The role of non-curative surgery in incurable, asymptomatic advanced gastric cancer

Abstract

Background: Although general agreement exists on palliative surgery with intent of symptom palliation in advanced gastric cancer (AGC), the role of non-curative surgery for incurable, asymptomatic AGC is hotly debated. We aim to clarify the role of non-curative surgery in patients with incurable, asymptomatic AGC under the first-line chemotherapy.

Methods: A total of 737 patients with incurable, asymptomatic advanced gastric adenocarcinoma between January 2008 and May 2012 at the Sun Yat-sen University Cancer Center were retrospectively analyzed, comprising 414 patients with non-curative surgery plus first-line chemotherapy, and 323 patients with first-line chemotherapy only. The clinicopathologic data, survival, and prognosis were evaluated, with propensity score adjustment for selection bias.

Results: The median overall survival (OS) outcomes significantly favored non-curative surgery group over first-line chemotherapy only group in entire population (28.00 versus 10.37 months, P = 0.000), stage 4 patients (23.87 versus 10.37 months, P = 0.000), young patients (28.70 versus 10.37 months, P = 0.000) and elderly patients (23.07 versus 10.27 months, P = 0.031). The median OS advantages of non-curative surgery over first-line chemotherapy only were also maintained when the analyses were restricted to single organ metastasis (P = 0.001), distant lymph node metastasis (P = 0.002), peritoneal metastasis (P = 0.000), and multi-organ metastasis (P = 0.010). Significant OS advantages of non-curative surgery over chemotherapy only were confirmed solid by multivariate analyses before and after adjustment on propensity score (P = 0.000). Small subsets of patients with surgery of single metastatic lesion after previous curative gastrectomy, and with surgery of both primary and single metastatic sites showed sound median OS.

Conclusions: There is a role for non-curative surgery plus first-line chemotherapy for incurable, asymptomatic AGC, in terms of survival. Randomized controlled trials are warranted to fill a gap in knowledge about the value of metastectomy and patient selection strategies.

Figure 1. Kaplan-Meier curves of non-curative surgery group and first-line chemotherapy only group, in entire population.

HR, hazard ratio; CI, confidence interval.

Figure 2. Kaplan-Meier curves of non-curative surgery and first-line chemotherapy only groups, by stages and inclusion criteria.

Figure 1A shows subgroup analysis according to stage, with stage 4 first-line chemotherapy only group as reference. Stage 4 here included metastatic and recurrent gastric cancer. Figure 1B shows subgroup analysis according to inclusion criteria, with the inclusion criteria (6) as reference. HR, hazard ratio; CI, confidence interval.

Figure 3. Kaplan-Meier curves of non-curative surgery group and first-line chemotherapy only group, by metastasis types.

HR, hazard ratio; CI, confidence interval.