Introduction

Gastric cancer is the fifth most common type of cancer and is the fourth-leading cause of cancer-related death1. Its incidence and mortality rates are notably high in East Asia. While systemic chemotherapy has prolonged survival of patients with advanced gastric cancer (AGC), median overall survival (OS) remains suboptimal2. Several chemotherapeutic agents such as trifluridine/tipiracil hydrochloride (FTD/TPI), irinotecan, nivolumab, and trastuzumab deruxtecan are beneficial as third- or later-line treatment for AGC3. Notably, in a large Japanese cohort study of 10,581 patients with AGC receiving palliative systemic chemotherapy, only 2390 patients (22.6%) underwent third-line chemotherapy4. This finding highlights the need for further developing treatment strategies in later lines of therapy.

FTD/TPI is an oral medication comprising a nucleoside antitumor component, trifluridine, and a thymidine phosphorylase inhibitor, tipiracil. The TAGS trial of FTD/TPI demonstrated improvement in OS over placebo for heavily pretreated patients with AGC5. Preclinical data showed that the combination of FTD/TPI and bevacizumab, a specific angiogenesis inhibitor against vascular endothelial growth factor (VEGF)-A, enhanced antitumor effects compared with FTD/TPI alone6,7. Ramucirumab, an anti-VEGF-receptor 2 monoclonal antibody, is an established standard of care for AGC, as evidenced by the results of the REGARD and RAINBOW trials8,9. Several single-arm phase II trials of FTD/TPI combined with ramucirumab exhibited promising antitumor activity and feasible safety profile in pretreated patients with AGC10,11.

Nivolumab, an immune checkpoint inhibitor targeting programmed cell death-1 (PD-1), has been approved for the primary treatment of AGC and is widely used in clinical practice. Results of a prospective observational study suggested that FTD/TPI monotherapy immediately following nivolumab has a synergistic antitumor effect in AGC, as evidenced by an objective response rate (ORR) of 10.9%12. Additionally, the simultaneous inhibition of PD-1 and VEGF pathways can synergize antitumor effects in AGC13,14,15,16. However, substantial data on the combination therapy of FTD/TPI and ramucirumab following immune checkpoint inhibitors are lacking. Therefore, this retrospective study aimed to evaluate the efficacy and safety of FTD/TPI plus ramucirumab versus FTD/TPI monotherapy in patients with AGC in later-line treatment, most of whom had a history of anti-PD-1 therapy.

Materials and methods

Patients

This retrospective study included consecutive patients with AGC who received FTD/TPI plus ramucirumab or FTD/TPI monotherapy as third-line or later treatment at Aichi Cancer Center Hospital between August 2018 and May 2023. The primary inclusion criteria were as follows: (1) histologically confirmed, unresectable, or recurrent gastric cancer; (2) Eastern Cooperative Oncology Group Performance Status (ECOG PS) of 0–2; (3) adequate bone marrow, hepatic, and renal function; (4) ability to maintain adequate oral intake; (5) history of treatment with two or more regimens; and (6) at least one treatment course of FTD/TPI plus ramucirumab or FTD/TPI monotherapy. This study was approved by the Institutional Review Board of Aichi Cancer Center Hospital (No. IR051103) and conformed to the guidelines of the Declaration of Helsinki. The requirement for written informed consent for this study was waived by the Institutional Review Board of Aichi Cancer Center Hospital because of the retrospective, non-interventional design and the available opt-out option on the institution’s website.

Procedures

FTD/TPI plus ramucirumab therapy involved administering oral FTD/TPI at a dose of 35 mg/m2 twice daily on days 1–5 and 8–12 of each 28-day treatment cycle combined with ramucirumab administered intravenously at a dose of 8 mg/kg, repeated every 2 weeks. FTD/TPI monotherapy included oral FTD/TPI at a dose of 35 mg/m2 twice daily on days 1–5 and 8–12 of each 28-day cycle. Dose modifications and treatment interruptions were performed at the discretion of each clinician, guided by established clinical trials5,10.

Evaluation of treatment

Clinical data were retrospectively collected from patient medical records. Patients were categorized into FTD/TPI plus ramucirumab and FTD/TPI monotherapy groups for outcome evaluation. Tumor response in patients with measurable lesions was assessed by each clinician according to the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.117. ORR was defined as the proportion of patients with measurable lesions who exhibited either a complete or partial response, as determined by investigators. Disease control rate (DCR) refers to the proportion of patients who achieved complete response, partial response, or stable disease. Progression-free survival (PFS) was calculated from the date of the first administration of the study treatment to the date of disease progression, as indicated by imaging findings, clinical progression, or death owing to any cause. OS was defined from the date of study treatment initiation to the date of death because of any cause or the last follow-up. Adverse events were assessed according to the National Cancer Institute Common Terminology Criteria for Adverse Events version 5.018.

Statistical analysis

The data cut-off date was November 15, 2023. PFS and OS were estimated using the Kaplan–Meier method, and the stratified log-rank test was utilized to compare variables among patients with respect to survival. Cox proportional hazards regression analysis was employed for survival analysis across different patient groups. The variables included in the multivariate Cox proportional model were selected based on factors with P-values < 0.2 in the univariate analysis. Age (≥ 65 vs. < 65 years), sex (male vs. female), ECOG PS (≥ 1 vs. 0), histology (diffuse vs. intestinal), history of gastrectomy (yes vs. no), lymph node metastasis (yes vs. no), liver metastasis (yes vs. no), peritoneal metastasis (yes vs. no), number of metastatic sites (≥ 2 vs. 1), number of prior chemotherapy regimens (≥ 3 vs. 1–2), duration of prior ramucirumab (≥ 3 months vs. < 3 months), ramucirumab-free interval (≥ 3 months vs. < 3 months), and anti-PD-1 inhibitor-free interval (IFI) (< 60 days vs. ≥ 60 days) were incorporated as confounders in the multivariate analysis of PFS and OS. To determine the potential benefits of continuation or re-challenge of ramucirumab and the synergistic effect of anti-PD-1 inhibitor and ramucirumab, cut-off values of three for ramucirumab-free interval and 60 days for IFI were adopted as previously described19,20. Exploratory efficacy analyses based on subgroups of liver metastasis (LM) and IFI were conducted, and all outcomes were compared. Statistical analyses were performed using R software version 4.1.0 (R Project for Statistical Computing, Vienna, Austria). All tests were two sided, and P-values < 0.05 indicated statistically significant differences.

Ethics approval

The study was conducted in accordance with the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of Aichi Cancer Center Hospital (No. IR051103).

Informed consent

The requirement for written informed consent for this study was waived by the Institutional Review Board of Aichi Cancer Center Hospital due to the retrospective study design without intervention, with an opt-out opportunity provided on the institution’s website.

Results

Patients

Among 106 patients, 36 and 70 were treated with FTD/TPI plus ramucirumab and FTD/TPI monotherapy, respectively. Table 1 summarizes patient characteristics. The proportions of patients with ECOG PS of 0 (P = 0.013) and a ramucirumab-free interval of ≥ 3 months (P = 0.028) were significantly higher in the FTD/TPI plus ramucirumab group than in the FTD/TPI monotherapy group. No significant differences in LM and IFI were observed between the two groups. Almost all patients had previously received fluoropyrimidine (100.0% vs. 100.0%), platinum (88.9% vs. 100.0%), taxane (88.9% vs. 100.0%), ramucirumab (86.1% vs. 97.1%), or anti-PD-1 antibody (91.7% vs. 85.7%).

Table 1 Patient characteristics.
Full size table

Efficacy outcomes

At the data cut-off date, the median follow-up time after initiating study treatment was 11.3 months (interquartile range, 8.2–19.0 months). A significantly higher proportion of patients in the FTD/TPI plus ramucirumab group achieved an objective response compared with that in the FTD/TPI monotherapy group (ORR, 25.8% vs. 5.0%, P = 0.007) (Table 2, Fig. 1). DCR was numerically better in the FTD/TPI plus ramucirumab group (58.1% vs. 38.3%, P = 0.081). Kaplan–Meier curves indicated a significantly longer median PFS in the FTD/TPI plus ramucirumab group (median PFS, 2.9 vs. 1.8 months; hazard ratio [HR]: 0.52, 95% CI 0.34–0.79; P = 0.001) (Fig. 2A). A numerical difference in OS favoring the FTD/TPI plus ramucirumab group was also observed (median OS, 7.9 months vs. 5.0 months; HR: 0.68, 95% CI 0.44–1.07; P = 0.089) (Fig. 2B). Multivariate analysis showed a significantly longer PFS in the FTD/TPI plus ramucirumab group (HR: 0.61, 95% CI 0.39–0.95; P = 0.030) (Table 3) but no significant survival difference between the two groups (HR: 0.97, 95% CI 0.60–1.58; P = 0.913) (Table 4).

Table 2 Tumor responsea.
Full size table
Figure 1
figure 1

Waterfall plot displaying the percentage changes from baseline in the sum of measurable lesions in the (A) FTD/TPI monotherapy and (B) FTD/TPI plus ramucirumab groups. FTD/TPI, trifluridine/tipiracil hydrochloride; PR, partial response; SD, stable disease; PD progressive disease.

Full size image
Figure 2
figure 2

Kaplan–Meier curves of (A) progression-free and (B) overall survival. FTD/TPI, trifluridine/tipiracil hydrochloride; RAM, ramucirumab; PFS, progression-free survival; OS, overall survival.

Full size image
Table 3 Univariate and multivariate analyses for progression-free survival.
Full size table
Table 4 Univariate and multivariate analyses for overall survival.
Full size table

Outcomes according to LM and IFI

Kaplan–Meier curves for PFS and OS, stratified by LM and IFI, are shown in Fig. 3. The highest ORR was achieved for LM in the FTD/TPI plus ramucirumab treatment group (ORR, 33.3%) (Supplementary Table 1). Greater treatment benefits on adding ramucirumab were observed in patients with LM (median PFS, 4.1 vs. 1.7 months; HR: 0.37, 95% CI 0.18–0.78; P = 0.009) compared with those observed in patients without LM (median PFS, 2.5 vs. 2.1 months; HR: 0.69, 95% CI 0.41–1.17; P = 0.170). Similarly, a numerically better survival benefit was observed in patients with LM (median OS, 9.7 vs. 4.3 months; HR: 0.47, 95% CI 0.20–1.08; P = 0.067) than in those without LM (median OS, 7.5 vs. 5.6 months; HR: 0.82, 95% CI 0.48–1.41; P = 0.472). The interaction P-values for PFS (P = 0.079) and OS (P = 0.288) were not significant. The highest ORR was observed in patients with IFI < 60 days undergoing FTD/TPI plus ramucirumab therapy (ORR, 46.7%) (Supplementary Table 2). The addition of ramucirumab within 60 days of anti-PD-1 inhibitor therapy showed better PFS (median PFS, 4.1 vs. 1.9 months; HR: 0.42, 95% CI 0.22–0.78; P = 0.004) and OS (median OS, 11.2 vs. 5.2 months; HR: 0.46, 95% CI 0.25–0.97; P = 0.043) than those after 60 days of anti-PD-1 inhibitor therapy (median PFS, 2.5 vs. 1.8 months; HR: 0.62, 95% CI 0.34–1.11; P = 0.099: median OS, 4.6 vs. 5.0 months; HR: 0.96, 95% CI 0.59–1.55; P = 0.860). The interaction P-values for both PFS (P = 0.331) and OS (P = 0.162) were not significant. No differences in outcomes were observed according to ramucirumab-free interval, duration of prior ramucirumab, and treatment pattern of prior ramucirumab (Supplementary Tables 3–5). The analysis stratified by prior use of immune checkpoint inhibitors showed better PFS in the FTD/TPI monotherapy group with prior use of immune checkpoint inhibitors than without (Supplementary Table 6).

Figure 3
figure 3

Kaplan–Meier estimates of (A) progression-free and (B) overall survival according to LM and (C) progression-free and (D) overall survival according to IFI. FTD/TPI, trifluridine/tipiracil hydrochloride; RAM, ramucirumab; PFS, progression-free survival; OS, overall survival; LM, liver metastasis; IFI, anti-PD-1 inhibitor-free interval; mo., months; HR, hazard ratio; CI, confidence interval.

Full size image

Safety

Disease progression was the most common reason for discontinuing study treatment in both the groups (91.7% in the FTD/TPI plus ramucirumab group vs. 94.2% in the FTD/TPI monotherapy group), and 2.8% and 2.9% of patients, respectively, discontinued study treatment because of adverse events. After such discontinuation, the proportions of patients receiving subsequent chemotherapy (38.2% vs. 42.6%) and best supportive care (61.8% vs. 57.4%) were similar between the two groups.

Dose reductions of FTD/TPI on initiating study treatment, as determined by each physician, were observed in six patients (16.7%) in the FTD/TPI plus ramucirumab group and 10 patients (14.3%) in the FTD/TPI monotherapy group. During the study treatment, 22 patients (61.1%) required dose reductions because of (N = 12, 33.3%), decrease appetite (N = 6, 16.6%), fatigue (N = 4, 11.1%), nausea (N = 1, 2.7%), and thrombocytopenia (N = 1, 2.7%) in FTD/TPI plus ramucirumab group, whereas 26 patients (37.1%) experienced dosing reductions because of neutropenia (N = 17, 24.2%), decreased appetite (N = 4, 5.7%), fatigue (N = 2, 2.9%), thrombocytopenia (N = 1, 1.4%), nausea (N = 1, 1.4%), and rash (N = 1, 1.4%). No Granulocyte Colony Stimulating Factor preparations as primary prophylaxis were administered in both groups.

Table 5 lists the adverse events occurring during study treatment. The proportion of patients with any grade of decreased appetite was higher in the FTD/TPI plus ramucirumab group than in the FTD/TPI monotherapy group (80.6% vs. 55.7%, P = 0.018). Any grade of fatigue was also numerically more common in the FTD/TPI plus ramucirumab group (69.4% vs. 48.6%). Conversely, any grade of anemia was observed numerically less often in the FTD/TPI plus ramucirumab group compared with that in the FTD/TPI monotherapy group (86.1% vs. 98.6%).

Table 5 Adverse events.
Full size table

The incidence of grade 3 or higher adverse events tended to be higher in the FTD/TPI plus ramucirumab group than in the FTD/TPI monotherapy group, although the difference was not statistically significant (72.2% vs. 64.3%, P = 0.833). The most common grade 3–4 adverse events were neutropenia (61.1% vs. 47.1%), leukopenia (33.3% vs. 35.7%), decreased appetite (22.2% vs. 8.6%), anemia (13.9% vs. 22.9%), and thrombocytopenia (13.9% vs. 10.0%). The proportion of febrile neutropenia was higher in the FTD/TPI plus ramucirumab group (13.8% vs. 2.9%, P = 0.018). No treatment-related deaths were observed in either group.

Discussion

To the best of our knowledge, this is the largest retrospective study to demonstrate the clinical benefit of FTD/TPI plus ramucirumab over FTD/TPI monotherapy in heavily pretreated patients with AGC. The significant improvements in ORR and PFS in the combination therapy group, though not paralleled by an increase in OS, suggest a notable therapeutic advantage, especially in patients with LM and a short IFI of < 60 days. This observation suggests that certain subgroups of patients with AGC might derive more benefit from this combination therapy.

In the present study, patients in the FTD/TPI plus ramucirumab group showed clinical improvement compared with those in the FTD/TPI monotherapy group. The outcomes of the FTD/TPI monotherapy group (DCR, 38.3%; median OS, 5.0 months) are consistent with those reported in the phase III TAGS trial, suggesting that our cohort is representative of the broader patient population with AGC receiving this treatment (DCR, 44%; median OS, 5.7 months)5. Moreover, the addition of ramucirumab in our study (ORR, 25.8%; median PFS, 2.9 months) showed a pattern of clinical benefit comparable with that observed in earlier phase II trials and retrospective studies (ORR, 0–16%; median PFS, 2.9–5.3 months)10,11,21. Despite these improvements in ORR and PFS, we did not observe a corresponding increase in OS. This phenomenon, in which improvements in intermediate endpoints do not translate into survival benefits, mirrors findings from other studies, including the RINDBeRG trial, which evaluated the addition of ramucirumab to irinotecan22. One possible explanation for the lack of OS benefit in our study could be the higher proportion of patients with favorable baseline characteristics, including ECOG PS of 0 and longer interval without previous ramucirumab exposure, in the combination therapy group. In addition, a substantial proportion of patients (40%) received subsequent chemotherapy. Thus, despite short-term efficacy based on ORR and DCR, these do not translate into a survival benefit. An ongoing randomized phase II trial comparing FTD/TPI plus ramucirumab with FTD/TPI monotherapy will provide further insight into this strategy23.

When optimizing ramucirumab treatment for patients with AGC with poor prognosis, identifying clinicopathologic predictors of efficacy is crucial. Our study results showed that adding ramucirumab to FTD/TPI numerically improved PFS and OS in patients with LM, consistent with previous findings that VEGF inhibitors benefit patients with LM across various cancers24,25,26. This result is supported by our previous analysis of 1355 patients with AGC, in which we observed a significant OS improvement in cases with LM post-ramucirumab approval25. These results are consistent with those from major trials such as the RAINBOW25,26. Preclinical data showed an association between anti-VEGF discontinuation and enhanced liver metastasis, indicating a strong correlation between VEGF and liver metastasis27. The mechanism by which VEGF inhibitors are effective against LM remains unclear; however, the unique tumor microenvironment in LM may enhance the effectiveness of VEGF inhibitors28.

Our study also explored the impact of prior anti-PD-1 inhibitor use and found only modest improvements in outcomes12,29. However, a more pronounced difference was noted in patients with a shorter IFI of < 60 days. One plausible explanation for this difference could be that therapeutic levels of anti-PD-1 inhibitor present at chemotherapy initiation do not persist30. No significant differences were observed in outcomes based on the ramucirumab-free interval or previous treatment patterns, consistent with previous efficacy evaluations of VEGF inhibitors19,22. This suggests that the timing of anti-PD-1 inhibitor therapy relative to chemotherapy initiation might be crucial.

In our study, a higher proportion of patients in the combination group experienced a decreased appetite (80.6% vs. 55.7%), consistent with data from a phase III trial of FTD/TPI plus a VEGF inhibitor for colorectal cancer7. The addition of a VEGF inhibitor to FTD/TPI increased the risk of severe neutropenia without significantly affecting the frequency of febrile neutropenia7,8,31. However, we observed that febrile neutropenia was more common in the combination therapy group (13.8%), with a rate higher than those previously reported for AGC treatment but without any treatment-related deaths10,21. This increased toxicity trend in the combination therapy group might have influenced treatment choices, especially for patients with high bleeding risks, potentially introducing selection bias. For instance, patients with massive ascites were more likely to receive monotherapy. The occurrence of neutropenia is associated with improved survival outcomes in patients treated with FTD/TPI32,33. These findings underscore the critical need to develop effective management strategies, including dose modifications and granulocyte colony-stimulating factor use, tailored to individual patient needs among those receiving FTD/TPI plus ramucirumab therapy.

The retrospective, small sample size, non-randomized, study design is a limitation, potentially introducing selection bias. This is particularly relevant given that treatment decisions were made at the discretion of the treating physicians. Additionally, the fact that a significant proportion of patients in both the groups received subsequent lines of chemotherapy suggests that other factors may have influenced OS.

In conclusion, the study results provide evidence of the clinical benefits of FTD/TPI plus ramucirumab with respect to ORR and PFS, as well as acceptable toxicity, in heavily pretreated patients with AGC. These findings highlight the potential of this combination therapy as a treatment option and underscore the need for further research, ideally through randomized controlled trials, to confirm these results and refine treatment strategies for this patient population.