Gastric cancer with positive peritoneal cytology: survival benefit after induction chemotherapy and conversion to negative peritoneal cytology

Gastric cancer is a leading cause of cancer death worldwide, with an estimated 5-year survival rate of 26% in Europe [1]. According to the GLOBOCAN 2018 estimates of cancer incidence, gastric cancer is the 5th most common neoplasm and the 3rd most deadly cancer worldwide, with 783,000 deaths in 2018 [2]. Due to the lack of national screening programs in Western countries, many patients are diagnosed with advanced stage disease and up to 20% present with peritoneal metastasis upon open or laparoscopic exploration [3, 4]. Peritoneal metastases have a dismal prognosis, and most patients receive palliative chemotherapy only [5]. In contrast, the relevance of microscopic disease, detectable only as CY+ remains unclear [6]. There is an ongoing debate if these patients should undergo gastrectomy [7, 8], and the available national guidelines are not consistent regarding treatment recommendations. For example, the National Comprehensive Cancer Network (NCCN) gastric cancer guidelines define CY+ as metastatic (M1) disease and recommend a palliative treatment [9], similar to the Japanese Gastric Cancer Association, which classifies positive peritoneal cytology as M1 and does not recommend surgery [10, 11]. In contrast, European guidelines, e.g., the German S3, or the ESMO guidelines are less detailed and remain unclear, probably due to the lack of data in Western patients [12, 13]. Another open question is how to deal with patients who had CY+ and converted to CY− after neoadjuvant treatment. Reports from Japan show an improved survival of this subgroup and therefore recommend a more radical treatment after conversion to CY− [14, 15]. The literature is scarce for Western patients [16], while novel treatment options for patients with locally advanced disease are available, e.g., the radical cytoreductive surgery (CRS) with hyperthermic intraperitoneal chemotherapy (HIPEC) [17] or the pressurized intraperitoneal aerosol chemotherapy (PIPAC) [18]. In this study, we analyzed the overall survival of gastric cancer with a focus on patients with CY+.

Peritoneal metastasis is one of the most common forms of metastasis in patients with gastric cancer reported in the literature between 10 and 20% [3, 4, 19]. Therefore, an accurate diagnosis remains crucial for prognostic evaluations and therapeutic decisions. Staging laparoscopy and peritoneal cytology are recommended tools that allow to detect occult peritoneal spreading, especially in patients in an early stage of the disease, presenting in almost 11% of the cases with a positive peritoneal cytology [20, 21]. In this study, we could confirm the prognostic impact of peritoneal disease and we could demonstrate that patients who converted to CY− after systemic treatment have a better cumulative survival compared with patients demonstrating persisting CY+. However, the high incidence of peritoneal recurrence despite response to neoadjuvant chemotherapy remains a problem and underlines the need for better local control and novel strategies.

The role of surgery in patients with conversion after neoadjuvant systemic treatment is controversially reported in the current literature. A systematic review revealed only three studies with neoadjuvant systemic treatment in patients with positive peritoneal cytology [6], and the time point of peritoneal cytology and inclusion criteria varied considerably among these three studies. For example, one study reported 39 patients with positive peritoneal cytology without peritoneal metastasis with a median survival of 12.8 months [22]. In this study, peritoneal cytology was performed only prior to neoadjuvant systemic treatment and not thereafter. Therefore, a difference between patients with or without conversion after neoadjuvant systemic treatment was not analyzed. Another group reported 48 patients with repeated cytology, and among them, 27 converted to negative cytology, while 21 did not [8]. They observed no benefit of surgery after conversion to negative lavage cytology. In two other studies, the benefit of surgery in patients after conversion to negative lavage cytology was clearly demonstrated [14, 16]. These two studies highlight the potential risk of developing CY+ during neoadjuvant systemic treatment in patients who were initially CY−. Unfortunately, there are no known risk factors predicting poor response to neoadjuvant treatment. Such patients would profit from an early response evaluation and discontinuation of neoadjuvant systemic treatment [23]. Some authors suggest that non-responders would profit from early surgery. However, in our study, we observed that patients without conversion to CY− after systemic treatment had a very bad prognosis, on average worse than patients with macroscopic peritoneal cancer. Despite the relatively small number of patients in our series and the absence of statistical difference between the two groups, this finding likely reflects the probably unfavorable and rapid tumor biology in this situation and should call for caution.

Regarding neoadjuvant treatment, in the past decades, many advances in perioperative chemotherapy for gastric cancer have been made. For example, two clinical trials which represented landmarks regarding the topic were the MAGIC trial and the French FNCLCC/FFCD 9703 study [24, 25]. Despite a great effort, the outcome for gastric cancer remained insufficient. Furthermore, several studies reported improved survival rates with the addition of taxane-based regimens in patients with advanced gastric cancer as well as with peritoneal metastasis. For example, Wang et al. described an overall response rate of 50% for taxane-based combination chemotherapy in patients with advanced gastric cancer [26]. Moreover, Fujiwara et al. described a highly effective and well-tolerated combination of intraperitoneal and systemic chemotherapy based on docetaxel, showing that 59% of the patients displayed major response and 56% showed negative results on peritoneal cytology and no macroscopic peritoneal metastasis after treatment [27]. Another interesting work by Fushida et al. shows an overall response rate of 22% and a conversion to CY− in 81% of the patients after administration of intraperitoneal docetaxel plus S-1 for gastric cancer with peritoneal carcinomatosis [28]. In one randomized phase II trial, docetaxel in combination with 5-fluorouracil demonstrated a better overall response rate compared to ECF in patients with advanced cancer (37.8% vs. 35%) [29]. In another phase II randomized trial, the addition of cisplatin to docetaxel and 5-fluorouracil showed a favoring effect with improved response rates of docetaxel in combination with cisplatin and 5-fluorouracil over ECF or docetaxel with 5-fluorouracil, although the study was not powered sufficiently to show superiority (36.6% vs. 25% vs. 18.5%) [30]. Recently, Al-Batran and his group set a new cornerstone with the FLOT4 trial, demonstrating the superiority of FLOT over ECF/X as perioperative chemotherapy in locally advanced gastric cancer (complete and subtotal regression rate of 37% vs. 23% p = 0.02) [31]. In the setting of gastric cancer with CY+, there is to our knowledge a lack of prospective randomized data about the superiority of one chemotherapeutic regimen compared to another so far. In our series, we observed a conversion rate from CY+ to CY− with FLOT of 82% (9/11) and no patient converted under ECF (0/3). Based on the trend towards a higher conversion rate in our series and on the available data about advanced gastric cancer, we conclude that patients with gastric cancer with CY+ should benefit from the FLOT regimen in the neoadjuvant setting. We are aware of our limited numbers of patients with CY+, which is itself a rare condition and we think that further multicentric studies with larger cohorts of patients should be performed to achieve stronger results.

Another interesting observation is the recurrence pattern, which obviously depends on the peritoneal cytology status. It is well known that advanced stage gastric cancer is an independent prognostic factor of peritoneal recurrence, which is reported in the current literature ranging from 8 to 62% [14, 32‐34]. In contrast, early-stage gastric cancer shows more often hematogenous recurrence patterns [35], and similarly, patients with CY− seem to have less peritoneal recurrence [14, 34]. In our series, recurrence rates in patients without peritoneal disease are comparable with the current literature [14]. We would like to highlight that patients with CY+ at any time had a higher risk to develop peritoneal recurrence, which indicates the need for adjuvant or additive therapy. Even for the patients who did convert to CY−, the recurrence rates are higher compared to the negative group. Although the mechanisms governing the occurrence of peritoneal metastasis and peritoneal recurrence remain unclear, we can postulate that based on the “seed and soil theory” the higher rate of peritoneal recurrence in the converted group is likely due to tumor cells that seed in the peritoneum and migrate through the basement membrane to the subperitoneal space, surviving therefore after neoadjuvant treatment and remaining undetected during peritoneal lavage [20]. Indeed, there is a meta-analysis of 20 randomized controlled trials which demonstrates the benefit of an adjuvant intraperitoneal chemotherapy after radical surgery in advanced gastric cancer, although a separate analysis investigating the role of adjuvant intraperitoneal chemotherapy in patients with CY+ and no visible peritoneal metastasis was not performed [36]. Recently, a phase II trial using a more aggressive approach combining CRS, gastrectomy, and HIPEC demonstrated encouraging results in patients with minimal peritoneal metastasis from gastric cancer or positive lavage cytology [37]. Some authors recommend even prophylactic HIPEC at the time of resection for advanced stage gastric cancer without peritoneal disease, highlighting the importance of reducing the risk of occult peritoneal dissemination during surgery in order to prevent future recurrence [38]. Currently, we may speculate that additive locoregional treatment in patients with CY+ might be considered in all patients undergoing resection, irrespective of their conversion status.

Another important issue that should be considered is the predicting factors that meant to achieve the best outcome possible. Our findings suggest that conversion of CY+ to CY− could be a prognostic factor for a better outcome. In the current literature, we can find some other important predictors that should be considered in the therapeutic decision-making. For example, peritoneal carcinomatosis itself appears to be a major driver of complications and dismal outcome for surgical gastrojejunostomy and should call for caution in the surgical strategy [39]. Moreover, an interesting study by Chen including 518 patients highlights the importance of the body mass index as another predicting factor for patients who have gastric cancer with peritoneal dissemination, especially in those who received palliative chemotherapy [40]. Other studies showed for example that lymphopenia and thrombocytosis were predictive factors for peritoneal seeding [41, 42]. Furthermore, recent studies try to identify new predicting factors such as the systemic immune-inflammation index aiming for a better oncological outcome and quality of life of our patients [43].

We would like to acknowledge the limitations of this study, which is a monocentric retrospective study. Furthermore, our analysis is based on a small number of patients with CY+ and without macroscopic peritoneal disease.

In conclusion, the overall survival of patients who convert to CY− after neoadjuvant chemotherapy may be superior compared to non-converted patients. Surgical treatment in well-selected patients should therefore be considered. However, peritoneal recurrence remains frequent despite conversion, urging for a better local control.