To the best of our knowledge, this is the largest cohort study to evaluate the efficacy and safety of nab-PTX plus RAM compared with PTX plus RAM as second-line treatment for patients with AGC. Our study indicated that the combination of nab-PTX plus RAM has a similar efficacy and safety profile to PTX plus RAM in patients with AGC. Although only a single-arm phase II trial has assessed the efficacy and safety of nab-PTX plus RAM, this regimen may be an option for previously treated patients with AGC. This alcohol-free regimen is linked to shorter infusion time and reduced rate of hypersensitivity reactions [13].
There were no significant differences in PFS and ORR between the nab-PTX plus RAM and PTX plus RAM groups. As real-world data, the efficacy observed in the PTX plus RAM group in our study was comparable to that recorded in the RAINBOW study [6]. The PFS and ORR in the nab-PTX plus RAM group were relatively inferior to those reported in the phase II trial of nab-PTX plus RAM for patients with AGC, showing a median PFS of 7.6 months and an ORR of 54.8% [13]. However, a higher proportion of patients who received previous platinum containing regimen and/or <6 month of duration of first-line chemotherapy, and had peritoneal metastasis were included in our study. The difference in patient characteristics may have led to lower ORR and shorter median PFS compared with those noted in the clinical trial. In terms of OS, there was no significant difference observed between the two groups. However, a higher proportion of patients who received subsequent anti-PD–1/PD-L1 therapy in the nab-PTX plus RAM group contributed to the plateau of the Kaplan–Meier curve at the long-term follow-up compared with that of the PTX plus RAM group. In a subgroup analysis, PFS in patients with moderate/massive ascites tended to be better with nab-PTX plus RAM than PTX plus RAM. The ABSOLUTE trial, which demonstrated the non-inferiority of weekly nab-PTX to weekly PTX for patients with AGC, also suggested an increased efficacy of weekly nab-PTX in patients with ascites or peritoneal metastasis [11, 16]. While the reason for this remains unclear, a multicenter randomized phase II P-SELECT trial of nab-PTX plus RAM versus PTX plus RAM as second-line therapy for AGC patients with peritoneal dissemination (WJOG10617G, jRCTs031180022) is underway and may confirm this observation.
The general safety profile of nab-PTX plus RAM was manageable and comparable to that of PTX plus RAM. Although anemia and fatigue (any grade) were more frequent in the nab-PTX plus RAM group, there was no difference in grade ≥3 of those adverse events between the two groups, and none of the patients discontinued treatment due to these adverse events. The incidence of sensory neuropathy was also significantly higher in the nab-PTX plus RAM group. However, only 3.5% and 1.4% of patients were forced to discontinue nab-PTX and PTX, respectively. In the phase II trial of nab-PTX plus RAM, 76.7% of patients experienced grade ≥3 of neutropenia. In our study, a relatively lower proportion of patients (56.6%) experienced this adverse event. The initial dose reduction of nab-PTX in 47.8% of patients in our study may have resulted in the lower frequency of grade ≥3 neutropenia; however, importantly, the median RDI was similar to that observed in the phase II trial [13]. These findings indicated that appropriate dose modification enables treatment continuation, irrespective of nab-PTX plus RAM or PTX plus RAM. Of note, there was no occurrence of hypersensitivity reactions in the nab-PTX plus RAM group with premedication of only chlorpheniramine. Two patients experienced grade 3 hypersensitivity reactions in the PTX plus RAM group and required emergent hospitalization despite adequate premedication. The incidence of infrequent hypersensitivity reactions in patients receiving nab-PTX was consistent with that noted in the ABSOLUTE trial [12]. Nab-PTX was suitable for shorter-time infusion without premedication. The incidence of specific TRAEs related to RAM was similar in both groups, and there were no unexpected TRAEs observed in our study.
This study had several limitations. Firstly, this was a non-randomized retrospective study performed in a single institution with a limited sample size. Secondly, there was a shorter follow-up time in the nab-PTX plus RAM group compared with that of the PTX plus RAM due to the approval of nab-PTX in 2017. Finally, all patients enrolled in this study were Japanese. Recently, a single-arm phase II trial of nab-PTX plus RAM for patients with AGC conducted in the United States of America suggested acceptable safety profiles [17]. These data support the results of our study for the application of this regimen to all patients, regardless of race.