Long‐term treatment outcomes in gastric cancer with oligometastasis

Abstract Aim While surgery is essential for curative treatment of gastric cancer with oligometastasis, its target, timing, and possibility of combination with other treatments are unclear. We herein investigated the clinical course and long‐term outcomes of gastric cancer with oligometastasis in the real world setting to determine the optimal therapeutic strategy. Methods The present study retrospectively analyzed 992 patients who received any treatment for metastatic or recurrent gastric adenocarcinoma at Tokyo Metropolitan Komagome Hospital between 2007 and 2019. Oligometastasis was defined as any one of the following: liver metastases (HEP) <3; lung metastases (PUL) <3; unilateral adrenal gland metastasis (ADR); para‐aortic lymph node metastasis (PALN); or one, distant, lymph node metastasis, excluding the regional lymph nodes (LYM). Overall survival was compared by the characteristics and treatments for the oligometastasis, and univariate and multivariate analyses were used to identify the prognostic factors of overall survival. Results Ninety‐seven patients (9.8%) with the following metastasis sites were enrolled: HEP (n = 27), PUL (n = 2), ADR (n = 3), PALN (n = 55), and LYM (n = 10). The median survival time of the cohort was 22.8 months, and the five‐year overall survival rate was 28.4%. On multivariate analysis, chemotherapy for the initial treatment (hazard ratio [HR]: 0.438; p = 0.048), distal gastrectomy and/or metastasectomy (HR: 0.290; p = 0.001), and R0 resection (HR: 0.373; p = 0.005) were identified as independent, positive factors of overall survival. Conclusion The long‐term outcomes of gastric cancer in patients with oligometastasis may improve if treatment is begun with chemotherapy rather than surgery.

having a small number of metastases distant from the primary tumor.
Its curability has been much discussed owing to the challenges posed by its local and systemic aspects that depend on the biological characteristics of the primary organ affected. 2Oligometastasis in gastric cancer is relatively rare, occurring only occasionally in the lungs, liver, and non-regional lymph nodes.4][5][6][7] Based on the accumulation of such data, the 6th edition of the Japanese gastric cancer treatment guidelines, published in 2021, provide a weak recommendation for surgery as means of improving long-term outcomes in patients with oligometastasis, especially of the para-aortic lymph node at station #16 a2/b1 or a single liver metastasis. 8wever, such recommendations are subdivided into organspecific ones because of the different levels of invasiveness and difficulties with surgery related to the anatomical location of the tumor and therefore cannot be generalized to other cases until comprehensive, organ-independent evidence for the treatment of oligometastasis of gastric cancer is established.Moreover, although previous studies in Europe reported that a combination of local and systemic therapy had the potential to improve long-term survival in patients with gastric cancer with oligometastasis, 9,10 it remains unclear whether surgery or chemotherapy should be the initial treatment.2][13] The REGATTA trial, 13 an international, open-label, randomized, phase-3 trial, found that upfront gastrectomy conferred no survival benefit on patients with advanced gastric cancer with a non-curable, single-organ metastasis (e.g., liver, para-aortic lymph node, peritoneum); in fact, it worsened compliance with chemotherapy and long-term outcomes, especially in patients with an upper third tumor, most of whom normally undergo a total gastrectomy.These findings suggested that initiating treatment with chemotherapy may be recommended for Stage IV gastric cancer irrespective of the organ with the metastasis.The present study aimed to investigate the clinical course and long-term outcomes of gastric cancer with oligometastasis to determine the optimal therapeutic strategy.

| Patients
The patients were retrospectively chosen from an institutional cancer registry and surgical database containing 992 patients who received any form of treatment, such as surgery, chemotherapy, or radiation therapy, for metastatic or recurrent gastric adenocarcinoma at Tokyo Metropolitan Komagome Hospital between 2007 and 2019.

| Evaluations
The tumors were evaluated using a combination of imaging modalities, such as computed tomography (CT), ultrasonography, magnetic resonance imaging, and radio-isotope inspection.Diagnostic laparoscopy was performed if peritoneal dissemination was suspected.
Clinical tumor depth, nodal status, and pathological response to chemotherapy were assessed using the Japanese Classification of Gastric Carcinoma (JCGC) 15th edition of the Japanese Gastric Cancer Association (JGCA). 14

| Treatment
In the absence of an established treatment strategy for gastric cancer with oligometastasis, the cancer board of Tokyo Metropolitan Komagome Hospital, which consists of gastric surgeons, physicians, oncologists, and pathologists, determined through discussion whether treatment should be started with surgery or chemotherapy and whether chemotherapy should be continued or surgery performed instead based on tumor progression and the patient's physical condition.Patients with metastatic gastric cancer generally received a 5-fluorouracil (5-FU)-based regimen as the first-line treatment in accordance with the recommendations of the Japanese gastric cancer treatment guidelines of the JGCA. 8If a patient had previously received a 5-FU-based regimen or lacked tolerance for it, a taxane or irinotecan (CPT-11)-based regimen was administered.
Curative surgery was considered for patients who were expected to achieve R0 resection via gastrectomy and/or metastasectomy.If the patient received upfront chemotherapy, surgery was provided to (1) those who were scheduled to receive a few cycles of neoadjuvant chemotherapy and surgery; and (2) those who had a clinical response to chemotherapy (non-progressive disease, non-PD) on CT (performed every few months) and were intolerant of the adverse events associated with the chemotherapy.On the other hand, a palliative gastrectomy was provided to patients experiencing symptoms of the primary tumor, such as difficulty with oral food intake or hemorrhaging from tumor, even if the chemotherapy had resulted in PD.The standard lymphadenectomy for advanced gastric cancer was D2.D1 and D3 lymphadenectomy were performed for palliative surgery and para-aortic lymph node metastasis, respectively.Patients with metachronous oligometastasis with a previous gastrectomy underwent a metastasectomy alone.Postoperative surgical complications were defined as Clavien-Dindo classification grade 2 or higher occurring within 30 postoperative days. 15e patients were followed up for at least 5 years with physical examinations, blood tests, and chest-abdominal CT to assess for tumor progression or recurrence.

| Statistical analyses and ethics
Overall survival (OS) was defined as the duration from the initiation of treatment to the patient's death from any cause or the date of the last follow-up.The Kaplan-Meier method was used to calculate the 5-year OS and median survival time (MST).The log rank test was performed to compare survival rates by the characteristics of, and treatments for, oligometastasis.

| Patient population
In total, 992 patients received treatment for gastric adenocarcinoma with a distant metastasis at the study center between 2007 and 2019.Figure 1 shows a flow diagram of the patients.Of the total cohort, 764 and 228 patients had synchronous and metachronous metastatic gastric cancer, respectively.The present study enrolled 97 patients (9.8%) after excluding 610 patients with a diagnosis of P1 and/or CY1, 179 patients with multiple organ metastases, 71 patients with liver metastases >4, four patients with lung metastases >4, 27 patients with a bone metastasis, and four patients with a brain metastasis.Table 1 shows the clinical characteristics of the patients.As Table 2 shows, the included patients had the following F I G U R E 1 Flow diagram of the patients.ADR, unilateral adrenal gland metastasis; CY1, peritoneal lavage cytology-positive; HEP, liver metastases ≤3; LYM, single station of distant lymph node metastasis excluding the regional lymph nodes; P1, peritoneal dissemination; PALN, para-aortic lymph node metastasis; PUL, lung metastases ≤3.

| Treatment
Table 3 shows the patients' treatment details.The majority of the patients were administered chemotherapy as best available regimen, singly or in combination with surgery.Fifty patients received chemotherapy alone, and 47 received surgery with or without chemotherapy.Of the latter, 21, 17, and nine patients underwent a total gastrectomy, distal gastrectomy, and metastasectomy alone, respectively.Postoperative complications more serious than Clavien-Dindo grade 2 occurred in 11 patients (23.4%).The incidence rate of postoperative complications was 23.5%, 33.3%, and 0% in patients who underwent a distal gastrectomy, total gastrectomy, and metastasectomy alone, respectively (p = 0.142).R0 resection was done in 42 patients, or 43.3% of the total cohort.Of five patients who underwent R1/2 resection, one was found to have received a

| Comparison of patient background and short-term outcomes by the initial treatment
Table 4 shows a comparison of patient background and short-term outcomes in patients with chemotherapy or surgery as the initial treatment.The clinical nodal status was more progressive in patients with upfront chemotherapy than upfront surgery.The predominant site of oligometastasis was PALN followed by HEP in the upfront chemotherapy group while it was HEP followed by PALN in the upfront surgery group.Upfront chemotherapy did not increase the postoperative complication rate (chemotherapy first: 14.3% vs. surgery first: 30.8%; p = 0.164).

| Survival analyses
Figure 2 shows the Kaplan-Meier curve for OS in the total cohort. The

TA B L E 3 (Continued)
The present study aimed to clarify the clinical course and long-term outcomes in patients with gastric cancer with oligometastasis to determine the optimal therapeutic strategy.

TA B L E 4
Comparison of patient background between initial treatment with chemotherapy and surgery.
7][18] In particular, the patients in the present study with oligometastasis who received chemotherapy followed by surgery, underwent a distal gastrectomy and/or metastasectomy, and achieved R0 resection showed remarkable, long-term survival, suggesting that a different therapeutic strategy is needed for patients with oligometastasis deriving from extensively metastatic or recurrent gastric cancer.
Several studies have investigated oligometastatic gastric can- Although AIO-FLOT was a small-scale study and not an RCT, its findings suggested that even metastatic gastric cancer is curable by surgery after chemotherapy if the extent of the metastasis is limited.A large-scale, nationwide, population-based study in the Netherlands reported the outcomes of local and/or systemic therapy in 594 patients with esophagogastric cancer with oligometastasis. 9Of the total cohort, 83, 22, and 489 patients underwent local treatment alone, both local and systemic therapy, and systemic therapy alone, respectively.The study concluded that local treatment alone (MST: 16.0 months; HR: 0.52; 95% CI: 0.31-0.90)or combination with systemic therapy (MST: 22.7 months; HR: 0.42; 95% CI: 0.22-0.82)improved OS to a greater degree than systemic therapy alone (MST: 8.5 months) in patients with esophagogastric cancer with oligometastasis.While these results had similar tendencies to ours in terms of the benefits of surgery, the MST was lower on the whole than in our study, a discrepancy which may derive from differences in the study population.In particular, it should be noted that the Dutch study included 101 patients (17.0%) with squamous cell carcinoma, with the liver as the chief site of the metastasis (32.3%), followed by extra-regional lymph nodes (20.9%).In the present study, PALN (56.7%) was the chief site of the metastasis, followed by the liver (27.8%).
Neo-adjuvant chemotherapy was individually developed for gastric cancer with para-aortic lymph node (PAN) metastasis in Japan because PAN was originally considered to be a regional lymph node.The JCOG0405, a prospective, multicentric, phase 2 study, 3 demonstrated a survival benefit of neoadjuvant chemotherapy with S-1 and cisplatin followed by a gastrectomy and D3 lymphadenectomy in patients with gastric cancer with a bulky, lymph node metastasis and/or para-aortic lymph node metastasis.This finding was recently highlighted as representative evidence of the efficacy of surgery after chemotherapy for oligometastatic gastric cancer and was incorporated into the recommendations of the latest Japanese gastric cancer treatment guidelines. 8This may be one reason accounting for the better OS observed on multivariate analysis in the patients initiating treatment with chemotherapy; the predominant metastatic site was PALN in the latter while it was HEP in the upfront surgery group.| 67 Recently, multicentric studies were conducted in Europe to establish a therapeutic strategy for gastric cancer with oligometastasis.Moreover, the multicentric randomized controlled trial, AIO-FLOT 5 (the RENAISSANCE trial), which aims to confirm the superiority of peri-operative chemotherapy with FLOT and surgery to chemotherapy with FLOT alone in patients with oligometastatic adenocarcinoma of the stomach or esophagogastric junction, is currently underway. 19In addition, the OligoMetastatic Esophagogastric Cancer (OMEC) project, which consists of five consecutive studies aimed at developing a multidisciplinary, European consensus statement on the definition, diagnosis, and treatment of oligometastatic esophagogastric cancer, has been launched. 20The project includes a plan for a prospective, international, multicentric trial as its final stage.The results of these studies will have important implications for this field.
Our study revealed that OS in patients with gastric cancer with oligometastasis was associated with the initiation of treatment with chemotherapy, R0 resection, and distal gastrectomy and/or metastasectomy.The CONVO-GC-1, 21 an international, retrospective cohort study of conversion surgery for Stage IV gastric cancer, also reported that patients with metastatic gastric cancer who underwent R0 resection after systemic chemotherapy were able to patients and the severity of weight loss were generally worse in patients who received a total gastrectomy than another type of gastrectomy. 22,23Moreover, the incidence of postoperative complications may affect the differences in long-term outcomes among the various types of surgery.Previous studies have reported that postoperative complications impact survival in patients with gastric cancer, [24][25][26] and the findings of the present study also demonstrated poor OS in patients with postoperative complications.
Univariate and multivariate Cox proportional hazards models were used to analyze the hazard ratio (HR) for OS.Variables with p < 0.1 on univariate analysis were included in multivariate analysis.For all the tests, two-sided p < 0.05 was considered to indicate statistical significance.Statistical analyses were performed using SPSS Statistics, ver. 25 (IBM, Chicago, IL, USA).The present study was approved by the Institutional Review Board of Tokyo Metropolitan Komagome Hospital.
cer.A German research team conducted the AIO-FLOT3 (Arbeitsgemeinschaft Internistische Onkologie-Fluorouracil, Leucovorin, Oxaliplatin and Docetaxel) trial, a prospective, non-randomized, phase 2 trial enrolling 252 patients with resectable or metastatic gastric or gastroesophageal junction cancer. 10They divided the eligible patients into a resectable arm (n = 65), limited metastatic arm (n = 60), and extensive metastatic arm (n = 127).Patients received perioperative chemotherapy with a median of eight (1-15) cycles of FLOT and underwent gastrectomy if restaging showed a chance of R0 resection.Survival analysis of the patients in the limited metastatic arm showed favorable survival, with an MST of 22.9 months in comparison with the patients in the extensive metastatic arm, who had an MST of 10.7 months (HR: 0.37; 95% CI: 0.25-0.55).
TA B L E 5