BevRam-GC01 study protocol: a phase I trial of bevacizumab plus ramucirumab and paclitaxel in advanced gastric cancer

Abstract

Background

Plasma vascular endothelial growth factor-A (VEGF-A) concentrations markedly increased immediately after ramucirumab administration, and the high VEGF-A concentrations were significantly associated with worse outcomes in advanced gastric cancer. In a preclinical study, combination therapy with anti-VEGF-A and anti-vascular endothelial growth factor receptor 2 (VEGFR2) antibodies showed superior tumor suppression compared with monotherapy, suggesting dual VEGF-A/VEGFR2 blockade may be a promising therapeutic strategy.

Design

The BevRam-GC01 trial is a phase I trial evaluating the tolerability and safety of bevacizumab biosimilar (5 or 10 mg/kg) plus ramucirumab (8 mg/kg) and paclitaxel (80 mg/m²) in patients with advanced gastric cancer refractory to first-line fluoropyrimidine–platinum chemotherapy. The study includes a safety part and an expansion part, in which, after confirming tolerability, patients will be randomized into three arms, including a control group, to obtain proof of concept and determine the optimal BEV-BS dose. Primary endpoints are dose-limiting toxicities in the safety part and adverse events in the expansion part. Key secondary endpoints include objective response rate and day 8 free VEGF-A suppression rate. Biomarker analysis will clarify the synergistic and complementary effects of anti-VEGFR2 and anti-VEGF-A antibodies on tumor biology, including angiogenesis, proliferation, and antitumor immunity, through multi-omics analysis and immune monitoring.

Graphical abstract

(A) The mechanism of dual blockade of the VEGF-A/VEGFR2 pathway using antibody-based agents. Plasma VEGF-A concentrations markedly increased immediately after RAM administration, and excess VEGF-A may have several potential detrimental effects, including the induction of an immune-cold TME and tumor proliferation via alternative signaling such as VEGFR1 and NRP1. BEV-BS may neutralize excess VEGF-A and thereby may help counteract the detrimental effects induced by ramucirumab. (B) The study schema of BevRam-GC01. The study includes a safety part and an expansion part, in which, after confirming tolerability, patients will be randomized into three arms, including a control group, to obtain proof of concept and determine the optimal BEV-BS dose through evaluation of safety and efficacy. Arrow color indicates signaling strength: red = signal transduction, cream = residual signal transduction, and blue = suppressed signal transduction. AEs, adverse events; BEV-BS, bevacizumab-biosimilar; BP, blood pressure; DCR, disease control rate; DLTs, dose-limiting toxicities; ECOG PS, Eastern Cooperative Oncology Group performance status; FU, fluoropyrimidine; No. of mets. orgs, number of metastatic organs; NRP1, neuropilin-1; ORR, objective response rate; OS, overall survival; PAC, paclitaxel; PFS, progression-free survival; RAM, ramucirumab; TME, tumor microenvironment; VEGF-A, vascular endothelial growth factor-A; VEGFR1, vascular endothelial growth factor receptor 1; VEGFR2, vascular endothelial growth factor receptor 2.

Highlights

• •Plasma VEGF-A concentrations markedly increased after ramucirumab administration and were associated with shorter survival.
• •Preclinical study revealed that neutralization of excess VEGF-A by anti-VEGF-A antibody showed superior tumor suppression.
• •This study evaluates a novel therapeutic concept of dual inhibition of the VEGF-A/VEGFR2 pathway with antibody agents.
• •Biomarker analysis is expected to assess the tumor biology, including angiogenesis, proliferation, and antitumor immunity.

Description of protocol

Introduction

According to global statistics in 2022, gastric cancer accounted for 968 350 new cases and 659 853 deaths worldwide, remaining the fifth most common cancer and the fourth leading cause of cancer-related death, which indicates persistently high incidence and mortality.¹ Although therapeutic advances have extended survival, the prognosis for patients with advanced gastric cancer (AGC) remains poor, with a median overall survival (OS) of ∼1 year globally and around 1.5 years in Japan and South Korea.²

The first-line standard treatment of AGC is the combination of a fluoropyrimidine and a platinum agent (FP).3, 4, 5 In human epidermal growth factor receptor 2 (HER2)-positive gastric cancer, the addition of trastuzumab is the standard first-line treatment, based on the results of the ToGA trial.⁶ Recently, the KEYNOTE-811 trial demonstrated the efficacy of adding pembrolizumab to trastuzumab plus FP, with particularly enhanced treatment efficacy observed in patients with combined positive score (CPS) ≥1.⁷ Accordingly, the addition of pembrolizumab is recommended for this population with CPS ≥1. In HER2-negative gastric cancer, first-line treatment with FP in combination with nivolumab is recommended for patients with CPS ≥5 based on the results of the CheckMate 649 trial,⁸ while combination therapy with pembrolizumab has shown efficacy in the KEYNOTE-859 trial, regardless of programmed death-ligand 1 (PD-L1) expression status.⁹ Furthermore, for CLDN18.2-positive gastric cancer, combination therapy of zolbetuximab with chemotherapy has shown to be effective in the SPOTLIGHT and GLOW trials.^10,11 Thus, first-line therapy for AGC is becoming increasingly precise, whereas second-line treatment remains less developed. In second-line treatment, the RAINBOW trial, which evaluated ramucirumab, an anti-vascular endothelial growth factor receptor 2 (VEGFR2) antibody, in combination with paclitaxel, demonstrated superior progression-free survival (PFS) and OS compared with paclitaxel alone; however, the outcomes were still suboptimal, with a median PFS of 4.4 months and a median OS of 9.6 months.¹² Given that a high implementation rate of second-line therapy is strongly correlated with post-first-line survival and OS in AGC, developing more effective second-line treatments is essential to achieve further improvements in patient outcomes.¹³

In phase I trials of ramucirumab, circulating vascular endothelial growth factor-A (VEGF-A) concentrations were observed to markedly increase immediately after administration.14, 15, 16 We previously reported that plasma VEGF-A concentrations on cycle 1 day 8 increased by 14.5-fold compared with baseline, and that patients with high VEGF-A concentrations had significantly worse outcomes than those with low concentrations.¹⁷ In a mouse xenograft model, a combination of anti-mouse VEGF-A antibody and the anti-mouse VEGFR2 antibody (DC101) demonstrated superior tumor growth inhibition compared with either monotherapy.¹⁸ In this preclinical study, although plasma total VEGF-A, comprising both free VEGF-A and VEGF-A bound to anti-mouse VEGF-A antibody, increased, the free VEGF-A, measured after removal of VEGF-A bound to anti-mouse VEGF-A antibody by immunoprecipitation, was suppressed. Several clinical studies have also reported that plasma total VEGF-A increases during bevacizumab treatment, whereas free VEGF-A (not bound to bevacizumab) is suppressed, and lower free-to-total VEGF-A ratio correlates with treatment efficacy.^19,20 These data suggest that maintaining low concentrations of free VEGF-A may be critical for therapeutic benefit in antiangiogenic therapy. Moreover, in the aforementioned preclinical study, transcriptomic analysis revealed that the expression of angiogenesis-related genes was suppressed in both DC101 monotherapy and combination therapy groups, whereas the genes related with stem cell-like properties and histone modifications were up-regulated in the DC101 monotherapy group but suppressed in the combination therapy group,¹⁸ suggesting that co-administration of the anti-VEGF-A antibody may help counteract the detrimental effects associated with anti-VEGFR2 monotherapy by neutralizing excess VEGF-A. Furthermore, recent preclinical and clinical data suggest that antiangiogenic agents may contribute to immune remodeling through the normalization of abnormal tumor vasculature.²¹ In light of these results, we hypothesized that dual inhibition of the VEGF-A/VEGFR2 pathway using anti-VEGF-A and anti-VEGFR2 antibodies could serve as a novel therapeutic option for AGC. Accordingly, we designed a phase I investigator-initiated clinical trial to evaluate the combination of ramucirumab and paclitaxel with bevacizumab biosimilar (BEV-BS).

Objective

The objectives of this study are to explore safety, tolerability, pharmacodynamics, and antitumor activity of BEV-BS plus ramucirumab with paclitaxel combination therapy in patients with AGC, to obtain clinical proof of concept (POC) for dual blockade of the VEGF-A/VEGFR2 pathway, and to determine the optimal dose of BEV-BS.

Study design and treatment

The BevRam-GC01 trial is a prospective, open-label, randomized, single-center, phase I study (Table 1). This trial has been registered with the Japan Registry of Clinical Trials (jRCTs 2031240512), with launch in January 2025, and is currently ongoing. The study design is illustrated in Figure 1. This study consists of a safety part and an expansion part. The safety part will evaluate tolerability of the triple combination therapy based on a 3 + 3 design, and the expansion part will assess its safety and efficacy. Two dose levels of BEV-BS are prepared: 5 mg/kg (level 0) and 10 mg/kg (level 1). Patients received paclitaxel (80 mg/m² intravenously (i.v.) on days 1, 8, and 15) and ramucirumab (8 mg/kg i.v. on days 1 and 15) in combination with BEV-BS, which was administered i.v. at the assigned dose level, synchronized with the ramucirumab schedule during the first four cycles, to focus on early pharmacodynamic effects while mitigating potential cumulative toxicities associated with dual blockade of the VEGF-A/VEGFR2 pathway. If dose level 0 is tolerable, dose level 1 will be initiated, which is the same as dose level 0 except that BEV-BS is administered at 10 mg/kg. If dose level 1 is tolerable, the expansion part will be conducted with three arms, randomized in a 1 : 1 : 1 ratio to receive dose level 0, dose level 1, or a control arm receiving paclitaxel plus ramucirumab at standard doses. If dose level 1 is not tolerable, randomization will be conducted between two arms: dose level 0 and the control arm. If dose level 0 is not tolerable, the study will proceed to dose level −1, which is the same as dose level 0 except that BEV-BS is administered at 2.5 mg/kg. If dose level −1 is tolerable, randomization will be conducted between two arms: dose level −1 and the control arm. If dose level −1 is not tolerable, the study will be terminated. A stratification factor for randomization is the number of metastatic organs (≥2 versus <2), which was most affected to plasma VEGF-A concentration on cycle 1 day 8.¹⁷ Three dose levels including the initial dose are set for dose reduction: 80, 70, and 60 mg/m² for paclitaxel; and 8, 6, and 5 mg/kg for ramucirumab. No dose reductions are set for BEV-BS. Dual blockade of the VEGF-A/VEGFR2 pathway raises concerns regarding hypertension, proteinuria, thromboembolism, and gastrointestinal perforation. Therefore, for safety purposes, all patients will be hospitalized until cycle 1 day 8 and closely monitored, with scheduled safety visits on day 22 during the first two cycles. Before study initiation, information was shared and a multidisciplinary collaborative framework was established involving gastric cancer surgeons, cardio-oncology specialists, and the emergency medicine department.

Table 1.

Specifications table

Subject area	Medicine and dentistry
More specific subject area	Gastroesophageal adenocarcinoma
Name of your trial in progress	BevRam-GC01 study protocol: a phase I trial of bevacizumab plus ramucirumab and paclitaxel in advanced gastric cancer
Reagents/tools	Bevacizumab biosimilar (BEV-BS) (Celltrion Inc., Incheon, Republic of Korea)
Trial design	The BevRam-GC01 trial is a prospective, open-label, randomized, single-center, phase I trial to evaluate the tolerability and safety of bevacizumab biosimilar (BEV-BS) (5 or 10 mg/kg, intravenously (i.v.) on days 1 and 15) in combination with ramucirumab (8 mg/kg, i.v. on days 1 and 15) + paclitaxel (80 mg/m2, i.v. on days 1, 8, and 15), to obtain proof of concept (POC), and to determine the optimal BEV-BS dose. BEV-BS will be administered through cycle 4. The study includes a safety part and an expansion part, in which, after confirming tolerability based on a 3+3 design, patients will be randomized into three arms, including a control group, to obtain POC and determine the optimal BEV-BS dose through evaluation of safety and efficacy. Primary endpoints are the incidence of dose-limiting toxicities in the safety part and adverse events in the expansion part. Key secondary endpoints include objective response rate and day 8 free VEGF-A suppression rate in cycle 1. Key inclusion criteria are as follows: (i) age between 18 and 75 years, (ii) ECOG performance status is 0 or 1, (iii) refractory or intolerant to first-line chemotherapy fluoropyrimidine–platinum chemotherapy, (iv) histologically confirmed unresectable or recurrent gastric or esophagogastric cancer, (v) urine protein level of 1+ or lower, (vi) blood pressure is adequately controlled. The minimum sample size is 27, considering feasibility: 6 subjects (3 per arm × 2 arms) in the safety part and 21 (7 per arm × 3 arms) in the expansion part.
Trial registration	jRCT2031240512
Ethics	This trial was approved by the Institutional Review Board of the Cancer Institute Hospital of JFCR (2024-0063). Before enrollment, written informed consent will be obtained from all patients.
Value of the trial in progress	•The BevRam-GC01 study is the first trial evaluating the tolerability and safety of dual inhibition of the VEGF-A/VEGFR2 pathway using anti-VEGF-A and anti-VEGFR2 antibodies. •A novel therapeutic concept of dual inhibition of the VEGF-A/VEGFR2 pathway with antibody agents is expected to be established, with potential applicability to other solid tumors. •Biomarker analysis will be held to evaluate not only angiogenesis but also tumor biology, including proliferation and antitumor immunity, through multi-omics analysis and immune monitoring.

ECOG, Eastern Cooperative Oncology Group; VEGF-A, vascular endothelial growth factor-A; VEGFR2, vascular endothelial growth factor receptor 2.

Figure 1.

The study schema of BevRam-GC01. The study includes a safety part and an expansion part, in which, after confirming tolerability, patients will be randomized into three arms, including a control group, to obtain proof of concept and determine the optimal BEV-BS dose through evaluation of safety and efficacy. AEs, adverse events; BEV-BS, bevacizumab biosimilar; BP, blood pressure; DCR, disease control rate; DLT, dose-limiting toxicity; ECOG PS, Eastern Cooperative Oncology Group performance status; FU, fluoropyrimidine; No. of mets. orgs, number of metastatic organs; ORR, objective response rate; OS, overall survival; PAC, paclitaxel; PFS, progression-free survival; RAM, ramucirumab; VEGF-A, vascular endothelial growth factor-A.

Measurement of plasma VEGF-A and evaluation of free VEGF-A concentration

Blood samples will be collected in 5 ml EDTA-2 Na tubes at baseline and on cycle 1 day 8. After collection, the blood will be centrifuged at 3000 g for 10 minutes at room temperature, and 2 ml of plasma will be stored at –80°C until VEGF-A measurement. VEGF-A concentrations will be measured using ELISA kits (R&D Systems, Minneapolis, MN) according to the manufacturer’s protocol. To measure free VEGF-A, plasma samples will be immunodepleted to remove VEGF-A bound to BEV-BS using Dynabeads Protein G (Thermo Fisher Scientific, Waltham, MA) before measurement, following previously described protocols.²² The day 8 free VEGF-A suppression rate will be calculated as:

$$\text{Suppression}\text{rate}=\frac{\text{Day}8\text{free}\text{VEGF}-\mathrm{A}-\text{Day}1\text{VEGF}-\mathrm{A}}{\text{Day}1\text{VEGF}-\mathrm{A}}$$

Eligibility criteria

Patients with histologically confirmed unresectable or recurrent gastric or esophagogastric cancer, who are refractory or intolerant to first-line chemotherapy with a combination of FP, are being enrolled. Key additional inclusion criteria are as follows: (i) age: 18-75 years, (ii) Eastern Cooperative Oncology Group performance status (ECOG PS): 0 or 1, (iii) presence of measurable lesions, (iv) urine protein level ≤1+, and (v) adequately controlled blood pressure (see Table 2 for full eligibility criteria). Considering safety with dual blockade of the VEGF-A/VEGFR2 pathway, these key inclusion criteria were more stringently defined than those used in the RAINBOW trial.

Table 2.

Eligibility criteria

Inclusion criteria 1.Age between 18 and 75 years at the time of informed consent. 2.Eastern Cooperative Oncology Group performance status of 0 or 1. 3.Documented progression or intolerance to chemotherapy containing a fluoropyrimidine (5-FU, capecitabine, or S-1) and a platinum agent (cisplatin or oxaliplatin), with a history of only one prior regimena. Use of agents such as nivolumab, trastuzumab, or zolbetuximab in the first-line setting is allowed. 4.Histologically (including cytology) confirmed unresectable advanced or recurrent gastric cancer (including gastroesophageal junction cancer), diagnosed as adenocarcinoma (papillary adenocarcinoma, tubular adenocarcinoma, poorly differentiated adenocarcinoma), signet-ring cell carcinoma, or mucinous adenocarcinoma, according to the Japanese Classification of Gastric Carcinoma, 15th edition (General Types of Malignant Epithelial Tumors). 5.At least one measurable lesion confirmed by contrast-enhanced CT of the chest, abdomen, and pelvis (slice thickness ≤5 mm) within 28 days before enrollmentb. 6. Adequate organ function based on laboratory tests carried out within 14 days before enrollment, without administration of granulocyte colony-stimulating factor or blood transfusion within 14 days before sampling: (a) Absolute neutrophil count ≥1500/mm3 (b) Platelet count ≥100 000/mm3 (c) Hemoglobin ≥9.0 g/dl (d) AST and ALT ≤3 × ULN, or ≤5 × ULN in case of liver metastases (e) Total bilirubin ≤1.5 × ULN (f) Serum creatinine ≤1.5 mg/dl, or creatinine clearance ≥50 ml/min (Cockcroft–Gault or 24-hour urine collection) (g) PT-INR ≤1.5 or prothrombin time ≤1.5 × ULN (h) aPTT ≤1.5 × ULN (i) Urinary protein negative (−), trace (±), or 1+ (dipstick or automated analysis) 7.Well-controlled blood pressure: at enrollment, controlled with only one antihypertensive agent (excluding fixed-dose combination antihypertensives), with systolic BP ≤140 mmHg and diastolic BP ≤90 mmHg. 8.Expected survival of at least 3 months from the date of enrollment. 9.All clinically significant toxicities from prior therapy must have resolved to Grade ≤1 (NCI CTCAE version 5.0), except for alopecia, dysgeusia, hyperpigmentation, nail changes, or other non-serious/non-life-threatening toxicities. Endocrine disorders managed with hormone replacement therapy are allowed if stable. 10.Agreement to use adequate contraception during the protocol treatment and for 6 months after completion. 11.Ability to provide peripheral blood samples for VEGF-A measurement (reference: 7 ml per collection). 12.Written informed consent provided by the patient before study participation.

Exclusion criteria 1.Histopathological diagnosis corresponding to a special type of malignant epithelial tumor, as defined in the 15th edition of the Japanese Classification of Gastric Carcinoma. 2.Prior treatment with paclitaxel, ramucirumab, or bevacizumab. 3.Contraindication to paclitaxel, ramucirumab, or bevacizumab. 4.History of major surgery, including stoma creation or bypass surgery, within 28 days before enrollment. 5.Receipt of radiotherapy within 28 days before enrollment (if the irradiated site includes measurable lesions, they may be considered measurable lesions if progression is confirmed after irradiation). 6.Evidence of central nervous system (brain, spinal cord, meninges) metastasis. 7.Receiving warfarin or antiplatelet agents, or taking direct oral anticoagulants with an intact primary lesion. 8.History of active bleeding from gastric cancer or gastrointestinal ulcer within 3 months before enrollment. 9.History of gastrointestinal perforation or gastrointestinal fistula. 10.History of deep vein thrombosis, pulmonary embolism, portal vein thrombosis, or other serious venous thromboembolism within 3 months before enrollment (venous port-related thrombosis or superficial venous thrombosis is not considered serious). 11.History of arterial thromboembolism (myocardial infarction, unstable angina, cerebral infarction, transient ischemic attack, etc.) within 6 months before enrollment. 12.History of hypertensive encephalopathy or hypertensive crisis. 13.Presence of deep ulcerative lesions (including those at the primary tumor site) with risk of perforation. 14.Complication of inflammatory bowel disease requiring treatment (immunomodulators, immunosuppressants, biologics, or surgery, etc.) within 12 months before enrollment. 15.Serious comorbidities (e.g. interstitial pneumonia or pulmonary fibrosis, heart failure, renal failure, hepatic failure, diabetes mellitus). 16. Presence of or history within 6 months before enrollment of the following cardiac diseases/conditions: (a) History of congestive heart failure (b) High-risk, uncontrolled arrhythmia (c) Clinically significant valvular heart disease 17.Presence of massive ascites continuously extending from the pelvic cavity into the upper abdomen on CT imaging, or history of paracentesis within 3 months before enrollment, or requirement for regular paracentesis. 18.Inability to maintain adequate oral intake. 19.Active infection (fever ≥38°C). Positive for HIV antibody, HBs antigen, or HCV antibody (subjects who are HCV antibody-positive but HCV RNA-negative are not excluded). 20.Positive for HBs antibody or HBc antibody and HBV DNA. Subjects are eligible if, however, after antiviral treatment (e.g. nucleoside/nucleotide analogues), HBV DNA is <1.3 log IU/ml (2.1 log copies/ml). 21.Active double cancer (synchronous or metachronous malignancy with a disease-free interval of <5 years). Lesions of carcinoma in situ are not considered active double cancer. 22.Pregnant or breastfeeding women (subjects are ineligible even if breastfeeding is interrupted), or women who may be pregnant. 23.Conditions that may make compliance with study procedures and follow-up difficult (e.g. psychiatric disorders interfering with daily life, difficulty in understanding the informed consent document, or subjects residing abroad planning to return home). 24.Any other condition that, in the opinion of the investigator, makes the subject unsuitable for participation in this study.

ALT, alanine aminotransferase; aPTT, activated partial thromboplastin time; AST, aspartate aminotransferase; BP, blood pressure; CT, computed tomography; HBc antibody, hepatitis B core antibody; HBs antigen, hepatitis B surface antigen; HBV, hepatitis B virus; HCV, hepatitis C virus; HIV, human immunodeficiency virus; NCI CTCAE, National Cancer Institute – Common Terminology Criteria for Adverse Events; PT-INR, prothrombin time – international normalized ratio; ULN, upper limit of normal; VEGF-A, vascular endothelial growth factor A.

^aPatients who relapse during or within 179 days after perioperative or adjuvant chemotherapy containing a fluoropyrimidine and a platinum agent are eligible. Patients who relapse ≥180 days after adjuvant chemotherapy with S-1 monotherapy or docetaxel/S-1, and subsequently become refractory or intolerant to fluoropyrimidine- and platinum-based chemotherapy, are also eligible.

^bMeasurable lesion: lesions with the longest diameter ≥10 mm (non-lymph nodes) or short axis ≥15 mm (lymph nodes) on CT scans. Lesions in irradiated fields can be considered measurable if progression is documented after radiotherapy. Note: in patients with contrast allergy or impaired renal function, non-contrast CT or MRI is acceptable.

Endpoints and assessments

The primary endpoints are the incidence of dose-limiting toxicities (DLTs) in the safety part and adverse events (AEs) in the expansion part, respectively. Key secondary endpoints include the objective response rate (ORR) and the day 8 free VEGF-A suppression rate in the expansion part. Secondary endpoints include the incidence of AEs in the safety part, OS, PFS, and disease control rate (DCR) in the expansion part. The incidence of DLTs will be assessed based on the predefined DLTs criteria during the first cycle (Table 3). If the incidence rate of DLTs is <16.7%, the dose level will be considered tolerable. The severity of AEs will be graded according to the National Cancer Institute Common Terminology Criteria for AEs (version 5.0). DLTs will be assessed in the DLTs-assessable population, defined as subjects who were enrolled in the safety part and have received the protocol-specified treatment. The incidence of AEs will be evaluated in the safety analysis set, defined as subjects who were enrolled in the expansion part and received the protocol-specified treatment with available safety data. ORR, DCR, PFS, and OS will be assessed in the following analysis sets: (i) the full analysis set (FAS), consisting of subjects who were enrolled in the expansion part and randomized, and received at least one dose of the study treatment; (ii) the per protocol set (PPS), consisting of subjects in the FAS who sufficiently complied with the protocol without major protocol deviations; and (iii) the POC analysis set, defined as subjects in the PPS who had available VEGF-A data in cycle 1.

Table 3.

Definition of dose-limiting toxicities

Hematologic

•Grade 4 neutropenia lasting ≥7 days

•Grade ≥3 febrile neutropenia

•Grade 4 anemia

•Grade ≥3 thrombocytopenia

Non-hematologic

•Uncontrolled hypertension (systolic ≥160 mmHg or diastolic ≥100 mmHg) persisting ≥2 weeks despite appropriate management

•Grade ≥3 non-hematologic toxicity (excluding hypertension) deemed clinically significant or life-threatening by the investigator

•Inability to receive bevacizumab biosimilar and ramucirumab on day 15 in cycle 1

•Inability to receive paclitaxel on both day 8 and day 15 in cycle 1

Computed tomography or magnetic resonance imaging will be carried out within 4 weeks before the initiation of treatment, repeated every 6 ± 2 weeks during the first six cycles, and every 8 ± 2 weeks thereafter. Tumor assessment will be conducted using the Revised Response Evaluation Criteria in Solid Tumors guideline (version 1.1). ORR is defined as the proportion of patients who achieve a complete response (CR) or partial response (PR) as their best overall response (BOR), confirmed at least 4 weeks after the initial response. DCR is defined as the proportion of patients who achieve CR, PR, or stable disease as their BOR. PFS is defined as the time from the date of enrollment to either the first objective evidence of disease progression or death from any cause. OS is defined as the time from the date of enrollment to death from any cause.

The POC assessment in this study is exploratory and hypothesis-generating and is not intended for formal hypothesis testing or confirmatory comparisons between treatment arms. To obtain clinical POC, the day 8 free VEGF-A suppression rate will be assessed in a POC analysis set. POC will be considered achieved if the point estimate of the ORR in either of the experimental arms exceeds that of the control arm, and the point estimate of the day 8 free VEGF-A suppression rate in the corresponding arm is lower than that in the control arm, or if the ORR point estimate exceeds that of the standard treatment group but the day 8 VEGF-A suppression rate does not, the principal investigator will determine whether clinical POC has been achieved, taking into account the 95% confidence intervals of the point estimates and the distribution of day 8 VEGF-A suppression (Table 4). In addition, the optimal dose of BEV-BS will be determined based on an integrated descriptive review of toxicity, VEGF-A modulation, and preliminary efficacy.

Table 4.

Exploratory POC assessment framework

Experimental arm	ORR point estimation	Day 8 free VEGF-A suppression rate	Interpretation
BEV-BS 5 mg/kg	Higher	Lower	POC achieved
BEV-BS 5 mg/kg	Higher	Not lower	POC considered based on integrated assessment
BEV-BS 5 mg/kg	Not higher	Any	Not achieved
BEV-BS 10 mg/kg	Higher	Lower	POC achieved
BEV-BS 10 mg/kg	Higher	Not lower	POC considered based on integrated assessment
BEV-BS 10 mg/kg	Not higher	Any	Not achieved

BEV-BS, bevacizumab biosimilar; ORR, objective response rate; POC, proof of concept; VEGF-A, vascular endothelial growth factor-A.

Translational research

In this trial, in addition to plasma samples for VEGF-A measurement, serial endoscopic biopsy specimens and peripheral blood mononuclear cells will be collected. Biomarker analyses will include the evaluation of angiogenesis, proliferation, and tumor immunity using multiplex immunohistochemistry (IHC), multi-omics approaches, and multiparameter flow cytometry, thereby assessing the synergistic and complementary effects of combining anti-VEGFR2 and anti-VEGF-A antibodies on these tumor-biological processes.

Sample size calculation

Considering feasibility, the minimum number of subjects required to evaluate the tolerability, and safety of the triple combination therapy, the sample size is set at 6 subjects (3 per arm × 2 arms) for the safety part, and 21 subjects (7 per arm × 3 arms) for the expansion part, for a total of 27 subjects. If the sample size for the triple combination in the expansion part is 14 subjects, and if the frequency of a major AE, which is expected to have an increased risk with the addition of BEV-BS, is ∼40%, the confidence interval width for a two-sided 95% confidence interval would be estimated with a precision of approximately ±27.5%. The enrollment period has been set to 1.5 years, and the follow-up period has been set to 0.5 years from enrollment of the last patients.

Statistical analysis

For categorical outcomes including ORR, the frequency, proportion, and, if applicable, the two-sided 95% confidence interval based on the Clopper–Pearson method will be estimated. Continuous outcomes, including the day 8 free VEGF-A suppression rate, will be summarized using descriptive statistics, and, if applicable, the two-sided 95% confidence interval for the mean will be estimated. For survival outcomes, the Kaplan–Meier method will be used to estimate the median survival time, survival proportions at each time point, hazard ratio, and their 95% confidence intervals in a descriptive and exploratory manner. No formal powered comparisons are intended in this phase I setting. All statistical analysis will be carried out using SAS version 9.4 (SAS Institute Inc., Cary, NC). Details of the statistical analyses will be described in a predefined statistical analysis plan finalized before data lock.

Discussion

Angiogenesis plays an important role not only in fetal development and wound healing, but also in the growth, invasion, and metastasis of solid tumors. The principal factors driving angiogenesis are VEGF-A and its receptor VEGFR2. In tumors, aberrant angiogenesis leads to increased vascular permeability due to the formation of abnormal blood vessels.²³ Consequently, elevated interstitial fluid pressure within the tumor results in reduced delivery of anticancer agents and hypoxia, thereby contributing to resistance to chemotherapy and radiotherapy.24, 25, 26 Furthermore, the up-regulation of anaerobic glycolysis affects tumor immunity by reducing cytotoxic T cells in the tumor microenvironment (TME).^27,28 Accordingly, overexpression of VEGF-A and VEGFR2 in tumor tissues, as well as elevated circulating VEGF-A concentrations, have been reported to be poor prognostic factors, and therapeutic strategies targeting these factors have been developed in several types of cancers, including colorectal, non-small-cell lung, and renal cell cancer.29, 30, 31 In gastric cancer, the AVAGAST trial failed to demonstrate the efficacy of bevacizumab as a first-line treatment in combination with FP.³² In this trial, regional differences in bevacizumab efficacy were observed; specifically, higher efficacy was noted in Pan America and Europe compared with Asia, where 90% of patients were from Japan and Korea. Subsequent biomarker analyses suggested that high circulating VEGF-A concentrations may serve as a predictive marker of bevacizumab, and demonstrated that circulating VEGF-A concentrations were significantly lower in Asian patients compared with those in other regions.³³ Therefore, circulating VEGF-A concentration is suggested to be involved not only in prognosis but also in the therapeutic efficacy of bevacizumab. Given the regional differences in bevacizumab efficacy and circulating VEGF-A concentrations in the AVAGAST trial, as well as the single-center Japanese design of this study, future multicenter or global studies are required to ensure broader generalizability.

Our previous data suggest that circulating VEGF-A increased after ramucirumab administration, and high VEGF-A concentration on cycle 1 day 8 was associated with worse survival.¹⁷ The exact mechanisms by which high VEGF-A is associated with shorter survival under ramucirumab treatment remain unclear. Because excess VEGF-A can bind to other receptors, such as VEGFR1 or neuropilin-1 (NRP-1), it may activate alternative pathways. VEGFR1 has been reported to be involved in tumor proliferation and is expressed in 76% of gastric cancers.³⁰ Therefore, it is hypothesized that elevated VEGF-A in response to VEGFR2 inhibition may contribute to tumor proliferation through VEGFR1-mediated signaling. In addition, VEGF-A may further promote immune suppression via VEGFR1 or NRP-1 expressed in the TME. Specifically, VEGF-A stimulates VEGFR1 on myeloid-derived suppressor cells and NRP-1 on regulatory T cells, thereby promoting immune tolerance.^34,35 Furthermore, preclinical experiments suggest that tumor-promoting pathways, including stem cell-like properties and histone modifications, were up-regulated in the DC101 (anti-mouse VEGFR2 antibody) monotherapy group. In contrast, these pathways were suppressed in the combination therapy group, which included an anti-mouse VEGF-A antibody.¹⁸ These data suggest that excess VEGF-A in response to ramucirumab may contribute to these detrimental effects and support the notion that targeting excess VEGF-A with a neutralizing antibody could be beneficial (Figure 2).

Figure 2.

The scheme of dual blockade of the VEGF-A/VEGFR2 pathway using antibody-based agents. VEGF-A binding to VEGFR2 promotes angiogenesis, immune-cold TME, and tumor proliferation. BEV-BS and RAM suppress VEGFR2 signaling but RAM induces excess VEGF-A. Excess VEGF-A can bind to VEGFR1 or NRP-1 and may promote detrimental effects via alternative signaling. Combination of BEV-BS and RAM may strongly suppress these signalings. Arrow color indicates signaling strength: red = signal transduction, cream = residual signal transduction, and blue = suppressed signal transduction. BEV-BS, bevacizumab biosimilar; NRP1, neuropilin-1; RAM, ramucirumab; TME, tumor microenvironment, VEGF-A, vascular endothelial growth factor-A; VEGFR1, vascular endothelial growth factor receptor 1; VEGFR2, vascular endothelial growth factor receptor 2.

In this clinical trial, we aim to establish the dual blockade of the VEGF-A/VEGFR2 pathway. When the targets of different agents overlap, combination therapy doses often require adjustment.³⁶ For example, in a phase I trial of sorafenib plus bevacizumab, high incidences of hypertension (67%), diarrhea (18%), and hand–foot skin reaction (HFSR) (79%) were observed, and 74% of patients required a sorafenib dose reduction to 200 mg daily, while a favorable response rate of 46% was observed in ovarian cancer.³⁷ The incidence of toxicities with ramucirumab monotherapy, such as HFSR (<1%) and diarrhea (14%) in any grade, is much lower than that with sorafenib (HFSR 21% and diarrhea 39%), although hypertension is more frequent with ramucirumab (16%) than with sorafenib (5%).^38,39 Unlike tyrosine kinase inhibitors, monoclonal antibodies exhibit lower off-target toxicities, which may contribute to their distinct AE profiles,⁴⁰ thereby suggesting that the addition of BEV-BS to ramucirumab is expected to result in only a minimal increase in toxicity. Based on these considerations, the dual blockade strategy of the VEGF-A/VEGFR2 pathway using antibody-based agents was proposed. In this trial, we aim to achieve two objectives. The first is to establish a novel therapeutic concept of the dual blockade of the VEGF-A/VEGFR2 pathway using antibody-based agents. If this approach is established, it is expected to be broadly applicable to other cancers such as colorectal cancer in which antiangiogenic agents are used. The second objective is to elucidate how the dual blockade of the VEGF-A/VEGFR2 pathway using antibody-based agents influences angiogenesis, other tumor-promoting pathways, and tumor immunity relative to VEGFR2 inhibition by ramucirumab. In particular, if neutralization of excess VEGF-A improves antitumor immunity through remodeling of the TME, this strategy could potentially hold promise for combination with immunotherapy in the future.

Conclusion

The BevRam-GC01 trial is the first clinical study to evaluate the tolerability and safety of BEV-BS in combination with ramucirumab and paclitaxel in patients with AGC. This clinical trial, which employs the dual blockade of the VEGF-A/VEGFR2 pathway using antibody-based agents, may establish a new therapeutic concept in the field of angiogenesis inhibitors. Furthermore, through accompanying translational research, the synergistic and complementary effects of combining anti-VEGFR2 and anti-VEGF-A antibodies on various biological processes—including angiogenesis, proliferation, and tumor immunity—are expected to be clarified.

CRediT author statement

Takeru Wakatsuki: conceptualization, methodology, resources, writing—original draft, visualization project administration, funding acquisition. Tetsuo Mashima: conceptualization, methodology, investigation, writing—original draft, visualization. Naoki Miyazaki: methodology, formal analysis, writing—original draft. Hiroki Osumi: resources, writing—review and editing. Shota Fukuoka: resources, writing—review and editing. Akira Ooki: writing—review and editing. Keitaro Shimozaki: writing—review and editing. Mariko Ogura: writing—review and editing. Koichiro Yoshino: writing—review and editing. Shohei Udagawa: writing—review and editing. Eiji Shinozaki: writing—review and editing. Keisho Chin: writing—review and editing. Noriyuki Ishida: methodology, formal analysis, writing—review and editing. Hiroyuki Seimiya: writing—review and editing, supervision. Kensei Yamaguchi: conceptualization, methodology, resources, writing—review and editing, project administration, funding acquisition, supervision.