Long-Term Oncological Outcomes of Granulocyte Colony-Stimulating Factor (G-CSF) Treatment in Gastrointestinal Cancers: A Systematic Review and Meta-Analysis

Abstract

Background: Granulocyte-colony stimulating factor (G-CSF) prophylaxis is widely used in gastrointestinal (GI) cancers. The use of G-CSF in GI cancers has not previously been investigated systematically in a meta-analysis. Thus, we systematically reviewed the literature to describe the G-CSF use and potential influence on long-term oncological outcomes in GI cancers.

Method: The literature search of this systematic review and meta-analysis was conducted in PubMed, Embase, Cochrane Library and Web of Science. The PRISMA-P guidelines were followed. Studies that reported data on patients with GI cancers undergoing oncological treatment with G-CSF prophylaxis were included. Outcomes of interest were overall survival (OS), progression-free survival (PFS) and adverse events (AE), specifically neutropenia grade III/IV. A time-to-event random-effects meta-analysis was conducted. Risk of bias was assessed using the Newcastle-Ottawa Scale and the Cochrane Risk of Bias Tool for Randomized Controlled Trials (RoB) tool.

Findings: In total, 2452 articles were screened for eligibility. Ultimately, 13 studies were included with a total patient number of 2673. The included studies indicated a positive association between OS and G-CSF prophylaxis (HR 0.72, 95% CI: 0.56-0.91, I²: 54%, low quality evidence). No significant relation between G-CSF use and PFS was found in the pooled analyses (HR 0.74, 95% CI: 0.51-1.08, I²: 73%, moderate quality evidence). However, a positive effect of G-CSF use was found in the retrospective cohorts reporting data on PFS (HR 0.50, 95% CI: 0.32-0.77, I²: 0%). A marked drop in neutropenia grade III/IV rates was observed in patients treated with G-CSF (risk ratio (RR) 0.46, 95% CI: 0.28-0.77, I²: 72%, high quality evidence).

Interpretation: G-CSF prophylaxis provides a reduction in neutropenia grade III/IV in patients with GI cancers (high level of certainty) and a favorable OS (low certainty), while PFS is unaffected (moderate certainty). Studies on PFS and G-CSF use are nonetheless limited.

Keywords: G-CSF; adverse events; filgrastim; gastrointestinal cancer; peg-filgrastim; survival.

Figure 1

PRISMA flow-chart. n represents the sample size. Created in BioRender. Fischer, O. https://BioRender.com/cp76rte (accessed on 23 March 2025).

Figure 2

Forrest plot on all studies reporting OS [57,59,60,61,62,64,65,67,68,69]. (A) and subgroup analysis regarding study design (B) applying a random effect model on time-to-event data. CI = confidence interval; HR = hazard ratio; OS = overall survival; TE = ln(HR); seTE = standard error for ln(HR).

Figure 3

Forrest plot on all studies reporting PFS [57,59,61,62,64,65,67,69]. (A) and subgroup analysis regarding study design (B) applying a random effect model on time-to-event data. CI = confidence interval; HR = hazard ratio; PFS = Progression-free survival; TE = ln(HR); seTE = standard error for ln(HR).

Figure 4

Forrest plot on all studies reporting AE [57,59,61,62,64,65,67,68]. (A) and subgroup analysis regarding study design (B) applying a random effect model on time-to-event data. CI = confidence interval; HR = hazard-ratio; AE = adverse event(s); TE = ln(HR); seTE = standard error for ln(HR).