Background: Cisplatin-based chemotherapy is frequently used to treat advanced gastric cancer (GC). Although it leads to increased overall survival (OS) when added to single agents or chemotherapy doublets, toxicity is also generally increased. The purpose of this meta-analysis study was to compare the efficacy of chemotherapy with and without cisplatin in patients with advanced GC.
Methods: Randomised trials that compared first-line cisplatin-based chemotherapy with regimens in which cisplatin was replaced by other agents were identified by electronic searches of PubMed, EMBASE, the Web of Science, and the Cochrane Central Register of Controlled Trials. Meta-analysis was performed using a fixed or random effects model. OS, reported as a hazard ratio (HR) and a 95% confidence interval (CI), was the primary outcome measure.
Results: Fourteen trials (5 phase III and 9 phase II), including 2,981 patients, were identified. Overall, chemotherapy regimens without cisplatin significantly improved OS (HR, 0.79; 95% CI, 0.68-0.92; p = 0.003), progression-free survival (PFS) (HR, 0.77; 95% CI, 0.66-0.90; p = 0.001), and response rate (RR) (OR, 1.25; p = 0.004) when compared to cisplatin-containing regimens. A subgroup analysis according to histology, site of the primary tumour and extent of disease was not possible due to lack of data.
Conclusions: Compared with cisplatin-based doublets and triplets, combinations in which cisplatin was replaced by new drugs improved outcome and RRs in randomised trials for advanced GC and therefore should be strongly considered in the metastatic setting. A limitation of this meta-analysis is that we cannot identify a subgroup of patients (according to histology, site of primary tumour or burden of metastatic disease) which could derive greater benefit from cisplatin-free chemotherapy.