Tumor-specific antigens

Tumor-specific antigens (TSAs) comprise gene products that are uniquely expressed in tumors, such as point-mutated ras oncogenes, p53 mutations, and products of ribonucleic acid (RNA) splice variants and gene translocations. Three mutations at codon 12 represent the vast majority of ras mutations, which are found in approximately 20% to 30% of some human tumors.¹ Although the ras protein is not found on the cell surface, one can envision vaccine therapy directed against peptide–MHC complexes on the cell surface. Indeed, there have been clinical trials in pancreatic carcinoma that target ras.^{2, 3} B-cell lymphomas overexpress a single immunoglobulin (Ig) variant on their cell surface; therefore, each B-cell lymphoma displays a unique target for immunotherapy.^4–6 The gene products of RNA splice variants and deoxyribonucleic acid (DNA) translocations also represent unique fusion proteins that can be specific targets for immunotherapy, including c-erb-B2 RNA splice variants and the bcr/abl product of DNA translocation of chronic myelogenous leukemia.

Several viruses are associated with the etiology of some cancers. An excellent example of this is the connection between human papillomavirus (HPV) and cervical cancer. This has led to FDA approval of the HPV vaccine for prevention of cervical cancer.⁷

Tumor-associated antigens

TAAs can be categorized into three major groups: oncofetal antigens, oncogene products, and tissue-lineage antigens (Table 1). Oncofetal antigens, normally found during fetal development, are greatly downregulated after birth. This class of antigens, which includes prostate-specific membrane antigen (PSMA), carcinoembryonic antigen (CEA), and the cancer mucin MUC-1, are often overexpressed in tumors compared with normal tissues. The MUC-1 TAA is overexpressed in the majority of human carcinomas and several hematologic malignancies. Much attention has been paid to the hypoglycosylated variable number of tandem repeats (VNTR) region of the N-terminus of MUC-1 as a vaccine target. While previous studies have described MUC-1 as a tumor-associated tissue differentiation antigen, studies have now determined that the C-terminus of MUC1 is an oncoprotein, and its expression is an indication of poor prognosis in numerous tumor types.⁸

BAGE, B melanoma antigen; CEA, carcinoembryonic antigen; EMT, epithelial–mesenchymal transition; gp100, glycoprotein 100; HCV, hepatitis C virus; HPV, human papillomavirus; MAGE-A3, melanoma-associated antigen-A3; MUC-1, mucin 1; NY-ESO, New York esophageal carcinoma antigen 1; OCT-4, octamer-binding transcription factor 4; PAP, prostatic acid phosphatase; PSA, prostate-specific antigen; SOX-2, (sex-determining region Y)-box-2; STn-KLH, sialyl-Tn-keyhole limpet hemocyanin; TERT, telomerase reverse transcriptase; VEGF-R, vascular endothelial growth factor receptor; WT-1, wild-type 1.

Oncogene and suppressor gene products, such as nonmutated HER2/neu and p53, are analogous to oncofetal antigens in that they can be overexpressed in tumors and may be expressed in some fetal and normal tissues. Similarly, telomerase, an enzyme important in cellular replication and chromosomal stability, is overexpressed in malignant cells as compared with most normal cells. Epitopes derived from human telomerase have been reported and presumably may be overexpressed by neoplastic cells.⁹

Tissue-lineage antigens such as prostate-specific antigen (PSA) and the melanocyte antigens MART-1/Melan A, tyrosinase, gp100, and TRP-1/gp75 are usually expressed in a tumor of a given type and the normal tissue from which it is derived. Tissue-lineage antigens are potentially useful targets for immunotherapy if the normal organ/tissue in which they are expressed is not essential, such as the prostate, breast, or melanocyte.

Whole-tumor-cell vaccines

The major advantage of using whole-tumor-cell vaccines is that tumor-cell-mutated TSAs and TAAs, some identified and most as yet undefined, may be present in the vaccine preparation. The two types of whole-tumor-cell vaccine platforms being examined clinically are (1) autologous (i.e., manipulating the tumor from a patient into a vaccine for that patient) and (2) allogeneic, in which tumor cells from other patients, usually from established tumor cell lines, are used. Preparing an autologous tumor cell vaccine requires an enormous amount of effort, as fresh tumor needs to be obtained at surgery and prepared in a similar manner for each patient. Due to the considerable variability of tumors among patients, and costs associated with custom vaccine production, most current whole-tumor-cell vaccine approaches have centered on the use of allogeneic cell lines and cell banks.^21–26 Moreover, the cell lines used in the preparations can be infected with vectors that express cytokine genes such as granulocyte-macrophage colony-stimulating factor (GM-CSF)²⁷ or other immunostimulatory transgenes²⁸ to enhance the immunogenicity of the tumor cell. Studies employing tumor cell lines transfected to express GM-CSF have met with some success in treating pancreatic cancer and other tumor types.^{24–27, 29, 30}

Peptides

Most peptides used to induce an immune response are from TSAs or TAAs and are approximately 8 to 11 amino acids in length. These peptides can bind with the appropriate class of the MHC molecule on the APC surface. These MHC-restricted responses are thus effective only if the appropriate MHC allele is present in a patient. The most studied MHC restriction element in the human population is the MHC class I allele, human leukocyte antigen (HLA)–A2, which is found in approximately 50% of the Caucasian population.

Using peptides as immunogens has some advantages: (1) preparation is relatively easy and affordable; (2) because the immunogen is extremely well-defined, the immune response can be analyzed in several ways and quantitated; and (3) peptides can be modified to be more immunogenic in the form of peptide agonists. The same property that makes a peptide vaccine attractive—its specificity—can also be a disadvantage. For example, a CTL (cytotoxic T-lymphocyte) epitope peptide may induce a CTL response that is short lived because of the lack of help provided by helper peptides. Moreover, peptides are useful only in the vaccination of patients who have that specific allele. Some peptide vaccines under study include HPV,³⁵ ras,^{36, 37} HER-2/neu,³⁸ MAGE,³⁹ MART-1, tyrosinase,⁴⁰ gp100,^{41, 42} CEA,⁴³ MUC-1,⁴⁴ and PSMA,^{45, 46} among others.⁴⁷

Vectors

Vectors are among the more flexible means of vaccine delivery. Several major platforms of both viral and bacterial vectors are now in use in the clinic (Table 2). Each vector category and type has its own potential advantages and disadvantages. Review articles have been published on the potential merits of these vectors.^57–63 In general, the advantage of a vector-based vaccine is that (1) multiple genes (including genes for costimulatory molecules and cytokines) can be inserted into some types of vectors; (2) the relative cost of production is low compared to the preparation and purification of proteins or whole-tumor-cell vaccines; (3) many vectors have the ability to infect professional APCs so that the antigens they express can be processed; and (4) viral and bacterial vectors are composed of prokaryotic proteins that act as natural vaccine adjuvants in stimulating the host immune response. The disadvantage of some, but not all, vectors is the development of host-induced immunity to the vector itself, thus limiting its continued use.

Dendritic cell vaccines

The dendritic cell (DC) is considered the most potent APC and, therefore, one of the most attractive means of immunization.^{93, 94} DCs can be employed by (1) loading with a peptide, protein, or anti-Id Ab, (2) infecting with a viral vector, (3) loading with apoptotic bodies from tumor cells, or (4) fusing with a tumor cell.^95–98 The major disadvantages of this approach are the great cost and effort involved. Large amounts of peripheral blood mononuclear cells (PBMCs) must be obtained from patients via leukapheresis. The PBMCs must then be cultured for several days in the presence of cytokines such as GM-CSF, IL-4, and/or TNF-α, and then reinfused into the patient. This must be done for each patient. A successful strategy, approved by the FDA, involves immunization with autologous DCs loaded ex vivo with a recombinant fusion protein consisting of the tumor antigen prostatic acid phosphatase (PAP) linked to GM-CSF for patients with metastatic prostate cancer, as is discussed below.

Combination of vaccines

Although each of the various methods of immunization described has advantages and disadvantages, the most effective immunization protocol may involve priming with one type of immunogen and boosting with another. This method may be advantageous because (1) two different arms of the immune system may be enhanced by using two different modalities (i.e., CD4⁺ and then CD8⁺ T cells); (2) one methodology may be more effective in priming naive cells while another modality may be more effective in enhancing memory cell function; and (3) some of the most effective methods of immunization, like the use of recombinant vaccinia virus or adenoviruses, can be used only a limited number of times because of host anti-vector responses. These vectors may be most effective when used as priming agents, followed by boosts with other agents. Numerous preclinical and clinical studies demonstrate the advantages of diversified prime-and-boost protocols.^{47, 66, 67, 99–103}

Prostate cancer

PAP, which is expressed on over 95% of prostate cancer cells, has been used as a target of the vaccine sipuleucel-T, also called Provenge® (PAP-GM-CSF-pulsed APCs). After early clinical trials of the sipuleucel-T vaccine demonstrated safety, larger trials were conducted in asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer (mCRPC). A Phase III trial¹³⁷ was conducted with overall survival (OS) as the endpoint, enrolling more than 500 patients. No change in time to progression (TTP) was seen, but OS was improved in the vaccine arm (25.8 months vs 21.7 months; p = 0.032). In April 2010, the FDA approved sipuleucel-T for the treatment of minimal or nonsymptomatic mCRPC. A second prostate cancer vaccine has also been evaluated in this same prostate cancer population. This “off-the-shelf” platform (PROSTVAC) consists of a recombinant vaccinia prime and multiple fowlpox booster vaccinations. Each vector contains transgenes for PSA and three costimulatory molecules (B7-1, ICAM-1 and LFA-3, designated TRICOM).^{138, 139} A randomized placebo-controlled 43-center Phase II trial enrolled 125 minimally symptomatic mCRPC patients.¹⁴⁰ Similar to the experience with sipuleucel-T, PROSTVAC did not alter TTP, yet improved median OS relative to placebo (control vector), 25.1 versus 16.6 months. Patients receiving PROSTVAC also had a 44% reduction in death rate compared to the control cohort (p = 0.0061).¹⁴⁰ The median OS in a second single-arm Phase II study¹⁴¹ was 26.6 months, similar to the larger PROSTVAC randomized trial. A global Phase III placebo-controlled trial of PROSTVAC in patients with minimally symptomatic prostate cancer (n = 1298) has now completed accrual. The endpoint is OS.

A recent review¹⁴² analyzed the immune impact induced by PROSTVAC (PSA-TRICOM) in prostate cancer patients. Of 104 patients tested for T-cell responses, 57% demonstrated an increase in PSA-specific T cells 4 weeks after vaccine, and 68% of patients mounted post-vaccine immune responses to TAAs not present in the vaccine (antigen spreading/antigen cascade). Measurements of systemic immune response to PSA may underestimate the true therapeutic immune response (as this does not account for cells that have trafficked to the tumor) and does not include antigen spreading/antigen cascade (as described below). Furthermore, although the entire PSA gene is the vaccine, only one epitope of PSA is evaluated in the T-cell responses. Because this vaccine is directed at generating a cellular/Th1 immune response, less than 0.6% of patients tested had evidence of PSA antibody induction following vaccine. Active surveillance is becoming an increasingly utilized approach in men diagnosed with “low-risk” prostate cancer. A randomized, multicenter, placebo-controlled study of PROSTVAC vaccination versus active surveillance in patients with newly diagnosed clinically localized prostate cancer is under way. The endpoints of this study will be the changes in immune infiltrate and tumor extent at the 12-month biopsy versus initial biopsy.

GVAX, a GM-CSF-secreting vaccine, is an admixture of two prostate cancer cell lines that have been transduced with a replication-defective retrovirus for GM-CSF expression. In two separate multicenter Phase II trials, patients with asymptomatic CRPC given GVAX had a median survival of 26.2–35.0 months.^{24, 143} By estimating predicted survival based on a commonly used nomogram,¹⁴⁴ patients in both trials exceeded anticipated survival by > 6 months.¹⁴⁵

Nineteen HLA-A2-positive patients with rising PSA without detectable metastatic disease or local recurrence received a vaccine containing 11 HLA-A*0201-restricted and two HLA class II synthetic peptides derived from prostate carcinoma tumor antigens subcutaneous (s.c.) for 18 months or until PSA progression. PSA doubling time of 4/19 patients (21%) increased from 4.9 to 25.8 months during vaccination.¹⁴⁶ A Phase II adenovirus/PSA vaccine trial was conducted in patients with recurrent or hormone refractory prostate cancer. The majority of the patients demonstrated anti-PSA T-cell responses above pre-injection levels. Sixty-four percent of the patients demonstrated an increase in PSA doubling time.¹⁴⁷ Clinical trials have also been carried out using peptide-pulsed DCs loaded with peptides of human PSMA,^148–150 as well as DCs transfected with PSA tumor RNA.¹⁵¹ In these studies, antigen-specific immune responses have been observed and some decreases in serum PSA have also been noted.

Melanoma

The vast majority of vaccine clinical trials have been conducted in patients with melanoma because (1) interferon (IFN) and IL-2 have both shown clinical responses in melanoma; (2) melanoma lesions often are readily accessible and thus can be studied for immune infiltrate and cells from melanoma lesions can be grown in culture to obtain both tumor cells and tumor-infiltrating lymphocytes (TILs); and (3) numerous melanoma-associated antigens have been identified.

Clinical trials have been conducted using various peptides derived from melanoma-associated antigens, including tyrosinase, MART-1, gp100, Melan-A (a modified gp100 epitope¹⁵²), and members of the MAGE family.^152–157 Vaccination-induced immune responses as well as clinical benefits were reported. Newer approaches have combined these peptides with GM-CSF emulsified in Montanide ISA-51.¹⁵⁸ Another peptide approach is vaccination with autologous tumor-derived heat shock protein/peptide complexes.¹⁵⁹ Immune responses and clinical responses have also been observed using melanoma peptide vaccination in conjunction with anti-CTLA-4,¹⁶⁰ melanoma peptide-pulsed DCs,^{94, 154} tumor RNA-transfected DCs,¹⁶¹ and intratumoral administration of autologous DCs.¹⁶² A randomized Phase III study showed no benefit using an autologous peptide-pulsed DC vaccine compared to dacarbazine in first-line treatment of patients with metastatic melanoma.¹⁶³

A randomized Phase III trial was conducted involving 185 patients with stage IV or locally advanced stage III cutaneous melanoma, expression of HLA*A0201, and suitability for high-dose IL-2 therapy. Patients were randomly assigned to receive IL-2 alone or gp100 plus IFA, followed by IL-2. The vaccine–IL-2 group, as compared with the IL-2-only group, had a significant improvement in overall clinical response (16% vs 6%, p = 0.03).The median OS was also longer in the vaccine–IL-2 group than in the IL-2-only group (17.8 months vs 11.1 months; p = 0.06).¹⁶⁴ In a multicenter randomized trial, 175 patients with stage IV melanoma were enrolled into four treatment groups. Best clinical response was partial response in 7/148 evaluable patients (4.7%) without significant difference among study arms.¹⁶⁵ In another trial, three HLA class I peptides were administered in a factorial 2 × 2 design: peptide vaccine alone, or combined with GM-CSF, or combined with IFN-α2b, or combined with both. At a median follow-up of 25.4 months, the median OS of patients with vaccine immune response was significantly longer than that of patients with no immune response (21.3 months vs 13.4 months; p = 0.046).¹⁶⁶

Vitespen is an autologous, tumor-derived heat shock protein gp96 peptide vaccine. Patients (n = 322) with stage IV melanoma were randomly assigned 2 : 1 to receive vitespen or physician’s choice of a treatment. OS in the vitespen arm was statistically indistinguishable from that in the physician’s choice arm.¹⁶⁷ A randomized trial provided evidence that a DC vaccine is associated with longer survival compared with a tumor cell vaccine, and is consistent with previous data suggesting a survival benefit from this patient-specific immunotherapy.¹⁶⁸ A Phase II trial of a DC vaccine for metastatic melanoma patients with mainly the HLA-A24 genotype was investigated; OS analysis revealed a significant survival prolongation effect in patients receiving the DC vaccine.¹⁶⁹ Studies were also carried out to detect a pretreatment gene expression signature predictive of response to MAGE-A3 vaccine in patients with metastatic melanoma. An expression signature was associated with the patient’s clinical response.¹⁷⁰ A Phase II trial investigated a peptide vaccination against survivin, a protein crucial for the survival of tumor cells, in patients with treatment-refractory stage IV metastatic melanoma. Survivin-specific T-cell reactivities strongly correlated with tumor response and patient survival.¹⁷¹

Several clinical studies have employed autologous melanoma cells as vaccine. Clinical responses were observed when these melanoma-cell vaccines were preceded by cyclophosphamide.^{172, 173} Clinical responses were also observed when autologous melanoma cells were transduced with GM-CSF and given as vaccine.^{174, 175} In addition, intratumoral administration of recombinant vaccinia encoding GM-CSF in patients with cutaneous melanoma has been reported to induce antitumor immune responses.^{31, 32, 176} Several randomized Phase II trials have been carried out using either autologous^{177, 178} or allogeneic^{179, 180} melanoma vaccine in patients with resected melanoma. In these trials, there is evidence of improved disease-free survival (DFS) in the vaccine arm versus control or no-treatment groups. Recent research indicates that immunization with hybrids of tumor cells and APCs can induce protective immunity and rejection of established tumors in various rodent models.^{95, 181, 182} Novel clinical trial strategies incorporating DC-based vaccines have recently been reported.^183–185 Purified hybrids from the fusion of DCs and tumor cells (dendritomas) have been shown to be safe and to induce both immunological and clinical responses when combined with low-dose IL-2.¹⁸⁴

T-VEC is an intralesional oncolytic immunotherapy comprising an HSV-type 1 engineered to make GM-CSF. In the randomized Phase III trial, 436 patients with stage IIIB-IV melanoma were randomized 2 : 1 to receive intralesional T-VEC versus s.c. GM-CSF. T-VEC significantly improved the proportion of patients with duration of response ≥ 6 months (16% vs 2%) and an overall response rate (26% vs 6%) versus s.c. GM-CSF. Many patients remained in remission with a median follow-up of over 18.4 months.^{186, 187}

Colorectal carcinoma

A prospective randomized-controlled Phase III clinical trial was carried out in 254 patients with stage II and stage III colon cancer using autologous tumor cell vaccine with BCG postsurgery. A 5.3-year median follow-up showed 40 recurrences in the control group and 25 in the vaccine group. The vaccine showed a statistically significant recurrence-free survival (RFS) in patients with surgically resected stage II colon cancer, but not stage III colon cancer.¹⁸⁸ Further follow-up demonstrated a significant survival benefit in the vaccine arm over surgery alone.¹⁸⁹

Several clinical trials have been conducted employing CEA-based vaccines. Results of vaccination in patients with recombinant vaccinia virus,⁶⁹ recombinant avipox virus,^{66, 190, 191} or a prime with recombinant vaccinia (rV)-CEA and multiple boosts with avipox-CEA⁴⁷ have demonstrated the generation of CEA-specific immune responses in patients with advanced gastrointestinal (GI) carcinomas and other CEA-expressing carcinomas. A trial using both vaccinia- and avipox-CEA vaccines containing TRICOM has suggested improved survival in patients receiving the prime-and-boost regimen plus GM-CSF; 40% had stable disease (SDs) for at least 4 months.^{47, 79, 192, 193} In a multicenter Phase II study, colorectal cancer patients received PANVAC vaccine (rV-, rF-CEA-MUC1-TRICOM) post-metastasectomy for liver and lung metastases.¹⁹² Patients received either PANVAC or PANVAC-infected DCs and there was no statistical difference in clinical outcome in the two arms. Vaccinated patients (n = 74) were compared to a contemporary “control” (not randomized) group of colorectal cancer patients (n = 161) undergoing metastasectomy and receiving standard-of-care therapies. Both this control group and the PANVAC-vaccinated groups had similar prognostic characteristics and TTP for both groups was virtually identical with a median TTP of approximately 24 months.^{20, 192} However, the OS of PANVAC-vaccinated patients was 95% at 2 years and 90% at 40 months, compared to the contemporary control group with an OS of 75% at 2 years, and 47% at 40 months. The OS values of the control group are similar to 3–5 year survival data in this population from five other clinical studies.^194–199 This is yet another example (as seen with ipilimumab, sipuleucel-T, and PROSTVAC) of no increase in TTP versus control with an increase in OS.

CEA peptide-pulsed DCs have also been employed in clinical studies using a modified CEA agonist peptide.²⁰⁰ Two of 12 patients with advanced colorectal cancers experienced tumor regression, one patient had a mixed response, and two had SD. Clinical responses correlated with the expansion of CEA-specific T cells. Twenty-six patients who had undergone resection of colorectal metastases were treated with intranodal injections of an autologous tumor lysate- and control protein-pulsed DC vaccine. Patients were randomized to receive DCs that had either been activated or not activated with CD40L. All patients were followed up for a minimum of 5.5 years. Patients with evidence of a vaccine-induced, tumor-specific T-cell-proliferative or IFN-γ response 1 week after vaccination had a markedly better RFS at 5 years (63% vs 18%, p = 0.037) than nonresponders.²⁰¹

Clinical studies have also been carried out using anti-Id antibodies directed against anti-CEA MAbs. These studies have demonstrated the induction of immune responses,^{202, 203} slowed disease progression, and prolonged survival.²⁰⁴ MVA containing the gene for an oncofetal antigen found on many tumors (5T4) has been tested in a variety of cancers (TroVax, Oxford, UK). In a trial of 22 patients with mCRC, immune responses appeared to correlate with disease control, which was seen up to 18 months in patients with immune responses.^205–207

Pancreatic carcinoma

Allogeneic whole-tumor-cell vaccines modified to secrete GM-CSF have also been employed in patients with pancreatic cancer. Evidence of vaccine-induced immune responses were observed as measured by delayed-type hypersensitivity.^{25, 208} A 60-patient Phase II study of the same vaccine in the adjuvant setting showed that post-chemotherapy induction of mesothelin-specific CD8 cells correlated with progression-free survival.²⁰⁹ The median OS was about 26 months compared with 21 months for chemotherapy alone in this same patient population at the same institution.²¹⁰ GVAX and CRS-207 are cancer vaccines that have been evaluated in pancreatic ductal adenocarcinoma (PDA). GVAX is composed of two irradiated GM-CSF-secreting allogeneic PDA cell lines administered 24 h after treatment with low-dose cyclophosphamide (Cy) to inhibit regulatory T cells. CRS-207 is recombinant live-attenuated, double-deleted Listeria monocytogenes, engineered to express mesothelin. GVAX induces T cells against a broad array of PDA antigens, and mesothelin-specific T-cell responses have been shown to correlate with survival. Mesothelin is a TAA overexpressed in most PDAs. In a prior study, patients with previously treated advanced PDA who received Cy/GVAX had better induction of mesothelin-specific CD8⁺ T cells than those treated with GVAX alone. Median survival was 4.3 and 2.3 months, respectively. Previously treated patients with metastatic pancreatic adenocarcinoma were randomly assigned at a ratio of 2 : 1 to two doses of Cy/GVAX followed by four doses of CRS-207 or Cy/GVAX alone. In a prespecified per-protocol analysis of patients who received at least three doses (two doses of Cy/GVAX plus one of CRS-207 or three of Cy/GVAX), OS was 9.7 months versus 4.6 months. Enhanced mesothelin-specific CD8⁺ T-cell responses were associated with longer OS, regardless of treatment arm.²¹¹

Breast cancer

Sialy1-Tn (STn) is a glycopeptide contained in carcinoma-associated mucin. Patients were randomized to receive vaccine [STn coupled to keyhole limpet hemocyanin (KLH)] with or without cyclophosphamide. Patients treated with vaccine and cyclophosphamide were reported to have lived significantly longer²¹² than those treated with the same vaccine without cyclophosphamide. In addition, STn KLH vaccine has been administered to breast and ovarian cancer patients after an autologous stem cell transplant following high-dose chemotherapy. Vaccinated patients appeared more likely to survive and less likely to relapse, but further studies are required.²¹³ A multicenter, double-blinded, randomized Phase III trial of STn conjugated to KLH vaccine was completed in 1028 women with metastatic breast cancer who had nonprogressive disease or no evidence of disease after first-line chemotherapy. The women treated with concomitant endocrine therapy and STn-KLH had longer TTP and OS than the control group of women who received KLH alone. Moreover, of the women who received endocrine therapy, those who had a median or greater antibody response to the STn-KLH vaccine had significantly longer median OS than those who had a below-median antibody response.²¹⁴

Several clinical trials have been carried out using MUC-1 as a target, including the use of MUC-1 peptides,^215–217 recombinant vaccinia virus encoding MUC-1 and IL-2,²¹⁸ MVA expressing human MUC-1,²¹⁹ and mannan–MUC-1 fusion protein.^{220, 221} In studies carried out with the HER2/neu peptides with or without GM-CSF, patients with breast cancer were able to generate HER2-specific immune responses.^222–224 A Phase I/II clinical trial was carried out vaccinating breast cancer patients with E75, an HLA-A2/A3-restricted HER2/neu peptide, and GM-CSF. The 24-month analysis DFS was 94.3% in the vaccinated group and 86.8% in the control group (p = .08). In subset analyses, patients who benefited most from vaccination had lymph-node-positive, HER2 IHC 1+-2+, or grade 1 or 2 tumors and were optimally dosed.²²⁵ Fusion-cell (DCs and tumor cells) vaccination of patients with metastatic breast and renal cancer was shown to induce both immunological and clinical responses.²²⁶ An anti-Id antibody that mimics the human milk-fat globule antigen has also been utilized as a breast cancer vaccine in conjunction with autologous stem cell transplantation (ASCT). In that study, the 3-year OS rate was 48%, whereas the progression-free survival rate was 32%.²²⁷ A HER2 peptide-based vaccine (nelipepimut-S) has been tested in a Phase II trial of 195 patients with localized breast cancer at high risk of recurrence. HLA-A2-/A3-positive patients received vaccine with GM-CSF and those that were not HLA-A2 or -A3 positive were controls. The 5-year DFS was 89.7% in vaccinated patients versus 80.2% in nonvaccinated patients. The 5-year DFS was 94.5% in optimally dosed patients. A Phase III study has been initiated.²²⁸

Lung cancer

Three Phase III studies in patients with non–small cell lung cancer (NSCLC) failed to meet their primary endpoints. A MAGE-A3 peptide-based vaccine was evaluated in stages IB, 2 and 3A tumor-resected NSCLC patients whose tumors expressed the MAGE-A3 gene.^{229, 230} The L-BLP25 vaccine consists of a peptide of the tandem repeat region of the MUC-1 gene. A Phase III trial employing this vaccine after chemoradiation in patients with unresectable stage 3 NSCLC did not meet its primary endpoint of increased OS.²³¹ Lucanix is a vaccine composed of four TGF-β2 antisense modified, irradiated allogeneic NSCLC cell lines. In a Phase III trial in stages IIIA, IIIB, and IV, for those NSCLC patients who did not progress after front-line chemotherapy, a considerable increase in OS was not observed in the vaccine arm; however, the time elapsed between randomization and the end of front-line chemotherapy had a significant impact on survival (p = 0.002).^{232, 233} Other vaccines in patients with NSCLC are also being evaluated. A modified vaccinia virus (MVA) encoding the MUC-1 and IL-2 genes was assessed in combination with first-line chemotherapy in patients (n = 74) with advanced NSCLC. Six-month PFS was 43.2% in the vaccine plus chemotherapy group versus 31.5% in the chemotherapy-alone group.²³⁴ A Phase II study of a telomerase peptide vaccine in NSCLC patients (n = 23) after chemoradiation has also been evaluated.²³⁵ Immune responders recorded a median PFS of 371 days versus 182 days for nonresponders.

Ovarian/cervical cancer

Clinical findings have also been reported in patients with advanced ovarian cancer using a vaccine consisting of an MAb directed against the tumor antigen CA125.²³⁶ A Phase II study examining an anti-idiotypic antibody vaccine that functionally imitated the CA125 antigen reported survival benefit; median OS of the vaccine group was 19.4 months versus 4.9 months for the control group.²³⁷

The purpose of another Phase II single-arm clinical trial was to evaluate whether pretreatment with low-dose cyclophosphamide improves immunogenicity of a p53-synthetic long peptide vaccine in patients with recurrent ovarian cancer. SD was observed in 20.0% (2/10) of patients.²³⁸ Two independent consecutive studies were conducted of combinatorial immunotherapy comprising DC-based autologous whole-tumor-antigen vaccination in combination with anti-angiogenesis therapy. Thirty-one patients had recurrent progressive stage III and IV ovarian cancer. Vaccination was well tolerated and elicited tumor-specific T-cell responses against various ovarian tumor antigens in both studies, and a clinical benefit of 65% correlated with the immune response with some experiencing prolonged PFS.²³⁹ A Phase II clinical trial of recombinant vaccinia-NY-ESO-1 (rV-NY-ESO-1) followed by booster vaccinations with recombinant fowlpox-NY-ESO-1 (rF-NY-ESO-1) in 22 epithelial ovarian cancer (EOC) patients with advanced disease who were at high risk for recurrence/progression was conducted. Median PFS was 21 months, and median OS was 48 months. CD8⁺ T cells derived from vaccinated patients were shown to lyse NY-ESO-1-expressing tumor targets. These data provide preliminary evidence of clinically meaningful benefit for diversified prime and boost recombinant poxviral-based vaccines in ovarian cancer.²⁴⁰

Studies were undertaken to determine the toxicity, safety, and immunogenicity of a human papillomavirus 16 (HPV16) E6 and E7 long peptide vaccine administered to end-stage cervical cancer patients. The HPV16 E6 and E7 long peptide-based vaccine was well tolerated and capable of inducing a broad IFN-γ-associated T-cell response even in end-stage cervical cancer patients.²⁴¹ Vulvar intraepithelial neoplasia is a chronic disorder caused by high-risk types of HPV, most commonly HPV type 16 (HPV-16). Twenty women with HPV-16-positive, grade 3 vulvar intraepithelial neoplasia were vaccinated with a mix of long peptides from the HPV-16 viral oncoproteins E6 and E7.²⁴² One-half of a group of 20 patients with HPV16-induced vulvar intraepithelial neoplasia grade 3 displayed a complete regression after therapeutic vaccination. There was a strong correlation between a defined set of vaccine-prompted specific immune responses and the clinical efficacy of therapeutic vaccination.²⁴³

Leukemia/lymphoma

The variable regions of B-cell receptor Igs on lymphomas are excellent targets for vaccines. These Ids have now been used successfully in the treatment of B-cell lymphomas.^{50, 244–247} An Id protein vaccine was used in 25 patients after a chemotherapy-induced second complete clinical remission in follicular lymphoma (FL). After vaccination, tumor-specific humoral or cellular immune responses were found in 20 of 25 patients.²⁴⁶ All of the responders with enough follow-up maintained a second CR longer than the first CR, whereas the five patients who did not mount an immune response had a second CR that was, as expected, shorter than the first CR. Vaccination with hybridoma-derived autologous tumor Ig Id conjugated to KLH and administered with GM-CSF induces FL-specific immune responses. A double-blind multicenter controlled Phase III trial of this vaccine was conducted. Of 234 patients enrolled, 177 (81%) achieved CR/CRu after chemotherapy and were randomly assigned. For the 177 randomly assigned patients, median DFS between Id-vaccine and control arms was 23.0 months versus 20.6 months, respectively (p = 0.256). For patients who received Id vaccine, median DFS after randomization was 44.2 months for Id-vaccine arm versus 30.6 months for control arm. Vaccination with patient-specific hybridoma-derived Id vaccine after chemotherapy-induced CR/CRu may prolong DFS in patients with FL.²⁴⁸

A multiple myeloma vaccine has been developed whereby patient-derived tumor cells are fused with autologous DCs, creating a hybridoma that stimulates a broad antitumor response. A Phase II trial was conducted in which patients underwent vaccination following ASCT to target minimal residual disease. Seventy-eight percent of patients achieved a best response of CR or PR and 47% achieved a CR/near CR (nCR). Remarkably, 24% of patients who achieved a partial response following transplant were converted to CR/nCR after vaccination and at more than 3 months post transplant, consistent with a vaccine-mediated effect on residual disease. The post-transplant period for patients with multiple myeloma provides a unique platform for cellular immunotherapy in which vaccination with DC/myeloma fusions resulted in the marked expansion of myeloma-specific T cells and cytoreduction of minimal residual disease.²⁴⁹ Clinical responses have also been observed using Id vaccine in patients with multiple myeloma.^{52, 250}

For patients with chronic myelogenous leukemia, vaccination with bcr-abl oncogene breakpoint fusion peptides has generated specific immune responses.²⁵¹ A study investigated the immunogenicity of Wilms tumor gene product 1 (WT1)-peptide vaccination in WT1-expressing acute myeloid leukemia (AML). Vaccination consisted of GM-CSF and WT1 peptide and KLH. AML patients (n = 17) received a median of 11 vaccinations. Objective responses in AML patients were 10 SDs including four SDs with more than 50% blast reduction and two with hematologic improvement. An additional four patients had clinical benefit after initial progression, including one complete remission and three SDs.²⁵² A Phase I/II trial investigated the effect of autologous DC vaccination in 10 patients with AML. Two patients in partial remission after chemotherapy were brought into complete remission after intradermal administration of full-length WT1 mRNA-electroporated DCs. In these two patients and three other patients who were in complete remission, the AML-associated tumor marker returned to normal after DC vaccination, compatible with the induction of molecular remission. Clinical responses were correlated with vaccine-associated increases in WT1-specific CD8⁺ T cell frequencies.²⁵³

Other tumor types

Intravesical BCG has been used successfully in the treatment of bladder cancer, implicating immune mechanisms.²⁵⁴ Intravesical administration of wild-type vaccinia virus has now been shown to be safe, and three of four patients treated were disease free at 4-year follow-up.²⁵⁵ The response of renal carcinoma to high-dose IL-2 and IFN-alpha also implicates immune mechanisms in therapeutic responses.²⁵⁶ An autologous RCC tumor-cell vaccine that was genetically modified to overexpress B7-1 to provide costimulation to tumor-reactive T cells was employed.²⁵⁷ In this single-arm Phase II study, 66 patients enrolled and 39 received at least one dose of vaccine and low-dose IL-2. Best responses were CR (3%), PR (5%), SD (64%), and PD (28%). A post hoc analysis suggested that lymphocytic infiltration of the vaccine site determined by biopsy directly correlated with survival (28.4 months versus 17.8 months, p = 0.045). A Phase I trial was conducted to evaluate the feasibility, safety, and efficacy of a DC-based vaccination in patients (n = 21) with metastatic RCC. Autologous mature DCs were pulsed with HLA-A2-binding MUC1 peptides. In six patients, regression of the metastatic sites was induced after vaccinations with three patients achieving an objective response (one complete response, two partial responses, two mixed responses, and one SD). An additional four patients were stable during the treatment for up to 14 months.²⁵⁸

Vaccination of pediatric solid tumor patients with tumor-lysate-pulsed DCs has been shown to expand specific T cells and mediated some tumor regression.²⁵⁹ In addition to DC fusion vaccination of patients with metastatic breast and renal cancer, clinical trials have examined vaccination of glioma patients with DC/glioma fusions.^{226, 260} Early clinical trials in patients with primary brain tumors have shown immune responses and objective responses with DC vaccines,²⁶¹ autologous formalin-fixed²⁶² or TGF-β-modified²⁶³ whole-tumor-cell vaccines. A pilot study of s.c. vaccinations with glioma-associated antigen (GAA) epitope peptides in HLA-A2-positive children with newly diagnosed brainstem gliomas (BSGs) and high-grade gliomas (HGGs) was undertaken. Twenty-six children were enrolled, 14 with newly diagnosed BSGs treated with irradiation and 12 with newly diagnosed BSGs or HGGs treated with irradiation and concurrent chemotherapy. Five children had symptomatic pseudoprogression, which responded to dexamethasone and was associated with prolonged survival. Only two patients had progressive disease during the first two vaccine courses; 19 had SD, two had partial responses, one had a minor response, and two had prolonged disease-free status after surgery.²⁶⁴

Considerations in the analysis of vaccine clinical trial results

It has now been demonstrated that appropriate vaccine delivery systems and vaccine strategies can elicit immune responses to TAAs in patients with a range of cancers and, in some cases, can mediate prolonged survival, prolonged disease-free interval, drops in serum tumor markers, and/or regression of metastatic disease. Patients with advanced cancers are perhaps the least appropriate population to demonstrate the efficacy of a cancer vaccine; thus, future clinical trials should be carried out in patients with less or with low tumor burden metastatic disease or earlier in the disease process. Optimal use of vaccines should occur immediately prior to, during, or immediately following adjuvant therapy. Several such studies are ongoing. The overall goal is the use of vaccine regimens in combination with front-line cancer therapies, thereby reducing the interval between disease diagnosis and the initiation of vaccine; this would bring the use of cancer vaccines closer to their original intention—use in patients with minimal disease.

Vaccination plus other immunotherapies

The recent clinical studies employing MAbs directed against the checkpoint inhibitor PD-L1/PD-1 immune suppressive axis have placed cancer immunotherapy into consideration as a mainstream component of cancer therapy. Results with checkpoint inhibitors have been most impressive in the treatment of the majority of patients with metastatic melanoma and subsets of patients with lung carcinoma.^266–271 The use of the PD-L1/PD-1 checkpoint inhibitors has also shown some activity in a minority of patients with other tumor types (e.g., bladder cancer) and extremely low levels of activity in other cancers such as colorectal and prostate cancer. The major correlate of clinical response in virtually all of these studies with the PD-L1/PD-1 checkpoint inhibitors is the presence of T cells in the tumor. It would appear that these TILs are inducing PD-L1 on tumor cells and consequently becoming anergized, thus recapitulating the process of the body’s defense against autoimmunity. The administration of an anti-PD-L1/PD-1 agent would “release the brakes” on T cells by reducing or eliminating this immunosuppressive phenomenon, leading to the activation of the TILs and subsequent tumor cell destruction (Figure 2). One potential reason for the lack of success of some tumor vaccine trials could be due to this upregulation of PD-L1 on tumor cells in response to the presence of TILs. It would thus appear that the use of a therapeutic vaccine prior to or at the same time as the use of an anti-PD-L1/PD-1 agent would induce the presence of T cells in the tumor, which would subsequently be activated by inhibiting the PD-L1 expression on the tumor, leading to tumor cell lysis (Figure 2). Carefully controlled randomized clinical studies will be required to validate this hypothesis.

Preclinical studies have shown that vaccine plus checkpoint inhibition can act synergistically. The addition of anti-CTLA4 MAb to CEA-TRICOM vaccination was shown to increase the avidity of T-cell responses, resulting in enhanced antitumor activity of the vaccine.^{138, 272} When used in combination with GVAX vaccine, blockade of both PD-1 and CTLA-4 resulted in reversal of CD8⁺ TIL dysfunction and enhanced antitumor activity.²⁷³

In a Phase I clinical study,²⁷⁴ 30 metastatic prostate cancer patients received PROSTVAC vaccine in combination with increasing doses of ipilimumab (anti-CTLA4). Fifty-eight percent of patients had PSA declines. While this trial was not randomized, median OS for patients receiving the combination was 34.4 months compared to 26.3 months for patients receiving PROSTVAC alone in another trial (Halabi-predicted survival¹⁴⁴ was similar for patients in both trials at about 18 months). A recently completed trial in mCRPC patients using ipilimumab plus radiation showed a median OS of 17.3 months. Also, no additional toxicity above that observed with ipilimumab alone was seen in the vaccine plus ipilimumab-treated patients. In another study, a DNA/peptide vaccine induced T-cell responses in melanoma patients that were “boosted” following CTLA-4 blockade.²⁷⁵ CTLA-4 MAb was also infused into nine previously immunized advanced cancer patients²⁷⁶; evidence of increased tumor immunity was seen in some melanoma and metastatic ovarian cancer patients.

There are also several ways in which vaccine and MAbs can be used in combination. These include (1) the use of therapeutic antibodies such as herceptin or rituxan in combination with vaccine, in which each acts independently, (2) the use of antitumor antibody cytokine fusion proteins such as antitumor antibody/IL-12 to enhance T-cell activity at the site of tumor,^277–281 and (3) the use of MAbs or fusion proteins directed against regulatory T cells.^282–284 Vaccines can also be employed in combination with adoptive transfer of antigen-specific T cells. In preclinical studies, point-mutated ras-specific T cells were adoptively transferred into immunodepleted hosts. Donor antigen-specific T-cell responses were greatest in recipient mice that received peptide boosts (with or without IL-2). These results indicate that a vaccine administration after T-cell transfer was more obligatory than exogenous IL-2 to sustain adoptively transferred T cells.²⁸⁵

Immunogenic cell death and immunogenic modulation

Certain cancer therapeutic regimens have the ability to trigger cancer cell death, which leads to stimulating endogenous immune responses against the tumor; this is termed “immunogenic cell death” (Figure 1).^286–293 The cardinal signs of immunogenic cell death are (1) calreticulin exposure on the surface of dying cells, (2) the release of high-mobility group box 1 (HMGB1), (3) the release of ATP and, most importantly, (4) cell death.²⁸⁹ Each of these molecules acts on DCs to facilitate the presentation of tumor antigens to the immune system.²⁸⁷ Calreticulin, a chaperone and calcium regulator, when exposed on the surface of a dying cell, serves as a phagocytic signal to DCs.^{291, 294} When HMGB1, a non-histone chromatin-binding protein, is released from dying cells, it engages TLR-4 on DCs, leading to DC maturation.²⁸⁹ The secretion of ATP by dying cells binds receptors on DCs, further supporting T-cell activation.²⁸⁸ In addition, tumor cells that survive therapy have been shown to alter their biology to render them more sensitive to immune mediated killing; this phenomenon is termed “immunogenic modulation” (Figure 1).^295–302 Immunogenic modulation encompasses a spectrum of molecular alterations in the biology of the cancer cell that independently or collectively make the tumor more amenable to CTL-mediated destruction. These include (1) downregulation of anti-apoptotic/survival genes,^{297, 298, 303} (2) modulation of antigen-processing machinery (APM) components,^{299, 302} and (3) calreticulin translocation to the cell surface of the tumor (Figure 1).^{265, 299} One can envision that these immunogenic consequences of anticancer therapy, ranging from immunogenic cell death to immunogenic modulation, can thus be harnessed to achieve optimal synergy with therapeutic cancer vaccines and other immunotherapy regimens, thereby maximizing the clinical benefit for patients receiving combination therapies.

Vaccine plus radiation

Radiation therapy is standard-of-care for multiple malignancies, aimed at direct tumor destruction. However, systemic disease, treatment resistance, or the need for sublethal dosing to minimize normal tissue toxicity, often translates into surviving tumor cell populations and disease progression.^{304, 305} Radiation can induce a continuum of immunogenic alterations in dying and/or surviving tumor cells (Figure 1). Lethal irradiation has been reported to induce immunogenic cell death.³⁰⁶ Although immune responses in cancer patients receiving radiation alone are often weak and rarely translate into protective immunity, the immunogenic effects of radiation therapy can be exploited to promote synergistic clinical benefit for patients receiving combination regimens with immunotherapy.^{304, 305, 307} It has been demonstrated that the use of relatively low doses of external beam radiation, insufficient to kill tumors, induces immunogenic modulation, altering those tumor cells to render them more susceptible to T-cell-mediated lysis. Cell-surface expression of MHC class I molecules and calreticulin on tumors was increased in a radiation-dose-dependent manner, ultimately rendering tumor cells more susceptible to T-cell-mediated killing.^{265, 298, 302} It has been demonstrated in preclinical studies that radiation combined with a cancer vaccine elicits greater tumor antigen-specific CD8⁺ T-cell responses and/or reduction in tumor burden than either modality alone.^{298, 308–310} Importantly, the induction of CD8⁺ T cells specific for multiple TAAs not encoded by the vaccine was observed after the combination therapy (antigen cascade). This polyclonal T-cell response functionally mediated the regression of TAA-negative metastases at distal s.c. or pulmonary sites.³¹⁰

These findings have translated into promising clinical findings for patients receiving radiation therapy plus immunotherapy.^{304, 305, 307, 311, 312} A Phase I study evaluated the combination of radiation therapy plus injection of an autologous DC vaccine in 14 patients with advanced-stage/metastatic hepatoma.³¹¹ Patients received a single fraction of radiation followed by two intratumoral injections of autologous immature DCs. This combination resulted in tumor-specific immune responses in 7/10 assessable patients, and six patients had objective clinical responses. Utilizing poxviral vaccines, patients in a randomized Phase II study with localized prostate cancer who were treated with radiation and a poxviral-based vaccine had a significant increase in PSA-specific T-cell responses compared to patients receiving radiation alone.³¹² In a multicenter Phase II study, patients with mCRPC (n = 44) were randomized to receive ¹⁵³Sm-EDTMP (Quadramet®, an FDA-approved radiopharmaceutical targeting bone metastasis) alone or in combination with PSA-TRICOM vaccine. TTP significantly improved (p = 0.03) with combination therapy (3.7 months) compared to ¹⁵³Sm-EDTMP alone (1.7 months).³¹³ In addition, several ongoing clinical trials are investigating this combination strategy,³¹⁴ including a Phase II study of sipuleucel-T plus radiation therapy,⁶² a Phase II study of sipuleucel-T plus stereotactic ablative body radiation,⁶³ and a pilot study of sipuleucel-T plus high-dose single-fraction radiation⁶⁴ in hormone-refractory prostate cancer.

Vaccine plus chemotherapy

If cancer vaccines are to be used early in the disease process, they would most likely need to be used in combination with certain chemotherapeutic regimens. While counterintuitive, it has recently been shown that vaccine therapy may not only be compatible with certain chemotherapies, but also may actually be synergistic (Figure 1).^{296, 315} Various chemotherapy agents have been shown to induce immunogenic modulation in tumors of diverse origin by upregulating various immune-relevant proteins on the surface of cancer cells, including multiple TAAs, calreticulin, adhesion molecules such as ICAM-1, and MHC class I proteins.^{295, 298–300, 303, 316, 317} These phenotypic changes translated into increased murine and/or human tumor sensitivity to CTL-mediated lysis in vitro after exposure to sublethal doses of chemotherapy with cisplatin/5-FU³¹⁷ docetaxel,²⁹⁹ paclitaxel²⁹⁹ and cisplatin plus vinorelbine.³¹⁷ These preclinical findings and others have translated into various hypothesis-generating clinical trials, now with encouraging preliminary data. In a randomized Phase II trial, metastatic breast cancer patients who received docetaxel combined with the therapeutic cancer vaccine PANVAC (poxvirus-based vaccine encoding the tumor antigens CEA and MUC-1, along with three T-cell costimulatory molecules; TRICOM) attained a superior PFS relative to those receiving docetaxel alone (6.6 months vs 3.8 months).³¹⁸ A recent report³¹⁹ demonstrated that the administration of docetaxel to patients with metastatic prostate or breast cancer, as well as that of cisplatin plus vinorelbine to NSCLC patients, significantly increased the ratio between effector T cells and Tregs and reduced the immunosuppressive activity of the latter in the majority of patients. These studies provide the rationale for the selective use of active immunotherapy regimens in combination with specific standard-of-care therapies to achieve the most beneficial clinical outcome among carcinoma patients. Several things are important in considering the use of chemotherapy with vaccine: (1) the combined use of vaccine and chemotherapy early in the disease process should not be confused with the use of vaccines following multiple regimens of different chemotherapeutic agents in the advanced disease setting, where the immune system would most likely be impaired, (2) not all chemotherapeutic agents will be compatible with vaccine, and (3) dose scheduling of vaccine when used with chemotherapy may be extremely important. Obviously, future studies will be required to optimize the combined use of vaccine and chemotherapy. A recent review on combining antineoplastic drugs with tumor vaccines has been published.³²⁰

Vaccine plus hormone therapy

The use of vaccines with anti-androgen therapy represents an intriguing possibility for the treatment of prostate cancer. It has previously been shown that androgen-ablative therapy induces T-cell infiltration in the prostate. T-cell infiltration was readily apparent after 1 to 3 weeks of therapy.³²¹ The immunogenic modulation potential of androgen deprivation has been described with a second-generation androgen-receptor antagonist (ARA), enzalutamide, using the murine TRAMP model of prostate carcinoma.³²² In vitro exposure of TRAMP-C2 prostate tumor cells to enzalutamide significantly enhanced cell-surface expression of Fas and MHC class I, consequently improving their sensitivity to immune-mediated lysis. These immunomodulatory properties of enzalutamide were exploited upon combination with a therapeutic cancer vaccine targeting a transcription factor associated with the metastatic process. Combination treatment significantly increased antigen-specific T-cell proliferation and OS compared to the levels observed in mice receiving no treatment or enzalutamide alone. In addition, enzalutamide or abiraterone, an inhibitor of androgen biogenesis, were both able to render androgen receptor positive LNCaP human prostate tumor cells more sensitive to T-cell-mediated lysis.³¹⁶ Findings from these and other studies provided a rationale for the use of ADT in combination with active immunotherapy. A clinical trial has been conducted in prostate cancer patients who had received prior hormonal therapy and had increasing serum PSA without radiographic evidence of metastases. The trial was randomized for a second-line hormone therapy (nilutamide) versus vaccine with a crossover at disease progression so that each arm would receive combination vaccine plus hormone therapy. Time to treatment failure and survival at 4+ years³²³ was prolonged in patients initially receiving vaccine alone and then receiving vaccine plus nilutamide. Further studies are obviously needed to validate such observations and are under way (NCT01867333, NCT01875250).

Vaccine plus small-molecule inhibitors

A number of recent studies have indicated that antiangiogenic tyrosine kinase inhibitors (TKIs) target multiple components of the tumor microenvironment and are an ideal class of agents for synergizing with cancer vaccines.³²⁴ TKIs are well known to modulate tumor endothelial cells, leading to vascular normalization; however, these agents have also been recently shown to decrease tumor compactness and tight junctions, thereby reducing solid tumor pressure and allowing for improved perfusion of collapsed vessels and increased tumor oxygenation.³²⁵ In addition, select TKIs are capable of inducing immunogenic modulation (Figure 1). The alteration of the immune landscape, direct modification of tumor cells, and improved vascular perfusion lead to improved antitumor efficacy when antiangiogenic TKIs are combined with immunotherapy. These data support the clinical combination of multitargeted antiangiogenic TKIs, including, but not limited to, cabozantinib,³²⁶ sunitinib,^{325, 327–329} and sorafenib,³²⁵ as well as to other antiangiogenic therapies, such as the anti-VEGF antibody bevacizumab, with cancer vaccines for improved treatment of solid tumors.

Dose scheduling of vaccine with other therapies

Arguably the most unique feature of cancer vaccine therapy is a vaccine’s ability to initiate a dynamic process of host immune responses that can be exploited in subsequent therapies (Table 4). In a Phase I study, patients with advanced-stage progressive cancer received a vaccine directed against cytochrome P4501B1. Most patients who developed immunity to vaccine but required salvage therapy on progression showed marked responses to their next treatment regimen; most of these responses lasted >1 year.³³¹ In another study in patients with extensive-stage small cell lung cancer,³³² a high rate of objective clinical responses to chemotherapy immediately followed vaccine therapy. These clinical responses were also closely associated with induction or augmentation of immune response to vaccine.

Source: Adapted from Ref. 330.

Three randomized clinical trials in prostate cancer provided further evidence of this phenomenon. In the first trial, patients with metastatic CRPC were randomized to receive vaccine alone or vaccine plus weekly docetaxel.³³³ Patients on the vaccine-alone arm were allowed to cross over to receive docetaxel at time of progression. After vaccine, median PFS on docetaxel was 6.1 months compared with a PFS of 3.7 months with the same docetaxel regimen and patient population at the same institution. Similar findings were observed using the sipuleucel vaccine.³³⁴ In a randomized multicenter study, patients in both the vaccine and placebo arms received docetaxel at progression. There was a statistically significant (p = 0.023) increase in OS with docetaxel in patients who had prior vaccine versus placebo.

In a Phase II trial,³³⁵ patients with nonmetastatic CRPC and rising serum PSA were randomized to receive either vaccine or nilutamide, an ARA. After 6 months, patients with rising PSA were allowed to cross over to a combination of both therapies. Median time to treatment failure was similar in the vaccine and ARA arms. However, for patients who received vaccine first and then received vaccine plus ARA, time to treatment failure was 13.9 months from the initiation of ARA and the time to treatment failure from the initiation of any therapy was 25.9 months. Of the initial randomized population, for those patients who first received nilutamide alone or nilutamide and then vaccine, 5-year OS was 38% versus a median OS of 59% for patients who first received vaccine alone or nilutamide plus vaccine.³³⁶

With traditional cytotoxic agents, it is widely believed that improved TTP is a prerequisite for improved OS. A recent study³³⁷ evaluated tumor regression and growth rates in four chemotherapy trials and one vaccine trial in patients with mCRPC. Cytotoxic agents affect the tumor only during the period of administration; soon after the drug is discontinued, due to drug resistance or toxicity, antitumor activity ceases and the growth rate of the tumor increases (Figure 3). The mechanism of action and kinetics of clinical response with vaccine therapy appears to be quite different (Table 4).³³⁷ Therapeutic vaccines do not directly target the tumor, but rather target the immune system. Immune responses often take time to develop and can potentially be enhanced by continued booster vaccinations; any resulting tumor-cell lysis can lead to cross-priming of additional TAAs, thus broadening the immune repertoire via a phenomenon known as antigen cascade or epitope spreading (Figure 4). This broader, and perhaps more relevant, immune response may also take some time to develop. Although a vaccine may not induce any significant reduction in tumor burden, vaccines as monotherapy have the potential to apply antitumor activity over a long period of time, resulting in a slower tumor growth rate (Figure 3). This deceleration in growth rate may continue for months or years and, more importantly, through subsequent therapies. This process can thus lead to clinically significant improved OS, often with little or no difference in TTP and a low rate of or lack of objective response³⁴⁰; thus, treating patients with lower tumor burden with vaccine may result in far better outcomes. It is hypothesized that the combined use of vaccine and cytotoxic therapy may result in both tumor regression (via the cytotoxic therapy) and reduced tumor growth rate (via vaccine therapy) (Figure 3).^{337, 338} Early clinical trials with vaccine may thus well have been terminated prematurely with the observance of tumor progression before sufficient vaccine boosts could be administered. This phenomenon has actually led to modifications in how vaccine clinical trials are now designed and to “new immune response criteria” for immunotherapy.³⁴¹

Antigen cascade

The phenomenon of antigen cascade (Figure 4) is now evolving as an important facet in the evaluation of cancer vaccines. This was first defined using the term epitope spreading, in which a given peptide was used as a vaccine and the host immune response post vaccination was not only directed to that epitope, but also to other epitopes of the same tumor antigen.^{308, 352, 353} This phenomenon has now been expanded using the term antigen cascade, in which a given antigen is used as a vaccine and the host immune response post vaccination is not only directed against the antigen in the vaccine but also to other antigens in the tumor. The explanation for these phenomena is that as a consequence of some tumor-cell destruction induced by the vaccine, APCs will engulf tumor fragments and then cross-present those tumor fragments to T cells, thus initiating the antigen cascade phenomenon. It has also been shown that 1/8 prostate cancer patients receiving radiation developed antibodies to an array of TAAs, while 15/33 patients receiving PROSTVAC plus radiation developed such antibodies.^{354, 355} Antigen cascade has now been observed in preclinical models³⁵⁶ and several trials have shown that antigen cascade correlated with antitumor activity.^{142, 357–361} This phenomenon of antigen cascade is also a mechanism by which the host will direct an immune response to the many mutated gene products in the tumor. Some studies have also suggested improved clinical outcomes for patients who demonstrated a broadened immune response to epitopes not found in the vaccine.^{359, 362–364}