Background and objectives. Peritoneal relapse (PR) is the most common pattern of gastric cancer (GC) recurrence after radical treatment. Currently, a variety of adjuvant intraperitoneal chemotherapy methods are being tested for their efficacy in reducing the level of PR. 

Methods. The radical treatment results of 226 patients with localized and locally-advanced GC have been analyzed. To select a group of patients with indications for adjuvant intraperitoneal therapy, a study of independent predictive factors and the development of a predictive PR nomogram for gastric cancer was completed.

Results. As a result of the analysis of about three dozen potential factors in mono- and multivariate analysis, the impact on PR risk was confi rmed by 4 independent predictive factors, namely: serosal invasion and its size (HR 9.36, p <0.001), morphological type according to Lauren (HR 5.3, p <0.001), index of regional lymph node involvement (HR 2.23, p = 0.015) and localization of the tumor in the stomach (HR 3.98, p <0,001).

Conclusions. A predictive PR risk nomogram of gastric cancer after radical surgical treatment has been developed based on the identifi ed independent factors, and it is of great clinical importance as a tool for segregating patients who require adjuvant intraperitoneal chemotherapy.

Extensive dissection of retroperitoneal lymph node (LN) collectors during D2 lymph node dissection inevitably leads to the crossing of a large number of lymphatic ducts with an increased risk of lymph leakage into the abdominal cavity and a potential risk of intraperitoneal dissemination of free malignant cells. The increased risk of intraperitoneal recurrence of gastric cancer (GC) after advanced lymph node dissection remains unclear. The aim of our study was to investigate an advanced lymph node dissection during surgical treatment of gastric cancer as a separate potential predictor of intraperitoneal relapse.

 The level of intraperitoneal recurrence and peritoneal recurrence free survival (PRFS) in 226 patients with localized and locally advanced GC after radical surgery with different levels of lymph node dissection were analyzed.

 The average number of removed LN was 22.94 ± 11.46 (from 4 to 67). There was a statistically significant difference (χ2 = 110.75; p = 0.0001) in the number of removed LN during lymph node dissection of different levels (D0, D1, D1 +, D2). In the long term the overall level of GC progression did not differ in patients with D0,1 (n = 110) and D1+,2 (n = 116) lymph node dissections - 40% and 38,79%, respectively (χ2 = 0.03; p = 0,85). The level of loco-regional recurrences (in regional groups of LN) in subgroup D0,1 was twice as high as in subgroup D1+,2 - 14.55% and 7.76%, respectively (χ2 = 2.67; p = 0.102). Tendency (p = 0.32) to increase the overall 5-year survival by 8.3% with D1+,2 was recorded. No significant increase in the level of intraperitoneal recurrences after advanced lymph node dissection was detected: in subgroup D0,1 they were recorded in 23 (20.9%) patients, in subgroup D1+,2 - in 22 (19%) (χ2 = 0.13; p = 0.72). The level of 2-year PRFS was the same in both subgroups: 77.2% with D0,1 and 82.8% with D1+,2 (p = 0.6, HR 0.36, 95% CI 0.5-3.68). No effect of  D1+,2 on the level of PRFS was found in the subgroup of patients with diffuse/mixed GC and pN+ (p = 0.62), and in the subgroup of patients with high (> 0.2) index of regional LN lesion (p = 0.8). At the same time, a significant increase of 5-year overall survival of patients with diffuse/mixed GC type with D1+,2 lymph node dissection was found (p = 0.048), and loss of efficacy of this level of lymph node dissection in cases of gastric serous membrane area lesion > 5 cm2. 

Carrying out D1+,2 lymph node dissection during surgical treatment of GC does not increase the risk of intraperitoneal recurrence in the long term. The effect of D1+,2 lymph node dissection on overall survival is offset by intraperitoneal relapse in patients with diffuse/mixed type of GC and serous membrane lesion area > 5 cm².