Clinical Features, Prognosis, and Prognostic Factors of Remnant Gastric Cancer Patients Who Received Potentially Curative Gastrectomy.

Introduction
Gastric cancer (GC) is the fifth most prevalent cancer and fourth leading cause of cancer-related deaths worldwide (1,2). The standard treatment for resectable GC is gastrectomy with or without perioperative adjuvant treatment (3,4). Among GC, remnant gastric cancer (RGC) is relatively rare occurring in 1-3% of GC (5-7). RGC is defined as GC develops at the remnant stomach after gastrectomy for benign or malignant disease.
Generally, RGC is technically difficult for surgery due to adhesion of remnant stomach to around organs (8,9). Moreover, RGC surgery increased postoperative surgical complications and prolonged postoperative stays due to difficulty of surgery. However, clinical features and clinical course of RCG are not fully evaluated. If the physicians understand the clinical feature of RGC, the physicians might select and provide more optimal surgery and perioperative care.
In addition, for primary GC, there were several prognostic factors detected and reported in previous studies. However, there are limited studies that evaluated the prognostic factors of RGC (10-12). RGC has already received lymphadenectomy, thus prognostic factors might be affected by previous surgery and prognostic factors which reported might be changed (13).
Considering these, to optimize the treatment of RGC, it is necessary to clarify the clinical features and prognostic factors of RGC. Therefore, we aimed to clarify the clinical features and prognostic factors of RGC.

Patients and Methods
Patients. The patients were selected from the consecutive database of the Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital, Department of Gastric Surgery, Tokyo, Japan, according to the following criteria: (i) histologically-proven gastric adenocarcinoma (ii) patients who underwent curative gastrectomy for gastric cancer as a primary treatment and archived R0 or R1 resection between 2005 and 2025. We excluded the patients who archived R2 resection.
Surgical procedures and adjuvant chemotherapy. In principle, all patients received distal, proximal, or total gastrectomy with nodal dissection for gastric cancer. D1 or a D1+ lymphadenectomy is indicated for cT1N0 tumors, and D2 is applied for cN+ or cT2-T4 tumors, regardless of the approach. Spleen-preserving D2 total gastrectomy was permitted in this study (14). Pathological stage II disease was treated with S-1 monotherapy, while pathological stage III disease was treated with S-1 plus docetaxel or capecitabine plus oxaliplatin (15-17).
Patient follow-up. At a minimum, patients receive physical examinations and hematological tests every three-twelve months for five years. Tumor marker levels (carcinoembryonic antigen and CA19-9) were checked every three-twelve month for five years. Computed tomography (CT) was performed every 6-12 months for 5 years. Endoscopy was performed 1, 3, and 5 years after surgery.
Evaluation and statistical analysis. The progression of tumors was evaluated by the 15th edition of the Japanese Gastric Cancer Classification. Differences between the resection margin positive group and resection margin negative group was analyzed by a χ2 test and Mann-Whitney U test. Univariate and multivariate logistic regression analyses were performed to identify risk factors for resection margin positive. In the multivariate analysis, we fitted linear regression models. To select a model, we used backward elimination. Overall survival (OS) curves were calculated by the Kaplan-Meier method. Univariate and multivariate analyses of survival were conducted using a Cox proportional hazards model. Statistical significance was defined as p<0.05. All statistical analyses were conducted using SPSS (v27.0 J Win; SPSS, Chicago, IL, USA). This study was approved by the IRB of the Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital.

Results
Patient characteristics. A total of 2,013 patients were eligible for the present study. Among 2,013 patients, 45 patients were RGC. In the present study, 45 patients were categorized as the RGC group, and 1968 patients were non-RGC group. Table I shows preoperative patients background between RGC group and non-RGC group. When comparing the patients’ background between RGC group and non-RGC group, there were significant differences observed in median age (72 years vs. 69 years, p=0.042), preoperative body mass index (BMI) (21.3 vs. 22.5, p=0.004), and clinical lymph node metastasis (11.1% vs. 29.4%, p=0.007). RGC patients were older and lower BMI than non-RGC patients.
Clinical course and survival between remnant gastric cancer and non-remnant gastric cancer. Table II shows the surgical and pathological findings between RGC group and non-RGC group. In the surgical findings, the conventional approach is more frequent in RGC group than non-RGC group (73.3% vs. 49.0%, p<0.001). In addition, RGC group had longer operation time (280 min vs. 238 min, p<0.001) and greater intraoperative bleeding (265 ml vs. 130ml, p=0.018). In the pathological findings, pathological lymph node metastasis was significantly fewer in RGC group than non-RGC group (22.2% vs. 40.1%, p=0.015), while number of harvest lymph node was significantly fewer in RGC group than non-RGC group (20 vs. 40, p<0.001). Pathological T factors, lymphatic invasion, and vascular invasion were almost similar between two groups. Medina hospital stay was significantly longer in RCG than non-RGC (15 days vs. 12 days, p<0.001).
Survival and prognostic analysis of remnant gastric cancer. Figure 1 shows overall survival (OS) curves between the RGC group and non-RGC group. Three- and five- years OS was 83.4% and 78.5% in RGC group, 85.3% and 81.0% in non-RGC group, respectively. There are no significant differences (p=0.529). In the prognostic analysis of RGC, pathological N factor and lymphatic invasion were significant prognostic factors in univariate analysis. In the multivariate analysis, lymphatic invasion is an independent prognostic factor for OS [hazard ratio (HR)=10.245, 95% confidence interval (CI)=1.150-91.281, p=0.037] (Table III). Figure 2 shows OS curves between the RGC with lymphatic invasion group and RGC without lymphatic invasion group. The three- and five- years OS was 67.7% and 67.7% in RGC with lymphatic invasion group, 100% and 92.9% in RGC without lymphatic invasion group, respectively.

Discussion
The aim of the present study was to clarify the clinical features of RGC and identify the prognostic factors of RGC. A major finding was that RGC patients were elderly and had surgical difficulties, while OS was almost similar between RGC and non-RGC patients. In addition, lymphatic invasion was an independent prognostic factor for OS in RGC patients. Thus, although the prognosis for RGCs themselves is not poor, RGC patients with lymphatic invasion have a poor prognosis and special attention is needed for these patients.
In the present study, RGC patients were elderly and had a lower BMI than non-RGC patients. In addition, RGC patients had longer operation time, greater intraoperative blood loss, longer postoperative hospital stays, and fewer number of harvested lymph nodes than non-RGC patients. Similar results were observed in the previous reports (18,19). These findings might be the result of the previous effect of gastrectomy. Interestingly, there are some discrepancies of lymph node metastasis evaluation between RGC and non-RGC. In the present study, the incidence of preoperative lymph node metastasis was higher in RGC group than non-RGC group. In contrast, the incidence of pathological lymph node metastasis was higher in the no-RGC group than the RGC group. This discrepancy might be difficulty of preoperative lymph node evaluation even imaging studies improvement and lower number of harvested lymph node in RGC than non-RGC. Therefore, preoperative lymph node evaluation might overestimate, and pathological lymph node might underestimate in RGC patients. It is widely accepted that lymph nodes metastasis is one of the strongest prognostic factors of GC. Thus, careful attention is needed to perioperative lymph node evaluation in RGC patients.
We found that the survival of RGC patients was almost similar to non-RGC patients. Previously, the survival of RGC patients was poorer than non-RGC patients due to advanced tumor stage with late diagnosis, poor nutritional status, and increased postoperative surgical complications with surgical difficulties. Kung et al. compared OS and recurrence free survival (RFS) between RGC (n=105) and non-RGC (n=2,622) between 1988 and 2012 (20). Five-years OS and RFS were 51.2% and 46.2% in RGC, while 54.5% and 54.1% in non-RGC. There were significant differences in both OS and RFS (p=0.035 and p=0.043, respectively). On the other hand, recent studies demonstrated similar results to our study. Galata et al compared OS between RGC (n=95) and primary gastric cancer (PGC) (n=1,345) between 972 and 2,014 patients using propensity score-matched analysis (21). They reported that estimated 5-years OS was 36.4% in RGC and 38.6% in PGC, respectively. There were no significant differences between RGC and PGC (p=0.772 and p=0.369). Moreover, Yan et al. compared the OS and disease-free survival (DFS) between RGC (n=76) and PRG (n=32,763) between 1988 and 2020 using propensity score-matched analysis (22). Five-years OS and ten-year DFS were 60.0% and 56.7% in RGC, while53.3% and 48.3% in non-RGC. There were no significant differences between RGC and PGC (p=0.65 and p=0.28). Changes of survival of RGC might be due to early diagnosis with improvement of imaging studies, introduction of perioperative nutritional management, and development of minimal invasive surgery. Actually, Takahasi et al. compared OS of RGC according to historical periods as prior periods (1970-1990) and latter periods (1991-2012), there was significant differences between prior periods and latter periods (23). The OS of RGC was significantly poor in prior periods than latter periods. These results support our speculations.
Lymphatic invasion is one of the independent prognostic factors for RGC patients in the present study (HR=10.245, 95% CI=1.150-91.281, p=0.037). A similar result was observed in the previous study. Kano et al. evaluated the prognostic factor of RGC in 95 RGC patients (13). They reported that lymphatic invasion was one of the independent prognostic factors. The possible reason why lymphatic invasion was an independent prognostic factor for RGC as follows. It is well known that lymph node plays an important role in preventing cancer cells from invading the lymph vessels and spreading throughout the body in various malignancies. However, RGC patients received lymphadenectomy in previous gastrectomy. Thus, RGC patients do not have the lymph node which prevents invading cancer cells. Therefore, cancer cells easily spread over the whole body via lymphatic vessel. On the other hand, previous studies demonstrated that lymph node metastasis, tumor invasion status, and postoperative surgical complications were the independent prognostic factors for RGC. Among various prognostic factors, metastasis lymph node to harvest lymph node ratio (LNR) is interesting and promising prognostic factors for locally advanced RGC. Although our study did not show the survival benefit of LNR due to including early stage RGC, LNR might improve the underestimation of lymph node metastasis. We will focus on this issue in future study.
Study limitations. First, the present study is retrospective single cohort study. In addition, our Institution is a specialized cancer center. Therefore, there might be the patient’s selection bias in the present study. Second, there might be time bias. Our study patients included between 2005 and 2025. During these periods, treatment strategies were changed. In the resectable GC setting, several adjuvant treatments were introduced as results of pivotal phase III studies (15-17). Moreover, in the advanced/unresectable setting, immune-check point inhibitors were also introduced in the treatment strategies. Thus, the clinical impacts of RGC might be changed by these factors. Third, the clinical impacts of splenectomy for RGC are unclear in the present study. At the present, tumor locates in greater curvature and tumor invasion is T2 or more need splenectomy. Recently, some researchers focused on the splenectomy for locally advanced RGC. In the present study, 8 RGC patients received splenectomy, while survival between the RGC patients with splenectomy and without splenectomy were almost similar (data not shown). Considering this, our study results need to be validated in another cohort. However, our study results included more than 2,000 GC cases and the incidence of RGC rate was 2.2% which are similar to other large cohort series (5-7). Thus, we believe that our study results have some clinical impacts of RGC treatment strategy.
In conclusion, although there have existed surgical, perioperative diagnosis, and tumor evaluation difficulties in RGC treatment, the survival of RGC patients was almost similar to that of non-RGC patients. On the other hand, the RGC patients with lymphatic invasion were poorer than those without lymphatic invasion. Thus, an optimal treatment strategy needs to be established for RGC according to risk factors.

Conflicts of Interest
The Authors declare no conflicts of interest in association with the present study.

Authors’ Contributions
TA and HC made substantial contributions to the conception and design. TA, HC, KH and HS made substantial contributions to the acquisition of data, or the analysis and interpretation of data. TA and HC were involved in drafting the manuscript or revising it critically for important intellectual content. All authors gave final approval of the version to be published. Each author participated sufficiently in the work to take public responsibility for appropriate portions of the content; and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All authors read and approved of the final manuscript.

Acknowledgements
The Authors express their sincere gratitude to Ms. Emiko Saito for her excellent data management.

Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.