM2
MUC-1		CD36
NY-ESO
p53
PAGE-4
PAP
PSA
PSCA
PSMA
Ras
sTn
TARP
VEGF

Approaches to Inducing Tumor-Specific Immunity

Several modalities that seek to reduce tumor burden by inducing tumor-specific immunity are currently under investigation in clinical trials. A review by Fournier and Schirrmacher recently described the state of the art in 2008 for cancer vaccines and autologous cell-based therapies (summarized in Table 2).¹⁰ Unfortunately, each of these classes of immunotherapies has had disappointing Phase III trial results. However, careful attention to clinical trial design, primary endpoint selection, patient population selection, combinational therapy approaches and vaccine delivery could potentially lead to successful treatment strategies for a variety of malignancies.

Table 2.

Partial list of current and potential cancer vaccine modalities

Tumor cells	Protein/molecule-based	Antigen-presenting cells	Vector-based
Autologous	Protein (HSP)	Unmodified	DNA/RNA
Allogeneic	Peptide	Apoptotic body loaded	Yeast
Irradiated	Glycopeptide	Gene modified
Oncolysates	Glycolipid mAb fusion proteins	Peptide pulsed RNA transfected	Bacterial: Salmonella
Gene modified:	Anti-idiotype mAb	Tumor lysate loaded	Listeria
GM-CSF		Tumor cell fusion	Lactobacillus
Costimulatory molecules
Lymphotactin			Viral: Vaccinia Adenovirus RNA replicons Avipox Fowlpox Canarypox MVA

Approaches to Enhancing Generation of TAA-Specific Immunity

TAAs are weakly immunogenic due to host immunotolerance, as well as the tumor’s immunosuppressive microenvironment and low expression of costimulatory molecules.^11,12 Immunomodulators provide an opportunity to enhance the immunogenicity of TAAs, leading to the generation of TAA-specific adaptive immunity (Table 3). The following sections review the clinical applications of such immunomodulators and their potential for improving the efficacy of current cancer vaccine strategies.

Table 3.

Partial list of immunomodulators used for cancer vaccine modalities

Type	Example	Proposed mechanism of action
Cytokine	GM-CSF	mDCs activation → IL-1 & IL-12 → T-cell activation → IFNy16–18
	IL-7	Expansion of CD8+ & CD4+ T cells31
	IL-12	T-cell activation → IFNγ137
	IL-21	Enhanced cytolytic activity of NK & CD8+ T cells → IFNγ34,138
TLR agonist	Imiquimod	pDC activation → IFA-α → mDC maturation → IL-12 → T-cell activation63,64
	CpG	pDC activation → IFA-α → mDC maturation → IL-12 → T-cell activation68
	MPL	mDC activation → IL-12 → T-cell activation → IFNγ68
Costimulation	B7–1	B7–1 → bind CD28 → T-cell activation43
	ICAM-1	ICAM-1 → bind LFA-1 → T-cell activation43
	LFA-3	LFA-3 → bind CD2 → T-cell activation43
	Anti-CTLA-4	Block CTLA-4 binding to B7–1 → reduced costimulatory inhibition43
Chemotherapy	Doxorubicin	Tumor apoptosis → DC activation → T-cell immunity85
Radiotherapy	EBRT	Increased MHC I expression → enhanced CTL-mediated killing93
Ablative therapy	RFA	DC maturation → enhanced TAA presentation → T-cell immunity99
	Cryoablation	Antigen depot → enhanced APC antigen presentation → T-cell immunity102
NSAID	Celecoxib	Increased MHC expression → enhanced TAA presentation → T-cell immunity118
Treg modulation	Denileukin diftitox	Reduced Tregs prevents T-cell downregulation119
Classical adjuvant	Montanide ISA 51 (IFA)	Antigen depot → enhanced APC antigen presentation → T-cell immunity74

Cytokines

TNFα, IFNα and IL-2 are FDA-approved cytokines currently used for the treatment of soft tissue sarcoma, renal cell carcinoma and/or melanoma.^13–15 Other FDA-approved cytokines used to enhance lymphocyte and/or platelet recovery following chemotherapeutic regimens include granulocyte-macrophage colony-stimulating factor (GM-CSF), granulocyte-colony-stimulating factor, erythropoietin and IL-11. Several of these cytokines, along with others, have demonstrated the ability to enhance APC or T-cell function. Increased APC function correlates with heightened levels of MHC, cell adhesion, and costimulatory molecules, which can lead to improved antigen presentation and induction of adaptive immune responses. Specifically, GM-CSF activates dendritic cells (DCs), which leads to production of IL-1 and IL-12 and activation of T-cells (Table 3).^16–18 In the production of the sipuleucel-T vaccine, an autologous DC-based therapy, GM-CSF is used to activate and mature DCs in vitro to enhance antigen uptake prior to patient reinfusion.¹⁹ A Phase III trial of sipuleucel-T demonstrated a significant increase in median survival of 4.5 mon in patients with hormone-refractory prostate cancer (25.9 vs. 21.4 mon; n = 127).²⁰ However, as the primary endpoint of progression-free survival was not met, a similar Phase III trial measuring overall survival as the primary endpoint is ongoing, with reports of a survival advantage in patients receiving sipuleucel-T treatment.^21,22 In another modality, GM-CSF (100 or 500 μg subcutaneously) has been used in combination with a follicular lymphoma-KLH anti-idiotype fusion antibody to induce tumor-specific CD4⁺ and CD8⁺ T-cell responses in 19 of 20 patients with follicular lymphoma,²³ and is currently being evaluated in a Phase III trial in this patient population.²⁴ Using a different vaccine platform, GM-CSF in combination with protein immunogens has also demonstrated an enhanced antigen-specific immune response.^23,25 GM-CSF-secreting irradiated allogeneic whole tumor cells (GVAX) have also been used in the treatment of prostate cancer.²⁶ A Phase I/II trial of GVAX in 80 patients with metastatic prostate cancer demonstrated a dose-dependent survival advantage that did not reach statistical significance.²⁷ However, two Phase III trials using GVAX alone or in combination with docetaxel and prednisone were terminated due to a low probability of meeting the primary survival endpoint.^28,29 A review by Parmiani et al. suggested that the adjuvant use of high-dose GM-CSF (≥100 μg) and repeated dosing may expand myeloid-derived suppressor cells and thus dampen the tumor-specific adaptive immune response.³⁰ Overall, the use of GM-CSF in combination with various vaccination strategies has demonstrated some clinical benefit.

Several other cytokines, including IL-2, IL-7, IL-12, IL-15 and IL-21, have a critical role in enhancing T-cell function.^31–34 The most studied of these, IL-2, has demonstrated clinical responses in metastatic melanoma and metastatic renal cell carcinoma.^15,35 In vitro, IL-2 induces proliferation of CD8⁺ T-cells capable of lysing tumor cells.³⁶ However, the combination of IL-2 with current peptide-based or tumor cell-based cancer vaccination strategies has yet to provide significant clinical benefit. Specifically, in patients with metastatic melanoma or metastatic renal cell carcinoma, multivalent autologous tumor lysates bound to silica beads in combination with IL-2 infusion (1.75 × 10⁶ IU/m²/day) provided no survival benefit compared to vaccination alone.³⁷ Similarly, other clinical trials using autologous tumor cell or peptide vaccines in combination with IL-2 have not yet demonstrated a significant clinical benefit.^38,39 Thus, the potential of IL-2 to enhance immunity to TAAs delivered by peptide or tumor cell lysate vaccination has not yet been realized, but determining the optimal dose, route and schedule for the combined use of cytokines and cancer vaccines may result in clinical benefit.

A different approach to the use of IL-2 in cancer vaccination utilizes a modified vaccinia Ankara (MVA) virus carrying mucin (MUC)-1 and IL-2 transgenes (MVA/MUC1/IL-2; TG4010).⁴⁰ A clinical study in prostate cancer patients with evidence of prostate-specific antigen (PSA) progression evaluated the significance of the timing of MVA/MUC1/IL-2 vaccinations.⁴¹ Patients in Arm 1 received 6 weekly vaccinations, then were vaccinated once every 3 w; patients in Arm 2 were vaccinated once every 3 w. A significant improvement in PSA doubling time was observed in Arm 1 compared to Arm 2. Finally, the use of IL-2 to expand lymphocytes for adoptive transfer is being utilized in clinical trials with evidence of clinical efficacy.⁴²

Overall, optimizing the use of cytokines to enhance generation of tumor-specific immunity is a challenging and ongoing process, but increased understanding of proper dosage and routes of administration, coupled with investigation of additional cytokines such as IL-7, IL-12, IL-15 and IL-21, could improve the clinical efficacy of cancer vaccines.

Costimulation

Efficient activation of T-cells requires two signals. The first signal is mediated through the presentation of a peptide-MHC complex on the surface of the APC to the T-cell receptor on the surface of the T-cell. The second signal involves binding of a T-cell costimulatory molecule on the surface of the APC to its ligand on the surface of the T-cell. One of the most studied costimulatory molecules is B7–1 (CD80), which interacts with either of two T-cell ligands, leading to upregulation (CD28) or downregulation (CTLA-4) of T-cell function (Table 3).⁴³ In a clinical setting, an allogeneic breast cancer cell line vaccine expressing B7–1, used in combination with GM-CSF or BCG, was evaluated in 30 women with metastatic breast cancer.⁴⁴ Following vaccination, increased CD8⁺IFNγ⁺ PBMCs were isolated, but no clinical responses were reported. Another trial used a canarypox viral vector expressing carcinoembryonic antigen (CEA) and B7–1 (ALVAC-CEA/B7–1) in combination with chemotherapy (FOLFIRI) in 118 metastatic colorectal cancer patients.⁴⁵ Vaccination with ALVAC-CEA/B7–1 increased CEA-specific T-cell responses, but clinical responses observed were similar to other trials using the FOLFIRI chemotherapy regimen alone. Kaufman et al. intratumorally injected a recombinant vaccinia (rV) virus expressing B7–1 (rV-B7–1) in 12 patients with metastatic melanoma.⁴⁶ Following vaccination, gene expression profiles from fine needle aspirations of melanoma lesions found an increase in CD8 and IFNγ expression in patients who exhibited a partial response, compared to patients with stable disease or progressive disease, suggesting an association between adaptive immune activation and clinical response. Another clinical trial of 28 patients with metastatic androgen-independent prostate cancer used docetaxel in combination with a vaccination strategy of rV-PSA admixed with rV-B7–1, followed by a similar recombinant fowlpox (rF) vector (rF-PSA) with GM-CSF.⁴⁷ PSA velocities stabilized or decreased in nine of nine patients who received vaccine in combination with docetaxel and in eight of 11 vaccinated patients who received docetaxel upon disease progression, compared to 13 of 17 patients in a historical control who received docetaxel alone. Induction of PSA-specific T-cell responses was reported, as well as induction of tumor-specific immunity to separate antigens (PSMA, PAP and MUC-1). The generation of immunity to additional TAAs not expressed by the vaccine (antigen cascade) may provide insight as to why immunity to the vaccine-delivered antigens alone often does not correlate with clinical responses, as antigen cascade may be necessary in order to induce this effect. Overall, cancer vaccine strategies incorporating B7–1 have demonstrated the generation of TAA immunity in patients with various malignancies.

Additional costimulatory molecules such as intercellular adhesion molecule (ICAM)-1 and lymphocyte function-associated antigen (LFA)-3 have been used in combination with B7–1 (the triad designated TRICOM) in cancer vaccines for various malignancies.⁴⁸ Delivering TRICOM via a poxviral vector leads to increased expression of these three costimulatory molecules on the surface of infected cells, which enhances T-cell costimulation and activation. In a preclinical CEA-transgenic mouse model, vaccination with an rV vector containing CEA and TRICOM (rV-CEA/TRICOM) increased the avidity of CEA-specific cytotoxic T lymphocytes (CTLs) 100-fold compared to rV-CEA only.⁴⁹ This enhancement of T-cell function demonstrates how adequate costimulation can enhance the quality of immune responses to TAAs. In a Phase I study of 15 patients with metastatic prostate cancer, an rV-PSA/TRICOM prime was followed by boosts of rF-PSA/TRICOM with or without GM-CSF.⁴⁸ Of six HLA-A*0201 patients, four demonstrated an increase in PSA-specific T-cell ELISPOT responses, and PSA velocities decreased in nine of 15 vaccinated patients. In another trial, 25 patients with metastatic carcinoma were vaccinated with rV-CEA/MUC1/TRICOM followed by boosts with a similar fowlpox vector, both in combination with GM-CSF.⁵⁰ Immune responses to MUC-1 or CEA were observed in nine of 16 vaccinated patients, and one patient with clear cell ovarian cancer and one with metastatic breast cancer demonstrated clinical responses radiographically. Together, these data demonstrate that increasing the costimulatory signal with TRICOM can enhance the generation of TAA-specific immunity and potentially lead to clinical benefits in cancer patients.

As described above, the costimulatory molecule B7–1 expressed by APCs can either activate (bind CD28) or downregulate (bind CTLA-4) T-cell responses. Ipilimumab is a CTLA-4-specific antibody used to reduce binding of B7–1 to CTLA-4 and thus potentiate T-cell activation.⁵¹ Clinical trials using ipilimumab as a monotherapy in metastatic melanoma patients or in combination with GM-CSF in metastatic prostate cancer patients have demonstrated some clinical responses and expansion of activated CD8⁺ T-cells.^52,53 A study combining ipilimumab with a gp100 peptide vaccine in 14 patients with metastatic melanoma reported that 11 patients demonstrated gp100-specific immune responses following in vitro sensitization and significant increases in CD3⁺CD4⁺ and CD3⁺CD4⁻ (presumed CD8⁺) T-cells following treatment; two patients had complete clinical responses and one had a partial clinical response.⁵⁴ Blocking CTLA-4 downregulation of T-cell responses in combination with vaccination is being further evaluated in several ongoing clinical trials.

Another approach to enhancing TAA-specific immunity is to activate APCs through costimulation. CD40 is one such target expressed on DCs, B-cells and monocytes.⁵⁵ Ligation of CD40 expressed on APCs leads to activation, whereas ligation of CD40 expressed by some carcinomas and melanoma can lead to apoptotic cell death.^55,56 A CD40 agonist monoclonal antibody (mAb) (CP-870,893) was used in a Phase I dose escalation trial of 29 patients with advanced solid tumors.⁵⁷ This study reported a dose-related increase in CD86⁺CD19⁺ B-cells following treatment, with four of 15 melanoma patients showing objective partial responses at restaging (43 d). Targeting CD40 has the benefit of APC activation and a direct cytolytic effect on tumors, and could prove beneficial for the treatment of various malignancies. Combining this CD40 mAb with autologous cell transfer or cancer vaccines may further improve patient outcomes.

Toll-Like Receptor Agonists

Among the primary signaling mechanisms of the innate system that discriminates between self and non-self are pattern-recognition receptors. Expressed on APCs, these receptors recognize pathogen-associated molecular patterns such as bacterial lipopolysaccharide (LPS), bacterial CpG motifs, viral DNA, dsRNA, mannose and yeast.⁵⁸ TLRs comprise one such class of pattern-recognition receptors. Several synthetic TLR agonists are under clinical investigation, and one TLR-7 agonist (imiquimod) is FDA-approved for superficial basal cell carcinoma, genital warts and actinic keratosis.⁵⁹ Direct application of topical imiquimod has demonstrated clinical responses in various cutaneous metastases of the skin.^60,61 This antitumor activity is likely directed through TLR-dependent (TLR-7/TLR-8 activation of APCs) and independent (tumor apoptosis) pathways.⁶² One immune-mediated mechanism of action for imiquimod is TLR-7 activation of plasmacytoid DCs (pDCs), leading to IFNα production that activates and matures myeloid-derived DCs (mDCs), leading to production of proinflammatory cytokines and enhanced antigen presentation that activates adaptive immunity (Table 3).^63,64 Imiquimod has been used clinically in combination with peptide/protein-based vaccines for malignant melanoma and prostate cancer.^65–67 In a trial of nine patients with malignant melanoma, vaccination with NY-ESO-1 protein in combination with topical imiquimod at a healthy skin extremity generated NY-ESO-1-specific CD4⁺ T-cell responses in six of eight patients.⁶⁶ A Phase I/II trial of 19 hormone-sensitive prostate cancer patients following biochemical failure compared a peptide vaccine (11 HLA-A*0201 peptides and two HLA class II peptides) alone or in combination with imiquimod, GM-CSF, hyperthermia, or a MUC-1-mRNA/protamine complex adjuvant.⁶⁷ Of four patients who received vaccine plus imiquimod, three demonstrated an increase of PSA doubling time and one showed an interim decline followed by a rise in PSA. Only four of 15 patients comprising all other groups reported similar PSA responses. Together, these reports demonstrate the ability of a TLR-7 agonist (imiquimod), used in combination with peptide/protein-based vaccines, to enhance the generation of TAA-specific immunity and lead to clinical benefits.

Bacterial CpG DNA motifs that signal through TLR-9 have been used in combination with various vaccination strategies to treat infectious diseases and cancer.⁶⁸ Several synthetic CpG oligodeoxynucleotides signal through TLR-9 and lead to B-cell activation, pDC production of IFNα, and/or mDC production of IL-12 (Table 3).⁶⁸ One TLR-9 agonist (PF-3512676) used in a Phase II trial of stage IIIB/IV non-small cell lung cancer (NSCLC) in combination with platinum-based chemotherapy demonstrated a trend in enhanced overall survival that did not reach statistical significance.⁶⁹ Additionally, two Phase III trials in NSCLC have been discontinued due to lack of benefit of PF-3512676 over standard-of-care chemotherapy.⁷⁰ Another vaccine using a TLR-9 agonist (VaxImmune™) in combination with a MAGE-A3 TAA (ASCI; GSK1572932A) is currently being evaluated in a Phase III trial of patients with early-stage (IB, II, IIIA) NSCLC.⁷¹ In a Phase II trial of 182 stage IB/II NSLC patients randomized 2:1 to receive MAGE-A3 vaccine versus placebo, endpoints for disease-free interval, disease-free survival, and overall survival were improved in the vaccine arm, but not to significant levels.⁷² Further, microarray analysis of tumor biopsies revealed a genetic signature associated with a high risk of postoperative relapse, which seemed to correlate with improved clinical efficacy of the vaccine.⁷³ More information about patient and tumor biology will help to identify the most appropriate patient populations for immunotherapy. Oil emulsions such as montanide ISA-51 (IFA) form an antigen depot and facilitate the slow release of peptide antigens to reduce the induction of immunological tolerance caused by presentation of antigen through nonprofessional APCs.⁷⁴ This approach, in combination with TLR agonists such as CpG, has demonstrated induction of TAA-specific T-cell responses in melanoma patients.^75,76

Another TLR agonist of interest is monophosphoryl lipid (MPL) A, which targets TLR-4.⁷⁷ MPL is a derivate of LPS and induces a Th1-like cytokine response similar to LPS.⁷⁸ L-BLP25 (Stimuvax®) is a liposomal vaccine that incorporates MPL and carries TAAs targeting the extracellular core protein of MUC-1. In a phase IIB trial, 171 patients with stage IIIB/IV NSCLC were randomized to either low-dose cyclophosphamide administered 3 d before weekly L-BLP25 vaccinations plus best supportive care, or best supportive care alone.⁷⁹ At a 2-y follow-up, a 4.4-mon survival advantage was noted in the vaccine arm, but did not reach statistical significance. A subgroup of patients with stage IIIB locoregional NSCLC appeared to respond better to L-BLP25 vaccination than patients with more progressive disease. L-BLP25 is currently being investigated in combination with cyclophosphamide versus placebo in a Phase III trial of patients with stage III NSCLC.⁸⁰ Targeting patients with less advanced disease may prove beneficial for this and other vaccination strategies.

Chemotherapy

Given the potential for lymphopenia and immunosuppression, using chemotherapeutic agents to enhance tumor-specific immune responses would appear to be counterintuitive. However, a growing body of evidence suggests that drugs commonly used in cancer chemotherapy can augment the antitumor effects of immunotherapeutic modalities. One mechanism proposed for this observed enhancement of the immune response is chemotherapy-induced apoptosis of tumor cells. The induction of tumor apoptosis by certain cytotoxic agents can not only activate DCs, but also provide them with an increased supply of tumor-specific antigens for presentation and cross-presentation to T-cells.^81–83 Furthermore, certain cytotoxic drugs can modulate tumor phenotype or induce immunogenic cell death, thus enhancing immune-mediated cell killing. In murine cell lines, 5-fluorouracil treatment and doxorubicin-induced caspase-dependent apoptosis enhanced CTL lysis of tumor cells.^84,85 This ability of specific cytotoxic drugs to enhance the immunogenicity of tumor cells makes the combination of chemotherapy and immunotherapy attractive.

Another benefit to combining cytotoxic agents with immunotherapeutic approaches is the onset of homeostatic T-cell proliferation following drug-induced lymphopenia. The phenomenon of homeostatic T-cell proliferation was demonstrated in a pilot trial of patients with metastatic melanoma who were treated with expanded tumor-infiltrating lymphocytes (TILs) and IL-2 after a lymphodepleting regimen of cyclophosphamide and fludarabine.⁸⁶ Results of this trial suggested that conditioning with nonmyeloablative chemotherapy before adoptive transfer of activated tumor-specific T-cells can enhance tumor regression and increase the rate of objective clinical responses. These findings correlated with those of Muranski et al. who reported that lymphodepletion enables the host to accommodate transferred T-cells and increases T-cell survival compared to competing cell populations.⁸⁷

While these studies demonstrate the benefit of combining lymphodepletion by cytotoxic agents with passive immunotherapy (adoptive T-cell transfer), others show that combining active immunotherapy (vaccination) with chemotherapy has clinical promise as well. As described above, docetaxel plus a recombinant poxvirus-based vaccine (rV-PSA + rV-B7–1) increased antigen-specific T-cell responses to PSA, as well as to cascade antigens derived from the tumor.⁴⁶ Similar results were found when patients with extensive small cell lung cancer were vaccinated with a DC-based p53 vaccine in combination with standard-of-care chemotherapy.⁸⁸ Specifically, 16 of 28 patients (57%) demonstrated p53-specific T-cell responses to vaccination. In addition, the authors observed a high rate of objective clinical responses to chemotherapy (61.9%) immediately following vaccination, which was closely associated with the induction of immunologic responses. These clinical trials show that chemotherapy can be administered with immunotherapy without inhibiting the generation of vaccine-specific immune responses. These studies also suggest that, despite the potency of cytotoxic anticancer drugs, certain agents have immunomodulatory activity and can thus be successfully combined with different immunotherapeutic approaches to potentiate tumor-specific immunity.

Radiotherapy

Radiotherapy, like chemotherapy, has traditionally been considered immunosuppressive and thus an unlikely candidate for combination with immunotherapeutic approaches. However, the combination of radiation and immunotherapy holds particular promise as a strategy for cancer treatment. Evidence suggests that this combination may prevent cancer cells from escaping immune recognition through two main mechanisms: (a) radiation-induced tumor-cell death increases the source of tumor antigens for immune responses, and (b) radiation modulates tumor-cell phenotype, allowing for efficient immune-cell access and increased sensitivity to T-cell killing.⁸⁹

Because of their low expression of costimulatory molecules, tumor cells often do not generate potent immune responses.^11,12 However, previous studies have demonstrated that irradiation of tumor cells can induce recognition and phagocytosis signals for DCs and cause the release of “danger” signals required for DC activation.^90,91 Furthermore, nonlethal doses of radiation have the ability to modulate tumor-cell phenotype to enhance T-cell-mediated killing.^92,93 In addition, antigens released by dying tumor cells following radiation can be presented and cross-presented by DCs, thereby activating tumor-specific immune responses. Radiation-treated tumor cells could therefore serve as an in situ autologous tumor vaccine, augmenting antitumor immunity that could eradicate residual primary tumor cells as well as distant metastases.^89,94 These preclinical studies have demonstrated the synergism of radiotherapy and immunotherapy in enhancing tumor-specific immune responses and immune-mediated tumor killing, and have provided the rationale for their combined use in the clinic.

A recent clinical trial combined a recombinant cancer vaccine with standard definitive radiotherapy in patients with localized prostate cancer.⁹⁵ Patients were assigned to receive definitive radiotherapy with or without vaccine to determine if the vaccine could induce an immune response in the presence of tumor irradiation. The vaccine regimen consisted of a priming vaccination of rV-PSA admixed with rV-B7–1, followed by monthly booster vaccinations with rF-PSA. Patients received local radiotherapy either alone or in combination with vaccine (between the fourth and sixth vaccinations). Of 17 patients in the combination arm who completed all eight vaccinations, 13 (76%) had at least three-fold increases in circulating PSA-specific T-cells, while no detectable PSA-specific T-cell responses were noted in the radiotherapy-only arm.⁹⁵ Patients in the combination arm also showed evidence of antigen cascade through generation of T-cell responses to prostate-associated antigens other than PSA, providing indirect evidence of immune-mediated tumor killing. Chi et al. also attempted to show enhanced tumor-specific immunity by combining vaccine with radiation therapy. In a Phase I trial, patients with refractory hepatoma received 8 Gy of external beam radiation followed by intratumoral injection of immature DCs on days 2 and 24.⁹⁶ Results of this study reported seven of ten evaluable patients had increases in α-fetoprotein-specific T-cell ELISPOT responses. Several of these patients also showed partial clinical responses, reinforcing the relationship between induction of antitumor immunity and reduction of disease burden. Taken together, these studies demonstrate that radiation can enhance tumor-specific immune responses and, in combination with immunotherapeutic approaches, may increase tumor cell killing compared to either modality alone.

Ablative Therapy

Another approach to inducing immunogenic tumor cell death is radiofrequency ablation (RFA), which converts radiofrequency waves into heat to achieve local temperatures sufficient for tissue destruction.⁹⁷ A clinical study of 20 patients with hepatocellular carcinoma (HCC) demonstrated a significant increase in circulating T-cell ELISPOT responses to tumor tissue following RFA treatment and an increase in circulating natural killer (NK) cells (CD3⁻CD56⁺).⁹⁸ Another trial using PBMCs from 19 patients with HCC demonstrated that RFA significantly increased maturation markers (CD83, CD86) of activated monocytes and monocyte-derived DCs.⁹⁹ Further, exposure of patient-derived, HCC-specific T-cell lines (CD4⁺ or CD8⁺) to RFA-treated HCC tumor increased T-cell IFNγ production. This could be due to increased antigen presentation, increased expression of costimulatory molecules, and/or reduced immunosuppression by the tumor cells. A clinical trial of 14 lung cancer patients undergoing RFA demonstrated a significant increase in total white blood cells three days post-RFA, a significant increase in proinflammatory plasma cytokines and chemokines three days post-RFA, and a significant decrease in circulating regulatory T-cells (Tregs) (CD4⁺CD25 hⁱCD69⁻) 30 d post-RFA.¹⁰⁰ Overall, these clinical reports demonstrate that RFA can enhance DC maturation, increase frequency of tumor-specific T-cells, and reduce Treg numbers, making RFA and immunotherapy an appealing combinational approach for cancer therapy.¹⁰¹

Cryoablation is a type of ablative therapy that dates back to the 1840s.¹⁰² Early clinical studies using cryoablation demonstrated regression of untreated metastatic lesions and indications of antitumor immunity.^103–106 More recent clinical studies have demonstrated increases in serum cytokines and generation of tumor-specific T-cells following cryoabaltion.^107,108 The simplicity of cryoablation, coupled with minimal negative effects on the immune system, make this procedure an attractive candidate for combination with immunotherapy.

High-Intensity Focused Ultrasound

High-intensity focused ultrasound (HIFU) is a recently developed and promising technique that uses a transducer to generate high-frequency ultrasound waves focused on a single point in the tumor. HIFU can deliver a large amount of energy that is subsequently converted to thermal energy, resulting in tumor cell destruction with minimal damage to surrounding tissue.¹⁰⁹ HIFU is receiving increasing interest for the clinical management of patients with various types of cancer, including prostate, breast, liver, pancreas, kidney, bone and soft tissue.^110,111

Wu et al. investigated whether tumor antigens expressed on breast cancer cells are preserved after HIFU treatment and could serve as a potential source of antigens to stimulate the immune system.¹¹² Primary lesions in 23 patients with biopsy-proven breast cancer were treated with HIFU and then analyzed for changes in expression of tumor antigen. Results demonstrated that several tumor antigens remained in the tumor debris after HIFU ablation, potentially indicating the generation of a TAA source for DCs. A more recent study investigated the status and function of TILs after HIFU ablation of breast cancer.¹¹³ In this trial, 48 patients were randomized to receive HIFU prior to modified radical mastectomy or surgery alone. Compared to surgery alone, treatment with HIFU and surgery significantly increased tumor-infiltrating T-cells, B-cells and NK cells. Additionally, HIFU treatment significantly increased Fas-ligand⁺, granzyme⁺ and perforin⁺ TILs compared to surgery alone, indicating the enhanced ability of lymphocytes to induce apoptosis or cytolytic death of tumor cells. These studies suggest that HIFU-treated tumor can serve as a source of tumor antigens, and that HIFU can increase the number and function of TILs.

Nonsteroidal Anti-Inflammatory Drugs

Nonsteroidal anti-inflammatory drugs (NSAIDs) have been gaining significant interest among oncologists as more is learned about their potential antitumor effects. A growing body of data suggests that NSAIDs can interfere with complex processes related to tumorigenesis, angiogenesis, metastasis and inhibition of apoptosis.^114–116 Attention has focused on cyclooxygenase-2 inhibitors, including celecoxib (Celebrex®) and indomethacin, to confirm their antiangiogenic and proapoptotic functions, as well as to determine their ability to initiate tumor-specific immune responses.

One study treated colorectal carcinoma patients preoperatively with oral NSAIDs and evaluated MHC and TIL gene expression.¹¹⁷ In this study, genes belonging to the MHC locus, including those responsible for peptide loading, antigen presentation and cytolytic activity, were upregulated in response to short-term preoperative NSAID treatment. Furthermore, NSAID pretreatment increased infiltration of CD4⁺ and CD8⁺ T-cells in the tumor stroma and decreased expression of molecules associated with immunosuppressive Tregs (FOXP3, IL-10). Similar findings were noted in a study that investigated the activity of celecoxib in patients with newly diagnosed, untreated nasopharyngeal carcinoma.¹¹⁸ Tumor biopsies were obtained before and after a 14-d course of celecoxib. Results of this study showed MHC class II DMB increased expression two-fold, the greatest change in expression level of any of the 35 genes examined. Clearly, utilizing NSAIDs in cancer treatment regimens is still in the beginning stages. However, the ability of these drugs to inhibit angiogenesis, induce apoptosis and upregulate genes crucial for initiating an immune response may enhance the function and antitumor effects of other immunotherapeutic modalities.

Regulatory T-cells

The ability of tumors to evade immune system surveillance enables tumor cells to grow and survive. CD4⁺CD25⁺FOXP3⁺ Tregs are important suppressors of active tumor-specific immune responses in cancer.¹¹⁹ Depleting Tregs and reducing their functionality has been shown to improve endogenous antitumor immunity and the efficacy of active immunotherapy in animal models for cancer, suggesting that inhibiting Treg function could also improve on the limited successes of immunotherapy in human cancers.^119–122 Denileukin diftitox (Ontak®) is a recombinant fusion protein consisting of IL-2 and diphtheria toxin targeting IL-2 receptor (CD25). It has been approved by the FDA to treat cutaneous T-cell leukemia/lymphoma, which is characterized by large numbers of malignant CD4⁺CD25⁺ T-cells.^123,124 Based on the antitumor effects of Treg reduction in leukemia/lymphoma patients, Curiel et al. designed a Phase 0/I clinical trial in patients with any advanced-stage epithelial carcinoma to test if denileukin diftitox would reduce Tregs, and thereby reduce immune suppression and boost antitumor immunity in these patient populations.¹²⁵ Results of this study demonstrated that denileukin diftitox decreased the number of circulating Tregs and was associated with improved cellular immunity in patients with ovarian, lung, breast or pancreatic cancer. Additional reports confirmed that denileukin diftitox reduced Treg numbers and improved tumor-specific T-cell responses in individuals with renal cell carcinoma and melanoma.^126,127 Importantly, the enhanced anti-tumor immune responses following Treg reduction also resulted in regression of melanoma metastases, confirming clinical benefit.¹²⁶ Numerous clinical trials are currently recruiting patients to further evaluate the clinical applicability of reducing Tregs, including Phase II trials in patients with metastatic pancreatic cancer, ovarian cancer, and advanced refractory breast cancer.^128–130 These and other studies will demonstrate how reducing Tregs can enhance current cancer vaccination strategies.

Daclizumab is an mAb targeting CD25 that was FDA-approved in 1997 for renal allograft rejection.¹³¹ This antibody blocks interaction of IL-2 with CD25, preventing T-cell proliferation, and thus could have a potential role in the treatment of T-cell leukemia or lymphoma.¹³² Recently, the combination of daclizumab and cytotoxic depsipeptide significantly increased the survival of mice bearing T-cell leukemia tumors.¹³³ This demonstrates another approach to targeting CD25 for combinational cancer immunotherapy.

Cyclophosphamide is a cytotoxic drug with demonstrated ability to reduce lymphocyte populations without being myeloablative.¹³⁴ Clinical studies have also demonstrated that low metronomic doses of cyclophosphamide can reduce Treg populations without reducing CD8⁺ T-cell populations.¹³⁵ However, single high doses of cyclophosphamide appear to be ineffective for Treg reduction.¹³⁶ Thus, proper dosing and scheduling of cyclophosphamide treatment in combination with cancer vaccination could likely enhance the generation of tumor-specific cytolytic immunity.

Conclusions

Initial steps in the development of cancer vaccines have led to the identification of TAAs and the development of novel vaccine strategies. Elucidation of the mechanisms involved in antigen recognition, processing and presentation by APCs, along with identification of signals that activate or suppress T-cells, has led to the discovery and exploitation of immunomodulators to enhance the generation of tumor-specific immunity. Combining cancer vaccines with traditional cytotoxic agents or radiation therapy provides an opportunity to improve patient survival and quality of life. Additional knowledge acquired from basic, translational and clinical studies will most likely translate the science of tumor immunology into clinically relevant modalities for a range of cancers.

Abbreviations:

APC

antigen-presenting cell

BCG

bacillus calmette-guerin

CEA

carcinoembryonic antigen

CTL

cytotoxic T lymphocyte

DC

dendritic cell

FDA

Food and Drug Administration

FOLFIRI

folinic acid/fluorouracil/irinotecan

GM-CSF

granulocyte-macrophage colony-stimulating factor

HCC

hepatocellular carcinoma

HIFU

high-intensity focused ultrasound

ICAM

intercellular adhesion molecule

LFA

lymphocyte function-associated antigen

LPS

lipopolysaccharide

mAb

monoclonal antibody

mDC

myeloid-derived dendritic cell

MHC

major histocompatibility complex

MPL

monophosphoryl lipid

MUC

mucin

MVA

modified vaccinia ankara

NK

natural killer

NSAID

nonsteroidal anti-inflammatory drug

NSCLC

non-small cell lung cancer

PAP

prostatic acid phosphatase

PBMC

peripheral blood mononuclear cell

pDC

plasmacytoid dendritic cell

PSA

prostate-specific antigen

PSMA

prostate-specific membrane antigen

rF

recombinant fowlpox

RFA

radiofrequency ablation

rV

recombinant vaccinia

TAA

tumor-associated antigen

TIL

tumor-infiltrating lymphocyte

TLR

toll-like receptor

Treg

regulatory T-cell