Retrospective observational study of salvage line ramucirumab monotherapy for patients with advanced gastric cancer

Gastric cancer is the third leading cause of death due to cancer worldwide [1]. Chemotherapy is the standard treatment for patients with advanced (unresectable or recurrent) gastric cancer (AGC). Although recent developments in chemotherapies for AGC have prolonged patients’ survival, their prognoses remain poor. The commonly used regimen for first-line treatment of AGC is combination chemotherapy, including fluoropyrimidine and platinum, which achieves a median progression-free survival (PFS) of ~ 6 months and a median overall survival (OS) of ~ 1 year [2‐4].

In a randomized phase III “Ramucirumab Monotherapy for Previously Treated Advanced Gastric or Gastroesophageal Junction Adenocarcinoma” (REGARD) trial, administering ramucirumab, an anti–vascular endothelial growth factor receptor (anti-VEGFR) monoclonal antibody, as a second-line treatment prolonged OS compared to the placebo (hazard ratio [HR] = 0.774; 95% confidence interval (CI) = 0.605–0.991) [5]. In addition, in a “Ramucirumab plus Paclitaxel versus Placebo plus Paclitaxel for Previously Treated Advanced Gastric or Gastroesophageal Junction Adenocarcinoma” (RAINBOW) trial, combination chemotherapy, including ramucirumab and paclitaxel as a second-line treatment, prolonged OS compared to the placebo and paclitaxel regimen (HR = 0.807; 95% CI = 0.678–0.962) [6]. Therefore, ramucirumab and paclitaxel doublet therapy is currently considered a standard second-line treatment, and ramucirumab monotherapy is also considered a second-line treatment option for AGC.

In clinical practice, some patients with AGC refractory to taxanes (e.g., patients who had an early relapse or progression during or soon after taxane-containing perioperative treatment) underwent ramucirumab monotherapy as third- or later-line treatment. Recent evidence has demonstrated that a taxane-containing regimen is beneficial in a perioperative setting. Al-Batran et al. [7] showed that a combination regimen, including fluorouracil, leucovorin, oxaliplatin, and docetaxel (FLOT), for resectable gastric cancer patients in a perioperative setting prolongs OS compared to a combination regimen including epirubicin, platinum, and fluorouracil. Yoshida et al. [8] reported that S-1 and docetaxel as adjuvant therapy prolonged relapse-free survival compared to S-1 alone. For patients who have an early relapse, despite undergoing docetaxel-containing perioperative chemotherapy, second-line chemotherapy without taxane (e.g., irinotecan or nivolumab) is considered optimal because of cross-resistance between docetaxel and paclitaxel [9, 10]. In such situations, ramucirumab monotherapy would be administered as third- or later-line treatment.

There are few reports on the efficacy and safety of ramucirumab monotherapy as third- or later-line treatment. Therefore, this retrospective, observational study evaluated the efficacy and safety of ramucirumab monotherapy as third- or later-line treatment for AGC refractory or intolerant to fluoropyrimidines and taxanes.

As second-line treatment for AGC, the REGARD trial showed a median PFS of 2.1 months and a median OS of 5.2 months [5]. A Japanese trial on ramucirumab monotherapy as second-line treatment showed a median PFS of 1.6 months and a median OS of 7.3 months [16]. This favorable OS, despite a PFS similar to that in this study, might be obtained by patient selection and subsequent treatments. We believe that the efficacy of ramucirumab monotherapy as third- or later-line treatment, as shown in this study, is comparable to previous studies that reported ramucirumab monotherapy as second-line treatment.

Paulson et al. [17] reported retrospective observational data on ramucirumab monotherapy for AGC patients in the United States. They studied 115 patients who underwent ramucirumab monotherapy, 38.3% of whom underwent ramucirumab monotherapy as third- or later-line treatment. The authors reported that the median duration of therapy was 1.87 months; the median OS and RR were not reported. Murahashi et al. [18] retrospectively analyzed the efficacy and safety of ramucirumab monotherapy as second- or later-line treatment at a single institution. The authors reported that 15 of their 19 patients underwent ramucirumab monotherapy as third- or later-line treatment. Overall, the median OS, median PFS, RR, and grade 3 AE incidence were 12.9 months, 2.1 months, 7.7, and 10%, respectively. Our results were comparable to these reports, suggesting prolonged survival after salvage-line ramucirumab monotherapy for AGC.

The NLR is reportedly a poor prognostic factor in several cancers, such as colon, breast, pancreatic, and gastric cancer [19‐22], in a metastatic or an adjuvant setting. The NLR is an inflammation marker; a high NLR reflects enhanced systemic inflammation and is associated with reduced tumor-specific immunity, such as decreased tumor-infiltrating lymphocytes in a tumor [23]. Neutrophilia inhibits the cytolytic activity of immune cells [24] and increases the production of factors that promote tumor growth, such as VEGF and hepatocyte growth factor [25]. Lymphocytopenia is often observed in patients with advanced disease, indicating a state of immunosuppression [26]. From a pooled analysis of REGARD and RAINBOW trials, Fuchs et al. [27] reported that 12 clinical and laboratory factors, including a high neutrophil and low lymphocyte count, at the start of ramucirumab monotherapy are negative prognostic factors for OS. From Korean retrospective cohort data of ramucirumab mono- or combination therapy, Jung et al. [28] reported that the NLR and albumin are independent prognostic factors for the PFS and OS. Our results also supported these findings.

Although studies have evaluated several predictive biomarkers of antiangiogenic therapy, no reproducible biomarkers, including soluble VEGF-C, VEGF-D, VEGFR1, and VEGFR3, have been identified in the REGARD trial [29]. Some studies have suggested the NLR’s predictive value. Cho et al. [30] reported that AGC patients with low NLRs have higher DCRs for cytotoxic chemotherapy compared to patients with high NLRs. Similarly, the association between the NLR and the tumor response of chemotherapy has been reported in several neoplasms, such as colon, head and neck, and lung cancer [20, 31, 32]. The immunological response caused by chemotherapy could be a reason for these findings. Compared to patients with high NLRs, patients with low NLRs are likely to have the inflammatory reaction environment mediated by lymphocytes [30]. Ogata et al. [33] and Bagley et al. [34] reported that the NLR might predict the tumor response of nivolumab in gastric cancer and non–small cell lung cancer, respectively. Tada et al. [35] reported that ramucirumab suppresses tumor infiltration of effector regulatory T-cells (eTreg) and decreases programmed cell death protein 1 (PD-1) expression by CD8+ T-cells, subsequently enhancing antitumor immunity. Therefore, the NLR might also predict the ramucirumab-induced immunological antitumor effect.

In third-line treatment, the patients’ performance status is generally worse than second-line treatment. In the REGARD and Japanese prospective phase II trials, 28 and 66% of patients had ECOG PS = 0, respectively [5, 16], compared to 8% in our study; in addition, 15% of the patients had ECOG PS = 2. In the REGARD and Japanese prospective phase II prospective trials, all AEs (any grade) occurred in 94 and 97% of patients (74% in our study), respectively, while AEs of ≥ grade 3 occurred in 57 and 25% of patients (13% in our study), respectively. These findings suggest that ramucirumab monotherapy is also feasible as salvage-line treatment.

This study had a few limitations. It was retrospective in nature, had a small sample size, and had selection bias. The median duration from the start of first-line treatment to the start of ramucirumab monotherapy was 21.9 months, which was long compared to the average of OS of AGC patients, suggesting that we selected patients with slowly growing cancers or cancers highly responsive to prior chemotherapy. However, because we analyzed multicenter data, our findings seem generalized to patients in clinical practice. Secondly, we could not assess patient’s quality of life (QOL). QOL is one of the most important outcomes for the patients in salvage-line treatments. In phase III REGARD trial, ramucirumab monotherapy demonstrated the better QOL compared with best supportive care in second-line. In our study, toxicities of salvage-line ramucirumab were comparable to the second-line setting, and actually, 93% of patients could continue the treatment until disease progression. Therefore, we speculated salvage-line ramucirumab might have similar effect for QOL, however, further investigation is needed.

Ramucirumab monotherapy demonstrates acceptable efficacy and feasibility as third- or later-line treatment for AGC. We consider physicians might choose this therapy as a salvage-line treatment option for AGC refractory to the standard therapies.