A randomized phase II clinical trial of personalized peptide vaccination with metronomic low-dose cyclophosphamide in patients with metastatic castration-resistant prostate cancer

Abstract

This study investigated the effect of metronomic cyclophosphamide (CPA) in combination with personalized peptide vaccination (PPV) on regulatory T cells (Treg) and myeloid-derived suppressor cells (MDSC), and whether it could improve the antitumor effect of PPV. Seventy patients with metastatic castration-resistant prostate cancer were randomly assigned (1:1) to receive PPV plus oral low-dose CPA (50 mg/day), or PPV alone. PPV treatment used a maximum of four peptides chosen from 31 pooled peptides according to human leukocyte antigen types and antigen-specific humoral immune responses before PPV, for 8 subcutaneous weekly injections. Peptide-specific cytotoxic T lymphocyte (CTL) and immunoglobulin G responses were measured before and after PPV. The incidence of grade 3 or 4 hematologic adverse events was higher in the PPV plus CPA arm than in the PPV alone arm. Decrease in Treg and increase in MDSC were more pronounced in PPV plus CPA treatment than in PPV alone (p = 0.036 and p = 0.048, respectively). There was no correlation between the changes in Treg or MDSC and CTL response. There was no difference in positive immune responses between the two arms, although overall survival in patients with positive immune responses was longer than in those with negative immune responses (p = 0.001). Significant differences in neither progression-free survival nor overall survival were observed between the two arms. Low-dose CPA showed no change in the antitumor effect of PPV, possibly due to the simultaneous decrease in Treg and increase in MDSC, in patients under PPV.

Introduction

It has been recognized that tumor-associated immunosuppression contributes significantly to tumor progression and resistance to immunotherapies [1]. Regulatory T cells (Treg) are known to suppress the activation of T cells. Increased levels of Treg have been detected in the peripheral blood mononuclear cells (PBMC), tumor microenvironment, draining lymph nodes of patients with prostate cancer [2, 3], other solid tumors [4–6], and hematologic malignancies [7]. Myeloid-derived suppressor cells (MDSC) are also associated with immunosuppression via several mechanisms, including defective dendritic cell function and activation of Treg [8]. Circulating MDSC are increased in some cancer patients and correlated with advanced clinical stages [9].

Cancer vaccines without consideration of the host immune cell repertoires cannot efficiently induce beneficial antitumor immune responses. Thus, the new concept of personalized peptide vaccination (PPV) was developed [10, 11]. In PPV treatment, appropriate peptide antigens for vaccination are screened, and up to 4 peptides are selected as vaccine candidates for each patient. We have previously reported that PPV treatment increased cellular and humoral immune responses and decreased prostate-specific antigen (PSA) levels in some patients with metastatic castration-resistant prostate cancer (mCRPC) [12–15]. However, PPV mono-therapy may be unable to produce an immune response of sufficient potency to induce tumor regression, when immune tolerance and a large tumor burden are present.

Cyclophosphamide (CPA) is a bifunctional alkylating agent, and continuous low-dose (metronomic) CPA can show enhanced immune responses against a variety of antigens with depletion of Treg [16], as well as the inhibition of angiogenesis by up-regulating the endogenous angiogenesis inhibitor, thrombospondin-1, in tumor and perivascular cells [17]. These findings prompted us to conduct a phase II randomized trial of the combination of PPV with metronomic low-dose CPA in patients with mCRPC, to investigate how metronomic CPA may affect Treg and MDSC and whether it could improve the antitumor effect of PPV.

Patients and methods

Patient selection

The protocol was approved by the Kurume University Ethics Committees and was conducted in an outpatient setting at the Cancer Vaccine Center of Kurume University in Japan (clinical trial registration, UMIN000005329). All patients were Japanese and provided written informed consent before participating in the study. Eligible patients had histologically confirmed prostate cancer; were men aged ≥18 years; had Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1; at least one metastatic lesion by radiographic imaging [bone scan or computed tomography (CT)]; and had progressive disease (PD) defined as more than two consecutive increases in serum PSA levels, new metastatic bone lesion, or progressive objective lesion on CT, despite adequate androgen deprivation therapy (ADT). Other inclusion criteria were adequate bone marrow function, hepatic function, renal function, and a positive status for HLA-A2, HLA-A24, HLA-A3, HLA-A11, HLA-A31, or HLA-A26, as well as at least two positive immunoglobulin G (IgG) responses to the 31 candidate peptides before vaccination. Patients continued ADT during the study. Patients with the following were excluded: an acute infection; a history of severe allergic reactions; pulmonary, cardiac, or other systemic diseases; and other inappropriate conditions of enrollment, as judged by the clinicians.

Study design and treatment

This was an open-label, randomized phase II trial of PPV plus oral low-dose CPA in patients with mCRPC. Patients were randomly assigned at a 1:1 ratio to either PPV plus low-dose CPA (study arm) or PPV alone (control arm), using a minimization technique with the following stratification factors: age (<70 or ≥70 years old), PS (0 or 1), and baseline PSA (<100 or ≥100 ng/ml). The primary endpoint was the depletion of immunosuppressive cells including Treg and MDSC, by combination of oral low-dose CPA with PPV treatment. The secondary endpoints included overall survival (OS), progression-free survival (PFS), safety, and immune responses.

In the PPV plus low-dose CPA arm, patients were orally administered 50 mg of CPA (Endoxan^®, Shionogi & Co., Ltd., Japan) once a day for 49 days (7 weeks) during PPV treatment. Patients in both arms received 8 doses of PPV at 1-week intervals (1 cycle) followed by 8 doses at 2- or 3-week intervals until unacceptable toxicity or withdrawal of consent. Under PPV treatment, 2–4 peptides were selected by the levels of IgG titers specific to 31 HLA-matched candidate peptides (Table 1). Each peptide was emulsified in a 5-ml plastic syringe with Montanide ISA 51 Incomplete Freund’s Adjuvant (Seppic, Paris, France), and selected peptides (up to 4) in 1.5 ml of emulsion (3 mg/peptide) were injected subcutaneously.

Table 1.

Peptide candidates for personalized peptide vaccination

Peptide	Amino acid sequence	Protein of origin	HLA type
CypB129–138	KLKHYGPGWV	Cyclophilin B	A2, A3sup
Lck246–254	KLVERLGAA	p56 lck	A2
Lck422–430	DVWSFGILL	p56 lck	A2, A3sup
MAP432–440	DLLSHAFFA	ppMAPkkk	A2, A26
WHSC2103–111	ASLDSDPWV	WHSC2	A2, A3sup, A26
HNRPL501–510	NVLHFFNAPL	HNRPL	A2, A26
UBE43–51	RLQEWCSVI	UBE2 V	A2
UBE85–93	LIADFLSGL	UBE2 V	A2
WHSC2141–149	ILGELREKV	WHSC2	A2
HNRPL140–148	ALVEFEDVL	HNRPL	A2
SART3302–310	LLQAEAPRL	SART3	A2
SART3309–317	RLAEYQAYI	SART3	A2
SART293–101	DYSARWNEI	SART2	A24
SART3109–118	VYDYNCHVDL	SART3	A24, A3sup, A26
Lck208–216	HYTNASDGL	p56 lck	A24
PAP213–221	LYCESVHNF	PAP	A24
PSA248–257	HYRKWIKDTI	PSA	A24
EGFR800–809	DYVREHKDNI	EGF-R	A24
MRP3503–511	LYAWEPSFL	MRP3	A24
MRP31293–1302	NYSVRYRPGL	MRP3	A24
SART2161–169	AYDFLYNYL	SART2	A24
Lck486–494	TFDYLRSVL	p56 lck	A24
Lck488–497	DYLRSVLEDF	p56 lck	A24
PSMA624–632	TYSVSFDSL	PSMA	A24
EZH2735–743	KYVGIEREM	EZH2	A24
PTHrP102–111	RYLTQETNKV	PTHrP	A24
SART3511–519	WLEYYNLER	SART3	A3sup
SART3734–742	QIRPIFSNR	SART3	A3sup
Lck90–99	ILEQSGEWWK	p56 lck	A3sup
Lck449–458	VIQNLERGYR	p56 lck	A3sup
PAP248–257	GIHKQKEKSR	PAP	A3sup

The safety and immunologic effects of these 31 peptides had been confirmed in previous clinical trials [11], and all peptides were prepared under conditions of Good Manufacturing Practice using a Multiple Peptide System (San Diego, CA)

A3sup A3 supertype (A3, A11, A31, and A33), EGF-R epidermal growth factor receptor, EZH enhancer of zeste homolog, HLA human leukocyte antigen, HNRPL heterogeneous nuclear ribonucleoprotein L, MAP microtubule associated protein, MRP multidrug resistance-associated protein, PAP prostatic acid phosphatase, PSA prostate-specific antigen, PSMA prostate-specific membrane antigen, PTHrP parathyroid hormone-related peptide, SART squamous cell carcinoma antigens, UBE ubiquitin-conjugated enzyme variant Kua, WHSC Wolf–Hirschhorn syndrome critical region

Treg and MDSC

Suppressive immune subsets, Treg and MDSC, among the PBMC were examined by flow cytometry. For the analysis of Treg, PBMC (0.5 × 10⁶) suspended in PBS containing 2 % FBS were incubated with anti-CD4, anti-CD25, and anti-FoxP3 antibodies (Ab) using the One Step Staining Human Treg FlowTM Kit (Biolegend, San Diego, CA).

Treg were identified as positive for CD-4, CD-25, and FoxP3. For the analysis of MDSC, PBMC (0.5 × 10⁶) suspended in PBS containing 2 % FBS were incubated with anti-CD3-FITC, anti-CD56-FITC, anti-CD19-FITC, anti-CD33-APC, anti-HLA-DR-PE/Cy7, and anti-CD14-APC/Cy7 Ab. MDSC were identified as positive for CD33 in the cell subset negative for lineage markers (CD3, CD19, CD56, CD14) and HLA-DR. After the samples were processed on a FACS Canto II (BD Biosciences, San Diego, CA), data were analyzed using the Diva software package (BD Biosciences). The frequencies of Treg and MDSC in the mononuclear cell gate defined by forward scatter and side scatter were calculated.

IgG and CTL

Immune responses to the vaccinated peptides were assessed at week 7. IgG titers to the vaccinated peptide were measured using a Luminex system with a cutoff value of 10 fluorescence intensity units (FIU) in 100-fold diluted samples [18]. Peptide-specific CTL activity in PBMC was evaluated by interferon-γ (IFN-γ) ELISPOT assay using an ELISPOT reader (CTL-ImmunoSpot S5 Series; Cellular Technology Ltd., Shaker Heights, OH) [19] and was carried out in triplicate. If IgG titers to vaccinated peptides were twofold higher at week 7 than those in the pre-vaccination plasma, or more than 100 spots to the corresponding peptide in the PBMC at week 7 was observed by IFN-γ ELISPOT assay, these changes were considered to be positive immune responses.

Toxicity and efficacy

Toxicity was assessed by the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0 (NCI-CTCAE Ver. 4), and patients were monitored at each visit. Serum PSA tests and routine laboratory studies were performed every 8 vaccinations, and CT and bone scans were performed every 6 months or upon progression of symptoms. Efficacy was evaluated by the Response Evaluation Criteria in Solid Tumors (RECIST), and PD was defined as radiographic progression or clinical progression.

Statistical methods

The trial was designed to detect a change of ≥30 % reduction in Treg between PPV plus low-dose CPA and PPV alone arms measured at week 7. To detect changes with a power of 80 % and a type I error of 5 %, using a two-sided t test and 1:1 random assignment, 64 patients were required. After accounting for possible dropouts from the study, the inclusion of 70 patients was planned. Differences in the frequencies and changes of Treg and MDSC between the two arms were analyzed by Student’s t test. Analyses of secondary endpoints of PFS, OS, safety, and immune responses to PPV were based on an intention-to-treat (ITT) population. PFS and OS were calculated as the time in months from study enrollment to the date of event or to the date of last contact for censored observations. The Kaplan–Meier method was used for analyzing the time-to-event endpoints, and between-treatment comparisons for PFS and OS were conducted using the log-rank test with a two-sided significance level of 5 %. Hazard ratios (HRs) and 95 % CI were calculated by Cox proportional hazards. Two-sided statistical significance was defined as a p value <0.05, and statistical analyses were performed using SAS software version 9.1 (SAS Institute, Cary, NC).

Results

Patients

A total of 70 patients at the Cancer Vaccine Center of Kurume University in Japan were randomly assigned to receive either PPV plus low-dose CPA (n = 35) or PPV alone (n = 35) between May 2011 and April 2013. After the randomization, one patient in the PPV alone arm withdrew consent. Therefore, in the CPA plus PPV and PPV alone groups, 35 and 34 patients received treatment, respectively (Fig. 1). Baseline patients and disease characteristics were well balanced between the treatment arms (Table 2).

Fig. 1.

Study flowchart. CPA cyclophosphamide, PPV personalized peptide vaccination

Table 2.

Patient characteristics at baseline

Characteristics	PPV plus CPA (n = 35)		PPV alone (n = 35)		p
	No	%	No	%
Age (years)					0.63
Mean ± SD	68.6 ± 7.7		67.7 ± 8.0
Median (range)	68.6 (48–80)		67.7 (52–84)
ECOG PS					0.8
0	24	69	23	66
1	11	31	12	34
PSA (ng/ml)					0.33
Mean ± SD	190.9 ± 350.5		325.0 ± 720.0
Median (range)	190.9 (0.2–1880)		328.8 (1.5–1910)
Gleason score					0.45
≤7	7	20	10	29
≥8	27	77	25	71
Unknown	1	3	0	0
HLA typing to vaccinated peptide					0.21
HLA-A24	26	74	20	57
HLA-A2	3	9	8	23
HLA-A3 supertypea	6	17	7	20
Metastatic sites					0.73
Bone only	17	49	17	49
Bone and nodal/organ	15	43	13	37
Others	3	8	5	14
Prior chemotherapy					0.99
Estramustine phosphate	7	20	7	20
Docetaxel	5	14	6	17
Estramustine phosphate plus docetaxel	15	43	14	40
None	8	23	8	23
Treg (% of lymphocytes)					0.7
Mean ± SD	0.78 ± 0.8		0.94 ± 1.0
Median (range)	0.48 (0.13–3.71)		0.55 (0.05–4.64)
MDSC (% of lymphocytes and monocytes)					0.74
Mean ± SD	0.49 ± 0.52		0.47 ± 0.33
Median (range)	0.34 (0–2.52)		0.33 (0.04–1.25)

χ ² test and t test were used for categorical and continuous variables, respectively

CPA cyclophosphamide, ECOG PS Eastern Cooperative Oncology Group performance status, HLA human leukocyte antigen, MDSC myeloid-derived suppressor cell, PSA prostate-specific antigen, PPV personalized peptide vaccination, Treg regulatory T cell

^aHLA-A3 supertype included HLA-A3, HLA-A11, HLA-A31, and HLA-A33

Safety

No treatment-related deaths occurred in either arm. The most frequently reported adverse events (AE) in both arms were grade 1 or 2 injection site reactions (88.6 % in PPV plus low-dose CPA and 91.2 % in PPV alone). Hematologic AE were more common and more severe in the PPV plus CPA treatment than in the PPV alone treatment, with grade 3 or 4 lymphocytopenia (11.4 vs. 2.9 %), anemia (17.1 vs. 2.9 %), neutropenia (2.9 vs. 0 %), and thrombocytopenia (5.7 vs. 2.9 %). Other grade 3 or 4 AE including appetite loss (5.7 vs. 2.9 %), peripheral edema (2.9 vs. 0 %), increased alkaline phosphatase (17.1 vs. 5.9 %), and hyponatremia (2.9 vs. 0 %) were more common in PPV plus CPA than in PPV alone. The treatment of PPV with or without CPA was generally well tolerated, and all the toxicities were manageable.

Treg, MDSC, and immune responses

Circulating Treg and MDSC, as well as immune responses during the treatment were evaluated in 31 patients who received PPV plus CPA and in 28 patients who received PPV alone (Fig. 1). Mean frequencies of Treg in the PPV plus CPA arm at pre-treatment and 7 weeks were 0.78 ± 0.8 and 0.7 ± 0.85 % (p = 0.51), and those in the PPV alone arm were 0.94 ± 1.0 and 0.8 ± 0.77 % (p = 0.84), respectively. There were no differences in the frequencies of Treg during the treatment. Mean frequencies of MDSC in the PPV plus CPA arm significantly increased after treatment (0.49 ± 0.52 vs. 0.68 ± 0.79 %, p = 0.048), but there was no difference in those in the PPV alone arm (0.47 ± 0.33 vs. 0.53 ± 0.47 %, p = 0.69). In addition for testing changes of Treg and MDSC frequencies before and after PPV within each group, comparisons of % changes of Treg and MDSC among two experimental groups were performed. The % change of Treg from pre-treatment level in the PPV plus CPA arm showed a greater decrease after treatment than in the PPV alone group (−15.3 ± 38.7 vs. 87.6 ± 219.7 %, p = 0.036), but there was no significant difference in the % change of MDSC between the two arms (48.7 ± 100.2 vs. 49.1 ± 90.5 %, p = 0.99) (Fig. 2a, b). Pearson’s correlation analysis demonstrated that there was no correlation between % changes of Treg (r = 0.09, p = 0.55) or MDSC (r = 0.09, p = 0.56) frequencies and CTL responses measured by ELISPOT assay (Fig. 2c, d). In addition, there was no difference in the rates of reduction in Treg between the two arms (45 vs. 36 %, p = 0.63). The rates of reduction in MDSC between the two arms also did not differ (26 vs. 43 %, p = 0.27). In the PPV plus CPA arm, peptide-specific IgG or CTL responses were observed in 10 of 31 patients (32 %) or 9 of 31 patients (29 %), respectively. In the PPV alone arm, peptide-specific IgG or CTL responses were observed in 14 of 28 patients (50 %) or 9 of 28 patients (32 %), respectively. The results did not differ significantly by treatment arm in χ ² analysis. In addition, there was no difference in the rate of positive immune responses between the two arms.

Fig. 2.

The % change of Treg (a) and MDSC (b) from pre-treatment in the two arms. The % change of Treg in CPA + PPV significantly decreased compared with PPV alone (p = 0.036). No correlation was observed between mean changes of Treg (c) or MDSC (d) and CTL

Efficacy

Five patients (14 %) in each treatment arm showed a ≥50 % PSA decline during the treatment. On the basis of investigator-derived assessment of disease response and progression using RECIST criteria, PD was noted in 22 patients (63 %) in the PPV plus CPA arm and 22 patients (63 %) in the PPV arm. Stable disease was observed in 13 patients (20 %) in each arm. No complete responses or partial responses were observed in either of the arms. At data cutoff (February 15, 2015) with a median follow-up of 17.4 months (95 % CI 11–21.1), median PFS time was 6.5 months (95 % CI 1.9–3.5 months) for PPV plus CPA and 5.5 months (95 % CI 1.4–2.3 months) for PPV alone (Fig. 3a): This difference was not significant (HR 0.89; 95 % CI 0.48–1.65; p = 0.7). Median OS time was 17.1 months (95 % CI 9.4–19.6) for CPA plus PPV and 18.5 months (95 % CI 7.9–26.2) for PPV alone (Fig. 3b), but this difference was not significant (HR 0.97; 95 % CI 0.53–1.78; p = 0.92).

Fig. 3.

Kaplan–Meier estimates of PFS (a) and of OS (b). PFS progression-free survival, OS overall survival. Evaluation of OS for patients stratified by the reduction in Treg (c), MDSC (d), or positive/negative immune response (e). Patients with positive immune responses had longer survival than those with negative immune responses (p = 0.001)

We also performed an OS subgroup analysis based on groups stratified by the reduction in Treg or MDSC, or the status of positive immune responses. Median OS times for patients with a reduction in Treg and without a reduction in Treg were 19.2 months (95 % CI 11.0–33.2) and 19 months (95 % CI 12.2–25.1) (p = 0.31), respectively (Fig. 3c). On the other hand, patients with a reduction in MDSC showed a trend of longer survival than patients without such a reduction, for which the median OS times were 26.8 months (95 % CI 15.6 to not reached) and 17.4 months (95 % CI 11.0–21.1), respectively, but this difference was not significant (p = 0.06) (Fig. 3d). Patients with positive immune responses showed significantly longer survival than those with negative responses, with median OS times of 27.1 months (95 % CI 18.5 to not reached) and 15.4 months (95 %CI 9.4–19), respectively, (p = 0.001) (Fig. 3e).

Discussion

Several clinical trials of metronomic chemotherapy involving low-dose CPA regimens have shown low-toxicity and antitumor effectiveness in patients with advanced malignant tumors, including patients with mCRPC [20–22]. Previous studies found that the antitumor effect of CPA is associated with a reduction and functional impairment of Treg, which may improve immune responsiveness to immunotherapy [16, 23, 24]. On the basis of these results, we attempted to investigate the effect of metronomic low-dose CPA on PPV treatment for patients with mCRPC in this phase II randomized trial. Although a decrease in Treg was noted, the combination of PPV with metronomic low-dose CPA treatment had no significant impact on antigen-specific CTL activity, PFS, or OS.

One possible explanation for the negative results may be that increased MDSC may affect immune responses or clinical outcomes in this combination of immunotherapy with metronomic low-dose CPA. MDSC and Treg are considered as the two major contributors mediating tumor-induced immunosuppression [25]. In the present study, there was no difference in the changes of MDSC between the two arms, but the mean frequency of MDSC after treatment was significantly increased in the PPV plus CPA arm (p = 0.048). Indeed, several investigators reported that CPA-based chemotherapy induced an MDSC population [9, 26, 27]. CPA elevated the number and activation status of CD11b+ myeloid cells in vivo by increasing levels of IFN-γ and other proinflammatory cytokines in the serum [4, 28, 29]. These myeloid cells were thought to induce de novo FOXP3⁺ Treg [30]. It was reported that a direct cell–cell interaction or production of specific soluble factors, including IL-10 in the presence of TGF-β or arginase, causes the induction of Treg by MDSC [8, 31]. Our findings suggested that combined treatment with CPA did not affect clinical outcomes after PPV possibly due to the decrease in Treg induced by CPA that may have been compensated for by the increase in MDSC.

Another possible explanation for the lack of improved outcomes is that optimal dosing and schedules of CPA with immune therapies have not yet been determined, which may have affected the results of the CPA studies. In animal studies, immune responses have been reported with CPA administered within 1 week before vaccination [23, 32, 33]. Previous nonrandomized clinical trials also showed that CPA increased immunogenicity when administered 3 days before a cancer vaccine with semiquantitative immunologic endpoints [34, 35]. Other clinical trials identified negative effects of CPA pre-treatment on cellular immune responses to a breast cancer cell vaccine [36, 37]. On the other hand, Ghiringhelli et al. [24] reported that metronomic CPA (50 mg/day for 1 month) led to a selective reduction in circulating Treg while restoring innate and acquired T cell receptor-driven T cell responses in some patients. However, Ge et al. [38] reported that low-dose CPA treatment (50 mg/day for 3 months) selectively reduced the number of Treg for a short time, but was insufficient for maintaining Treg at low frequencies for longer periods in metastatic breast cancer patients. Thus, differences in CPA schedules may have contributed to the differences in our results compared with those previously reported.

Because the selected peptides and numbers of used peptides for PPV were similar between the CPA plus PPV and PPV alone groups, another possible explanation of the negative findings of the present study may have been due to the choice of antigens for PPV treatment, despite choosing the antigens based on preexisting host immunity prior to vaccination. Thus, the choice of suitable antigens for PPV remains to be examined.

In summary, the addition of low-dose CPA resulted in no change in the antitumor effect of PPV, possibly due to the simultaneous mitigation of Treg and increase in MDSC in patients treated with PPV.

Abbreviations

Ab

Antibodies

ADT

Androgen deprivation therapy

AE

Adverse events

CPA

Cyclophosphamide

CRPC

Castration-resistant prostate cancer

CT

Computed tomography

CTL

Cytotoxic T lymphocytes

ECOG

Eastern Cooperative Oncology Group

FBS

Fetal bovine serum

FIU

Fluorescence intensity units

HLA

Human leukocyte antigen

HR

Hazard ratio

IFN-γ

Interferon-γ

IgG

Immunoglobulin G

ITT

Intention-to-treat

mCRPC

Metastatic castration-resistant prostate cancer

MDSC

Myeloid-derived suppressor cells

OS

Overall survival

PBMC

Peripheral blood mononuclear cells

PBS

Phosphate-buffered solution

PD

Progressive disease

PFS

Progression-free survival

PPV

Personalized peptide vaccine

PS

Performance status

PSA

Prostate-specific antigen

TAA

Tumor-associated antigen

Treg

Regulatory T cells

Compliance with ethical standards

Conflict of interest

Noguchi M has served as an advisory board consultant for Green Peptide Co. Ltd. Itoh K has served as a consultant and received research funding from Taiho Pharmaceutical Company. Yamada A is a part-time executive of Green Peptide Co. Ltd. and has stock in this company. Moriya F, Koga N, Matsueda S, Sasada T, and Kakuma T declare no competing interests.