Gastric cancer is the fifth most common malignant neoplasm and the third leading cause of cancer-related death in the world [1]. Peritoneal metastasis is one of the most frequent recurrence sites in patients who received gastrectomy for advanced gastric cancer, and this is associated with a quite poor survival outcome [2,3]. In gastric cancer, lavage cytology-positive (CY1) is related to peritoneal metastasis and the 5-year OS rate was reported to be only 2% in 1999 [4]. Due to development of chemotherapy, 5-year OS rate improved to 26% in a 2012 report [5]. Most recently, the 5-year OS rate for patients with CY1 or P1a was reported as 21.5–22.3%, which still remains unsatisfactory [6]. Peritoneal metastasis is developed by the implantation of peritoneal free cancer cells released from primary tumors [7,8]. Extensive intraoperative peritoneal lavage (EIPL) treatment has been attracting attention, because EIPL might be effective to prevent peritoneal metastasis by removing intraperitoneal free cancer cells before their fixation to peritoneum and progression.

A RCT conducted by Kuramoto revealed that the 5-year overall survival rate of patients receiving EIPL plus intraperitoneal cisplatin was much better than that of patients who received intraperitoneal cisplatin or no treatment after gastrectomy for CY1P0 gastric cancer [9]. EIPL can mechanically reduce the number of free cancer cells in the abdominal cavity through a limiting-dilution–like effect. While intraoperative or perioperative chemotherapy may exert a synergistic effect with EIPL by biologically eliminating cancer cells, EIPL alone might improve survival by removing intraperitoneal free cancer cells and preventing peritoneal metastasis. However, it remained unclear what kind of population is the best candidate of EIPL.

Our group revealed that the combination of intra-operative stamp cytology and peritoneal lavage cytology (CY) improved the identification of individuals with high risk of peritoneal metastasis [10]. The stamp cytology method can detect free cancer cells exposed on the serosal surface, which may lead to peritoneal metastasis in the near future. Accordingly, we considered that patients with CY1 and/or stamp+ could be the best candidate of EIPL, and started the current clinical trial in 2017.

On the other hand, in 2019, phase 3 RCT conducted by Misawa was reported, in which there is no statistically significant difference in DFS or OS between patients who had EIPL after gastrectomy for advanced gastric cancer of T3 status or above [11]. In addition, a clinical trial (JLSSG0901) revealed the non-inferiority of laparoscopic gastrectomy for advanced gastric cancer in 2023 [12]. This resulted in the significant decrease of open surgery cases. EIPL technically takes time in laparoscopic surgery. Due to the situations above, we decided to stop our trial, and analyze the already enrolled patients.

This study aimed to clarify the effectiveness of EIPL in patients with CY1 and/or stamp-positive findings. Although open surgery has become less common, making it premature to draw definitive conclusions from this study, it represents the first attempt to evaluate EIPL in this specific population. We believe that the data obtained will provide important insights for developing future treatment strategies for advanced gastric cancer.

Because of the recent decline in the number of open abdominal surgery, we need to stop the trial at 13 cases enrollment. As shown in Fig 2A, B, Median RFS was 14.5 months (3-year RFS was 38%) and median OS was NR months (3-year OS was 57%), these results were relatively good compared with our prespecified endpoints, although we were not able to analyze the endpoints appropriately due to the lack of required cases. This suggests that EIPL might be effective to prevent peritoneal recurrence in selected cases.

Although the subgroup analysis suggested that EIPL may preferentially reduce peritoneal recurrence in biologically selected patients—particularly those with CY0 and stamp-positive disease—these findings are exploratory in nature. Because the primary subgroup result for peritoneal RFS was derived from a very small sample, the survival estimates are inherently unstable, increasing the vulnerability of the study to type I error. Therefore, these observations should be interpreted with caution, and validation in adequately powered, controlled trials will be essential to determine whether EIPL truly confers a peritoneal recurrence–reducing benefit in these specific patient populations.

Recently, impact of EIPL for prevention of peritoneal dissemination from gastric cancer has been examined. Misawa and colleagues evaluated the effect of EIPL for gastric cancer with cT3-4 [11], and they concluded that EIPL did not improve survival for this population. However, this result may be because the most patients in their study had negative findings on CY. Furthermore, although cT3-4 status in pre-operative diagnosis was inclusion criteria in their study, 18.0% of cases were diagnoses as pT1-2 postoperatively. The selection of peritoneal metastasis high risk patients should be critical to evaluate the effect of EIPL, but as mentioned above, only with preoperative imaging, accurate tumor depth diagnosis is difficult. On the other hand, our group reported that intra-operative stamp cytology can identify high risk of peritoneal metastasis [10]. In this study, some of pT3 cases were diagnosed as stamp + . We previously reported that the distance from the tumor invasion front to the serosa (DIFS) is useful for predicting peritoneal metastasis [15]. A positive stamp cytology result theoretically indicates a shorter DIFS. Therefore, we consider stamp cytology to be a clinically useful technique for identifying patients with a small DIFS (or those who are essentially T4a), who are at particularly high risk for peritoneal metastasis.

In this research, in order to recruit patients at high risk of peritoneal dissemination, we determined to include cases who are diagnosed as CY1 and/ or stamp + . Indeed, postoperative recurrence with perineal metastasis was observed in five cases (38.5%) and it was the most common recurrence site in this study. Depth of invasion were deeper than T3 in all cases. Accordingly, inclusion criteria we set seems to be appropriate in terms of the risk of peritoneal metastasis. In the subgroup analysis, the 3-year RFS of CY0 and stamp+ cases were 66.7%. Although the analyzed number of patients was small, this data is similar to the 3-year RFS rate of 67.7% of stage III gastric cancer in JACCRO GC-07 trial with the S-1 plus docetaxel (DS) group [16]. Considering that only half of stamp+ cases underwent DS treatment as adjuvant, EIPL might improve survival outcome of patients who did not undergo DS treatment. However, we cannot definitely conclude the effect of EIPL, since we did not compare the results between EIPL and non-EIPL group. In addition, DS therapy was not balanced between groups and no adjustment or stratification was performed; therefore, it may have acted as a confounding factor. As a result, our speculation that EIPL might have compensated for the limited use of DS therapy should be interpreted with caution. On the other hand, what we can conclude is that CY0, Stamp+ cases showed significantly better peritoneal recurrence free survival than CY1 cases. This data implies that the impact of stamp+ should be different from CY1, and rather similar to pT4a. This is quite important finding, and stamp cytology could be useful in future clinical study, when researchers would like to select only pT4a cases during operation.

Intraperitoneal chemotherapy (IPC) may be alternative option for decreasing peritoneal recurrence for high risk patients, although the evidence about IPC has not been definite. Kuramoto et al. demonstrated that IPC combined with EIPL was superior to surgery plus EIPL or surgery alone in patients with positive lavage cytology. The Phoenix GC trial indicated that IPC was promising for patients with Stage IV, although its superiority was not statistically proven. According to these data, IPC could lead to possible benefit of IPC as biological irrigation, instead of EIPL. In addition, it remains debatable whether normal saline is the best solution for EIPL. Akasaka et al reported that lactate Ringer’s solution is the better solution for reducing the risk of peritoneal metastasis. This issue should be investigated in future [17].

Complications related to EIPL was confirmed in one case; hyponatremia. After the treatment only with replenishment of electrolytes, hyponatremia was improved promptly. The other complications related to EIPL didn’t develop. Although perioperative electrolyte changes were statistically significant, the actual amount of change is not so much. In addition, it is suggested that secretion of ADH due to surgical stress may cause hyponatremia [18]. EIPL was performed without serious complications; however, the single-arm design and absence of a control group limit our ability to conclude its safety or attribute survival outcomes to EIPL.

There are several limitations to this study. First, the number of enrolled patients was too small. The extremely small sample size resulted in highly unstable survival estimates, which undermines the reliability of the findings. This limitation should be taken into account when interpreting the results. We considered several reasons for this slow enrollment. As described above, the number of laparoscopic surgeries increased during the study period. Although EIPL can be performed during laparoscopic surgery, stamp examination is technically difficult in the laparoscopic setting. Considering the potential benefits of stamp cytology, it may be necessary to develop new devices, such as small but durable slide glasses that can be introduced into the abdominal cavity through laparoscopic ports. Second, the number of patients with gastric cancer itself has been decreasing, largely due to the eradication of Helicobacter pylori. Finally, the COVID-19 pandemic also contributed to the reduced number of surgical procedures. In addition, although this was a prospectively designed study, some post-hoc survival analyses were performed, and the statistical power to detect differences between groups was clearly insufficient. Furthermore, because the study was terminated pragmatically rather than according to predefined stopping rules or an interim analysis plan, some degree of bias may have been introduced. This limitation should be taken into account when interpreting the results.

Although definite conclusion could not be drawn due to the lack of the required number of cases, we were able to perform EIPL safely, and EIPL might be effective for selective patients. Stamp cytology should be effective to diagnose real T4a cases intraoperatively, and useful technique for selecting T4a cases in future clinical trials.