Effect of omentum preservation on long-term prognosis of locally advanced gastric cancer: a systematic review and meta-analysis

Abstract

Background

The effect of omentum preservation (OP) on locally advanced gastric cancer (LAGC) remains controversial. This study aimed to investigate the long-term prognosis of LAGC patients with OP versus omentum resection (OR).

Methods

A comprehensive search of databases including PubMed, Web of Science, Embase, and Cochrane Library was conducted up until February 2024. Statistical analysis was performed using Stata 12.0 software. The primary outcome was to assess the impact of OP on the long-term prognosis of patients with LAGC, including overall survival (OS) and recurrence-free survival (RFS).

Results

A total of six case-control studies were included, encompassing a cohort of 1897 patients. The OP group consisted of 844 patients, while the OR group comprised 1053 patients. The study results showed that the OS (HR = 0.72, 95% CI: 0.58–0.90, P = 0.003) and 5-year RFS (HR = 0.79, 95% CI: 0.63–0.99, P = 0.038) in the OP group were superior to those observed in the OR group. Subgroup analysis indicated that 5-year OS (HR = 0.64, P = 0.003) and 5-year RFS (HR = 0.69, P = 0.005) in the OP group were also better than those in the OR group in Korea. However, the subgroup analysis conducted on stage T3-T4 tumors revealed no statistically significant differences in OS (P = 0.083) and 5-year RFS (P = 0.173) between the two groups.

Conclusion

Compared with OR, OP shows non-inferiority in patients with LAGC and can be considered a potential treatment option for radical gastrectomy.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12957-024-03521-3.

Introduction

Despite a global decline in gastric cancer (GC) incidence and mortality rates, GC remains the fifth most prevalent malignant tumor worldwide and ranks fourth as a leading cause of cancer-related deaths. According to the 2020 Global Cancer Statistics Report, there were approximately 1,089,000 newly diagnosed cases of cancer and 769,000 fatalities reported globally. Moreover, the incidence of GC exhibits significant regional disparities, with particularly high rates in East Asia [1]. In recent years, the treatment of GC has evolved into a comprehensive model involving surgical intervention and multiple adjuvant therapies, including chemotherapy, immunotherapy, radiotherapy, targeted therapy, and more [2]. However, radical gastrectomy remains the cornerstone of treatment for locally advanced gastric cancer (LAGC).

The radical gastrectomy procedure for GC involves removing the primary tumor as well as metastatic lymph nodes and surrounding infiltrating tissues, such as the omentum, to minimize the risk of distant metastasis and enhance curative efficacy. Currently, D2 lymph node dissection for LAGC has been recognized as a means to prevent lymph node metastasis. Peritoneal metastasis is one of the most common spreading sites of GC, and the prognosis is poor. To minimize clinical peritoneal implantation, it has been necessary to resect the entire omentum, including the bursa omentum, in previous treatments.

The structure and function of the omentum, however, are complex. On one hand, the omentum serves as an immune organ and contributes to immune defense [3]. On the other hand, it is also involved in infection, wound healing, and tissue regeneration [4]. At present, the effect of omentum preservation (OP) in early gastric cancer (EGC) has been demonstrated. For LAGC, previous meta-analyses have confirmed that OP exhibits reduced intraoperative blood loss compared to omentum resection (OR), along with a tendency towards shorter operation time, suggesting favorable short-term outcomes associated with OP [5–8]. Nonetheless, the long-term survival and recurrence rates regarding OP in LAGC have not reached a consensus. Therefore, this study aims to conduct a meta-analysis by systematically searching published articles on the role of OP in the long-term prognosis of LAGC.

Methods

This study adheres to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and utilizes the Assessing the Methodological Quality of Systematic Reviews (AMSTAR) tool for analysis. The scope of the literature search is from the inception of records in the electronic database until February 2024. Data were comprehensively and systematically collected from PubMed, Web of Science, Embase, and Cochrane Library. Furthermore, manual searches were conducted to ensure the inclusion of all relevant articles by scrutinizing the references cited in previous studies. The search terms used included combinations of “Stomach Neoplasms,” “gastric carcinoma,” “gastric cancer,” “carcinoma of the stomach,” “omentectomy,” and “omentum.” Two experienced reviewers independently searched the databases, and any discrepancies were resolved by a third reviewer. This study has been registered with PROSPERO under the number CRD42024516152.

Selection and criteria of study

Studies were included according to the PICOS principles if: (1) the patient was diagnosed with LAGC and underwent radical gastrectomy; (2) the experimental group underwent radical gastrectomy combined with OP; (3) the control group underwent radical gastrectomy combined with OR; (4) the primary or secondary outcomes included, but were not limited to, survival endpoints; (5) research types of articles were cohort studies, case-control studies, or randomized controlled trials (RCTs).

Exclusion criteria were defined as follows: (1) articles included EGC; (2) articles were deficient in completed data or the data was unquantifiable; (3) the language of the article was not English; (4) considering updated literature, those with a more comprehensive sample size or more recent ones were included.

Data extraction

The available data from all studies were independently extracted by two researchers based on a standardized data extraction table. The subsequent results were obtained through the above review process: study characteristics (country, study design, year, sample size, follow-up), patient characteristics [age, sex, body mass index (BMI)], tumor characteristics (location, size, TNM stage), treatment (type of surgery, surgical approach, extent of lymph node dissection, adjuvant chemotherapy), outcome measures [overall survival (OS), recurrence-free survival (RFS)].

Assessment of quality

The Cochrane bias risk assessment tool was utilized for RCTs, and the Newcastle-Ottawa Scale (NOS) was employed for case-control and cohort studies. The former consists of seven components: selection bias, implementation bias, measurement bias, follow-up bias, reporting bias, and other biases. Meanwhile, the latter evaluates risk from three perspectives: selection, comparability, and outcomes. In case of dispute, consensus was reached by consultation with the third reviewer.

Statistical analysis

The collected data in this meta-analysis was analyzed using Stata 12.0 software. The effect sizes for survival data were determined by calculating the hazard ratio (HR) along with its corresponding 95% confidence interval (CI). The chi-square test was employed to assess heterogeneity, and when I^2 < 50% or P > 0.05, it was deemed as exhibiting low heterogeneity; otherwise, it was considered to demonstrate high heterogeneity. In addition, due to potential heterogeneity (such as country, extent, and type of surgery, surgical approach) between studies, a random-effects model was used in this meta-analysis to improve the reliability of statistical results. Publication bias and sensitivity analyses were performed when the number of articles was more than 10. All outcome-related data were analyzed with two-sided tests. P < 0.05 was considered statistically significant.

Results

Studies selection

According to the designed literature retrieval process, a total of 2109 terms were queried. After excluding duplicate studies, 1567 studies remained. Upon reviewing the title and abstract, a total of 1546 studies were excluded, while 21 studies were found relevant to the topic. After thoroughly reading these 21 studies, it was determined that 15 did not meet the criteria: three included EGC, one lacked interesting outcomes, four had incomplete data, and seven were unavailable for statistical analysis. Ultimately, six studies were included in the meta-analysis [9–14]. The literature selection process is illustrated in Fig. 1.

Fig. 1.

Flow diagram describing inclusion and exclusion criteria

Study characteristics

All six studies included in this analysis are case-control studies, with five of them utilizing propensity score matching (PSM). These studies are conducted in East Asia, including one study from China, two from Japan, and three from Korea. The publication years of these studies range from 2013 to 2023, with sample sizes ranging between 108 and 526 subjects. Furthermore, all patients included in these analyses suffered from LAGC.

Pathological features of the patients

A total of 1897 patients are enrolled, with 844 cases in the OP group and 1053 cases in the OR group. Regarding tumor pT stage, four studies restrict patients to pT3 or pT4, and the other two studies additionally involve pT1 or pT2. In terms of surgical intervention, in two studies, all patients underwent laparoscopic distal gastrectomy. Regarding the dissection of lymph types, four studies incorporated D1 + and D2 surgery, while one study exclusively performed D2 surgery, and another study incorporated D2 + surgery based on D1 + and D2. The pathological features of the patients are presented in Table 1, with supplementary patient information provided in Supplementary Table 1.

Table 1.

Characteristics of all the studies included in the meta-analysis

Author	Song, M.		Seo, W. J.		Ri, M.		Lee, H.		Jeong, S. A.		Hasegawa, S.
Year	2023		2021		2020		2023		2023		2013
Country	China		South Korea		Japan		South Korea		South Korea		Japan
Group	OP	OR	OP	OR	OP	OR	OP	OR	OP	OR	OP	OR
Number	58	50	225	225	263	263	107	107	93	310	98	98
pT stage
1	0	0	0	0	48	47	0	0	0	0	0	0
2											34	30
3	25	22	111	100	215	216	61	61	0	0	30	34
4a	33	28	114	125			46	46	93	310	34	34
pN stage
0	20	18	73	75	145	148	56	56	44	70	41	39
1			47	42					21	53	23	25
2	38	32	42	42	118	115	51	51	28	187	17	18
3			63	66							17	16
Histology
Well/ moderately	29	24	68	67	92	97	36	36	17	64	58	61
Poorly/undifferentiated	29	26	157	161	135	127	71	71	76	246	40	37
Surgical type
TG	0	0	56	58	112	107	0	0	32	142	37	46
DG	58	50	169	167	151	156	107	107	61	168	61	52
Surgical approach
Open	0	0	60	69	NA	NA	0	0	2	254	84	98
Laparoscopy	58	50	165	156	NA	NA	107	107	91	56	14	0
Lymphadenectomy	D2	D2	D1+: 22 D2: 203	D1+: 25 D2: 200	≤D1: 8 D1+: 146 D2: 103 ≥D2: 6	≤D1: 11 D1+: 146 D2: 102 ≥D2: 4	D1+: 14 D2: 93	D1+: 1 D2: 106	< D2: 4 D2: 89	< D2: 7 D2: 303	D1 + a: 10 D1 + b: 3 D2: 83 D2+: 2	D1 + a: 9 D1 + b: 3 D2: 84 D2+: 2
Retrieved lymph node	36.0 ± 9.4	34.2 ± 8.9	42	42	39 (30–50)	41 (32–51)	37 (11–104)	37 (12–84)	NA	NA	NA	NA

NA, not available; OP, omentum preservation; OR, omentum resection; TG, total gastrectomy; DG, distal gastrectomy

Quality assessment

The quality assessment of the six studies based on NOS reveals that all the articles achieve scores ranging from 7 to 9, indicating a higher level of quality.

Analysis of survival outcomes

Regarding OS, which includes five studies, patients who underwent OP have a higher survival rate compared to those who received OR (HR = 0.72, 95% CI: 0.58–0.90, P = 0.003, I^2 = 9.1%, Fig. 2). In terms of five-year RFS, which includes four studies, patients who received OP also have a higher 5-year RFS than those who received OR (HR = 0.79, 95% CI: 0.63–0.99, P = 0.038, I^2 = 15%, Fig. 3).

Fig. 2.

Forest plot of overall survival with omentum preservation versus omentum resection(P = 0.003)

Fig. 3.

Forest plot of 5-year recurrence-free survival with omentum preservation versus omentum resection(P = 0.038)

Subgroup analysis

The subgroup analyses were conducted based on tumor stage and region. For T3-T4 stage (cT3-4 or pT3-4) tumors, the study results show no significant difference in OS (HR = 0.81, P = 0.083, Fig. 4a) and 5-year RFS (HR = 0.82, P = 0.173, Fig. 4b) between both groups. In terms of regions, the OP groups are better than the OR groups in both 5-year OS (HR = 0.64, P = 0.003, Fig. 5a) and 5-year RFS (HR = 0.69, P = 0.005, Fig. 5b) in Korea. Meanwhile, with two studies included, there is no statistical difference in OS between the two groups in 5-year Japan (HR = 0.75, P = 0.31, Fig. 6).

Fig. 4.

Forest plot of the prognosis of omentum preservation versus omentum resection in T3-T4 tumors: (a) overall survival (P = 0.083); (b) 5-year recurrence free survival (P = 0.173)

Fig. 5.

Forest plot of the prognosis of omentum preservation versus omentum resection for Korean subgroup: (a) 5-year overall survival (P = 0.003); (b) 5-year recurrence free survival (P = 0.005)

Fig. 6.

Forest plot of 5-year overall survival with omentum preservation versus omentum resection for Japanese subgroup (P = 0.31)

Discussion

Six case-control studies involving 1,897 patients were included. The primary outcomes assessed were long-term prognoses. For overall OS and RFS, the results showed that OP was superior to omentum resection (OR), confirmed in the Korean subgroup analysis as well. However, in Japan, subgroup analysis displayed no statistical difference between the two groups. Additionally, the results for T3-T4 tumors were not significantly different between the experimental and control groups. Collectively, OP is deemed advantageous for the long-term prognosis of patients diagnosed with LAGC.

Resectable GC is primarily treated by surgery, including for EGC and LAGC. For EGC, supported by evidence-based medicine, OP can effectively mitigate complications and intraoperative blood loss, yielding favorable long-term prognosis [15]. In contrast, the prognosis for LAGC is unsatisfactory, with a high recurrence rate. The metastatic patterns of GC include hematogenous metastasis, lymphatic metastasis, local recurrence, and implantation metastasis. A large RCT conducted in the Netherlands confirmed that D2 lymphadenectomy was superior to D1 lymphadenectomy in terms of recurrence and survival, and D2 lymphadenectomy remains the standard surgical procedure for LAGC today [16].

The most common type of GC implant metastasis is peritoneal metastasis, reported to account for more than 50% of cases [17]. Currently, cytoreductive surgery (CRS), hyperthermic intraperitoneal chemotherapy (HIPEC), and other technologies are used to treat and prevent peritoneal implantation [18–20]. The mechanism of peritoneal metastasis in GC can be described by Tiffin Paget’s “seeds and soil” theory [21]. Specifically, tumor cells undergo the following processes: (1) initially, tumor cells are spontaneously shed or released during surgery [22], facilitated by a diverse array of growth factors, cytokines, and proteases. This process involves the epithelial-to-mesenchymal transition (EMT), where E-cadherin plays a crucial role in maintaining cellular polarity [23]. (2) The survival of tumor cells in the hypoxic tumor microenvironment (TME) relies on the presence of hypoxia-inducible factor-1α (HIF-1α). HIF-1α stimulates the production of angiopoietin-like protein (ANGPT), epidermal growth factor (EGF), and other proangiogenic factors to facilitate oxygen delivery to tumor cells [24, 25]. (3) Tumor cells infiltrate the peritoneum by penetrating the mesothelium and lymphatic foramen within the peritoneal cavity. Previous research has demonstrated that gastric cancer cells express growth factors, such as transforming growth factor-β1 (TGF-β1) [26], and adhesion molecules, like cluster of differentiation 44 (CD44) [27], which stimulate mesothelial cells to enhance collagen and fibrin production. This process promotes omental fibrosis and facilitates tumor cell attachment to the mesothelium [28]. Another possible route of metastasis is through the lymphatic foramen located in milky spots of the omentum, where macrophage migration occurs. However, tumor cells can also invade subcutaneous spaces via this pathway [29]. (4) Adherent tumor cells must cross the basement membrane to invade the peritoneum. Increased expression of matrix metalloproteinases (MMPs) such as MMP-2 and MMP-7 during cancer metastasis disrupts peritoneal structure and correlates with GC dissemination [30–32]. (5) Successful invasion of tumor cells, through autocrine or paracrine growth factors such as vascular endothelial growth factor (VEGF), promote blood vessel formation and expand the invasion range [33]. Based on the above theory, considering the high rate of peritoneal metastasis in LAGC, it is necessary to remove the complete omentum after surgery to reduce recurrence. However, in our study, prophylactic OR did not improve the long-term prognosis of patients.

Our results indicate that OP was noninferior to OR in the long-term prognosis of LAGC. Possible conjectures are as follows. Firstly, the significance of OR is to prevent tumor implantation and lymph node metastasis. In fact, OP also removes part of the omentum and lymph nodes. The incision was made 3 to 4 cm from the gastroepiploic arch, and the omentum connected by short gastric vessels was preserved. In this process, complete lymph node dissection can be achieved. OR and OP have similar effects in preventing GC metastasis. Secondly, in addition to implant metastasis, blood and lymph node metastasis are also important routes of metastasis. Kakushima, N. et al. reported several cases of Krukenberg tumor arising from EGC confined to the mucosa [34]. Because the mucosa of EGC is limited, it is not easy to implant and metastasize, but it can spread through blood vessels and lymphatic vessels. Indeed, Krukenberg tumor most commonly occurs through retrograde lymph node metastasis as a type of metastatic tumor [35]. As previously mentioned, prophylactic OR aims to prevent peritoneal implantation metastasis; however, due to alternative modes of metastasis existing, OR may not be efficacious, particularly in advanced malignant tumors. Finally, Powless, C. A. et al. reported a study of early resectable cancer in which tumor cells were found in only 4% of patients with omental tissue microscopic biopsy [36]. Similar results were reported by Cecelia A Powless et al., with only 3.8% of patients up-staging the pathology by omentum biopsy [37]. In the resectable stage of GC, the detection rate of cancer cells in the omentum is low, and the effect of OR is not significant. However, in LAGC, especially after stage T4 when the tumor penetrates the serosa and invades adjacent structures, tumor cells tend to metastasize to distant sites, and the OR has little value.

LAGC was divided into stage II and stage III. Because of the potential positive correlation between the risk of peritoneal metastasis and tumor stage, a subgroup analysis was performed for stage T3-T4 tumors. The results revealed that OP did not have a discernible impact on long-term survival. Theoretically, there is heterogeneity in the prognosis of different tumor stages, but further subgroup analysis cannot be done due to the lack of data. But there are also studies that provide data based on the staging of T3-T4. Compared to patients with T1-T2, those with T3-T4 have a worse prognosis, are more prone to recurrence and metastasis, and require standard D2 lymph node dissection and greater omentum resection. The OP in stage T3-T4 LAGC is controversial. Therefore, we conducted a subgroup analysis for stage T3-T4. In addition, the treatment modality for LAGC is surgery combined with perioperative therapy. In theory, subgroup analysis should be conducted based on different adjuvant treatment modalities, but due to a lack of data, subgroup analysis cannot be performed, which may be one of the reasons leading to the high heterogeneity in our study.

Regarding long-term survival, OP showed an overall better outcome in this study (including within the Korean subgroup). This observation may be attributed to the functional significance of the omentum. Within the omentum are small gaps between mesothelial cells where fat-associated lymphoid clusters (FALCs) form by aggregating white blood cells; these FALCs serve as fundamental units for immune function within the omentum [38]. They filter peritoneal fluid to mount local immune responses against antigens and inflammatory stimulation [39]. In addition, a RCT focusing on short-term benefits of OP in EGC patients revealed reduced operation time, decreased intraoperative blood loss, and lower incidence of major complications such as anastomotic leakage [40]. Studies have reported that intraoperative complications are closely related to recurrence after radical gastrectomy for GC, and the 5-year survival rate of patients without complications is significantly higher than that of patients with complications [41]. This result further supports our view. In general, OR has a positive effect on the long-term prognosis of patients.

For long-term outcomes, previous meta-analyses have confirmed the noninferiority of OP in terms of OS and RFS [5–8, 42, 43]. In addition, the results from the group of Kong M. [43] and Chen M. [42]et al. further proved the beneficial effect of OP on OS. Noteworthily, our study showed that OP not only provides an OS benefit but also reduces the recurrence rate. Taking the time variable into account, we used HR and 95% CI, and heterogeneity was low. Consequently, these findings can be deemed more robust and reliable. Furthermore, our study also demonstrated that OP was noninferior to OR in subgroup analyses based on region and tumor stage.

However, our study has certain limitations. Firstly, our study was based on a retrospective design and the level of evidence was not high. Clinical studies on the long-term prognosis of LAGC based on OP are currently underway, but the data is not yet mature, such as the studies conducted by Sato, Y. [44] et al. and Lu, S. [45] et al. The results of these RCT studies are highly anticipated. Additionally, the sample size and number of studies included in our analysis were insufficient, which may introduce more bias and heterogeneity; therefore, we look forward to the publication of large-sample RCTs in the future. Finally, all the studies included in our analysis were conducted in Asia and had a strong ethnic and regional homogeneity among patients. Hence, our findings may not be directly applicable to regions outside Asia.

Conclusion

This study confirmed the feasibility of OP in LAGC. OP can improve the long-term prognosis of patients with LAGC (including OS and RFS). This finding may need to be further confirmed by large sample RCTs.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

GC

Gastric cancer

LAGC

Locally advanced gastric cancer

EGC

Early gastric cancer

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

International Prospective Register of Systematic Reviews

BMI

Body mass index

OS

Overall survival

RFS

Recurrence free survival

RCT

Randomized controlled trial

NOS

Newcastle-Ottawa Scale

HR

Hazard ratio

CI

Confidence Interval

PSM

Propensity score matching

OP

Omentum preservation

OR

Omentum resection

EMT

Epithelial-to-mesenchymal transition

TME

Tumor microenvironment

ANGPT

Angiopoietin-like protein

EGF

Epidermal growth factor

TGF-β1

Transforming growth factor-β1

CD44

Cluster of differentiation 44

MMP

Matrix metalloproteinase

VEGF

Vascular endothelial growth factor

FALCs

Fat-associated lymphoid clusters

IL-6

Interleukin- 6

TNF-α

Tumor necrosis factor-α

AMSTAR

Assessing the Methodological Quality of Systematic Reviews

Author contributions

XZ and WR contributed to drafting, conception, and design. WS contributed to manuscript writing. WS and TH contributed to data collection; WS and TH performed procedures and data analysis. TH polished this article. All authors contributed to manuscript revision, read, and approved the submitted version.

Funding

Not applicable.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.