A phase II trial of nivolumab for patients with platinum-refractory recurrent or metastatic salivary gland cancer

Abstract

Background

Salivary gland cancer (SGC) is rare and has various histological types. This rarity and heterogeneity have hindered elucidation of the therapeutic contribution of systemic therapy, including immune checkpoint inhibitors, to recurrent or metastatic SGC (RM-SGC). The purpose of this trial was to investigate the efficacy and safety of nivolumab for platinum-refractory RM-SGC.

Methods

This phase II trial for platinum-refractory RM-SGC was conducted at nine centers. Nivolumab 240 mg was administered intravenously every 2 weeks. The primary endpoint was the objective response rate (ORR), and secondary endpoints were disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and safety.

Results

Twenty-four patients were enrolled between March 2018 and January 2022. The main histological types were salivary duct carcinoma (n = 10), adenoid cystic carcinoma (n = 6), and adenocarcinoma not otherwise specified (n = 5). The ORR was 8.3% (2/24, 80% CI, 2.2–20.6), with two partial responses in patients with salivary duct carcinoma. The DCR was 29.2% (7/24), while all of the other 17 patients (70.8%) showed progressive disease on first disease evaluation at 12 weeks. With a median follow-up of 21.2 months, median PFS and OS were 3.0 months (95% CI, 2.8–3.2) and 25.0 months (95% CI, 10.9–39.1), respectively. There were no new safety concerns with nivolumab monotherapy.

Conclusions

This phase II trial of nivolumab for patients with platinum-refractory RM-SGC did not meet its primary endpoint of ORR. Although nivolumab may be worth further development in salivary duct carcinoma, these results may raise concerns over nivolumab monotherapy for RM-SGC.

Nivolumab failed to meet the primary endpoint, with an ORR 8.3%, but showed efficacy for salivary duct carcinoma, with an ORR of 20%. Adenoid cystic carcinoma did not show a response, with an ORR of 0%.

Introduction

Salivary gland cancer (SGC) is rare, and >20 histological types are described in the 5th Edition of the World Health Organization Classification [1]. SGCs can arise from the major salivary glands, which consist of the parotid, submandibular, and sublingual glands, and also the minor salivary glands, but most commonly develop from the parotid, submandibular glands. Regarding histological type, the most common types are salivary duct carcinoma (SDC), adenoid cystic carcinoma (AdCC), adenocarcinoma not otherwise specified (ANOS), and mucoepidermoid carcinoma (MEC) including low and high grade [2]. Largely due to this rarity and heterogeneity, the effects of systemic therapy—including the use of immune checkpoint inhibitors (ICIs)—for recurrent or metastatic SGC (RM-SGC) have not been well elucidated.

Nevertheless, some histological types, including SDC, show expression of androgen receptor (AR) or human epidermal growth factor receptor 2 (HER2), and target therapies for these are being developed and established as a treatment option. Specifically, the combination of androgen deprivation therapy and a luteinizing hormone–releasing hormone (LH-RH) agonist has shown efficacy in AR-positive SGC [3–5], while the combination of anti-HER2 antibody and docetaxel has shown efficacy in HER2-positive SGC [6, 7]. Secretory carcinoma, an extremely rare histological type, has a high frequency of neurotrophic tropomyosin receptor kinase (NTRK) gene fusion that is a potential therapeutic target [8]. The NTRK gene encodes the TRK protein, which is involved in neuronal differentiation and proliferation, and the presence of the fusion gene has been reported to promote cancer cell proliferation. Indeed, a TRK inhibitor for NTRK fusion-positive SGC including secretory carcinoma has shown a highly favorable response rate of 92% [9]. Furthermore, multi-targeted tyrosine kinase inhibitors have also shown efficacy, and targeted therapies using these, including lenvatinib and sorafenib, are recommended in the American Society of Clinical Oncology (ASCO) and National Comprehensive Cancer Network (NCCN) guidelines [10–13].

ICIs have recently become available for other cancer types [14–20]. As with systemic therapy, however, their effects in SGC have not been well elucidated due to its rarity and heterogeneity. Because of the relatively higher frequency and indolent nature of AdCC, clinical trials for ICIs are often conducted in AdCC and non-AdCC cohorts separately, and some studies have reported that ICIs might be less effective for AdCC [21–24]. Evaluation for histological types other than AdCC remains insufficient, however, and clinical trials that include non-AdCC should therefore be pursued.

Therefore, the purpose of this trial was to investigate the efficacy and safety of nivolumab for patients with platinum-refractory RM-SGC, including non-AdCC.

Patients and methods

Study design and participants

This trial was conducted as an open-label, single-arm, multicenter phase II trial of nivolumab monotherapy for patients with SGC at nine centers in Japan. Eligible patients were aged 20 years or older, and had histologically proven, platinum-refractory RM-SGC (defined as recurrence or disease progression during or within 6 months after platinum-based therapy) that was not indicated for local treatment; an Eastern Cooperative Oncology Group (ECOG) performance status of 0, 1 or 2; measurable lesions by Response Evaluation Criteria in Solid Tumours version 1.1 (RECIST v1.1); a life expectancy of at least 3 months; available tumor tissue for PD-L1 expression and biomarker analysis; and adequate organ function (neutrophil count ≥1000 cells/mm³; hemoglobin ≥8.0 g/dl; platelets ≥75 000/mm³; aspartate aminotransferase (AST) and alanine aminotransferase (ALT) ≤3.0 times the upper limit of normal (ULN) or ≤ 5.0 times the ULN in patients with liver metastases; total bilirubin ≤1.5 times the ULN; and creatinine ≤2.0 times the ULN). Patients were excluded if they had other known malignancy at the time of registration; active infection; active central nervous system metastases or carcinomatous meningitis; a history of ICIs; active autoimmune disease requiring treatment; received radiation therapy within 14 days before registration; been treated with systemic corticosteroids (prednisolone equivalent >10 mg/day; inhaled and topical steroids and hormone replacement therapy was acceptable) or other immunosuppressive therapies [use for adverse events (AEs) was acceptable]; or a history of interstitial pneumonia.

The study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The study protocol was reviewed and approved by the institutional review board of each site before study initiation, and all patients provided written informed consent. This trial was registered with the University Hospital Medical Information Network Clinical Trials Registry (UMIN-CTR; number UMIN000029636).

Procedures

Nivolumab 240 mg was administered intravenously every 2 weeks until disease progression, death, unacceptable toxicity, withdrawal of consent, or investigator’s decision. A computed tomography (CT) scan of the chest, abdomen and pelvis was performed within 4 weeks before registration and every 12 weeks after the date of registration. AEs were assessed according to the National Cancer Institute Common Terminology Criteria for AEs version 4.0 (CTCAE v4.0) and were evaluated at baseline and at each visit during the treatment period and for a minimum of 28 days after discontinuation of study treatment. HER2 status was assessed by immunohistochemistry (IHC) and by gene amplification using fluorescence in situ hybridization (FISH), with positivity defined as IHC 3+ or IHC 2+ and FISH-positive. AR status was assessed by IHC and defined as positive when at least 1% of cell nuclei in tumor tissue stained positively [3]. Histological diagnoses were evaluated centrally by two pathologists.

Outcomes

The primary endpoint was the objective response rate (ORR), defined as the proportion of patients with complete response (CR) or partial response (PR) according to RECIST v1.1. CR and PR required confirmation by consecutive imaging evaluation conducted at least 4 weeks after the initial recording date. Secondary endpoints were the disease control rate (DCR), progression-free survival (PFS), overall survival (OS), and safety. The DCR was defined as the proportion of patients with CR, PR, or stable disease (SD); PFS as the time from registration to the time of disease progression or death due to any cause; OS as the time from registration to the time of death from any cause; and safety endpoint by the presence of AEs identified from laboratory and physical findings. The severity of AEs was graded using CTCAE v4.0. Serious AEs were defined as those resulting in death, or in life-threatening, persistent, or significant disability or incapacity requiring hospitalization or prolonged hospitalization for treatment and which resulted in congenital anomalies or other medically significant events.

Statistical analysis

We set the lower threshold at 3.0% because a response rate of <3.0% is not considered clinically meaningful when considering clinical benefit and because no previous clinical trials have been conducted in patients receiving platinum-refractory RM-SGC. The expected ORR was set at 15%, based on ORRs in a phase III trial of nivolumab monotherapy in platinum-refractory squamous cell carcinoma of the head and neck [25] and in an SGC cohort in a phase II study of pembrolizumab monotherapy [23]. Significance level was set at a one-sided significance level of 10%, with a power of 80%, considering the rarity of the disease. Based on these assumptions, the required number of patients using a binomial test was calculated to be 23 patients. With allowance for patients dropping out of the trial, the number of registered patients was set at 24 patients.

The ORR and DCR were analyzed together with 95% confidence intervals (CIs). PFS and OS were analyzed by the Kaplan–Meier method, and median values and 95% CIs were estimated. Safety was evaluated in all patients who received at least one dose of nivolumab. The primary and secondary endpoints were assessed in all registered patients who received at least one dose of nivolumab. All analyses were done with SPSS statistics version 24.

Results

Patients

Twenty-four patients were registered at nine centers in Japan between March 2018 and January 2022. As of the data cutoff in January 2023, the median follow-up period was 21.2 months (range 2.5–45.6 months), and all patients had completed nivolumab monotherapy and experienced disease progression. All registered patients received at least one dose of nivolumab and were included in the efficacy and safety analyses.

Baseline patient characteristics are shown in Table 1. SDC and AdCC were the most common histological types, accounting for 41.7% and 25.0% of cases, respectively. The parotid gland was the most common primary site (66.7%). One line of prior chemotherapy was most common, and nivolumab monotherapy was commonly used in a second-line setting (75.0%). Of the total, 29.2% of cases were HER2-positive and 41.7% were AR-positive. At data cutoff, all patients discontinued study treatment with disease progression. Twenty-one patients (87.5%) who discontinued study treatment received subsequent treatment, including surgery (n = 1, 4.2%), chemotherapy (n = 15, 62.5%), AR-targeted therapy (n = 2, 8.3%), investigational drug (n = 2, 8.3%), and HER2-targeted therapy (n = 1, 4.2%). The other three patients (12.5%) did not receive subsequent therapy.

Table 1.

Patient characteristics

Characteristic	Patients (n = 24)
Median age (range), years	65.5 (34–78)
Sex
Male	16 (66.7)
Female	8 (33.3)
ECOG performance status
0	11 (45.8)
1	13 (54.2)
Histological type
Salivary duct carcinoma	10 (41.7)
Adenoid cystic carcinoma	6 (25.0)
Adenocarcinoma NOS	5 (20.8)
Mucoepidermoid carcinoma	2 (8.3)
Acinic cell carcinoma	1 (4.2)
Location of primary site
Parotid gland	16 (66.7)
Submandibular gland	5 (20.8)
Others	3 (12.5)
Platinum refractory	24 (100)
Prior lines for recurrent or metastatic disease
0	2 (8.3)
1	19 (79.2)
2	2 (8.3)
3	1 (4.2)
Prior radiotherapy	16 (66.7)
Prior surgery	20 (83.3)
Metastatic sites
Lung	18 (75.0)
Liver	5 (20.8)
Lymph node	10 (41.7)
Bone	5 (20.8)
Other	1 (4.2)
HER2-positive	7 (29.2)
AR-positive	10 (41.7)

NOS, not otherwise specified; ECOG, Eastern Cooperative Oncology Group; HER2, human epidermal growth factor receptor 2; AR, androgen receptor.

Efficacy

The primary endpoint of ORR was achieved in 8.3% (80% CIs, 1.1–15.6). Two cases of confirmed PR were observed in patients with SDC (ORR; 20.0%) versus 0% in the other histological types, including AdCC, ANOS, MEC, and acinic cell carcinoma (AcCC). The DCR was 29.2% (95% CIs, 11.0–47.4). An SD of 6 months or longer (long SD) was observed in five cases, three of which were ANOS. Seventeen patients (70.8%) showed PD on first disease evaluation at 12 weeks (Table 2).

Table 2.

Best overall response according to RECIST version 1.1

	All (n = 24)	SDC (n = 10)	AdCC (n = 6)	ANOS (n = 5)	MEC (n = 2)	AcCC (n = 1)
Best overall response, n (%)
CR	0	0	0	0	0	0
PR	2 (8.3)	2 (20.0)	0	0	0	0
SD	5 (20.8)	0	2 (33.3)	3 (60.0)	0	0
PD	17 (70.8)	8 (80.0)	4 (66.7)	2 (40.0)	2 (100)	1 (100)
ORR, n (%)	2 (8.3)	2 (20.0)	0	0	0	0
80% CIs	1.1–15.6	–	–	–	–	–
DCR, n (%)	7 (29.2)	2 (20.0)	2 (33.3)	3 (60.0)	0	0
95% CIs	11.0–47.4	–	–	–	–	–

CR, complete response; PR, partial response; SD, stable disease; PD, progressive disease; ORR, overall response rate; CIs, confidence intervals; DCR, disease control rate; SDC, salivary duct carcinoma; AdCC, adenoid cystic carcinoma; ANOS, adenocarcinoma, not otherwise specified; MEC, mucoepidermoid carcinoma; AcCC, acinic cell carcinoma.

Of the 23 patients who underwent at least one tumor evaluation after treatment with nivolumab, 7 patients (30.4%) experienced tumor regression in target lesions, of whom 3 had SDC. In addition, 2 of these 3 SDC patients had at least a 30% reduction in tumor size (Figs 1 and 2). With a median follow-up of 21.2 months, median PFS and OS were 3.0 months (95% CIs, 2.8–3.2 months) and 25.0 months (95% CIs, 10.9–39.1 months), respectively (Fig. 3A and B).

Figure 1.

Waterfall plot of tumor burden changes during nivolumab monotherapy. Change from baseline in target lesion size was evaluated according to response evaluation criteria in solid Tumors version 1.1. MEC, mucoepidermoid carcinoma; SDC, salivary duct carcinoma; AdCC, adenoid cystic carcinoma; ANOS, adenocarcinoma not otherwise specified; AcCC, acinic cell carcinoma.

Figure 2.

Spider plot of tumor burden changes during nivolumab monotherapy. Longitudinal changes in target lesion size from baseline. MEC, mucoepidermoid carcinoma; SDC, salivary duct carcinoma; AdCC, adenoid cystic carcinoma; ANOS, adenocarcinoma not otherwise specified; AcCC, acinic cell carcinoma.

Figure 3.

Kaplan–Meier analysis of survival. (a) Progression-free survival and (b) overall survival.

HER2-positive disease was found in seven patients (six in SDC and one in MEC). The ORR and DCR in these cases were 28.6% (2 PR) and 100% (2 PR and 7 SD), respectively. AR-positive disease was found in 10 patients (9 in SDC and 1 in ANOS), in whom the ORR and DCR were 20% (2 PR) and 30% (2 PR and 1 SD), respectively (Supplementary Table S1).

Safety

AEs of any grade occurred in 20 of 24 (83.3%) patients, of which 12.5% experienced grade 3 or 4 AEs. The most frequent AEs were fatigue (n = 7, 29.2%), AST increased (n = 7, 29.2%), and ALT increased (n = 6, 25.0%). Immune-related AEs occurred in 9 of 24 patients (37.5%), the most frequent of which were rash (n = 4, 16.7%), dry skin (n = 3, 12.5%), and hypothyroidism (n = 3, 12.5%). No grade 3 or higher immune-related AEs were observed, and no infusion reactions were observed (Table 3). No AEs led to the discontinuation of treatment or treatment-related death.

Table 3.

Adverse events

	All patients (n = 24)
All adverse events (AEs)	20 (83.3%)
Grade 3 or 4 AEs	4 (12.5%)
Immune-related AEs	9 (37.5%)
Adverse event	Any grade, n (%)	Grade 3–4, n (%)
Fatigue	7 (29.2)	0
AST increased	7 (29.2)	0
ALT increased	6 (25.0)	0
Creatinine increased	4 (16.7)	0
Stomatitis	4 (16.7)	0
Rash	4 (16.7)	0
Anemia	3 (12.5)	2 (8.3)
ALP increased	3 (12.5)	0
Anorexia	3 (12.5)	0
Dry skin	3 (12.5)	0
Diarrhea	3 (12.5)	0
Hypothyroidism	3 (12.5)	0
Hyponatremia	2 (8.3)	1 (4.2)
Amylase increased	2 (8.3)	0
Nausea	2 (8.3)	0
Vomiting	2 (8.3)	0
Pruritus	2 (8.3)	0
Hyperthyroidism	1 (4.2)	0

AST, aspartate aminotransferase; ALT, alanine aminotransferase.

Discussion

In this phase II trial to assess the efficacy and safety of nivolumab in patients with platinum-refractory RM-SGC, including non-AdCC, nivolumab monotherapy failed to demonstrate a clinical benefit, with an ORR of 8.3%. Furthermore, although 2 of 10 SDC patients achieved PR (ORR 20%), no patient with other histological types, including AdCC, showed a response (ORR 0%). Since platinum-refractory RM-SGC was investigated, long SD may be meaningful. Three of five SD cases were ANOS, and all of these had a long SD. Although ANOS showed no response, considering the clinical aggressiveness of ANOS, long SD may be clinically significant. In terms of safety, nivolumab was well tolerated and no new safety concerns were identified.

This clinical trial revealed differences in efficacy by histological type, with efficacy seen in SDC versus little or no efficacy in other histological types. Differences in efficacy by histological type have been reported [18, 19, 24], and our present results are considered to support these previous results. Further, because of the indolent nature of AdCC and its lower sensitivity to chemotherapeutic agents compared to other histological types, previous clinical trials have often been conducted in AdCC alone or with separate cohorts for non-AdCC disease. These trials also showed the limited efficacy of ICI in AdCC, as in our trial, versus a modest response in SDC [18, 19, 24].

Despite the overall limited efficacy, nivolumab showed promise specifically for patients with SDC, with an ORR of 20% (2/10). Efficacy of ICI in SDC has been confirmed in retrospective studies, suggesting the potential efficacy of nivolumab monotherapy for SDC as compared to other histologic types [26, 27]. Further, a similar trend was seen in a phase 2 trial of the combination of nivolumab and ipilimumab [24]. Nevertheless, new therapeutic approaches aimed at further efficacy should be pursued. In other cancer types, the combination of ICI and a vascular endothelial growth factor (VEGF) pathway inhibitor has been evaluated as a promising approach for improving efficacy. This combination is based on several rationales: first, in the immune-excluded type, immunosuppressive cells inhibit CD8⁺ T cell infiltration into the tumor [28], and it has been suggested that the VEGF plays an important role in this mechanism [29]; second, VEGF inhibitors have been reported to have various effects in mouse models, including activation of dendritic cells, suppression of regulatory T cells (Treg) and myeloid-derived suppressor cells (MDSCs), polarization of tumor-associated macrophages (TAMs) from M2-like to M1-like, and promotion of T cell infiltration into the tumor [30]; and third, a change in tumor immune microenvironment (TIME) from cold tumor to hot tumor with the use of an anti-PD-L1 antibody and anti-VEGF antibody in combination a mouse model has been reported [31]. These effects suggest that the combination of ICI and anti-VEGF antibody may have a synergistic effect, and clinical trials of this combination have indeed shown efficacy in several solid tumors [32–35]. With regard to SDC, one report described infiltration of CD8⁺ T cells into the stroma of SDC, but low expression of PD-L1 [36], suggesting that the TIME in SDC may be an immune-excluded type [37, 38]. In addition, SDC has been reported to have increased angiogenesis compared to other histologic types [39]. For these reasons, combination therapy with ICI and a VEGF pathway inhibitor might in the future be considered for SDC with an immune-excluded environment and increased angiogenesis.

In contrast to the modest SDC responses, our patients with AdCC did not show a response to nivolumab monotherapy in this trial, with an ORR of 0%. Previous studies have also shown limited efficacy of anti-PD-1 monotherapy or a combination of nivolumab plus ipilimumab in AdCC cohorts [21, 22, 24, 27, 28]. These results reproducibly confirm that ICI might not be efficacious for patients with AdCC. Several reports have investigated the reason for this. Gene expression analysis showed a lower immune score and lower immune cell infiltration into the tumor in AdCC compared to other histological types [39]. Further, TIME analysis using multiplex fluorescence immunohistochemistry showed little immune cell infiltration in either tumor or stroma, suggesting an immune-desert environment [36]. Therefore, new therapeutic strategies that overcome this resistance to immunotherapy are needed. In this regard, we have focused on antibody–drug conjugates (ADCs), which have been reported to cause immunogenic cell death as one mode of action and may enhance the immunogenicity of TIME [40, 41]. If the target protein for ADCs is present, the combination of ICIs and ADCs may be effective in tumors with an immune-desert environment through synergistic effects. In fact, the combination of ICIs and ADCs has been proven effective in urothelial carcinoma and is now recognized as standard therapy for these patients [42]. Accordingly, for AdCC, given the reportedly high expression of trophoblast cell-surface antigen-2 (Trop-2), a potential target of ADCs, and presence of an immune-desert environment [36, 43, 44], the combination of ICI and Trop-2 ADCs can be considered a therapeutic strategy for AdCC.

A second promising treatment is chimeric antigen receptor-T cells (CAR-Ts). CAR-Ts are gene-modified T cells that produce a protein that fuses the target recognition site of an antibody with the intracellular molecules necessary to activate the T cell. Although CAR-Ts requires sufficient expression of the target [45], common histological types of SGC are characterized by high Trop-2 [43, 44]. While the efficacy of ICI monotherapy is limited in AdCC with an immune-desert environment, CAR-T may be expected to be effective in AdCC. In addition to immunotherapeutic strategies, recent efforts to directly target MYB, a key oncogenic driver in AdCC, have also emerged. Notably, the MYB transcription factor inhibitor RGT-61159 has shown preclinical antitumor activity in patient-derived xenograft models, and a first-in-human study has been initiated for relapsed/refractory AdCC. These precision-oncology approaches, together with cell-based immunotherapy, may offer new therapeutic options for patients with AdCC who derive limited benefit from ICIs [46, 47].

This study has a few limitations. First, the number of patients in this phase II trial was necessarily small owing to the rarity of SGC. Because of the rarity of SGC, we did not limit histological types for inclusion and set a one-sided alpha of 0.1. Therefore, the results should be carefully interpreted. Although our results suggest the potential efficacy of nivolumab in SDC, no clinical benefit was seen in other histological types. Given that the number of patients with each histological type was very small, however, drawing conclusions is difficult. Solving this issue will require the establishment of an infrastructure that enables international collaborative trials for rare cancers, including SGC. Second, differences in antitumor activity by histological type may reflect differences in TIME. To clarify this issue, we are now conducting translational research, including next-generation sequencing, PD-L1 expression, IHC, and microbiome analysis of saliva and stool, to investigate the association between the efficacy of nivolumab and TIME in SGC using archival tissues collected in this phase II trial.

Conclusion

This phase II trial of nivolumab monotherapy for patients with platinum-refractory RM-SGC did not meet its primary endpoint of ORR. On evaluation by histological type, SDC showed a modest response to nivolumab, whereas the other histological types, including AdCC, showed no efficacy at all. Although nivolumab monotherapy may be worth further development in SDC, these results may raise concerns over nivolumab monotherapy for RM-SGC.

Author contributions

Yoshiaki Nagatani (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Writing—original draft), Naomi Kiyota (Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Supervision, Writing— original draft, Writing—review and editing), Tomoko Yamazaki (Writing—review and editing), Yukinori Asada (Writing—review and editing), Masaaki Higashino (Writing—review and editing), Hironaga Satake (Writing—review and editing), Shogen Boku (Writing—review and editing), Ari Nishimura (Writing—review and editing), Hirokazu Uemura (Writing—review and editing), Ichiro Ota (Writing—review and editing), Katsunari Yane (Writing—review and editing), Kaoru Tanaka (Writing—review and editing), Takuma Onoe (Writing—review and editing), Yuji Hirayama (Writing—review and editing), Takahiro Tsujikawa (Writing—review and editing), Hajime Fujiwara (Writing—review and editing), Hikari Shimoda (Writing—review and editing) Ken-Ichi Nibu (Writing—review and editing), Yoshinori Imamura (Writing—review and editing), Shiro Kimbara (Writing—review and editing), Taiji Koyama (Writing—review and editing), Yohei Funakoshi (Writing—review and editing), and Hironobu Minami (Supervision, Writing—review and editing)

Conflict of interest statement

Y.N. reports a relationship with Ono Pharmaceutical Co., Ltd., Bristol-Myers Squibb Co., MSD Co., Ltd., and Chugai Pharmaceutical Co., Ltd. that includes speaking and lecture fees. N.K. reports a relationship with Ono Pharmaceutical Co., Ltd., Bristol-Myers Squibb Co., AstraZeneca Co., Ltd., Boehringer Ingelheim, Bayer, GSK, AbbVie, Merck Biopharma, and Merck Sharp & Dohme that includes grants, honoraria, and advisory roles. T.Y. reports a relationship with Merck Biopharma that includes funding. H.S. reports a relationship with Ono Pharmaceutical Co., Ltd., MSD Co., Ltd., AstraZeneca Co., Ltd., Bristol-Myers Squibb Co., Chugai Pharmaceutical Co., Ltd., and Merck Bio Pharma, MSD Co., Ltd. that includes funding or grants and speaking and lecture fees. S.B. reports a relationship with Ono Pharmaceutical Co., Ltd., Bristol-Myers Squibb Co., MSD Co., Ltd., Chugai Pharmaceutical Co., Ltd., and AstraZeneca Co., Ltd. that includes speaking and lecture fees. K.T. reports a relationship with Ono Pharmaceutical Co., Ltd., Bristol-Myers Squibb Co., MSD Co., Ltd., Chugai Pharmaceutical Co., Ltd., AstraZeneca Co., Ltd., and Merck Biopharma that includes speaking and lecture fees. T.T. reports a relationship with Ono Pharmaceutical Co., Ltd., Bristol-Myers Squibb Co., Merck Biopharma, and Merck Sharp & Dohme that includes speaking and lecture fees and advisory roles. Y.I. reports a relationship with MSD Co., Ltd. that includes speaking and lecture fees. T.K. reports a relationship with Bristol-Myers Squibb Co. and Merck Biopharma that includes speaking and lecture fees. H.M. reports a relationship with Chugai Pharmaceutical Co., Ltd. that includes honoraria and funding. For other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Funding

This study was funded by a Grant-in-Aid for Scientific Research C (Grant number: 17K11384, 23K08963) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Data availability

The data sets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Prior presentation

Presented at the American Society of Clinical Oncology Annual Meeting, Chicago, IL, June 2–6, 2023 (abst 6092, Poster Session).