Introduction
Gastric cancer (GC) is the fifth most common cancer and the third leading cause of cancer death worldwide [
1]. The high mortality rate in GC is mainly due to the lack of specific early manifestations, subsequently leading to the late diagnosis and treatment [
2,
3].
For patients with advanced or recurrent GC combination chemotherapy regimens consisting of fluoropyrimidines and platinum, with trastuzumab, in HER2 positive cancer, or a third agent such as taxane or anthracycline are the gold standard in the first line treatment [
4‐
8]. However, the prognosis for these patients remains poor, with a median overall survival (OS) that does not exceed 10–12 months [
9,
10]. Furthermore, the majority of patients do not respond or relapse within a short time after the end of first-line therapy and only about one-third of patients receive second-line chemotherapy treatments [
11]. Various cytotoxic agents both as monotherapy and in combination have been extensively studied in the second-line setting with minimal benefit in these patients [
12,
13].
However, although clinical studies have shown that second-line therapy improves OS compared to BSC to a statistically significant extent in actual clinical practice, second-line regimen may not be offered to all patients, mainly due to poor PS experienced after first-line therapy [
14‐
16]. Furthermore, in translating the survival benefit from clinical trials to real life, regional ethnic differences must be considered. In fact, almost all Asian patients with metastatic GC receive second-line therapy while in Western countries, less than half of patients receive progressive second-line treatment after first-line therapy [
17].
More recently, the use of ramucirumab, a human monoclonal antibody (IgG1) vascular endothelial growth factor receptor-2 (VEGFR-2) antagonist, plus paclitaxel versus paclitaxel alone in second line showed to increase median OS (9.6 months vs 7.4 months) and progression free survival (PSF) (4.4 months vs 2.9 months) in all subgroups of the phase III (RAINBOW) trial [
18]. Ramucirumab plus paclitaxel demonstrated the highest objective response rate (ORR) of 28% reported in the second line setting compared with 16% in the paclitaxel alone group (
p = 0.0001). Ramucirumab alone versus best supportive care (BSC) confirmed to improve OS in the REGARD study [
19]. Based on these findings, paclitaxel plus ramucirumab became the standard second-line treatment for advanced GC. In this study, we retrospectively evaluated the efficacy of ramucirumab plus paclitaxel in patients with advanced GC, presenting, details of the association between partial tumor response and survival outcomes.
Discussion
A large part of patients with GC are initially diagnosed with unresectable or metastatic disease, and first-line chemotherapy guarantees a median overall survival often not exceeding 12 months [
21,
22].
To date, the RAINBOW trial demonstrated the highest second-line response rate in patients with advanced gastric or gastroesophageal junction (GEJ) adenocarcinoma. Significant improvements in PFS (HR 0.635; 95%CI 0.536, 0.752;
p < 0.0001), OS (HR 0.807; 95% CI 0.678–0.962;
p = 0.0169), and overall response rates (27.9%
p = 0.0001) have been reached [
18]. Moreover, a subgroup analysis in Western population, showed a median OS of 8.5 months for ramucirumab and paclitaxel and 5.9 months for paclitaxel alone. Median PFS was 4.2 and 2.8 months in the two group, respectively with an ORR of 26.8% in the combination arm (
p = 0.0004) [
23]. Subsequently, real-life studies on ramucirumab and paclitaxel as second-line therapy in metastatic GC have achieved results in terms of OS and PSF overlapping [
24,
25].
In our single institution study, median OS [8 months, range 7–10] and PFS [4 months, range 3–5] were in line with the pivotal trial. However, focusing on patients who achieved PR to treatment, a define benefit was recorded in terms of survival. In these patients, compared with patients that do not experienced PR, OS and PFS were 15 months (HR 0.43; 95% CI 0.18–0.98;
p = 0.04) and 11 months (HR 0.42; 95% CI 0.19–0.94;
p = 0.02), respectively (Figs.
1 and
2). At the multivariate analysis, we identified ECOG PS = 1 (HR, 1.24; 95% CI, 1.01–1.76;
p = 0.04) and presence of peritoneal metastases (HR, 2.03; 95% CI, 0.403.68;
p < 0.01) as factors correlated with worse survival outcomes, whereas PR (HR, 0.55; 95% CI, 0.44–0.927;
p = 0.04) has been correlated with higher survival. Thus, although second-line treatments have guaranteed an improvement in terms of survival, the prognosis of these patients remains poor and only a few benefit from it. This emphasizes the need for identifying predictive biomarkers to better select patient and direct it to second-line chemotherapy with ramucirumab and paclitaxel or clinical trials.
Previously, depth of response (DpR) has been correlated with post progression survival in subgroups of gastric cancer patients receiving second-line chemotherapy, indicating DpR as possible new predictor for efficiency [
26]. However, the predictive value of DpR is not sure, which may be related to other factors, such us the mutation status of Human Epidermal Growth Factor Receptor 2 and treatment methods, among others.
A recent prospective study has suggested the prognostic value of some circulating factors such as neutrophil-to-lymphocyte ratio (NLR) and myeloid-derived suppressor cells (MDSCs) on survival outcomes in GC patients receiving ramucirumab plus paclitaxel treatment [
27]. Another study suggested that the occurrence of high-grade neutropenia can predict response to treatment with ramucirumab and paclitaxel. In this analysis, patients who experienced grade ≥ 3 neutropenia had a PFS of 6.6 months (95% CI 3.3–8.4) and an OS of 11 months (95% CI 5.9–13.1) compared to 4.4 months (95% CI 3.9–5.2) and 8.7 months (95% CI 7.8–10.1) for patients with lower grade neutropenia [
28].
Natsume et al. identified a correlation between aberrant expression of placental growth factor (PlGF) and ramucirumab responders and non-responders. OS (
p = 0.046) and PFS (
p = 0.016) were significantly shorter in the PlGF-high group than in the PlGF-low group. Overall response rates were 50% and 0% in the PlGF-low and high group, respectively [
29].
However, despite the various efforts made, no predictive biomarkers have yet been identified and the mechanism underlying the response or resistance to the combination of ramucirumab and paclitaxel remains unclear [
30].
Moreover, Cascinu et al. reported the correlation between tumor response, and the symptom palliation in the intent to treat population of the RAINBOW study, as also observed in patients with metastatic breast cancer receiving chemotherapy [
31].
This study presents several limitations mainly due to the retrospective nature of the data collection, the limited number of patients included, and a single Oncologic Center involved. Moreover, the inclusion of patients with primary progressive disease, who have an extremely poor survival, in the subgroup which do not experience PR, amplifies the prognostic impact of PR itself. However, while aware that tumor response should be associated with improved survival, this may not necessarily occur [
32]. What we want to underline with this work is the statistically significant difference in OS and PFS that we observed between the patients who have achieved PR and who have not achieved PR, which further pushes us to continue looking for biomarkers capable of selecting patients guaranteeing the best therapeutic choice Figs.
1 and
2.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.