Abstract
Background/Aim: Ramucirumab plus paclitaxel has been widely used as a second-line chemotherapy for treating advanced gastric cancer. However, the real-world data of this regimen for older patients with gastric cancer (GC) remains unrevealed. The aim of this study was to clarify the feasibility and efficacy of this regimen for older patients with GC in a single-arm, phase II study. Patients and Methods: Patients aged ≥70 years having unresectable or recurrent GC who met the eligible criteria were enrolled. Paclitaxel was administered at a dose of 80 mg/m2 on days 1, 8, and 15, and ramucirumab was administered at a dose of 8 mg/kg on day 1 and day 15 of a 4-week cycle. Primary endpoint was the incidence of adverse events and secondary endpoints were response rate, progression-free survival, and overall survival. A total of 25 patients were enrolled in the full-set analysis. Results: Grade 3 or more adverse events were observed in 21 patients (84.0%). Neutropenia was most frequently observed (68.0%), followed by peripheral sensory neuropathy (12.0%), and febrile neutropenia (12.0%). Median progression-free survival and overall survival were 6.9 months and 13.4 months, respectively. Disease control rate was 88.0%, and response rate of patients with measurable lesions was 52.9%. Notably, no treatment-related deaths occurred. Conclusion: Ramucirumab plus paclitaxel as a second-line chemotherapy demonstrated acceptable oncological outcomes, despite the occurrence of frequent adverse events. It is necessary to carefully select patients and adjust treatment regimens in older patients with GC to safely administer chemotherapy and subsequently achieve satisfactory long-term outcomes.
Gastric cancer (GC), including gastroesophageal junction cancer, is the fourth leading cause of cancer-related deaths worldwide (1). Over the past decade, with the emergence and advancements of new chemotherapeutic agents for unresectable or recurrent GC, the overall survival (OS) has extended from 13.1 to 17.45 months (2-4). A randomized phase III trial, the RAINBOW trial, demonstrated a significant survival benefit of incorporating ramucirumab, a human IgG1 monoclonal antibody vascular endothelial growth factor-2 antagonist, alongside paclitaxel as a second-line chemotherapy for unresectable or recurrent GC (5).
As the global life expectancy continues to rise due to improvement in social conditions and advancements in medical knowledge and techniques, we often encounter older patients with advanced GC in our daily clinical practice (6). Although the ramucirumab plus paclitaxel regimen is now widely used for treating unresectable or recurrent GC as a second-line chemotherapy, real-world data regarding its application in older patients with advanced GC are currently lacking (7). Older patients often experience more frequent adverse events than younger patients, primarily due to the declining functions of organs, such as the bone marrow, heart, and kidney (8). Therefore, feasibility and effectiveness of ramucirumab plus paclitaxel chemotherapy for older adults presents a clinically significant question. Therefore, we conducted a prospective, non-randomized, open-label, single institute phase II trial to evaluate the feasibility of this regimen as a second-line chemotherapy for treating older patients with unresectable or recurrent GC in the Yokohama Clinical Oncology Group (YCOG 1601) (UMIN ID: 000023515).
Patients and Methods
Study design and patients. The present study was a non-randomized, open-label, single institute phase II registration trial focusing on older patients with unresectable or recurrent GC who were refractory or intolerable to prior treatment with a fluoropyrimidine-based regimen. Written consent form was obtained from each patient. This study was conducted according to the Declaration of Helsinki and approved by the ethical board of the Yokohama City University (Approval number: D1603002).
Eligibility criteria. All eligible patients had histologically confirmed gastric adenocarcinoma with metastasis or recurrence, were ≥70 years, had an Eastern Cooperative Oncology Group performance status (PS) 0 or 1, had received one prior treatment with a fluoropyrimidine-based regimen, exhibited no massive ascites extending to the upper abdomen, as assessed by contrasted computed tomography, had no metastasis in the central nervous system, maintained sufficient oral food intake, possessed adequate bone marrow function (white blood cell count ≥3,000 cells/mm3, absolute neutrophil count ≥1,500 cells/mm3, hemoglobin ≥8.0 g/dl, and platelet count ≥75,000 cells/mm3), adequate liver function (serum aspartate aminotransaminase ≤three times higher of upper limit of normal (ULN) and serum alanine aminotransaminase ≤three times of ULN, total bilirubin ≤2.0 mg/dl), adequate renal function (serum creatinine ≤1.5 mg/dl), and adequate coagulation function (prothrombin time ≤1.5 times of ULN and activated partial thrombin time ≤1.5 times of ULN). Patients were excluded if they had uncontrolled diabetes mellitus and were receiving hypoglycemic medication (blood sugar ≥130 mg/dl), deep vein thrombosis, pulmonary vein thrombosis, a history of myocardial infarction or cerebral vascular infarction, a history of gastrointestinal bleeding within three months, a history of gastrointestinal perforation or fistula within six months, uncontrolled hypertension or receiving anti-hypertensive medication, or were otherwise deemed unsuitable for the study by attending physicians. At the beginning of the study, the eligibility criterion for age was set at 75 years old or older; however, due to difficulties in patient recruitment, the protocol was revised to include patients aged ≥70 years.
Treatment procedure. Paclitaxel (Sawai, Osaka, Japan) was administered at a dose of 80 mg/m2 over 60 min via intravenous infusion on days 1, 8, and 15 of a 4-week cycle. Ramucirumab (Eli Lilly, Hyogo, Japan) was administered at a dose of 8 mg/kg over 60 min via intravenous infusion on days 1 and 15 of a 4-week cycle. This treatment was repeated until disease progression or the onset of unacceptable toxicity. Paclitaxel could be administered if neutrophil count was ≥1,200 cells/m3 on day 1 and ≥1,000 cells/m3 on days 8 and 15; platelet count was ≥75,000 cells/m3 on day 1 and ≥60,000 cells/m3 on days 8 and 15. The patient was required to have adequate liver function defined as serum aspartate aminotransaminase ≤three times the ULN (or five times, if liver metastasis existed), serum alanine aminotransaminase ≤three times the ULN (five times, if liver metastasis existed), total bilirubin ≤one and a half times of ULN. Ramucirumab administration could be suspended in cases where elective surgery was planned, wound healing was prolonged, with the onset of ≥grade 2 hypertension, or proteinuria >2 g and <3 g/day. Ramucirumab administration could be discontinued if patients experienced venous or arterial thrombosis, ≥grade 3 infusion reactions, ≥grade 3 gastrointestinal bleeding, gastrointestinal perforation or fistula, congested heart failure, reversible posterior leukoencephalopathy syndrome, symptomatic grade 2 or grade 3 uncontrollable hypertension, and proteinuria ≥3 g/day.
Endpoints and evaluation of treatment. The primary endpoint was the incidence of adverse events. The secondary endpoints were response rate, progression-free survival (PFS), and OS. Tumor response was evaluated at each cycle of the regimen using CT imaging. Toxicity was graded according to the Common Terminology Criteria for Adverse Events (version 4) (9). Measurable lesions were assessed following the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 and classified as complete response (CR), partial response (PR), stable disease (SD), nonCRnonPD if measurable lesions were absent, and progressive disease (PD) (10).
Statistical analysis. In order to achieve 80% power to detect the effectiveness of tumor response with a one-sided alpha of 0.05, we performed a sample size calculation. Based on the subgroup analysis of East Asians in the RAINBOW trial, the threshold was set at 19%, with an expected value of 41% (11). Under these assumptions, the estimated required sample size was 33, and we planned to enroll 35 patients. Continuous variables were expressed as median (interquartile range) or mean [standard deviation (SD)]. PFS was measured from day 1 of the first cycle until disease progression or death from any cause, and OS was measured from day 1 of the first cycle until the last follow up. All statistical analyses were performed using JMP® 16 (SAS Institute Inc., Cary, NC, USA). Categorical data were evaluated using the chi-square test. PFS and OS were calculated using the Kaplan–Meier method. Statistical significance was set at p<0.05.
Results
Patient characteristics. Between May 2017 and December 2021, 27 patients were assigned to receive the protocol treatment. Two patients met the exclusion criteria and were excluded from the analysis (Figure 1). Despite our initial plan to recruit more patients, we determined that further recruitment would be difficult even with an extension of the patient enrollment period. Table I summarizes the baseline characteristics of the 25 patients enrolled in this study. Among these patients, 15 (60.0%) had primary unresectable tumor and 10 (40.0%) had post-gastrectomy recurrences. Notably, none of the recurrent cases had undergone neoadjuvant chemotherapy before gastrectomy. The most frequent metastatic sites were the liver and lymph nodes. Two patients (8.0%) had peritoneal metastases. Nine patients (36.0%) exhibited an undifferentiated histological type, and seven (28.0%) were positive for human epidermal growth factor 2. The most commonly administered prior chemotherapy regimen was S-1 plus oxaliplatin (44.0%). Among the 21 patients who experienced disease progression or intolerability with ramucirumab plus paclitaxel regimen, 18 (85.7%) received a 3rd line chemotherapy, with the most frequent regimen being nivolumab (57.1%). Among the 15 patients who experienced failure with a 3rd line chemotherapy, 11 received a 4th line chemotherapy, with the most frequent regimen being nivolumab (20.0%). Two patients received a 5th line chemotherapy, and none underwent conversion surgery.
Flow diagram of patient selection process. Among the 27 patients initially assigned to the study, two were excluded due to deep venous thrombosis (DVT) prior to the treatment, resulting in the enrollment of 25 patients in the full analysis set.
Baseline patient characteristics.
Treatment-related adverse event. Table II summarizes the adverse events associated with the protocol treatment. Adverse events of any grade occurred in all 25 patients (100.0%), whereas ≥grade 3 adverse events occurred in 21 patients (84.0%). Notably, no treatment-related death was noted. The most frequent adverse event observed was the decrease in neutrophil count (76.0%), followed by appetite loss (68.0%), fatigue (68.0%), and peripheral sensory neuropathy (52.0%). With respect to ramucirumab-related adverse events, hypertension occurred in five (20.0%), proteinuria in four (16.0%), and gastrointestinal perforation in two (8.0%) patients, and one patient required emergent operations. Moreover, gastrointestinal bleeding occurred in one (4.0%) and protracted wound healing in one (4.0%). Details regarding dose reduction and skipping paclitaxel administration are shown in Figure 2. Dose reduction of paclitaxel ranged from 0% to 71.4%, and the paclitaxel administration was skipped in the range of 43.4% to 100% across all treatment cycles. We compared the incidence of adverse events between 70-74 (<75) and 75 or older (≥75) years age groups (data not shown). Although the differences were not statistically different, ≥grade 2 stomatitis and appetite loss tended to occur frequently in the ≥75 group (p-values were 0.138 and 0.158, respectively).
Treatment-related adverse events.
Dose reduction and skipping paclitaxel administration. The number and percentage of dose reduction and skipping paclitaxel administration are shown. C: Cycle.
Drug exposure. Table III summarizes the drug exposures of the enrolled patients. The mean (SD) durations of ramucirumab and paclitaxel were 23.4 (19.5) and 21.0 (14.8) weeks, respectively. The mean numbers of infusions of ramucirumab and paclitaxel were 9.9 (8.5) and 10.9 (7.7), respectively. The mean cumulative doses of ramucirumab (mg/kg) and paclitaxel (mg/m2) were 77.7 (67.5) and 795.2 (601.1), respectively. The mean relative dose intensities (%) of ramucirumab and paclitaxel were 84.2 (18.8) and 65.0 (18.7), respectively.
Summary of drug exposure.
Short-time results. Among the 17 patients with measurable lesions, 1 (5.9 %), 8 (47.1 %), and 6 (35.5 %) showed CR, PR, and SD, respectively. Figure 3 shows a waterfall plot of the maximum percentage reduction in tumor size from the baseline in patients with measurable lesions. The response and disease control rates were 52.9% and 88.2% in patients with measurable lesions (n=17). The disease control rate was 88.0% in the 25 enrolled patients.
Waterfall plot of maximum percentage in tumor size from baseline. Changes in tumor size assessed by Response Evaluation Solid Tumors version 1.1 are shown for each patient with measurable lesions. Dotted lines denote a 20% increase (criterion for progressive disease) or a 30% reduction (criterion for partial response).
Long-term results. The median follow-up period was 15.33 months. The median PFS was 6.87 months [95% confidence interval (CI)=3.73 to above range not reached], and the median OS was 13.43 months (95%CI=8.63-16.63) (Figure 4). Among the 18 patients who died during the observation period, 16 died due to disease progression, one died due to intestinal pneumonia of post-chemotherapy, and one case had unknown cause of death.
Progression-free survival and overall survival. (A) Progression-free survival: median survival time is 6.87 months (95% confidence interval=3.73 to the above limit not reached). (B) Overall survival: median survival time is 13.43 months (95% confidence interval=8.63-16.63).
Discussion
This non-randomized, open-label, single-institute phase II registration trial involving older patients with GC treated with ramucirumab plus paclitaxel revealed that 84.0% of the patients experienced ≥grade 3 adverse events. The most frequent ≥grade 3 adverse event was the decrease in neutrophil count (68.0%). The median PFS and OS were 6.87 and 13.43 months, respectively. The response rate in patients with measurable lesions was 52.9%. Despite the frequent occurrence of adverse events, no treatment-related deaths were noted, and the tumor response to this regimen was relatively good. Although patient recruitment fell short of the planned number, this study is the first to provide insights into the short- and long-term results of ramucirumab plus paclitaxel therapy in older patients with GC.
In the main and subgroup analyses involving an East Asian population conducted as a part of the RAINBOW trial, the median PFS and median OS in the ramucirumab plus paclitaxel group were 4.4 and 9.6 months in the main analysis and 5.5 and 12.1 months in the subgroup analysis, respectively. Although direct comparison of the results of the RAINBOW trial and those of the present study may not be straightforward, our study yielded notable results. First, the frequency of severe adverse events were notable. Second, the response and disease control rates were acceptable at 53.0% and 88.0%, respectively (27.9% and 80.0%, respectively, in the RAINBOW study). Third, median PFS and median OS in our study were also comparable to those in the RAINBOW trial at 6.87 and 13.43 months (5.5 and 12.1 months, respectively, in the subgroup analysis of East Asian patients).
At the beginning of this study, the cut-off age for patient eligibility was set at 75 years; however, it was later adjusted to 70 years due to challenges in patient recruitment. Notably, the definition of older patients differs among previous studies, with the cut-off age ranging from 70 to 80 years (6, 8, 12, 13).
Therefore, we believe that the adjusted cut-off value for older adults in our study is acceptable. The subset analysis revealed no superiority of the ramucirumab plus paclitaxel group over the paclitaxel group in patients aged ≥65 years in the RAINBOW trial (5). In contrast, the present study demonstrated comparable antitumor activity to the RAINBOW trial. Therefore, our findings provide valuable insights for the clinical management of older patients with advanced GC.
The patients included in this study underwent a rigorous selection process; however, ramucirumab plus paclitaxel may be considered an optional treatment for this population, provided the comprehensive management of adverse events. In the present study, although the duration of therapy and number of chemotherapeutic treatments were comparable to those of the RAINBOW trial, the cumulative dose and relative dose intensity of paclitaxel in our study patients were lower than those in all age-registered patients included in the RAINBOW trial (759.2 mg/m2 and 65.0% vs. 1067.9 mg/m2 and 84/0%). These results could be attributed to the necessary frequent dose reduction of paclitaxel for the older adults across all treatment cycles to avoid adverse events (5). Despite the adjustments made to the regimen, we encountered frequent severe adverse events, such as neutropenia. Furthermore, patients aged ≥75 years tended to show frequent stomatitis and appetite loss. Thus, it is essential to implement optimal dose reduction to ensure safe continuation of chemotherapy in older patients with GC. A previous study recommended initial dose reduction of nab-paclitaxel in patients with advanced GC (14). Such modifications should be accepted by the older population to ensure the safety of treatment.
In this study, the antitumor activity of ramucirumab plus paclitaxel in older patients with GC was comparable to that reported in the RAINBOW trial, despite the lower dose intensity. Therefore, the standard dose of each drug may be relatively high in older patients with several comorbidities. Even a lower dose of the drug may be sufficient for older patients with reduced drug metabolic ability. Therefore, it is necessary to establish an appropriate chemotherapeutic regimen for older patients to obtain favorable outcomes.
Study limitations. First, this was a single-arm phase II study involving a small sample of Japanese patients. Second, we were unable to recruit a sufficient number of patients to meet the estimated sample size requirements. Third, due to the difficulties in patient recruitment, the cut-off age for the definition of older patients was changed from 75 years to 70 years.
Conclusion
Ramucirumab plus paclitaxel as a second-line chemotherapy in older patients with GC demonstrated a relatively good tumor response and long-term survival, despite the occurrence of frequent adverse events. Thus, careful selection of patients and adequate modification of drug regimens in older patients with GC is crucial to ensure safe administration of chemotherapy and achieve satisfactory long-term outcomes.
Acknowledgements
The Authors thank the secretariat of YCOG for supporting this study.
Footnotes
Authors’ Contributions
SS, KC, and EI designed and drafted the article. TY, YA, KK, ST, KH, KT, and AH collected data and assisted in preparing the article. All Authors read and approved the final article.
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
Funding
This study was supported by YCOG research grant.
- Received October 21, 2023.
- Revision received November 9, 2023.
- Accepted November 10, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.










