The characteristics of cancer cells and immunological status against cancers differ widely among patients, even among those with the same histological types of cancer and identical MHC types. Subsequently, several types of personalized cancer vaccines are currently undergoing clinical trials. The first type uses autologous tumors or their components as vaccine sources, but no clear data showing clinical benefits for cancer patients have yet been reported.(20,21) In the second type of personalized vaccine, antigens or peptides are chosen by measuring their expression in autologous tumors. However, because the acquisition of appropriate cancer specimens is often very difficult, this type of protocol is not used widely.
The third type of vaccine attempts to account for the immunological diversity of individual patients. In this protocol, CTL precursors are measured in prevaccination PBMC, and then peptides for which higher CTL precursor frequencies are detected are administered (personalized peptide vaccine). As a translational clinical study, we conducted a series of clinical trials of personalized peptide vaccines for advanced cancer patients.(8–10,12,22–26) Pre-vaccination PBMC and plasma were provided to examine cellular and humoral immune responses to 14–25 or 16–23 peptide panels in HLA-A24+ or HLA-A2+ patients, respectively. The positively identified peptides (up to four per patient) were subsequently chosen for vaccination. One representative case of a patient with advanced HRPC is given in Fig. 4. Each of the 16 peptides was investigated for reactivity to prevaccination PBMC and plasma. PSA-248, PTHrP-102 and Lck-486 peptides were recognized by both the CTL (cut-off level: 50 ng/mL of peptide-specific interferon-γ production by the peptide-stimulated PBMC) and IgG (cut-off level: 5 fluorescence intensity units of peptide-specific IgG), whereas the SART3-109 peptide was recognized by IgG at the highest level. Therefore, these four peptides were administered to the patient as vaccines. The same assay was repeated using the postvaccination (6th) vaccination to evaluate immune responses to the vaccinated peptides. The scale of the horizontal line was changed due to the augmentation of peptide-specific CTL activity (two-fold) and the increased levels of antibody (IgG) reactive to the peptide (200-fold). Post-vaccination PBMC showed increased levels of CTL activity in response to SART3 and Lck peptides, whereas postvaccination plasma showed increased levels of IgG reactive to PSA-248 (750-fold increase), PTHrP-102 (730-fold) and SART3-109 (67-fold) peptides. As shown in this representative case, the personalized peptide vaccination achieved activation of both cellular and humoral responses to at least one of the four vaccinated peptides in the majority of patients, with definite clinical benefits for malignant gliomas and cervical cancers.(9,10) Kinetic studies of magnetic resonance imaging of four partial response (PR) cases with advanced malignant glioma are shown in Fig. 5.(10) An overall summary of the clinical responses of our personalized peptide vaccinations is given in Table 1. The clinical responses of advanced solid cancers (n = 72) to the personalized peptide vaccination consisted of eight PR (11.1%), 28 stable disease (SD) (38.9%) and 36 progressive diseases (PD) (50%), with an overall disease control (PR + SD) rate of 50%. There was no objective response for advanced lung, stomach or pancreatic cancer patients, or for metastatic melanoma patients, whereas the objective response was seen in advanced malignant glioma patients (two cases of grade 3 and three cases of grade 4 glioma), uterine cervical cancer and colon cancer patients. In contrast, the clinical responses of advanced solid cancers (n = 38) to our non-personalized peptide vaccination consisted of no PR, nine SD and 29 PD, with an overall disease control rate as low as 23.7% (Table 1). The overall survival of patients with advanced cervical or gastric cancer (scirrhous types) who received personalized peptide vaccination was significantly longer than that of these patients (Fig. 6).(27) In addition, measurement of cellular and humoral responses to the vaccinated peptides in the postvaccination samples was well correlated with the clinical responses if the patients received personalized peptide vaccination, but not if they received classical (non-personalized) types of peptide vaccination. For example, both CTL and IgG responses specific to the vaccinated peptides in the postvaccination samples were activated in all PR cases, a CTL response alone was activated in all SD cases, and no response was seen in all PD cases (Fig. 7).(10) Both the classical and personalized peptide vaccines were well tolerated, with the major adverse effects being a grade 1 or 2 inflammatory skin reaction at the injection site.
Figure 4. Screening of peptides and evaluation of immune responses. Pre-vaccination peripheral blood mononuclear cells (PBMC) and plasma were provided to examine cellular and humoral immune responses to 16 peptide panels in a HLA-A24+ patient with advanced hormone refractory prostate cancer (HRPC). Namely, each of the 16 peptides was investigated for reactivity to prevaccination PBMC and plasma. PSA-248, PTHrp-102 and LCk-486 peptides were recognized by both the cytotoxic T lymphocytes (CTL) (cut-off level: 50 ng/mL of peptide-specific interferon-γ production by the peptide-stimulated PBMC) and IgG (cut-off level: 5 fluorescence intensity units of peptide-specific IgG), whereas SART3-109 peptide was recognized by IgG at the highest level. Therefore, these four peptides administered to the patient as vaccines. The same assays were conducted using the samples after the sixth vaccination to evaluate immune responses to the vaccinated peptides. The scale of the horizontal line was changed due to the augmentation of peptide-specific CTL activity (two-fold) and IgG reactive to the peptide (200-fold).
Download figure to PowerPoint
Collectively, these results indicate that the personalized peptide vaccine is superior to the classical type of peptide vaccine for advanced solid cancers, but it is not sufficient to provide definite clinical benefits for the majority of advanced solid cancers, with the exception of malignant gliomas or uterine cervical cancers.