Re-evaluating the neutropenia–survival paradigm in irinotecan-based therapy for advanced gastric cancer: post-hoc analysis of the RINDBeRG trial

Background Chemotherapy-induced neutropenia (CIN) has been considered a favorable prognostic marker across multiple cytotoxic regimens for gastrointestinal malignancies. However, its prognostic relevance in irinotecan (IRI)-based therapy for advanced gastric cancer (AGC) remains unclear. Methods This post-hoc analysis used data from the phase III RINDBeRG trial comparing ramucirumab (RAM) plus IRI with IRI monotherapy as third- or later-line treatment for RAM-refractory AGC. Overall survival (OS) was evaluated according to the occurrence of grade 3 or higher neutropenia, defined as CIN. OS was also evaluated according to the baseline absolute neutrophil count (ANC), stratified by quartile-based cutoffs. Results Among the 393 patients in the safety population, 44.5% and 28.5% in the RAM + IRI and IRI arms developed CIN, respectively. Baseline characteristics were generally comparable, with some imbalances across baseline ANC groups. In the RAM + IRI arm, those with and without CIN had a median OS of 9.7 and 8.9 months, respectively (HR = 0.84; 95% CI, 0.62–1.13; p = 0.25). In the IRI arm, those with and without CIN had a median OS of 8.4 and 8.6 months, respectively (HR = 0.85; 95% CI, 0.60–1.19; p = 0.34). Higher baseline ANC was associated with shorter OS, particularly in the RAM + IRI arm (7.3 vs. 10.9 months; HR = 1.83; 95% CI 1.14–2.94; p = 0.01). Conclusions Unlike other cytotoxic regimens, CIN was not associated with survival in AGC patients receiving IRI with or without RAM. The clinical relevance of baseline ANC warrants further investigation. Trial registration jRCT2031200106.