A new peptide vaccine OCV-501: in vitro pharmacology and phase 1 study in patients with acute myeloid leukemia

Wilms’ tumor 1 (WT1) is a promising target of new immunotherapies for acute myeloid leukemia (AML) as well as for other cancers. OCV-501 is a helper peptide derived from the WT1 protein. OCV-501 induced OCV-501-specific Type 1 T-helper (Th1) responses dose-dependently and stimulated helper activity of the specific Th1 cells in peripheral blood mononuclear cells from healthy donors. OCV-501 also enhanced the increase in WT1-killer peptide-specific cytotoxic T lymphocytes. OCV-501 stimulated the OCV-501-specific Th1 clones in an HLA class-II restricted manner and formed a complex with HLA class-II protein. OCV-501-specific Th1 clones demonstrated significant OCV-501-specific cytolytic activity against OCV-501-pulsed B-lymphoblastoid cell line cells. Based on the pre-clinical results, phase 1 clinical trial was conducted. The result of this trial suggested that the subcutaneous administration of OCV-501 once weekly for 4 weeks at doses of 0.3, 1, and 3 mg in older patients with AML during complete remission was safe and well tolerated. The maximum tolerated dose was considered to be ≥3 mg. Of the nine subjects enrolled, neither relapse nor blast cells were observed during the study. Immunological responses were observed in OCV-501-specific delayed-type hypersensitivity test. This trial was registered at http://www.clinicaltrials.gov as NCT 01440920. Electronic supplementary material The online version of this article (doi:10.1007/s00262-017-1981-3) contains supplementary material, which is available to authorized users.


Introduction
Acute myeloid leukemia (AML) is the most common leukemia in older adults. Chemotherapy is a standard treatment for patients with AML and is usually divided into remission induction and consolidation therapy. Treatment efficacy and tolerability deteriorate markedly with advancing age. Complete remission (CR) rates in younger patients exceeded 70%, but declined to 50% in older patients [1]. One of the major obstacles to curing AML, particularly in older patients, is its propensity to relapse after the achievement of CR with chemotherapy or hematopoietic stem-cell transplantation [2]. Therefore, new therapeutic strategies for preventing relapse after consolidation therapy for AML are urgently needed. The graft versus leukemia effect associated with allogeneic hematopoietic stem cell transplantation strongly suggests that immunotherapy is a promising AML treatment [3,4]. Recent studies have identified several promising AML antigens as targets of immunotherapy [4]. Wilms' tumor 1 (WT1) antigen is acknowledged as a top-ranked among 75 cancer antigens [5].
The WT1 was first isolated from Wilms' tumor, a cancer of the kidney in children, as a tumor suppressor gene [6]. Extensive investigations by Oka and Sugiyama revealed that WT1 possesses oncogenic function and is strongly expressed in hematological malignancies and some solid cancers [7][8][9]. High-level expression of WT1 was an accurate predictor of poor disease-free and overall survival rates [10]. The WT1 antisense oligomers [11] and WT1-specific CTLs [12] inhibited the growth of leukemic cells without affecting normal cells, suggesting that WT1 plays an important role in leukemogenesis.
Clinical trials of cancer vaccines using synthetic WT1 peptide have been conducted in patients with AML as well as with solid tumors for more than a decade and some clinical responses and benefits have been observed [13][14][15]. The earlier generation of WT1 peptide vaccines was the HLA class I-binding short peptide (killer peptide), consisting of 8-9 amino acids, which can be easily synthesized as a drug candidate. CD8 + T cells recognize tumor-associated antigen (TAA)-derived killer peptides presented on cancer cell surfaces in association with HLA class I molecules, leading to cancer cell death [16,17]. Recently, it was reported [18,19] that the beneficial effects derived from WT1-killer peptides were short because of the induction of T cell tolerance. Repeated delivery of killer peptides led to the rapid loss of high-avidity peptide-specific CD8 + CTLs and CD4 + Type 1 T-helper (Th1) cells are required for secondary expansion and memory in CD8 + CTLs [20,21]. Therefore, to overcome poor clinical outcomes of cancer vaccination, helper peptides that elicit CD4 + Th1 cells should be considered [22,23].
OCV-501 (developed by Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) is a synthetic peptide consisting of the natural sequence derived from the WT1 gene product protein, without any modification or combination with other peptide sequences [24]. It can induce specific CD4 + Th1 cells from peripheral blood mononuclear cells (PBMC) from healthy donors, but does not directly induce CD8 + CTL with killer epitope-specificity (e.g. WT1-126, WT1-235, etc.). Therefore, OCV-501 is an HLA class IIrestricted WT1-helper peptide for therapeutic cancer vaccine. Here, we report in vitro pharmacological properties of OCV-501 and the phase 1 clinical trial in older patients with AML.

Study design
This was an open label, multi-center, phase 1 trial. A traditional 3 + 3 study design was used, with cohorts of three to six patients. If one of the three patients experienced a dose-limiting toxicity (DLT) in the cohort, up to three patients would be enrolled at the same dosage level. If two or more patients experienced a DLT, no further dose escalation would be performed and additional patients were to be enrolled at a lower dose, to confirm the maximum tolerated dose (MTD). The MTD was defined as the highest dose at which none of the first three patients or one of up to six patients of total experienced a DLT in the cohort.

Patients
Older patients (≥60 years) with AML participated in this study. The eligibility criteria were that patients must have achieved their first CR with an induction regimen and completed standard consolidation therapy, and have been identified as WT1 mRNA positive, with one of the following HLA class II types: HLA-DRB1*01:01, *04:05, *15:01, *15:02, *08:03, or *09:01. Patients with myelodysplastic syndrome apparently evolved into AML and patients with AML accompanied by t(15;17)(q22;q12), (PML/RARalpha) were excluded. Patients scheduled for bone marrow transplantation, taking immunosuppressants and adrenal cortical steroids exceeding the acceptable therapeutic doses, with autoimmune diseases or with a medical history of active autoimmune diseases, and immunocompetent patients were excluded. HLA genotyping was performed using a PCR-based typing method [27].

Drug administration
OCV-501 emulsified with Montanide ISA 51 adjuvant for injection (Seppic Inc., Paris, France) was administered subcutaneously weekly for 4 weeks (day 1, day 8, day 15, and day 22). In each cohort, the first administration of the subsequent patient was allowed only after the second administration of the first patient had been completed. End of treatment and post-treatment examinations were performed after 1 week (on day 29) and 4 weeks (on day 50) from the last administration of OCV-501, respectively. The trial consisted of 3 cohorts at a dose of 0.3 mg in cohort 1, 1 mg in cohort 2, and 3 mg in cohort 3 which is in the range of 0.1 to 10 mg/body that is generally known to have no sign of dose dependency in immunological evaluations [28]. Administration commenced with cohort 1 and progressed to cohorts 2 and 3, depending on the assessment for DLT in the preceding cohort.

Safety and efficacy assessments in patients
Safety and tolerability of OCV-501 were operationalized as adverse events, clinical laboratory test, Eastern Cooperative Oncology Group performance status, vital signs (blood pressure, pulse rate, body temperature), body weight, 12-lead ECG, pulse oximetry, chest X-ray, and observation of the administration site. These adverse events were defined as treatment-emergent adverse events (TEAEs) and determined using the National Cancer Institute Common Terminology Criteria for Adverse Events score, version 4.0. DLT was the primary endpoint and determined based on adverse events.
As the efficacy variables, outcomes were evaluated by relapse of AML when assessed according to the International Working Group response criteria [29]. To monitor the minimal residual disease, WT1 mRNA level in peripheral blood was measured on day 1 (screening) and day 29 (end of treatment) and then once a month until morphological relapse. Peripheral blood levels of WT1 mRNA were measured using a WT1 mRNA Assay Kit 'Otsuka' (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan). RNA was extracted using the QIAamp RNA Blood Mini Kit (QIAGEN GmbH, Hilden, Germany) and the PCR product level of WT1 was quantified. The cutoff value of WT1 mRNA for early detection of relapse of AML was 200-copy/μg RNA. If the value surpassed the cut-off value, the investigator judged the need for bone marrow examination based on the subject's condition.
WT1 mRNA level, delayed-type hypersensitivity (DTH), anti-OCV-501 Ab titer, and total IgG were determined as exploratory endpoints. The immune response to OCV-501 was analyzed using a DTH skin reaction test at the screening visit (baseline), on day 29, and day 31. An aqueous solution of OCV-501 (0.1 mg) for injection without Montanide was administered intra-dermally at approximately the center of the forearm flexors of the subject, and redness and induration were assessed at 48 h after administration. Anti-OCV-501 Ab titer was evaluated experimentally using anti-OCV-501 IgG (Otsuka Pharmaceutical Co., Ltd., Tokyo, Japan) in plasma by ELISA on day 1 and day 29. Total IgG was measured on day 1, day 29 and day 50 (post treatment) by immunoassay.

Statistical analysis
An ANOVA with repeated measures (crossover type) or Wilcoxon signed rank test was performed for proportion assessment between the groups. Dose dependencies were analyzed using the Dunnett test with randomized block design. A two-way ANOVA was performed on specific lysis data with the group and E/T ratio as factors. A twotailed t test was used and the level of significance was set at 0.05. SAS software (release 9.1, SAS Institute Japan) was used for all analyses. Statistical analyses in clinical study were not performed.

Ex vivo induction and activation of OCV-501-specific helper T cells
From day 7 to day 14, OCV-501-specific Th1 cells were significantly expanded in OCV-501 culture using human PBMC of 20 healthy donors as shown in Fig. 1a, but not in solvent culture (p = 0.0005). Among 18 of 20 PBMC cultures, OCV-501-specific Th1 cells were increased and showed HLA-class II restriction by blocking experiments with anti-HLA-DR Ab. However, these were blocked with neither anti-HLA-DQ Ab nor mouse IgG2a. The HLA-DR restriction was detected in 7 of 18 samples. HLA-DR-specific inhibition in a typical and representative data is shown in Fig. 1b. T-cell activation in the other 11 of 18 samples was not inhibited by anti-HLA-DR/anti-HAL-DQ antibodies. This suggests that OCV-501 would stimulate some types of its specific T-cells in HLA-DR-restricted, and others in HLA-DP-restricted manner. Significant dose-dependent T-cell activation was observed at 10 μg/mL and higher concentrations of OCV-501 in the 20 OCV-501-induced PBMC cultures (Fig. 1c). Helper activity of OCV-501 was confirmed using WT1-killer peptides. OCV-501 enhanced the increase in WT1-126-specific CTLs in 3 out of 3 samples (Fig. 1d) and WT1-235mu-specific CTLs in 6 of 8 samples (Fig. 1e) in the presence of PBMC-derived OCV-501-specific Th1 cells and WT1-killer peptide (WT1-126 or WT1-235mu)-specific CTLs.

Clinical trial
Between October 2011 and February 2013, 13 patients from 4 study sites in Japan were enrolled in this trial. Of the 13 patients from whom written informed consent was obtained, 4 were specified as screen failures (2 patients did not match the HLA types in inclusion criteria, 1 patient had Grade 3 lab test abnormality and 1 patient had a relapse of AML). Nine evaluable patients were enrolled in this study ( Table 2). The median age was 70 years (range, 62─74 years). Chemotherapy varied among these patients, however, all had completed the planned AML therapy at the time of vaccination and were in first CR according to standard criteria. All had evidence of a measurable WT1 transcript at screening. Each cohort consisted of 3 subjects and all subjects completed the planned 4 vaccinations with either a 0.3, 1 or 3 mg dose of OCV-501.

Safety assessment
Incidence of TEAEs is summarized in Table 3. There were neither deaths nor serious TEAEs during the treatment period. None of the subjects discontinued OCV-501 administration due to TEAE. Grade 3 TEAEs included lymphocyte count decreased and neutrophil count decreased (1 subject each) in the OCV-501 1-mg cohort and thrombocytopenia (1 subject) in the OCV-501 3-mg cohort. These events were considered unrelated to OCV-501. The incidence of drug-related TEAEs by system organ class and preferred term are presented in Table 3. Injection site reactions, including erythema, induration, mass, pain, and pruritus were observed in all subjects. However, all drugrelated TEAEs were Grade 1 or 2 in severity. No DLT was observed in any of the 3 cohorts and the MTD was considered to be ≥3 mg. This suggests that OCV-501 is safe and tolerable.

Efficacy outcomes
None of the 9 subjects with AML had a relapse from the time of screening (day-14 to day-1) to the time of end-oftrial (day 29). In the blood smear examination, no blast cells were observed in any of the 9 subjects. The percentage of myeloblasts in the 9 subjects was less than 5% before and after OCV-501 vaccination (Supplementary Table 1). Expressed levels of WT1 mRNA at screening (day 1) and at the end of treatment (day 29) were <50-950 and <50-2400 copies/µg RNA, respectively (Table 4). Decreases in WT1 mRNA were found in 2 subjects in the OCV-501 1-mg cohort and 2 subjects in the 3-mg cohort. All subjects were tested for a DTH response. Redness with induration of >5 mm diameter was confirmed in 1 subject each in the 0.3-mg cohort and 3-mg cohort and redness without induration was observed in 2 subjects in the 0.3 mg cohort and 1 subject each in the 1-mg cohort and 3-mg cohort (Supplementary Table 2). Anti-OCV-501 IgG was negative in all subjects at end of treatment (day 29) ( Table 4). Total IgG levels at screening (day 1), end of treatment (day 29), and post treatment (day 50) were 1097-1499, 1041-1601, and 1106-1725 mg/dL, respectively (Supplementary Table 3).

Discussion
OCV-501 is a synthetic helper peptide, which consists of the natural sequence derived from the WT1 protein [24]. It was demonstrated that OCV-501 significantly induced OCV-501-specific Th1 cells in PBMC from 20 healthy donors. Using both the induced primary Th1 cells and the established Th1 clones, OCV-501 also activated the specific Th1 cells in a dose-dependent manner. Furthermore, OCV-501 increased WT1-killer peptide-specific CTLs in the presence of OCV-501-specific Th1 cells. OCV-501-specific Th1 clones demonstrated significant OCV-501-specific cytolytic activity against OCV-501pulsed B-LCL cells. In addition, reports have shown that WT1 332 (a WT1 helper peptide that has the same amino acid sequence as OCV-501)-specific Th1 clone and -specific TCR-transduced CD4 + T cells were able to respond to WT1-transfected B-LCL and kill WT1 + leukemic cells, respectively [24,30,31]. It has been reported that HLA-DR molecules are highly expressed in AML cells [32]. Therefore, these pre-clinical results suggest that OCV-501 activates both direct and indirect antitumor (anti-leukemic) cellular immunity, including specific cytotoxic Th1 cells and WT1-peptide-specific CTL cells, however, it is necessary to confirm the cytolytic activity of OCV-501-specific Th1 cells against several AML-derived leukemic cell lines.
HLA class II-restriction of helper peptides is generally not as strict as the HLA class I-restriction of killer peptides, since helper peptides can bind to a variety of HLA class II molecules with loose recognition [33]. In this     [34]. This suggests that OCV-501 can be used not only for Japanese patients but also for worldwide cancer patients. Much attention has been paid to the development of helper peptides [22,23], since helper peptide vaccination induces and activates cancer-specific Th1 cells, which are the control tower for cancer immunity, and increases the antitumor effect by inducing the antibodies that stimulate proliferation and activation of CTLs, activation of APCs, and incorporation of cancer cells through IL-2 and IFN-γ production. Several approaches have been investigated to potentiate the clinical responses of killer peptide-based vaccines [35,36]. Mixture of killer and helper peptides and a long helper peptide containing class I and class II epitope regions demonstrated to respond to both CD4 + and CD8 + T cells [17,[37][38][39][40][41][42]. Potential clinical benefits of the multiple peptide vaccine were also observed in survival curves in patients with AML compared with unvaccinated controls [40]. These observations suggest that the multiple peptides including killer/helper epitopes seems to be one of the promising strategies for activating both CD4 + Th1 and CD8 + CTL responses. On the other hand, vaccination with a mixture of 6 HLA-DR-restricted melanoma helper peptides induced both specific Th1-dominant CD4 + T cell responses and Ab responses, associated with improved overall survival among patients with metastatic melanoma [43,44]. The survival strongly correlated with early Ab response and/or with early T-cell response [44]. There was a hierarchy of immunodominance of helper peptides and no evidence that peptide length or the type of source tumor antigen predicted immunodominance. These results would reveal that only one single immunodominant helper peptide can elicit both specific Th1-dominant CD4 + T cell responses and Ab responses, suggesting that helper peptide would be a potent cancer vaccine without combination with killer and/or helper peptides [45,46]. Additionally, easy formulation of a single helper peptide rather than multiple peptides vaccine should be an additional advantage of OCV-501 from drug development point of view. Also, OCV-501, single WT1-helper peptide, had helper peptide activities including wide-range HLA-class II restrictions and cytolytic CD4 + T cell-induction which might be stronger than that of the other TAA-helper peptides.
We conducted an open label, multi-center, phase 1 trial of OCV-501 to evaluate the safety and tolerability in older AML patients in CR. Nine patients were enrolled and all completed the study. All doses of OCV-501 administered subcutaneously 4 times every 4 weeks were well-tolerated and safe, and the MTD was considered to be ≥3 mg. Injection site reactions were observed in all patients at all dosage levels. These might have been due to the Montanide used as an adjuvant, the frequent side effects of which are well known to include inflammatory reactions, granulomas and ulcers at the injection site [47]. Clinical efficacies and immunological responses were unconfirmed in this clinical study. However, no patients relapsed during the study. All the patients provided informed consent to continue further vaccination with OCV-501 in an extension trial to evaluate the safety and efficacy of continuous administration of OCV-501. Immune response in DTH with induration of >5 mm diameter was observed in 2 patients, suggesting that these might be clues to clinical efficacy due to OCV-501 vaccination. Anti-OCV-501 IgG formation was found in rats and dogs in pre-clinical studies. Although anti-OCV-501 Ab was not detected during the study, this appeared in some patients in the follow-up period after the study (data not shown). These data might also support immune responses to OCV-501 suggesting the rationale to use the dose range of this study in future studies.
In conclusion, OCV-501, a WT1 helper peptide, induced OCV-501-specific Th1 responses dose-dependently and stimulated helper activity of the specific Th1 cells in PBMC from healthy donors in an HLA class II-restricted manner. OCV-501-specific Th1 clones showed significant OCV-501-specific cytolytic activity against B-LCL cells. In this phase 1 clinical trial, OCV-501, administered subcutaneously once a week for 4 weeks to older patients with AML, was well-tolerated and safe with a considerable MTD of ≥3 mg. Further clinical studies of OCV-501 in patients with AML should be considered to confirm its safety, and efficacy.