Neoadjuvant chemotherapy for gastric cancer. Is it a must or a fake?

Aim: To investigate the neoadjuvant chemotherapy (NAC) effect on the survival of patients with proper stomach cancer submitted to D2 gastrectomy.

Methods: We proceeded to a review of the literature with PubMed, Embase, ASCO and ESMO meeting abstracts as well as computerized use of the Cochrane Library for randomized controlled trials (RCTs) comparing NAC followed by surgery (NAC + S) with surgery alone (SA) for gastric cancer (GC). The primary outcome was the overall survival rate. Secondary outcomes were the site of the primary tumor, extension of node dissection according to Japanese Gastric Cancer Association (JGCA) performed in both arms, disease-specific (DSS) and disease-free survival (DFS) rates, clinical and pathological response rates and resectability rates after perioperative treatment.

Results: We identified a total of 16 randomized controlled trials comparing NAC + S ( = 1089) with SA ( = 973) published in the period from January 1993 - March 2017. Only 6 of these studies were well-designed, structured trials in which the type of lymph node (LN) dissection performed or at least suggested in the trial protocol was reported. Two out of three of the RCTs with D2 lymphadenectomy performed in almost all cases failed to show survival benefit in the NAC arm. In the third RCT, the survival rate was not even reported, and the primary end points were the clinical outcomes of surgery with and without NAC. In the remaining three RCTs, D2 lymph node dissection was performed in less than 50% of cases or only recommended in the "Study Treatment" protocol without any description in the results of the procedure really perfomed. In one of the two studies, the benefit of NAC was evident only for esophagogastric junction (EGJ) cancers. In the second study, there was no overall survival benefit of NAC. In the last trial, which documented a survival benefit for the NAC arm, the chemotherapy effect was mostly evident for EGJ cancer, and more than one-fourth of patients did not have a proper stomach cancer. Additionally, several patients did not receive resectional surgery. Furthermore, the survival rates of international reference centers that provide adequate surgery for homogeneous stomach cancer patients' populations are even higher than the survival rates reported after NAC followed by incomplete surgery.

Conclusion: NAC for GC has been rapidly introduced in international western guidelines without an evidence-based medicine-related demonstration of its efficacy for a homogeneous population of patients with only stomach tumors submitted to adequate surgery following JGCA guidelines with extended (D2) LN dissection. Additional larger sample-size multicentre RCTs comparing the newer NAC regimens including molecular therapies followed by adequate extended surgery with surgery alone are needed.