1 Introduction

Gastric cancer (GC) is currently the fifth most common cancer and the third leading cause of cancer deaths worldwide [1]. Significant geographic and ethnic variations in incidence and mortality reported across the globe suggest extensive differences in environmental and genetic risk factors as well as screening strategies, treatment, and prevention strategies [2,3,4]. Despite significant advancement in early detection and comprehensive treatments in the past three decades, the outcome of gastric cancer patients remains less satisfactory. The 3-year disease free survival for resectable gastric cancer is about 70% in Asian population and 60% in Caucasian [5, 6].

This article aims to provide a systemic review of the current progress and the remaining challenges of GC comprehensive treatment regarding surgical intervention, systemic chemotherapy, and preoperative comprehensive intervention.

2 Management of early gastric cancer

Early gastric cancer (EGC), defined as cancer confined to the mucosa or submucosa, is often difficult to be identified due to lack of specific clinical manifestations and its unspecific endoscopic appearance of flat shapes without ulceration [7]. A meta-analysis reported a 9.4% rate of missed GC diagnosis during upper gastrointestinal endoscopy [8]. Thus, quality improvement in endoscopy and application of highly sensitive and specific biomarkers are warranted for early detection. High-definition endoscopy with image enhancement is reported to outperform white light endoscopy alone for its superior performance in observing the micromorphology and the microvascular structure of mucosal lesions in EGC detection [9]. Meanwhile, a number of tumor markers were frequently used for early detection of GC, including carcinoembryonic antigen, carbohydrate antigens (CA) 19–9, CA72–4, CA125, pepsinogen, and α-fetoprotein [10]. Besides, multiple genetic alterations, including chromosome instability, single nucleotide mutations, as well as epigenetic alterations such as abnormal DNA methylation, microRNAs and long non-coding RNAs dysregulation are involved in GC initiation and progression [11]. However, no single one of these biomarkers is sufficiently sensitive and specific for GC early detection till now. With the rapid development of artificial intelligence in image recognition using endoscopic images and machine learning techniques integrating multi-omics datasets [12, 13], further studies on AI-assisted decision-making for EGC patient management will provide novel solutions.

3 Surgical management of gastric cancer

Surgical resection plays a pivotal role in the treatment for patient with GC. Removal of the tumor accompanied by appropriate lymphadenectomy has been shown to improve postoperative patient survival time. Surgical approaches mainly depend on disease stage, surgeon’s preference and experience. Progress in resection options is discussed below.

3.1 Wider application of endoscopic resection

Endoscopic resection (ER) of early-stage malignant and pre-malignant gastrointestinal lesions was initially performed as endoscopic mucosal resection (EMR) by Deyhle et al. in 1973 [14]. However, EMR was unable to completely resect larger-sized lesions at one time, potentially limiting appropriate histopathological evaluation to meet oncological principle. Gotoda et al. first reported endoscopic submucosal dissection (ESD) in 1999 [15]. ESD increased pathologically complete resection rates and reduced local recurrence compared to EMR. The conventional absolute indications for EMR/ESD are less than 2 cm, intramucosal, well-differentiated adenocarcinomas without ulceration [16]. Japanese researchers recently conducted a series of clinical trials to evaluate the efficacy and safety of expanded indications for ESD. The studies suggest ESD as an acceptable treatment for > 2 cm ulcer-negative differentiated-type GC, ≤3 cm ulcer-positive differentiated-type GC, ≤3 cm minimal submucosal invasive (≤500 μm from the muscularis mucosa) differentiated-type GC, and ≤ 2 cm ulcer-negative undifferentiated-type GC [17, 18].

Researchers also performed many clinical studies comparing ESD with surgery for EGC [19]. A meta-analysis of 18 retrospective studies showed a lower risk of procedure-related death (OR = 0.21, 95%CI: 0.07–0.68), a lower risk of overall complication (OR = 0.47, 95%CI: 0.34–0.63), as well as lower costs and better quality of life for ESD compared with surgical resection. Meanwhile, ESD showed a lower rate of en bloc resection (OR = 0.07, 95%CI: 0.03–0.21), histologically complete resection (OR = 0.07, 95%CI: 0.03–0.14) and a higher rate of local recurrence (OR = 5.42, 95%CI: 2.91–10.11). However, overall survival (OS) and disease-specific survival (DSS) between ESD and surgery were similar [19].The results support ESD as a safe procedure of less expensive, and faster recovery than surgery for EGC.

Accordingly, ESD provides an alternative option for patients with EGC. Nevertheless, due to the higher local recurrence risk and remaining challenges in managing postoperative bleeding and selecting patients with high risk of metastasis for additional surgery, tailored indications and surveillance strategies for ESD are needed.

3.2 Minimally invasive surgery

3.2.1 Laparoscopic gastrectomy

Since laparoscopic gastrectomy (LG) with lymphadenectomy was first reported in 1994, this minimally invasive surgical approach has steadily expand. The feasibility and safety of LG for EGC were evaluated and established by prospective clinical trials [19,20,21,22]. The effectiveness of LG regarding long-term oncological outcomes has also been confirmed in Japanese (JCOG0912) [20] and Korean (KLASS-01) [23] patients with stage I GC. The Chinese Laparoscopic Gastrointestinal Surgery Study (CLASS) Group CLASS02 study compared the safety of laparoscopic total gastrectomy (which is relatively complicated among LGs) vs open total gastrectomy for patients with stage I GC. The result showed comparable safety for laparoscopic total gastrectomy by experienced surgeons. LG for EGC is now accepted worldwide as alternative treatment for its safety, faster patient recovery, less blood loss, and fewer wound-related complications than open surgery. LG for advanced-stage GC remains controversial, with concerning shortcomings including difficulty in dealing with large tumors or bulky metastatic neoplastic lesions, the risk of intra-operative tumor spillage in serosa-positive GC, and a higher recurrence rate compared with open approaches [24]. Many large-scale randomized controlled trials from East Asia have reported acceptable surgical outcomes [25,26,27], however, long-term survival outcomes are still pending.

One limitation of current studies is a lack of cases with proximal advanced GC, which is predominant in Western countries. A study using data from a Japanese nationwide web-based data entry system for surgical procedures reported a significantly higher incidence of leakage for laparoscopy-assisted total gastrectomy comparing with the open approach (5.7% vs. 3.6%, p = 0.02) [28]. However, the recently reported CLASS-04 study, designed as a single-arm prospective multicenter clinical trial from China, showed that laparoscopic spleen-preserving total gastrectomy with D2 lymphadenectomy for advanced upper third GC had a similar incidence of postoperative complications compared to open distal gastrectomy. Thus, the applicability of LG for advanced GC still needs to be investigated with more high-quality evidence.

3.2.2 Robotic gastrectomy

Robotic gastrectomy (RG) is an innovative technology that overcomes the innate limitations of conventional laparoscopic surgery by offering advantages such as 3D stereoscopic high-definition view, less fatigue, complete tumor filtering, seven degrees of wrist-like motion, and a shorter learning curve [29]. Most current studies comparing LG and RG are single-center, small-sample studies, and the findings are inconsistent. A recent meta-analysis of 40 retrospective studies found that RG was associated with higher operating time, higher cost, less intraoperative blood loss, and a lower rate of surgical complications compared to LG [30]. Moreover, the RG group had a significantly increased number of retrieved lymph nodes, but similar resection margin distance and recurrence rate. Therefore, selecting between minimally invasive surgical approaches requires a comprehensive consideration of the clinical evidence, the surgeon’s experience, cost, and other factors.

3.3 Function-preserving surgery

Regardless of whether a surgery is minimally invasive or open approach, there has been an effort to preserve gastric function. The pylorus participates in gastric emptying and food retention. Operations preserving the pylorus could help maintain normal gastric functions, improve postoperative life quality, reduce the frequency of postoperative feeding, and reduce the incidence of diarrhea, dumping syndrome, and other complications [31]. Pylorus-preserving gastrectomy (PPG) is recommended for EGC located in the middle third of the stomach with a distal tumor border of at least 4 cm proximal to the pylorus [32]. For cT1N0M0 GC located in the upper third of the stomach, metastasis to lymph node stations No.4d/No.5/No.6 is rare. Thus, proximal gastrectomy is applicable for cases with remnant stomach over half the original size [33].

Compared with total gastrectomy (TG), proximal gastrectomy showed significant advantage in preserving gastric function, reducing anastomotic complications, and promoting postoperative nutritional status [34]. A Korean study comparing laparoscopic-assisted pylorus-preserving gastrectomy (LAPPG) with distal gastrectomy (LADG) reported similar 3-year recurrence-free survival rates (98.2% vs 98.8%) but lower rates of complications (7.8% vs 17.0%), suggested LAPPG as a better option than LADG in terms of nutritional advantage and complications [35]. Further investigations on short-term and long-term outcomes comparing LAPPG with LADG were performed in a Korean RCT (KLASS-04) [36].

Further challenges lie in improving digestive tract reconstruction after function-preserving surgery, which is highly associated with postoperative functional disorders. Chinese experts recently developed a consensus on digestive tract reconstruction after proximal gastrectomy, recommending seven types of reconstruction surgeries [37]. Of all reconstruction methods, esophagogastrostomy is the most prevalent, with advantages of shorter operation time and less blood loss. However, esophagogastrostomy also has the high incidence of esophageal reflux. Meanwhile, anti-reflux reconstructions like jejunal interposition and jejunal pouch interposition are not generally adopted because of their increased complexity and lack of clinical evidence [38].

3.4 Evolution in radical surgery

Based on a 15-year study involving a Dutch trial, D2 lymphadenectomy has become the cornerstone of surgical treatment for GC, especially in Eastern countries, due to its lower risk of local recurrence and GC-related mortality than D1 [39, 40]. Current research on improving the quality of radical surgery is focused on specific topics and technical variants including lymphadenectomies beyond D2, splenectomy and No.10 lymph node dissection, omentectomy, and membrane anatomy-guided surgery.

3.4.1 Lymphadenectomies beyond D2

Many studies investigate lymphadenectomies beyond D2, but only a few are prospective trials [41]. A Japanese randomized controlled trial (RCT) (JCOG9501) comparing gastrectomy with D2 vs D2 plus para-aortic nodal dissection (PAND) lymphadenectomy in patients with a curable clinical stage reported similar morbidity and survival rates between the procedures [42]. Controversial results regarding survival benefits were reported in studies on D2 lymphadenectomy plus No.14v nodal dissection. Zhang et al. showed poorer overall survival and increased complication rates in distal GC patients with No. 6 nodes metastasis who added No.14v lymph node dissection to D2 gastrectomy [43]. For patients undergoing total gastrectomy, whether to perform splenectomy has long been controversial [44]. A Japanese RCT (JCOG0110) found equivalent survival outcomes (5-year survival: 75.1% vs 76.4%) in splenectomy patients compared with non-splenectomy. However, this study found a 2.4% No.10 node metastases rate in the splenectomy group and a 3.5% No.10 node metastases rate in the non-splenectomy group. Seven of the eight patients with No. 10 metastasis developed poor prognosis (recurrence and death). Furthermore, following the JCOG0110 report, there has also been controversy regarding clearance of splenic hilum. Chinese researchers have launched two RCT (NCT02333721 and NCT04050787) to evaluate the prognosis of No.10 lymph node dissection in advanced proximal GC treated by laparoscopic D2 radical gastrectomy. Universally, lymphadenectomies beyond D2 for extra-regional lymph node metastasis is only suggested in experimental settings and in the setting of conversion surgery [32].

3.4.2 Omentectomy

The omentum has abundant blood vessels and lymphatic vessels which provide a susceptible environment for metastasis. Omentectomy was once considered an important part of radical gastrectomy for advanced GC, but different views have arisen in recent years. The OMEGA prospective cohort study reported that greater omentum lymph node metastasis incidence was only 5% in pathological specimens [45]. In a recent retrospective cohort study, the incidence of omental lymph node metastasis was 1.8% and only occurred in stage III-IVGC patients. Omental lymph node metastasis significantly related to tumor size, N stage, clinical stage, and venous invasion. The results suggest that total omentectomy should be avoided in many T3/T4 tumors. A Chinese RCT (NCT04108494) and a Japanese RCT (UMIN000005421) are currently analyzing survival outcomes related to omentectomy.

3.4.3 Mesogastrium excision

Total mesorectal excision or complete mesocolic excision for rectal and colon cancers, based on principles from embryology and membrane anatomy, have become standard surgical oncology procedures [46, 47]. Accordingly, gastrectomy with complete mesogastrium excision is imperative to reduce local-regional recurrence. Therefore, the concepts of en bloc mesogastrium excision (EME) or complete mesogastrium excision (CME) have been proposed recently [48]. EME is not directly applicable for GC due to the unique mesentery of the stomach compared to the colon and rectum. Dissection of the anterior membrane of the transverse colon mesentery and the pancreatic capsule, the core technique in the procedure, is especially difficult to perform in laparoscopic surgery [49]. The safety and feasibility of mesogastrium excision has been supported by several studies [50]. Further investigation on survival outcomes is currently ongoing within a Chinese phase III RCT (NCT01978444).

3.4.4 Non-surgical treatment of gastric cancer

Non-surgical treatment for GC is mainly divided into preoperative neoadjuvant, postoperative adjuvant, and palliative therapy. Although surgery is the only curative method for GC, a large number of clinical studies have confirmed that perioperative chemotherapy may improve the survival benefit. Meanwhile, close attention should be paid to the differences between Eastern and Western chemotherapy regimens, as well as surgical options [51].

4 Non-surgical treatment of gastric cancer

4.1 Adjuvant chemotherapy for resectable gastric cancer

An open parallel multi-center phase III RCT, the CLASSIC trial (NCT04135781), was undertaken in China, South Korea, and Taiwan. This trial recruited 1035 patients with stage II-IIIB GC who received curative intended D2 gastrectomy. These patients were randomly assigned to eight cycles of post-operative capecitabine plus oxaliplatin or surgical treatment alone. The 3-year disease-free survival rates were 74% for the chemotherapy group and 59% for the surgery only group, promoting the option of chemotherapy after radical D2 gastrectomy [52]. Another randomized study assessed the therapeutic effect of perioperative ECF (epirubicin, cisplatin, and infused fluorouracil) versus radical surgery alone group for locally advanced gastric cancer patients (ISRCTN93793971). The results indicate that progression-free survival (PFS) (HR = 0.66) and 5-year survival rate (36% vs 23%) improved significantly in the perioperative chemotherapy group, with much smaller and less advanced tumors compared with the surgery only group [53].

The FLOT regimen (fluorouracil, leucovorin, oxaliplatin combined with docetaxel) is the leading choice for neoadjuvant chemotherapy in Europe, while the SOX regimen (S-1 plus oxaliplatin) is preferred in Eastern countries. The open, randomized, phase II/III trial in Germany (NCT01216644) reported the efficacy of perioperative FLOT and ECF/ECX regimens for locally advanced, resectable GC. The FLOT group had better clinical benefits and longer median overall survival (OS) (50 vs 35 months) [5]. Another new phase II two-arm randomized clinical trial in China (NCT03636893) showed no significant differences in safety or efficacy between FLOT and SOX (S-1 plus oxaliplatin) regimens. However, SOX may help reduce tumor regression grading compared with FLOT (32.4% vs 20.0%) [54]. In addition, post-operative chemotherapy had a better 5-year OS than chemoradiotherapy in the CRITICS trial when adjusted for all known confounding factors [55].

4.2 Systemic therapy for unresectable gastric cancer

4.2.1 Cytotoxic therapies

The combination of fluoropyrimidine and platinum compounds is recognized as first-line therapy for Eastern and Western unresectable GC populations, despite differences in epidemiological or clinicopathological characteristics, biological behavior, etc. In East Asia, S-1 plus cisplatin (SP) is a recommended first-line chemotherapy regimen for metastatic or recurrent GC. A multicenter phase III study evaluated PFS rate between SOX and SP (NCT01671449) [56]. The results in PFS (SOX vs SP: median 5.6 months vs. 5.7 months; HR 0.85, 95% CI 0.67–1.07), overall response rates (SOX vs SP: 58% vs. 60%, p = 0.7) and OS (SOX vs SP: median 12.9 vs. 11.4 months; HR 0.86, 95% CI 0.66–1.11) indicated SOX was non-inferior to SP with different tolerance of toxicity profiles. Since oxaliplatin is more commonly administered than cisplatin due its lower toxicity and higher convenience, SOX can be a standard first-line regimen [57]. A retrospective study of advanced GC in China was designed to assess the effectiveness and safety of docetexal plus S-1 (DS) and SOX. Clinical outcomes and side effects were similar between the two groups. Thus, both DS and SOX regimens are used for advanced GC patients [58].

4.2.2 Targeted therapies

Numerous recent studies have focused on the molecular mechanism of carcinogenesis in malignancies such as breast, colorectal, and lung cancers. These findings prompted development of targeted therapies for GC applications, with targets including human epidermal growth factor receptor 2 (HER2), epidermal growth factor receptor (EGFR), and vascular endothelial growth factor (VEGF) and its receptors, which may revolutionize treatment and improve prognosis for GC patients.

HER2-targeted therapies

Trastuzumab, a monoclonal antibody against HER2, was added to chemotherapy as standard first-line treatment for HER2-positive advanced GC. Participants in the ToGA multi-center phase III randomised trial were randomly assigned to chemo (capecitabine or fluorouracil plus cisplatin) or chemo plus trastuzumab groups (NCT01041404). The two groups experienced similar adverse effects but different median OS (13.8 months in trastuzumab plus chemo vs 11.1 months in chemo alone) [59]. The optimal trastuzumab regimen is still being explored. RC48-ADC, a novel humanized anti-HER2 antibody, was developed in China (NCT02881190). This new antibody displayed positive and active anti-tumor effects which may provide new directions for advanced GC patients [60].

VEGF/VEGFR-targeted therapies

VEGF and its receptor-related target agents have gained increasing focus, with new studies on target therapies being carried out worldwide. Ramoxiuzumab can suppress tumor angiogenesis by specifically binding VEGFR-2 to inhibit its activation. Subgroup analysis in RAINBOW, a phase III trial of ramucirumab plus paclitaxel, was conducted for advanced GC patients in East Asia. Prolonged median overall survival and PFS were experienced, with 12.1 and 5.5 months, respectively, in the ramucirumab plus paclitaxel group compared to 10.5 and 2.8 months in the paclitaxel group [61].

Apatinib was the first oral tyrosine kinase inhibitor (TKI) that selectively targets VEGFR-2. As a third-line chemotherapy, apatinib improved median OS and PFS in metastatic Chinese GC patients with acceptable toxicity [62]. The combination of paclitaxel or S1 plus apatinib also had good therapeutic effects for advanced GC in vivo and in vitro [63]. An additional phase II trial confirmed that apatinib plus chemotherapy showed better survival benefits and higher rates of R0 resection and conversion for initial unresectable GC patients [60]. Thus, these results suggest that apatinib is effective and relatively tolerable for unresectable GC patients who have received at least first-line chemotherapy.

Tyrosine kinase inhibitors

Tyrosine kinase inhibitors (TKIs) are being investigated as the broadest category of targeted therapies. The most common TKIs target phosphorylation signaling cascades, including HER2, EGFR, VEGF, and MET, that control cellular survival processes. Although numerous TKIs are being investigated for GC, few have shown encouraging results. VEGF inhibitors such as Sunitinib and Vandetanib as well as HER2-targeting TKIs such as lapatinib, afatinib, dacomitinib, and neratinib have largely been found not to yield substantial clinical and survival benefits. Reasons of the unsatisfied outcomes could be incorrect individuation of patient subgroups as well as the presence of cross-talking intracellular signals leading to a compensating resistance due to the administration of a single kinase inhibitor. Thus, the utility of current TKIs in GC treatment is still questionable [64].

4.3 Immunotherapy

Immunotherapy is a promising treatment for GC in addition to surgery, chemotherapy, and radiotherapy. It mainly stimulates or mobilizes the body’s immune system to enhance anti-tumor immunity in the tumor microenvironment to inhibit and kill tumor cells. Clinically, biological immunotherapy is often combined with radiotherapy and chemotherapy after surgery.

Immune checkpoint blockade has emerged as one of the most promising therapeutic options in the history of cancer treatment [65]. Designed to interfere with inhibitory pathways that naturally constrain T cell reactivity, immune checkpoint blockade releases inherent limits on the activation and maintenance of T cell effector function. Checkpoint inhibitors targeting CTLA-4, PD-1, and PD-L1 have yielded unprecedented and durable responses in a significant percentage of cancer patients in recent years. In a phase 3 trial (CheckMate 649), researchers evaluated first-line PD-1 inhibitor-based therapies in previously untreated, unresectable, non-HER2-positive gastric, gastro-oesophageal junction, and oesophageal adenocarcinoma. The results showed significant improvements in OS (HR 0.71, 98.4% CI: 0.59–0.86) and PFS (HR 0.68, 98% CI 0.56–0.81) in patients treated with Nivolumab plus chemotherapy versus chemotherapy alone. These findings support Nivolumab plus chemotherapy as a new standard first-line treatment for these patients.

Chimeric antigen receptor (Car T) therapy is a novel type of treatment that utilizes and manipulates T-cell receptors to recognize and attack cancer cells. Clinical trials have shown significant antitumor activity in liquid malignancies such as neuroblastoma, chronic lymphocytic leukemia, and B cell lymphoma [66]. This therapy is now being investigated in solid tumors including GC.

Adapting immune therapies is an emerging and promising research field for GC treatment. Further studies are needed to find the most vulnerable population for these immune therapies.

5 Radiotherapy

With the advancement of radiotherapy (RT) technology, the use of RT in gastric cancer is becoming increasingly common. Although role of radiation therapy still is controversial due to the inconclusive and conflicting results from many prior studies [67].

Results from the INT0116 trial (adjuvant chemoradiotherapy Trial for Gastric Cancer) revive the interest in the role of radiotherapy in GC [68]. Among the novel radiotherapy techniques, intensity-modulated radiation therapy (IMRT) has been shown to yield greater dose conformity than three-dimensional conformal radiation therapy (3D-CRT), which allows a dose escalation or reduction of normal tissue exposure [69].

5.1 Preoperative RT

Preoperative RT is mainly used to shrink the tumor before surgery. Meanwhile, it may play important roles in suppression of micrometastasis and reduction of local recurrence [67]. Radiotherapy may be particularly beneficial for patients with esophagogastric junction (EGJ) squamous-cell carcinoma for their highly radiosensitive nature [70]. Hagen et al. reported a significantly increased rate of tumor complete resection (92% vs. 69%, P < 0.001) and median OS (49.4 months vs. 24 months) in the chemoradiotherapy–surgery group versus patients treated with surgery alone. Two ongoing phase III RCTs from China (NCT03013010 and NCT01815853), will provide additional evidence for preoperative chemoradiotherapy in GEJ cancer and locally advanced gastric. The current results of the clinical studies are promising, which indicating radiotherapy as a rational therapeutic option.

5.2 Postoperative RT

Although the results of INT0116 trial suggested that radiotherapy plus chemotherapy as postoperative treatment could significantly improves overall and relapse-free survival. The less precisely controlled surgical technique (only 10% D2 dissection, 36% D1 dissection, and 54% D0 lymphadenectomy) and backwardness of radiation technology might impede the validity [71]. In 2015, the final report of ARTIST trial from Korean showed that postoperative RT did not significantly improve the disease-free survival [72]. Another two phase III clinical trials from China and Korea during the same period found that the postoperative RT after D2 dissection increased local recurrence-free survival but did not improve OS [73, 74]. Considering the negative results of the trials from the east, where D2 dissection were performed as standard procedure, operation with D2 dissection plus postoperative chemotherapy might be sufficient.

Perioperative RT as an important adjuvant therapy to improves the PFS and reduces the rate of local recurrence is promising for carefully selected GC patients after further validation.

6 Enhanced recovery after surgery

In 1997, H. Kehlet first emphasized that multidisciplinary perioperative comprehensive intervention might reduce the undesirable sequelae of surgical injury with improved recovery and that it may lead to major reductions in postoperative morbidity and overall costs [75]. This concept of enhanced recovery after surgery (ERAS) gradually became accepted and became consensus for colon surgery in 2005 [76]. This consensus introduced that the core principles of ERAS in colon surgery are avoiding preoperative intestinal preparation and long fasting, keeping body temperature stable, and using non-opioid analgesics in surgery. The consensus also encouraged removing drainage tubes and mobility interventions earlier after operation [76]. The first ERAS consensus guidelines for gastrectomy appeared in 2014 to provide optimal recommendations for perioperative medical management, mainly for patients with GC undergoing distal or total gastrectomy [77]. In the specific surgical guidelines for GC, the ERAS concept differs from traditional surgery in avoiding nasogastric tubes and in the timing of oral feeding after surgery [77]. In a Japanese RCT, 148 patients with GC were randomly assigned to ERAS versus conventional groups. Researchers observed significantly shorter hospital stays (9 days vs 10 days; P = 0.037), lower rates of postoperative complications (4.1% vs 15.4%; P = 0.042), and lower total costs (JPY 1,462,766 vs JPY 1,493,930; P = 0.045) in the ERAS group, indicating that patients receiving the ERAS protocol may see short-term postoperative benefits [78].

The absence of nasogastric tubes is a key concept in the ERAS protocol which has proven critical in perioperative management. In 1991, researchers found that routine use of postoperative nasogastric drainage after major abdominal surgery did not prevent nausea and vomiting and that the tube caused moderate to severe discomfort in 83% of patients. In addition, an RCT of 161 patients with GC undergoing gastrectomy in China from 2007 to 2009 showed patients without nasogastric decompression had no significant difference in postoperative complications and other adverse events, but reduced incidence of nausea and vomiting compared with those with nasogastric decompression [79]. Meanwhile, several studies have shown that long-term use of nasogastric tubes may increase the risk of aspiration pneumonia [80,81,82].

Indwelling peritoneal drainage catheters following surgery may potentially increase the risk of infection. Therefore, most ERAS protocols recommend minimizing use of indwelling catheters and tubes. In 2019, Japanese researchers retrospectively analyzed patients who underwent laparoscopic gastrectomy for GC and found no significant difference in the incidence and severity of postoperative complications between the drain and no-drain groups [83]. Furthermore, the days after surgery until initiation of soft diet (6.3 ± 7.4 vs 4.9 ± 2.9 days, P = 0.036) and postoperative hospital stays (15.7 ± 12.9 vs 13.0 ± 6.3 days, P = 0.023) were significantly greater in the drainage group than in the non-drainage group [83]. Similar results from many retrospective and prospective trials show better short-term outcomes in patients undergoing gastrectomy without indwelling tubes, suggesting that indwelling tubes may provide no additional benefit and might even prove harmful [84,85,86].

In the ERAS protocol for gastrectomy, the choice of minimally invasive surgery with less trauma and faster recovery is essential [87]. A RCT among 171 elderly patients (age ≥ 65 y) showed the ERAS protocol was feasible and effective for elderly patients with GC undergoing laparoscopic total gastrectomy. Increased HLA-DR expression on monocytes and decreased CRP levels demonstrated improved immune function and suppressed inflammatory reaction in the ERAS group [88].These findings indicate that the combination of minimally invasive surgery and the ERAS protocol may expand indications for gastrectomy.

However, some researchers raised a controversial issue about whether patients see benefits besides reducing length of stay and saving resources [89]. This might be due to the limited follow-up time and sample size of some RCTs. More studies are warranted in tailoring the ERAS protocol for GC patients in the future.

7 Conclusion and perspectives

However, despite this progress, many challenges remain. Although clinical evidence is accumulating at an increasing rate, high-quality well designed muti-center prospective studies are still limited. Large differences in preventive and treatment strategies still exist between the East and the West: e.g., more than 50% of GC are diagnosed at an early stage in Japan in contrast with about 27% in the USA [90]. Endoscopic screening and surveillance as secondary preventive strategy for people with high-risk histology is performed in Japan and Korea successfully [90]. Meanwhile, for Western countries with low rates of GC, the balance between benefits, harms and costs of screening remains unclear [91]. The inconsistency in strategies across the world impedes generalization of treatment progress. More international and cross-disciplinary cooperation in GC research is needed in the future.