Abstract
Background/Aim: Irinotecan and trifluridine/tipiracil (FTD/TPI) are fourth-line treatment options after third-line nivolumab for advanced gastric cancer (AGC). However, the efficacy and safety of irinotecan and FTD/TPI in the fourth-line setting after third-line nivolumab remains unclear. This study aimed to evaluate the efficacy and safety of irinotecan and FTD/TPI in the fourth-line setting after third-line nivolumab. Patients and Methods: We identified 137 AGC patients treated with nivolumab as third-line treatment in our institute between October 2017 and July 2021. Of these, we recruited 19 AGC patients who initiated irinotecan and 23 AGC patients who initiated FTD/TPI in the fourth-line setting until September 2021. Results: The median overall survival was 5.83 months for irinotecan and 6.31 months for FTD/TPI. Median time-to-treatment failure was 2.07 months for irinotecan and 1.64 months for FTD/TPI. While the frequency of all-grade diarrhea was higher in irinotecan (36% vs. 17%), grade ≥3 neutropenia tended to be higher in FTD/TPI (21% vs. 35%). Conclusion: Irinotecan and FTD/TPI may be clinically useful as fourth-line treatments after nivolumab.
Gastric cancer is the sixth most common cancer and the third most common cause of cancer-related death worldwide (1). Although chemotherapy has been developed for advanced gastric cancer (AGC), the prognosis remains relatively poor. Human epidermal growth factor receptor 2 (HER2) is a transmembrane tyrosine kinase receptor protein and known as a therapeutic target. For HER2 positive gastric cancer, trastzumab is administered in combination with chemotherapy in first-line treatment. Median overall survival (OS) was reported as 10-14 months in human epidermal growth factor receptor 2 (HER2) negative AGC (2-6), and 14-17 months in HER2 positive AGC (7, 8). Currently, nivolumab, a programmed death-1 (PD-1) inhibitor, is being used to treat AGC. The efficacy of nivolumab monotherapy was examined in the ATTRACTION-2 trial, which is the first phase 3 study that demonstrated the efficacy and safety of nivolumab in patients receiving third- or later-line therapy (9). The TAGS study also showed the efficacy of trifluridine/tipiracil (FTD/TPI) compared with a placebo in the same lines (10). Irinotecan demonstrated a survival benefit in second-line therapy compared to best supportive care (11). In a phase III trial comparing irinotecan and paclitaxel as second-line therapy, crossover results showed an OS of 8-9 months, longer than those reported using either therapy alone (12). This suggests that irinotecan is effective in third-line treatment, and several retrospective studies were reported (13-15). As a result of these studies, nivolumab, irinotecan, and FTD/TPI are standard regimens in the third- or later-line setting for AGC in several guidelines (16, 17). Afterward, the efficacy of nivolumab as a first-line treatment was examined. The CheckMate 649 study showed the efficacy of nivolumab in combination with chemotherapy in the first-line setting, especially in patients whose tumors had a programmed death-ligand 1 (PD-L1) combined positive score (CPS) of five or more (6). When nivolumab is not given as first-line chemotherapy, it is a candidate treatment in third-line or later settings. Irinotecan and FTD/TPI are also candidate treatments in the third- or later-line setting, but nivolumab is preferred due to its lower toxicity.
Immune checkpoint inhibitors (ICI) are reported to enhance the efficacy of subsequent chemotherapy in various types of cancers (18-21). However, for AGC, enhancement of post-ICI chemotherapy remains controversial. Kato et al. reported that OS and time-to-treatment failure (TTF) of cytotoxic chemotherapy were not improved in AGC patients exposed to prior ICI compared to those not exposed to ICI before cytotoxic chemotherapy (22). Conversely, Sasaki reported that initial ICI prolonged progression-free survival of ramucirumab plus paclitaxel (23). Differences in modulation of tumor microenvironment by chemotherapeutic agents might affect the enhancement of chemotherapy administered after ICI, but the actual differences in the enhancement of subsequent chemotherapy are unclear.
In clinical practice, when nivolumab is administered to AGC patients as third line therapy, irinotecan and FTD/TPI are considered as candidates for fourth-line treatment. These chemotherapeutic agents are associated with the immune system. FTD/TPI was reported to have synergistic anticancer activity when combined with PD-1 inhibitors; this treatment increases the infiltration of cytotoxic CD8+ T cells and reduces regulatory T cells in microsatellite stable-type murine colorectal cancer cells (24). Irinotecan reduces the abundance of Foxp3+ regulatory T cells in the tumor immune-microenvironment and increases PD-L1 expression on tumor cells and tumor-infiltrating immune cells (25). These reports indicate that initial nivolumab treatment may enhance the therapeutic effect of these agents. However, clinical data regarding the use of these agents in the fourth-line setting following nivolumab are missing. The study aimed to evaluate the clinical outcomes of patients receiving irinotecan or FTD/TPI as fourth-line therapy after nivolumab third-line therapy.
Patients and Methods
Patients. We retrospectively examined AGC patients who received nivolumab as third-line treatment at Kanagawa Cancer Center Hospital between October 2017 and July 2021. Among these patients, we recruited AGC patients who initiated irinotecan or FTD/TPI in the fourth-line setting until September 2021. The inclusion criteria were as follows: histologically proven AGC; received irinotecan or FTD/TPI in the fourth-line setting after third-line nivolumab; Eastern Cooperative Oncology Group performance status (26) ≤2; adequate bone marrow (neutrophil ≥1,000/mm3, platelets ≥7.5×104/mm3), hepatic and renal function [aspartate transaminase/alanine aminotransferase ≤100 IU/l in patients without liver metastasis (≤200 IU/l in patients with liver metastasis), total bilirubin ≤1.5 mg/dl, serum creatinine ≤1.5 mg/dl].
Written informed consent for treatment was obtained from all patients. The present study was approved by our institutional review committee (approval number: 2023 epidemiologic study-125) and met the standards outlined in the Declaration of Helsinki.
Chemotherapy. Irinotecan 150 mg/m2 was intravenously administered every 2 weeks. FTD/TPI 35 mg/m2 was administered orally twice daily on days 1-5 and 8-12 every 4 weeks. Treatment was continued until disease progression, intolerable toxicity, or patient request. The dose reduction was decided by investigators based on adverse events or suspended administration due to toxicity.
Assessment. Tumor response was assessed according to the guidelines of the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (27). The overall response rate (ORR) was defined as the percentage of complete response (CR) or partial response (PR). Disease control rate (DCR) was defined as the percentage of CR, PR, or stable disease (SD). Computed tomography (CT) was performed every 2-3 months or when clinical signs suggested progressive disease. Adverse events were evaluated per Common Terminology Criteria for Adverse Events version 5.0 (28).
Statistical analysis. Continuous variables are presented as median and range, and categorical variables as number and percentage. Continuous variables were compared using the Mann–Whitney U-test, and categorical variables were compared using Fisher’s exact test in patient characteristics. Fisher’s exact test was also applied to compare tumor response. OS was defined as the duration from the initiation of chemotherapy to death from any cause. TTF was calculated as the period from the initiation of chemotherapy to treatment discontinuation or death from any cause. The data cut-off date was March 31, 2022. Kaplan–Meier curves of estimated OS and TTF were generated, and comparisons were performed using a log-rank test. Cox regression analyses for OS were performed. The fourth-line treatment (irinotecan vs. FTD/TPI) and univariate factors with a p-value <0.20 were applied to the multivariate Cox regression analysis to adjust patient backgrounds. A two-sided p-value <0.05 was considered significant. All statistical analyses were conducted using EZR software, version 1.32 (Saitama Medical Center, Jichii Medical University, Saitama, Japan) (29).
Results
Patient characteristics. We identified 137 AGC patients who received nivolumab as third-line treatment. Among these patients, nivolumab was discontinued in 124 patients until September 2021. Subsequent chemotherapy was administered to fifty-two patients. Nineteen patients received irinotecan and 23 patients received FTD/TPI in the fourth-line setting. The other regimens included cisplatin and irinotecan (n=3), capecitabine and oxaliplatin (n=2), S1 and oxaliplatin plus trastuzumab (n=1), ramucirumab plus paclitaxel (n=1), trastuzumab deruxtecan (n=1), fluorouracil, leucovorin, and irinotecan (FOLFIRI) plus panitumumab (n=1) and investigational drug (n=1). The patient treated with FOLFIRI plus panitumumab had both AGC and advanced colorectal cancer. Nivolumab was switched to FOLFIRI plus panitumumab due to the progression of colorectal cancer. Seventy-two patients did not receive subsequent chemotherapy after the failure of third-line nivolumab (Figure 1).
Patient flow diagram. FTD/TPI: Trifluridine/tipiracil; IRI: irinotecan; CDDP: cisplatin; XELOX: capecitabine and oxaliplatin; SOX: S1 and oxaliplatin; Tamb: trastuzumab; RAM: ramucirumab; PTX: paclitaxel; T-DXd: trastuzumab deruxtecan; Pmab: panitumumab.
Table I displays the patient characteristics. The proportion of liver metastasis was higher in the irinotecan group than in the FTD/TPI group. In contrast, the proportion of peritoneal metastasis was higher in the FTD/TPI group than in the irinotecan group. There were no significant differences in the other characteristics. Third-line nivolumab was discontinued due to disease progression in all study patients.
Patient characteristics.
Treatment efficacy. Thirteen patients in the irinotecan group and 15 patients in FTD/TPI group had measurable lesions according to RECIST version 1.1 criteria (27). Among patients with measurable lesions, an objective response was observed in 0 of 13 patients (0%) in the irinotecan group and 2 of 15 patients (13%) in the FTD/TPI group. Disease control was 3 of 13 patients in the irinotecan group (23%) and 4 of 15 in the FTD/TPI group (27%). Among all study patients with measurable lesions, CR was observed in one FTD/TPI group patient. The patient had recurrent disease in a mediastinal lymph node before nivolumab. The multiple mediastinal lymph node swelling was observed 6 weeks after the initiation of nivolumab, and these lymph node metastases were new lesions, therefore, nivolumab was terminated and switched to FTD/TPI. The swelling in multiple mediastinal lymph metastases shrank one month after initiating FTD/TPI (Figure 2). CR was not observed in the irinotecan group.
57-year-old male with recurrent gastric cancer with mediastinal lymph node metastases, who achieved complete response with trifluridine/tipiracil (FTD/TPI) in the fourth-line treatment after third-line nivolumab. A single mediastinal lymph node metastasis was found before nivolumab (dashed arrows) (A). Multiple new mediastinal lymph node metastases (solid arrows) were found 2 months after the initiation of nivolumab (B). Multiple mediastinal lymph node metastases (solid arrows) shrank and achieved complete response one month after the initiation of FTD/TPI (C).
At the data cut-off, 37 patients had died, and 41 patients had experienced treatment termination. Median OS was 5.82 months [95% confidence interval (CI)=3.19-9.40] in the irinotecan group and 6.31 months (95%CI=3.97-10.8) in the FTD/TPI group. The hazard ratio (HR) was 1.02 (95%CI=0.53-1.95, p-value=0.96) (Figure 3A). Median TTF was 2.07 months (95%CI=1.74-3.29) in the irinotecan group and 1.64 months (95%CI=1.28-2.89) in the FTD/TPI group. HR was 0.88 (95%CI=0.46-1.66, p=0.69) (Figure 3B). Subsequent chemotherapy was discontinued in all patients in the irinotecan group and 22 of 23 patients in the FTD/TPI group. The reasons for discontinuing irinotecan included disease progression in 18 patients and toxicity in one patient. Among those treated with FTD/TPI, disease progression occurred in 20 patients and toxicity in two patients. After the termination of fourth-line treatment, subsequent chemotherapy was given to 9 of 19 patients (47%) in the irinotecan group and 4 of 22 patients (18%) in the FTD/TPI group.
Overall survival and time-to-treatment failure of irinotecan and trifluridine/tipiracil (FTD/TPI) in the fourth-line setting after third-line nivolumab. (A) Overall survival. (B) Time-to-treatment failure. CI: Confidence interval.
The univariate and multivariate Cox regression analyses for OS are shown in Table II. Univariate Cox regression analysis identified three factors with p-value <0.20: i) histological type, ii) ascites, and iii) alkaline phosphatase levels. The multivariate Cox regression analysis included these three factors and the fourth-line treatment (irinotecan vs. FTD/TPI). The multivariate Cox regression analysis did not show any significant poor prognostic factors; however, patients with ascites had poor prognoses. Fourth-line treatment difference (irinotecan vs. FTD/TPI) was not a significant prognostic factor either in univariate analysis (HR=1.09, 95%CI=0.53-1.95, p=0.96) or multivariate analysis (HR=0.90, 95%CI=0.45-1.79, p=0.76).
Univariate and multivariate Cox regression analysis for overall survival.
Adverse events. Adverse events are summarized in Table III. The frequency of all-grade diarrhea was higher in the irinotecan group. In contrast, grade ≥3 neutropenia frequency was higher in the FTD/TPI group. One patient died of adverse events in the FTD/TPI group 22 days after initiation and grade 3 elevated total bilirubin, aspartate aminotransferase, and alanine aminotransferase were observed. The patient did not have metastatic sites that could cause liver damage in CT; therefore, we determined that liver failure caused the patient’s death. The cause of liver failure was not identified because the liver biopsy was impossible due to ascites and thrombo-cytopenia.
Adverse events during irinotecan and trifluridine/tipiracil.
Discussion
This is the first report evaluating the efficacy and safety of irinotecan and FTD/TPI following third-line nivolumab therapy. These agents showed similar efficacy and feasibility. Previous studies showed that median PFS and OS were 2.2-3.2 and 4.0-6.6 months in irinotecan, and 2.0 and 5.7 months in FTD/TPI, respectively (10, 13-15). Considering these results, the study population did not observe an obvious enhancement of the efficacy of irinotecan and FTD/TPI. However, the efficacy and safety were comparable to previous studies. These facts indicate that both agents can be treatment options in the fourth-line setting after nivolumab.
An observational study called REVIVE was reported recently (30). In that study, the efficacy of cytotoxic agents, such as irinotecan, FTD/TPI, and oxaliplatin-containing regimens, after nivolumab for AGC was evaluated. Median PFS and OS in subsequent chemotherapy after nivolumab were 2.9 months and 7.5 months, respectively, and the objective response rate was 16.8% in all study patients. The present study’s clinical outcomes were worse than those in the REVIVE study. It was reported that patients responding to prior ICI were likely to respond to subsequent chemotherapy (19). The present study included patients without response to prior nivolumab more frequently. The REVIVE study excluded patients whom physicians regarded as ineligible for recruiting. There was a possibility that physicians avoided recruiting patients with poor conditions. These factors might have led to worse clinical outcomes in the present study. On the other hand, our study population, including patients with poor conditions, might be closer to the real-world population. The present study showed the frequency of receiving subsequent chemotherapy after third-line nivolumab and subsequent fourth-line chemotherapy. The frequency of subsequent chemotherapy after first-line and second-line chemotherapy was recently reported, but data about late-line chemotherapy have not been well reported (31). Moreover, the number of treatment lines in subsequent chemotherapy after nivolumab was not reported in the REVIVE study. Our data are useful for understanding the current state of late-line treatment for AGC in clinical practice.
One patient achieved CR during FTD/TPI. There are two possible reasons for this. First, FTD/TPI after nivolumab showed a dramatic response in a case report. Therefore, FTD/TPI enhanced by nivolumab might contribute to achieving CR (32). Another possibility is pseudoprogression (33). In this pattern of responses, tumor size or count increases initially but decreases after the initial evaluation. In that case, it is difficult to judge pseudoprogression retrospectively because subsequent chemotherapy was initiated without a second evaluation. No biomarkers and clinical features are initially used to distinguish pseudoprogression from disease progression. The continuation of nivolumab is difficult in many cases because of tumor volume or patient conditions after tumor enlargement. Therefore, switching to subsequent chemotherapy may be reasonable at the timing of initial tumor enlargement.
Cytotoxic agents are reported to modulate the tumor microenvironment in different ways (24, 25), which suggests that the influence of prior nivolumab on the effect of these agents might vary. The efficacy of subsequent chemotherapy was reported to differ among cytotoxic regimens used after ICI in melanoma (34). In the present study, one patient achieved CR in FTD/TPI, but we could not find any remarkable difference in the total population between irinotecan and FTD/TPI. The univariate and multivariate analyses showed similar results. The difference in irinotecan and FTD/TPI were not evaluated, and neither univariate nor multivariate analyses were performed in the REVIVE study (30). In contrast, Sasaki et al. reported that enhancement of efficacy in subsequent chemotherapy by prior ICI was observed only in ramucirumab plus taxanes (23). Enhancement was not observed in irinotecan or taxanes. Ramucirumab, targeting vascular endothelial growth factor receptor 2, is reported as an immune modulator, increasing PD-L1 expression and CD8+ T cell infiltration and reducing immunosuppressive regulatory T cells in the tumor microenvironment of AGC (35). The combination of ramucirumab plus PD-1 inhibitors showed efficacy in AGC in several studies (36, 37). These factors indicate that the influence of prior ICI might be stronger on ramucirumab than on cytotoxic agents, and might not differ much among cytotoxic agents in AGC.
Irinotecan and FTD/TPI were feasible as fourth-line treatments after third-line nivolumab, and adverse events were consistent with those in previous clinical studies (10, 13-15). Nivolumab did not enhance the toxicity of subsequent chemotherapy. One patient died of liver failure during FTD/TPI. FTD/TPI was renally excreted, and liver failure was not reported in the TAGS study (10). In contrast, nivolumab sometimes induces liver toxicity. Treatment-related adverse events included aspartate aminotransferase elevation (3%) and alanine aminotransferase elevation (2%) in the ATTRACTION-2 study (9). Some immune-related adverse events (irAEs) occur several months after the initiation of nivolumab. Although it is difficult to make a firm conclusion, there is the possibility of liver failure caused by prior nivolumab in that patient. Late occurrence of irAEs should be noted even after nivolumab is terminated.
Nivolumab has been switched to the first-line, in combination with chemotherapy, from third-line treatment for HER2 negative AGC patients with CPS ≥5 after the CheckMate 649 study (6). Irinotecan and FTD/TPI are candidates for third-line therapy for these patients. Our data might also be suggestive in this situation. Nivolumab was reported to bind CD8 T cells for more than 20 weeks (38). Median PFS in second-line treatment in AGC was 2.3-4.4 months (12, 39), which shows that the influence of first-line nivolumab can continue until third-line treatment. Nivolumab is the standard third-line treatment for HER2 negative with CPS <5 or HER2 positive patients. For these patients, our data are still informative.
The present study has several limitations. First, this is a retrospective study at a single institution with a small number of patients. Thus, further evaluation is warranted in a larger cohort. Second, CPS, micro-satellite instability, and tumor mutation burden were not evaluated in all study patients, and immune status may influence clinical outcomes. Third, UGT1A1 was not evaluated in all study patients, and the toxicity of irinotecan is associated with UGT1A1 (40).
In conclusion, irinotecan and FTD/TPI showed similar efficacy and feasibility as fourth-line treatments after third-line nivolumab, which were comparable to previous studies (10, 13-15). Our findings can inform physicians deciding on the indication of irinotecan or FTD/TPI for AGC patients in this situation. Both agents can be treatment options in the fourth-line setting after third-line nivolumab. Further studies including a large number of patients are necessary to comfirm our results in clinical practice.
Acknowledgements
The Authors would like to thank Editage (www.editage.com) for English language editing.
Footnotes
Authors’ Contributions
KH, MF, NM, and AN participated in literature research and drafting of the article. KH, MF, KF, TH, MW, YI, SO, IH, HS, SN, KK, TY, T. Ogata, and T. Oshima participated in treating patients. KH, MF, and AN participated in analyzing the study data. FJ and SM edited the final version of the article. All Authors read and approved the final manuscript.
Conflicts of Interest
The Authors declare no conflicts of interest regarding the present study.
- Received March 19, 2023.
- Revision received April 1, 2023.
- Accepted April 3, 2023.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.









