Intraperitoneal chemotherapy for gastric cancer

Gastric cancer with peritoneal metastasis carries a poor prognosis and is considered incurable. Intraperitoneal chemotherapy (IPC) has been explored for preventing and treating peritoneal metastasis, but clinical trials show conflicting results and guidelines provide inconsistent recommendations. A comprehensive evaluation of intraperitoneal chemotherapy effects in gastric cancer is needed.

To evaluate the benefits and harms of IPC in gastric cancer for: (1) prophylactic IPC plus radical surgery versus radical surgery alone in people at high risk of peritoneal metastasis; and (2) therapeutic IPC plus cytoreductive surgery (CRS) versus CRS alone in people with confirmed peritoneal metastasis.

We searched CENTRAL, MEDLINE, Embase, two Chinese databases, three trials registries, and gray literature sources. We also checked references and contacted study authors. The latest search was conducted on 12 June 2025.

We included parallel randomized controlled trials (RCTs) comparing IPC with no IPC for prophylactic and therapeutic use. We excluded studies without postoperative systemic chemotherapy. We assessed trial trustworthiness using the trustworthiness of randomized controlled trials (TRACT) checklist.

Our outcomes were overall survival, serious adverse events (SAEs), surrogate endpoints for overall survival including disease-free survival (DFS) for prophylactic IPC and progression-free survival (PFS) for therapeutic IPC, quality of life (QOL), total adverse events (AEs), anastomotic leakage, and intra-abdominal abscess.

We assessed the risk of bias (RoB) for outcomes reported in the summary of findings tables using the Cochrane RoB 2 tool.

We synthesized results for each outcome using meta-analysis, by calculating hazard ratios (HRs) and risk ratios (RRs) with 95% confidence intervals (CIs) for survival outcomes and dichotomous outcomes. Where meta-analysis was not possible, we summarized results narratively. We used GRADE to assess evidence certainty.

We identified nine parallel RCTs involving 829 participants that met eligibility criteria. Seven studies (656 participants) evaluated prophylactic IPC and two studies (173 participants) evaluated therapeutic IPC. Among the included studies, seven studies were conducted in China. Follow-up ranged from 0.2 months to 83.5 months. We assessed most outcomes to have some concerns or a high risk of bias.

We are very uncertain about all results due to very low certainty evidence, downgraded for risk of bias, indirectness, and imprecision for each outcome. Comparison 1: prophylactic IPC plus radical surgery versus radical surgery alone in participants at high risk of peritoneal metastasis Six trials investigated hyperthermic intraperitoneal chemotherapy (HIPEC) effects; one examined normothermic intraperitoneal chemotherapy (NIPEC). No studies reported QOL, SAEs, or total AEs. IPC may increase overall survival (HR 0.66, 95% CI 0.48 to 0.91; 6 studies, 522 participants; very low-certainty evidence) but may have little to no effect on DFS (HR 0.85, 95% CI 0.40 to 1.82; 1 study, 134 participants; very low-certainty evidence). IPC may make little to no difference to anastomotic leakage (RR 1.68, 95% CI 0.43 to 6.58; 4 studies, 366 participants; very low-certainty evidence) and intra-abdominal abscess (RR 2.04, 95% CI 0.19 to 21.80; 1 study, 105 participants; very low-certainty evidence). Comparison 2: therapeutic IPC plus CRS versus CRS alone in participants with confirmed peritoneal metastasis Two trials investigated HIPEC effects. No studies reported on the incidence of total AEs or intra-abdominal abscess. IPC may increase overall survival (HR 0.52, 95% CI 0.28 to 0.96; 2 studies, 173 participants; very low-certainty evidence). IPC may make little to no difference to SAEs (RR 1.25, 95% CI 0.37 to 4.26; 1 study, 68 participants; very low-certainty evidence) and anastomotic leakage (RR 0.90, 95% CI 0.15 to 5.49; 2 studies, 126 participants; very low-certainty evidence). One study with 105 participants (follow-up ranging from 0.2 months to 65.4 months) suggests that IPC may increase median PFS from 3.5 months (95% CI 3.0 to 7.0) to 7.1 months (95% CI 3.7 to 10.5); P = 0.047; very low-certainty evidence. However, IPC may result in little to no difference in QOL measured by the European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Questionnaire Core 30 (QLQ-C30) and the EORTC Quality of Life Questionnaire-Stomach module (QLQ-STO22) (P = 0.102 for QLQ-C30 global health status; 1 study, unclear number of participants; very low-certainty evidence).

Current evidence for IPC is limited. Very low-certainty evidence suggests prophylactic and therapeutic IPC (primarily HIPEC) may improve survival and may have little to no effect on anastomotic leakage. Prophylactic IPC may have little to no effect on tumor recurrence and intra-abdominal abscess. Therapeutic IPC may have little to no effect on SAEs; limited evidence indicates IPC may delay tumor progression and may make little to no difference to QOL. These results are highly uncertain and require cautious interpretation. Therefore, it is not possible to draw definitive conclusions or outline specific implications for the routine use of IPC in individuals with gastric cancer based on the current evidence. Further high-quality, long-term RCTs - particularly involving non-Asian populations - are needed, along with more comprehensive data on safety and QOL. Additionally, a considerable number of studies are still ongoing, meaning effect estimates and conclusions may change when these findings become available.

This Cochrane review was funded (in part) by the foundation of the National Natural Science Foundation of China (No. 82472926), the Foundation of Science & Technology Department of Sichuan Province (2023YFS0060; 23ZDYF2812), the 1.3.5 Project for Disciplines of Excellence, West China Hospital, Sichuan University (ZYJC21006; 2023HXFH005), and the Postdoctor Research Fund of West China Hospital, Sichuan University (2025HXBH063).

Protocol (2009) available via doi.org/10.1002/14651858.CD008157 Updated protocol (2023) available via doi.org/10.1002/14651858.CD015698.