• Open Access

First clinical trial of cancer vaccine therapy with artificially synthesized helper/ killer-hybrid epitope long peptide of MAGE-A4 cancer antigen


  • This study was approved by the Institutional Review Boards of Hokkaido University School of Medicine and Institute for Genetic Medicine, Hokkaido University (Hokkaido, Japan) and registered at UMIN Clinical Trials Registry (http://www.umin.ac.jp/ctr/index-j.htm) as UMIN000003489 (Cancer vaccine study with MAGE-A4/Survivin helper peptide). Informed consent for publication was obtained from the patient.

To whom correspondence should be addressed.
E-mail: tak24@igm.hokudai.ac.jp


A patient with pulmonary metastasis of colon cancer was treated with artificially synthesized helper/killer-hybrid epitope long peptide (H/K-HELP) of MAGE-A4 cancer antigen. The patient was vaccinated with MAGE-A4-H/K-HELP combined with OK432 and Montanide ISA-51. There were no severe side-effects except for a skin reaction at the injection site. MAGE-A4-H/K-HELP induced MAGE-A4-specific Th1 and Tc1 immune responses and the production of MAGE-A4-specific complement-fixing IgG antibodies. Tumor growth and carcinoembryonic antigen tumor marker were significantly decreased in the final diagnosis. This is the first report that artificially synthesized MAGE-A4-H/K-HELP induces Th1-dependent cellular and humoral immune responses in a human cancer patient. (Cancer Sci 2012; 103: 150–153)

Case report

First clinical findings in one patient treated with an artificially synthesized MAGE-A4-helper/killer-hybrid epitope long peptide (H/K-HELP) cancer vaccine.

A 62-year-old Japanese man had a history of pulmonary metastasis and resected colon cancer without active infections (human immunodeficiency virus [HIV], hepatitis B virus [HBV] and hepatitis C virus [HCV]), severe heart diseases (New York Heart Association class 3 or 4) and autoimmune diseases (scleroderma, Sjogren’s syndrome, idiopathic thrombocytopenic purpura, multiple sclerosis and rheumatoid arthritis). The patient was enrolled in this trial and confirmed MAGE-A4 antigen expression and human leukocyte antigen (HLA)-phenotypes (HLA-A*310102, A*330301, and DRB1*150101).

The pulmonary metastatic colon cancer of the patient was shown to express MAGE-A4 cancer antigen by both immunohistological analysis and real-time PCR (Fig. 1a–e). MAGE-A4-H/K-HELP was artificially synthesized by conjugating MAGE-A4278–299 helper epitope with MAGE-A4143–154 killer epitope by a glycine linker (Fig. 1f). In a phase I study, the enrolled patients were treated with subcutaneous injection of MAGE-A4-H/K-HELP (1 or 10 mg) combined with OK432 and Montanide ISA-51 four times at 2-week intervals (Fig. 1g). In the present case, the patient was first treated with 1 mg MAGE-A4-H/K-HELP four times at 2-week intervals and thereafter treated with 10 mg dose-escalated MAGE-A4-H/K-HELP six times. Physical and hematological examinations were monitored before and after vaccination with MAGE-A4 H/K-HELP. As a result, it was confirmed that there were no severe side-effects, although a skin reaction occurred at the injection site (Grade 2 reaction, Common Terminology Criteria for Adverse Events [CTCAE] v4.0 criteria). The isolated CD4+ T cells from the patient’s PBMC did not produce significant levels of interferon (IFN)-γ by stimulation with H/K-HELP in the presence of antigen presenting cells (APC) before vaccination. However, the levels of IFN-γ-production by CD4+ T cells greatly increased in the patient after vaccination with 1 mg MAGE-A4-H/K-HELP (Fig. 2a), and the levels of Th1 response increased approximately 10 times after vaccination with 10 mg MAGE-A4-H/K-HELP (data not shown). The numbers of CD8+ Tc1 cells detected by IFN-γ-enzyme-linked immunosorbent spot (ELISPOT) assay also increased in the patient’s PBMC after 1 mg MAGE-A4-H/K-HELP vaccination (Fig. 2b), and the increased Tc1 response was also observed after 10 mg vaccination (data not shown). We determined that the epitope recognized by CD4+ T cells was exactly our identified helper epitope (MAGE-A4278–299) in H/K-HELP, while the CD8+ T cells recognized an unknown new killer epitope in a part of the helper epitope sequence of H/K-HELP (data not shown). Thus, MAGE-A4-H/K-HELP induced both Th1 and Tc1 responses in the patient irrespective of the patient expressing HLA DR1501 bound to MAGE-A4278–299 helper epitope but not HLA A24 bound to MAGE-A4143–154 killer epitope in MAGE-A4-H/K-HELP. Moreover, our vaccine protocol with MAGE-A4-H/K-HELP mainly induced MAGE-A4-peptide-specific IgG3 antibody (Ab) and slightly induced IgG1 Ab, both of which are Th1-dependent complement-fixing Ab (Fig. 2c,d). The levels of MAGE-A4-specific IgM also slightly but significantly increased at an early stage of the 1 mg vaccination, but the levels did not change after the 10 mg vaccination. In contrast to IgM, the levels of MAGE-A4-peptide-specific IgG3 and IgG1 Ab greatly increased after the 10 mg vaccination (Fig. 2d). However, no increase in IgG2 and IgG4 was observed after vaccination with H/K-HELP. We demonstrated that MAGE-A4-specific IgG Ab recognized MAGE-A4143–154 killer epitope in H/K-HELP (data not shown). Thus, it was demonstrated that the MAGE-A4278–299 helper epitope exactly stimulated CD4+ T cells and a part of the helper epitope peptide also activated CD8+ T cells, while the MAGE-A4143–154 killer epitope triggered the production of MAGE-A4-specific IgG Ab, indicating the long peptide vaccine, H/K-HELP, appeared to be beneficial for inducing both cellular and humoral immune responses in the cancer patient. In parallel with superior immune responses, tumor growth and serum levels of the carcinoembryonic antigen tumor marker were slightly decreased during cancer vaccine therapy with MAGE-A4-H/K-HELP (Fig. 3), and this patient’s clinical response was finally judged as stable disease (Response Evaluation Criteria in Solid Tumors [RECIST] v1.1 guidelines(1)). This is the first report that cancer vaccine therapy with artificially synthesized MAGE-A4-H/K-HELP induced Th1-dependent cellular and humoral immune responses in a cancer patient. So far, no severe side-effects were observed in six patients treated with MAGE-A4-H/K-HELP (three patients with 1 mg and three patients with 10 mg, CTCAEv4.0), and significant T-cell responses were demonstrated in 50% of patients (data not shown).

Figure 1.

 Treatment of a patient with pulmonary metastasis of colon cancer with MAGE-A4-H/K-HELP. Immunohistochemical staining with anti-pan MAGE-A4 mAb (57B) for (a) MAGE-A4 bladder cancer (negative control), (b) head and neck cancer (positive control), which revealed similar staining intensity to testis, and (c, d) a tumor tissue sample of the enrolled patient. (e) MAGE-A4 mRNA expression levels were assessed by real-time PCR. Normal colon and testis were used as a negative or positive control, respectively. (f) The amino acid sequence of artificially synthesized MAGE-A4-H/K-HELP used in this trial. MAGE-A4-H/K-HELP was artificially synthesized by conjugating MAGE-A4278–299 helper epitope (22 amino acid sequence in the right box) with MAGE-A4143–154 killer epitope (12 amino acid sequence in the left box) by a glycine linker. (g) Vaccination protocol for the patient in a phase I study. The patient was vaccinated with 1 or 10 mg MAGE-A4-H/K-HELP mixed with OK-432 (0.02KE) and Montanide ISA-51 four times at 2-week intervals. H/K-HELP, helper/killer-hybrid epitope long peptide; HLA, human leukocyte antigen.

Figure 2.

 Cellular and humoral immune responses in a patient vaccinated with MAGE-A4-H/K-HELP. (a,b) MAGE-A4-H/K-HELP-specific interferon (IFN)-γ production by CD4+ T cells or CD8+ T cells was assessed at various weeks after vaccination with MAGE-A4-H/K-HELP (1 mg) by ELISA (a) or enzyme-linked immunosorbent spot (ELISPOT) assay (b), respectively. *P < 0.05, against control survivin-H/K-HELP peptide. (c) The levels of antibodies (total IgG, •; IgG1, ○; IgG2, bsl00066; IgG3, △; IgG4, bsl00001; and IgM, □) against MAGE-A4-H/K-HELP peptide in the patient’s serum sample (400-fold diluted) was determined at 2, 4, 6 and 8 weeks after vaccination with 1 mg MAGE-A4-H/K-HELP. *P < 0.05, against 0 weeks after vaccination. (d) The levels of antibodies (total IgG, •; IgG1, ○; IgG2, bsl00066; IgG3, △; IgG4, bsl00001; and IgM, □) against MAGE-A4-H/K-HELP peptide in the patient’s serum sample (400-fold diluted) was determined at 2, 4, 6, 8, 10 and 12 weeks after vaccination with 10 mg MAGE-A4-H/K-HELP. H/K-HELP, helper/killer-hybrid epitope long peptide.

Figure 3.

 Clinical responses in a colon cancer patient treated with MAGE-A4-H/K-HELP vaccine. (a) The carcinoembryonic antigen (CEA) value in the patient’s serum was measured using ELISA on various days after vaccination. (b) Maximum tumor diameter was measured by computer tomography. H/K-HELP, helper/killer-hybrid epitope long peptide.


Cancer vaccine therapy using HLA class I-binding 8–9 amino acids short cancer peptide has been performed to induce cancer-specific cytotoxic T lymphocytes (CTL) in cancer patients.(2,3) Although cancer vaccine therapy with short peptide induced increased tetramer+ cancer-specific CTL and long stable disease, the vaccine therapy focused only on CTL activation appears to be suboptimal in conquering cancer.(4–6) This might be because of the existence of strong immunosuppressive tumor escape mechanisms and the lack of helper T-cell activation.(7,8)

Therefore, it is essential to develop an efficient method to overcome immunosuppression to combat cancer. We initially demonstrated the critical role of Th1 and Th2 immunity in the tumor-bearing host and have proposed that the introduction of Th1-dominant immunity is essential for inducing fully activated CTL and immunological memory.(8–11) Recently, it has been demonstrated that a mixture of various synthetic long peptides (SLP) derived from the naturally occurring sequence of human papilloma virus (HPV)-16 oncoproteins was superior to short tumor peptides in terms of inducing a complete or partial response in vulvar intraepitheral neoplasia.(12) Zwaveling et al.(13) also reported that HPV-16-derived 35 amino acid-long peptide eradicated the established HPV-16-expressing mouse tumor. Thus, long peptide vaccine containing both helper and killer epitopes appeared to be a rational strategy to activate Th1-dependent antitumor immunity.(8) In contrast to viral-related cancer antigenic long peptide, p53-long peptide vaccine induced no complete or partial response in a clinical trial of human cancers, although it induced significant T cell responses.(14,15) Moreover, the first clinical trial using a synthetic 15 amino acid peptide vaccine containing a naturally occurring combination of helper and killer epitopes of gp100175–189 exhibited no significant impact for therapeutic efficacy of melanoma.(16) Therefore, what kinds of long peptide induce a beneficial therapeutic effect against human cancer still remains unclear.

Here, we prepared an artificially synthesized long peptide, which conjugate MAGE-A4 class I-binding epitope and our defined helper epitope,(17) and applied it to a patient with pulmonary metastatic colon cancer. In contrast to short gp100175–189 peptide including helper and killer epitopes,(16) we successfully induced cancer-specific Th1/Tc1 cells and complement-fixing Ab (IgG1 and IgG3)(18) by 40 amino acid H/K-HELP. This discrepancy might be because artificially synthesized 40 amino acid-long peptide but not short peptide has a beneficial structure for inducing favorable dendritic cells (DC) presentation and subsequent activation of Th1 and Tc1 cells.

Bijker et al.(19) reported that SLP of class I-binding long cancer peptide was efficiently processed by professional APC and subsequently exhibited a sustained stimulating activity of DC to induce Th-dependent, tumor-specific CTL.(19) We have also confirmed that ovalbumin (OVA)-H/K-HELP was superior to short peptide in curing mice with OVA-expressing tumor (data not shown). In another clinical trial, we demonstrated that survivin-H/K-HELP induced a complete response in a breast cancer patient (data not shown). Thus, we believe that artificially synthesized H/K-HELP of cancer antigen will become a promising tool to induce Th1-dependent cellular and humoral immunity in cancer patients, as well as SLP derived from natural tumor-associated HPV-16 antigen peptide.(12)


The authors thank the patient for consenting to the publication of their clinical details. The authors are also grateful to the clinical research coordinators of Hokkaido University Hospital for their support with the treatment of the patient. This study is part of a “Development of Innovative Cancer Immunotherapy by Helper T cells”, sponsored by a Translational Research Promotion Project of the New Energy and Industrial Technology Development Organization (NEDO).

Disclosure Statement

All authors declare that they have no conflict of interest.