Immunological evaluation of personalized peptide vaccination for patients with histologically unfavorable carcinoma of unknown primary site

Abstract

The immunological characteristics of carcinoma of unknown primary site (CUP) are not well established due to inclusion of heterogeneous types of metastatic tumors with the absence of any detectable primary site. We evaluated the immune responses in patients with histologically unfavorable CUP during personalized peptide vaccination (PPV). Ten patients with histologically unfavorable CUP who had been treated by PPV after chemotherapy failure were analyzed. In PPV treatment, up to four human leukocyte antigen-matched peptides of a total 31 peptides were selected according to preexisting host immunity before vaccination and administered subcutaneously. Peptides derived from the Lck antigen were most often chosen for use among all patients. CTL responses were increased in 8 of the 10 and 5 of the five patients tested at the end of the first and second PPV cycles, respectively. Increases in humoral responses after vaccination, including IgG, IgG1, IgG3, IgA, and IgM, were observed against not only the vaccinated peptides but also the non-vaccinated peptides. Severe adverse events due to PPV were not observed. Median overall survival was 13.9 months (95 % CI 4.0–22.5 months). PPV activated both cellular and humoral immune responses to short peptides derived from CTL epitopes in the majority of CUP patients. PPV with Lck-derived peptides may be a feasible, new treatment modality for histologically unfavorable CUP patients due to its safety and strong ability to boost immune responses, although its clinical efficacy remains to be investigated in larger-scale trials.

Electronic supplementary material

The online version of this article (doi:10.1007/s00262-016-1887-5) contains supplementary material, which is available to authorized users.

Introduction

Carcinoma of unknown primary site (CUP) is one of the ten most frequent cancers worldwide, and the incidence of CUP is 3–5 % of all malignancies [1–4]. Metastasis of a malignancy without a detectable primary site is considered as CUP. Although a taxane/platinum regimen is initially recommended as first-line chemotherapy for patients with CUP, survival continues to be poor with a median survival time (MST) of approximately 5–9 or 12–14 months in the histologically unfavorable or favorable groups, respectively [1–7]. Despite these characteristics and the severity of prognosis, the underlying biology of CUP remains poorly understood [8], and there are two opposing hypotheses [9]. CUP is considered to be a single biological entity with the absence of a primary site and early metastatic disease. Another hypothesis is that CUP represents different types of tumors with a primary site. Nevertheless, CUP is a challenging heterogeneous tumor with an unmet research need.

Recently, the immunotherapeutic approach has emerged as an attractive strategy for the treatment of advanced cancer. We have devised personalized peptide vaccination (PPV) as a new immunological treatment. In PPV treatment, up to four peptides are selected for vaccination from 31 candidate peptides based on human leukocyte antigen (HLA)-class IA type and a level of peptide-specific immunoglobulin G (IgG) titer in pre-treatment plasma. We showed that PPV had favorable clinical and immune responses in advanced metastatic cancer patients with different primary sites including prostate, lung, breast, ovarian, colorectal, bladder cancers, and glioblastoma [10–12]. For further development of PPV as a new treatment of CUP, it is important to investigate the immunological characteristics of CUP patients. In this study, we investigated the immunological characteristics of patients with histologically unfavorable CUP and the immunological responses to PPV.

Materials and methods

Patients

Ten consecutive patients with histologically unfavorable CUP had been enrolled in 4 phase II PPV studies after failure of taxane/platinum-based chemotherapies at the Kurume University Cancer Vaccine Center between April 2009 and April 2013. These 4 trials are phase II clinical trials of PPV for patients with advanced cancer in whom standard treatments failed, with the only difference being the PPV treatment schedule. The Kurume University Ethical Committee approved the protocols of these phase II clinical trials, and they were registered in the UMIN Clinical Trials Registry (UMIN000001482 for 7 patients, UMIN000006927 for 1 patient, UMIN000010068 for 1 patient, and UMIN000014855 for 1 patient). All patients provided written informed consent to enroll in the clinical trials.

Eligibility

All patients were required to have a diagnosis of CUP after a standard evaluation that included a complete history, physical examination, blood counts, chemistry profile, computed tomography (CT) of chest, abdomen and pelvis, and directed radiologic findings of the symptomatic area. Key inclusion criteria of the protocols included age of 20–80 years; Eastern Cooperative Oncology Group (ECOG) performance status 0 or 1; adequate hematologic, hepatic, and renal function; HLA-A2, HLA-A24, HLA-A3, HLA-A11, HLA-A31, HLA-A33, or HLA-A26 positive status; positive IgG responses to two or more of the 31 candidate peptides [10]. Patient exclusion criteria included clinically significant pulmonary, cardiac, or other systemic diseases; concurrent infection; a history of severe allergic reactions; and other inappropriate conditions for enrollment.

Treatment

Under PPV treatment, 2–4 peptides were selected from the 31 candidate peptides based on results of HLA typing and a high titer of peptide-specific IgG in pre-treatment plasma. The 31 candidate peptides were applicable for the majority of the global population, and they were previously shown to be safe and capable of inducing tumor-reactive CTL activity in patients with cancer [10]. Preparation of all peptides was under the conditions of good manufacturing practice using a Multiple Peptide System (San Diego, CA). A maximum of four peptides (3 mg, 1.5 ml emulsion/per peptide) were subcutaneously administered with incomplete Freund’s adjuvant (Montanide ISA-51VG; Seppic, Paris, France) every week for 6 consecutive weeks in 8 patients, every week (4×) followed by biweekly (4×) administration in one patient (Pt No. 10), and weekly administration (4×) in the remaining patient (Pt No. 7). After the first cycle, the 4 antigen peptides were re-selected based on peptide-specific IgG titers, followed by biweekly injection in 8 patients, biweekly injection (4×) and monthly administration (4×) in one patient (Pt No. 10), and monthly administration (4×) in the remaining patient (Pt No. 7). These different administration regimens were due to the different schedules of each protocol.

Safety and efficacy

Adverse events during the study were assessed and categorized according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0 (NCI-CTC Ver. 4). The clinical responses of all patients were evaluated during and after study treatment until disease progression and/or death was documented, and responses were categorized by using the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 criteria.

Immune responses to the vaccinated peptides

Cytotoxic T lymphocyte (CTL) and IgG responses to the 31 candidate peptides were analyzed in blood samples before vaccination, after the first cycle of vaccination and after the second cycle of vaccination. The CTL responses were evaluated by interferon (IFN)-γ ELISPOT assay (MBL, Nagoya, Japan) using peripheral blood mononuclear cells (PBMC), and the IgG responses were measured by the Luminex system (Luminex, Austin, TX) using plasma as previously described [13, 14]. The levels of peptide-specific IgM, IgA, IgE, and IgG1-G4 titers in plasma were also measured by the same assay.

Antigen-specific T cell responses were determined by the difference between spot numbers (mean of duplicate samples) in response to the corresponding peptide and those in response to the control HIV peptide. Changes were considered significant if the spot number in the post-vaccination PBMC was more than twice that in the pre-vaccination or if at the post-cycle of vaccination, a non-detectable spot changed to more than 10 spots for at least one vaccinated peptide in the PBMC as observed by IFN-γ ELISPOT assay.

The levels of IgG, IgM, IgA, IgE, and IgG1-G4 were obtained as fluorescence intensity units (FIU), and if the levels in the post-vaccination plasma were more than twice those in the pre-vaccination plasma, the response was considered augmented.

Flow cytometry analysis

As immunosuppressive cells, myeloid-derived suppressor cells (MDSC), regulatory T-cells (Treg), B and T lymphocyte attenuator (BTLA), cytotoxic T lymphocyte antigen 4 (CTLA-4), programmed cell death 1 (PD-1), and anti-lymphocyte activation gene 3 (LAG-3) were measured by a FACS Canto II (BD Biosciences, San Diego, CA) using the Diva software package. Anti-CD4-FITC (clone RPA-T4), anti-CD-8-PerCP-Cy5.5 (clone RPA-T8), anti-CD33-FITC, anti-CD11b-FITC, anti-HLA-DR-PE/Cy7, anti-FOXP3-FITC, anti-CD-25-FITC, anti-CD-272 (BTLA), anti-CD-152 (CTLA-4), anti-CD-279 (PD-1), and anti-CD-223 (LAG-3) from BioLegend were used in this study for antibodies. A mixture of FITC-conjugated CD3/CD16/CD19/CD56 antibodies (BioLegend) was used to detect lineage negative cells.

Statistical analysis

Progression-free survival (PFS) and overall survival (OS) from the date of the first vaccination were examined by the Kaplan–Meier method, and the differences in the survival curves between the groups were compared by the log-rank test. Differences in quantitative and categorical variables were analyzed by the student’s t test and Chi-square test, respectively. All statistical analyses were performed using SAS 9.1 (SAS Institute, Cary, NC), and a two-sided P values <0.05 were considered significant.

Results

Patient characteristics

The clinical and pathological characteristics of the 10 patients with histologically unfavorable CUP are shown in Table 1. Subjects were 5 males and 5 females, and median age was 60 years old (39–74) with PS of 0 (n = 9) or 1 (n = 1). Pathologies were 5 adenocarcinomas (Ad), 2 squamous cell carcinomas (SCC), 1 undifferentiated carcinoma, 2 mixtures (Ad and undifferentiated carcinoma, poorly differentiated carcinoma and SCC). There were 2 suspected primary sites in the lungs and 2 in the colon, and one site in the pancreas, thymus, intrahepatic biliary tract, gastrointestinal tract, and an unknown site. The number of tumor sites was 2 (n = 5), 3 (n = 3), 4 (n = 1), and 7 (n = 1). Five patients had organ metastases. Before enrollment, all patients underwent at least one taxane/platinum regimen. No patients were classified as the histologically favorable subset, which was defined by recent publications [1–4].

Table 1.

Clinical and pathological characteristics of ten patients with histologically unfavorable CUP

Patient no.	Age	Gender	PS	Metastatic site (no. of tumors)	Pathology	Suspected primary site	No. of previous chemotherapy regimens
1	70	Male	0	Liver, LNs (3)	Adenocarcinoma undifferentiated carcinoma	Pancreas	1
2	62	Male	0	Thymus, pericardium, LN (3)	Undifferentiated carcinoma	Thymus	1
3	39	Female	0	Pleural region, LNs (3)	Squamous cell carcinoma	Lung	2
4	59	Male	0	LNs (2)	Adenocarcinoma	Intrahepatic biliary tract	3
5	73	Female	0	LNs (2)	Squamous cell carcinoma	Unknown	1
6	74	Female	0	LNs (2)	Adenocarcinoma	Gastrointestinal tract	1
7	63	Male	0	LNs, chest wall (4)	Poorly differentiated carcinoma squamous cell carcinoma	Unknown	6
8	48	Female	0	Lung, stomach (2)	Adenocarcinoma	Colon	1
9	49	Male	0	Brain, thyroid, LNs (7)	Adenocarcinoma	Lung	2
10	50	Female	1	Lung, bone (2)	Adenocarcinoma	Colon	2

CUP carcinoma of unknown primary site, LN lymph node, PS performance status

Peptides and peptide-specific immune responses

Table 2 shows the summary of selected peptides, peptide-specific CTL and IgG, and clinical responses during the study. All patients received 4 peptides with at least one Lck-derived peptide (9 patients during the first cycle and the remaining patient during the second cycle) due to higher IgG responses both before and after the vaccination. Furthermore, 2 Lck-derived peptides were administered during the first cycle in four patients, and five of the ten patients who continued the second cycle received at least one Lck-derived peptide. SART3-derived peptides were the second most used peptides for vaccination in six patients during the first cycle, followed by two SART3-derived peptides administered during the first cycle in two patients. Seven patients were treated by a combination of chemotherapy with PPV, and three patients were treated with PPV alone.

Table 2.

CTL, IgG, and clinical responses

Pt no.	Peptide	CTL (spots/well)			IgG (FIU)			Treat	BCR
		Pre-first cycle	Post-first cycle	Post-second cycle	Pre-first cycle	Post-first cycle	Post-second cycle
1	ppMAPkkk-432	134	ND	32	34	ND	10	PPV with chemo	SD
	HNRPL-501	ND	ND	ND	371	417	ND
	SART3-511	16	ND	ND	272	187	13
	Lck-90	49	ND	ND	398	323	18
	SART3-109	ND	ND	77	86	119	ND
2	Lck-246	ND	1428	ND	31	2993	25,579	PPV with chemo	PD
	WHSC2-103	ND	845	785	64	551	690
	WHSC2-141	ND	1702	ND	57	15,284	41,563
	SART3-309	20	809	ND	76	134	186
3	SART3-734	ND	ND	na	136	630	na	PPV	SD
	Lck-90	ND	ND	na	29	28	na
	SART3-109	ND	127	na	35	13	na
	ppMAPkkk-432	6	ND	na	3513	3162	na
4	SART2-93	23	ND	na	230	307	na	PPV with chemo	SD
	EGF-R-800	11	158	na	55	306	na
	Lck-486	17	268	na	114	20,274	na
	Lck-488	59	ND	na	331	597	na
5	SART2-93	ND	ND	81	87	70	69	PPV	PR
	MRP3-503	ND	71	ND	83	41	147
	Lck-486	71	300	155	51	25	1754
	Lck-488	18	ND	1755	97	39	535
6	PSA-248	ND	7	75	24	2322	15,436	PPV with chemo	SD
	MRP3-1293	36	12	ND	80	129	3069
	SART3-109	ND	ND	ND	335	386	23,328
	Lck-488	ND	35	ND	ND	75	46
7	SART3-302	ND	303	na	297	9074	na	PPV	SD
	SART3-511	ND	ND	na	1344	1719	na
	SART3-734	106	ND	na	23,125	22,952	na
	Lck-90	ND	257	na	291	287	na
8	Lck-246	215	602	na	26	21	na	PPV with chemo	PD
	ppMAPkkk-432	ND	ND	na	84	55	na
	UBE2 V-43	ND	25	na	511	24,498	na
	SART3-734	ND	22	na	104	75	na
9	SART2-93	ND	250	ND	35	38	86	PPV with chemo	PD
	Lck-486	ND	31	149	51	59	22,331
	Lck-488	182	ND	199	25	30	10,457
	PTHrP-102	ND	ND	14	23	21	28
	PAP-213	ND	ND	ND	20	24	11,571
	MRP3-1293	12	ND	34	16	14	16
10	EGF-R-800	ND	25	na	125	98	na	PPV with chemo	SD
	Lck-488	16	ND	na	48	153	na
	PTHrP-102	ND	199	na	59	33	na
	Lck-422	21	ND	na	39	ND	na

Underlined peptides were selected for the second cycle of vaccination. Positive CTL and IgG responses are shown in bold

BCR best clinical response, chemo chemotherapy, CTL cytotoxic T lymphocyte, FIU fluorescence intensity unit, na not applicable, ND not detectable, PD progressive disease, PPV personalized peptide vaccination, PR partial response, SD stable disease, Treat treatment

A pre-vaccination CTL response to at least one of the vaccinated peptides was detectable in all ten patients, although the levels were very low. Notably, CTL responses against the peptides derived from the Lck antigen, which is known to be aberrantly expressed in metastatic cancer cells [15–17], were detectable in pre-vaccination PBMCs in 6 of the 10 patients. At the end of the first and second cycles of vaccination, peptide-specific CTL responses were increased in 8 of the 10 and all 5 of the 5 patients, respectively. CTL responses against Lck-derived peptides were increased in 6 patients after the first vaccination.

After the first and second cycles of vaccination, IgG responses to at least one of the vaccinated peptides were increased in 7 of the 10 patients and in 4 of the 5 patients, respectively. Increased IgG responses to Lck-derived peptides were observed in 3 of the 10 and 3 of the 5 patients at the same points, respectively. We further investigated IgG1-4, IgA, IgM, and IgE responses to each of the 31 peptides in the pre- and post-first cycle vaccination plasma by a simultaneous assay to better understand the PPV-induced humoral responses (see Supplementary Table). As a result, increases in peptide-specific IgG1, IgG3, IgA, and IgM were observed not only for the vaccinated peptides but also for the non-vaccinated peptides in more than 5 patients. In contrast, increases in IgG4 or IgE responses were small or rarely observed, respectively.

Flow cytometry analysis

The mean frequency of immunosuppressive cells in PBMCs at pre- and post-first vaccination is shown in Fig. 1. Although the frequencies of MDSC, Treg, CD4+BTLA, CD+8BTLA, CD4+CTLA-4 and CD4+PD-1 in the PBMCs of one patient with negative CTL and IgG responses (Pt No. 1) were increased after the first cycle of vaccination, the differences in the mean frequency of immunosuppressive cells were not significant.

Fig. 1.

Mean frequency of suppressive immune cell subsets among both pre- and post-first cycle vaccination PBMCs: a, MDSC; b, Treg; c, BTLA; d, CTLA-4; e, PD-1; f, LAG-3. There were no significant differences in the mean frequencies of the suppressive immune cell subsets. Note MDSCs were identified as positive for CD33 in the cell subset negative for lineage markers (CD3, CD19, CD56, CD14) and HLA-DR; Tregs were identified as positive for CD-4, CD-25, and FoxP3

Safety and efficacy

Grade 1 or 2 injection site reactions (n = 10) were the most common adverse events. Grade 3 adverse events were anemia (n = 2), anorexia (n = 2), liver disfunction (n = 1), and gastrointestinal hemorrhage (n = 1). These grade 3 adverse events were due to disease progression or other causes, not the vaccinations themselves.

There were no patients with complete response (CR). One patient had a partial response (PR), 5 patients had stable disease (SD), and the remaining 4 patients had progressive disease (PD) as the best clinical response. Decrease in metastatic tumor size by CT or tumor markers during PPV was observed in 2 of the 9 patients and 1 of the 7 patients, respectively. CT findings and immune responses of the patient with PR (Pt No. 5) at pre- and post-first cycle PPV are shown in Fig. 2. This patient was a 73-year-old woman who progressed after chemo-radiation therapy for swelling of the mediastinal lymph nodes and supraclavicular lymph node, and was treated with PPV mono-therapy. After the first cycle of PPV, the mediastinal lymph node tumor size decreased from 35 × 25 to 15 × 15 mm with no change in size of the supraclavicular lymph node tumor; thus clinical response was judged as PR. In this patient, Lck-derived peptide-specific CTL and IgG responses were increased after the first cycle of PPV (Fig. 2); she continued PPV treatment and survived for 27.7 months. In all 10 patients with histologically unfavorable CUP treated by PPV, median PFS was 5.5 months (95 % CI 0.7–6.5 months) (Fig. 3a), and median OS was 13.9 months (95 % CI 4.0–22.5 months) (Fig. 3b).

Fig. 2.

Computed tomography (CT) findings and immune responses of a patient (Pt No. 5) with histologically unfavorable carcinoma of unknown primary site treated by personalized peptide vaccination (PPV). The mediastinal lymph node tumor size was reduced after the first cycle of PPV (a at pre-PPV and b at post-first cycle PPV). Lck-derived peptide-specific CTL (c) and IgG (d) responses were increased after the first cycle of PPV. Arrows indicate tumor site

Fig. 3.

Kaplan–Meier estimates of progression-free survival (a) and of overall survival (b). Dotted lines show 95 % confidence intervals

Discussion

Despite the development of immunotherapy for many malignancies, immunological evaluation and an immunological therapeutic approach for CUP remain to be established. To our knowledge, the present study is the first report investigating the immunological responses of PPV treatment in patients with histologically unfavorable CUP. The following immunological characteristics of CUP were observed in the present study. First, CUP patients showed potent immune responses to Lck-derived peptides after previously undergoing taxane/platinum-based chemotherapies that genetically suppress tumor immunity. Second, CTL and IgG responses were maintained before PPV, and increases in CTL and IgG responses were observed in the majority of patients with histologically unfavorable CUP after PPV. Third, PPV treatment for patients with histologically unfavorable CUP showed a better safety profile and longer survival of approximately 5–9 months than previously reported [1–7], although few patients were analyzed. Thus, in the patients with a poor prognosis, the metastatic and primary sites may have been destroyed by the immune system.

In the present study, Lck-derived peptides were most often chosen for use in the vaccination due to higher IgG responses before vaccination along with positive CTL activity in the pre-vaccination PBMCs in 6 of the 10 patients. CTL responses against Lck peptides were increased in 6 of the 10 patients post-vaccination. Furthermore, one patient with PR showed increases in Lck-derived peptide-specific CTL and IgG responses and a reduction in metastatic tumor size. These results indicated that one of the appropriate tumor-associated antigens (TAAs) for immunotherapy in patients with histologically unfavorable CUP may be the Lck antigen (p56^Lck). p56^Lck is a src family tyrosine kinase that contributes to the process of neoplastic transformation and the growth of TGF-β-initiated tumor cells, and is expressed in metastatic cancer cells [15–17]. Although our previous clinical trials showed the superiority of PPV over the vaccination using common peptides based on immune and clinical responses [18, 19], PPV with Lck-derived peptides might be suitable for patients with histologically unfavorable CUP.

Our previous PPV studies demonstrated that the PPV-induced CTL or IgG response increases were observed in only a portion of the patients (30–50 % for CTL or 50 % for IgG increases) in advanced metastatic cancer patients in whom standard treatments failed [10–13]. In the present study on previously treated CUP patients, CTL responses were increased in 8 of the 10 and 5 of the 5 patients tested after the first and second cycles of vaccination, respectively. However, some PPV-induced CTL responses decreased after the second cycle vaccination. These decreasing CTL responses may be, in part, due to the immunosuppressive cells or immune check point molecules expressed on TILs as well as tumor cells. To further improve the clinical efficacy of PPV, it is important to investigate the mechanism of the decrease in CTL response after PPV, and one approach for improvement may be a combination therapy of immune check point blockades with PPV such as CTLA-4 or PD-1 antibodies. In addition, humoral responses including IgG, IgG1, IgG2, IgG3, IgA, and IgM were observed not only against the vaccinated peptides, but also against the non-vaccinated peptides in the majority of patients tested. We have previously reported that such spread of humoral response to non-vaccinated peptides was a favorable factor for patients treated with PPV [20, 21]. These results suggested that CUP patients may have some common features, distinct from other advanced cancer patients, due to their strong immune responses to short peptides derived from the CTL-epitope, regardless of previously conducted chemotherapies.

Analysis of immune cell subsets among PBMCs in only 10 patients may not provide a definitive conclusion, but is important for better understanding the factors involved in the possible clinical benefits of PPV in patients with histologically unfavorable CUP. Although the frequencies of suppressive immune cell subsets in patients with no immune responses were increased, there was no significant difference in the mean frequency of these cell subsets. The roles of these suppressive immune cells in the immune responses to PPV remain to be examined.

In this study, a limited number of patients were investigated for the PPV-induced immunological responses in patients with histologically unfavorable CUP. Therefore, analysis of the clinical outcomes was not the main objective. However, it is important to further investigate the clinical outcome of these patients, due to the urgent need of development for a new treatment modality for CUP patients, whose prognosis continues to be poor with a MST of approximately 5–9 months in the histologically unfavorable group [1–7]. Furthermore, the prognosis of previously treated CUP patients has not been well studied. There were no CR patients, but 1 PR patient, 5 SD patients, and 4 PD patients. There were no severe adverse events due to PPV with a median OS of 13.9 months, better than the previously reported MST.

In conclusion, the present results indicated that PPV with Lck-derived peptides may be a feasible, new treatment modality for patients with unfavorable subsets of CUP due to its safety and strong ability to boost immune responses, although its clinical efficacy remains to be investigated in larger-scale trials.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

Ab

Antibody

Ad

Adenocarcinoma

BTLA

B and T lymphocyte attenuator

CR

Complete response

CT

Computed tomography

CTL

Cytotoxic T lymphocyte

CTLA-4

Cytotoxic T lymphocyte antigen 4

CUP

Carcinoma of unknown primary site

ECOG

Eastern Cooperative Oncology Group

FBS

Fetal bovine serum

FIU

Fluorescence intensity unit

HIV

Human immunodeficiency virus

HLA

Human leukocyte antigen

IFN-γ

Interferon-γ

IgG

Immunoglobulin G

LAG-3

Anti-lymphocyte activation gene 3

MDSC

Myeloid-derived suppressor cell

MRI

Magnetic resonance imaging

MST

Median survival time

OS

Overall survival

PBMC

Peripheral blood mononuclear cell

PBS

Phosphate-buffered solution

PD

Progressive disease

PD-1

Programmed cell death 1

PFS

Progression-free survival

PPV

Personalized peptide vaccine

PR

Partial response

PS

Performance status

SCC

Squamous cell carcinoma

SD

Stable disease

TAA

Tumor-associated antigen

Treg

Regulatory T cell

Compliance with ethical standards

Conflict of interest

Noguchi M. has served as an advisory board consultant for Green Peptide Co. Ltd. Itoh K. has received research funding from Taiho Pharmaceutical Company. Yamada A. is a part-time executive of Green Peptide Co. Ltd. and has stock. All other authors declare no competing interests.