Impact of surgical resection on the survival in Mexican patients with gastric cancer: A meta-analysis and systematic review

Background:

Gastric cancer (GC) is one of the most frequent cancer types in Mexico. The primary method used as a treatment is surgical resection. The role of surgery in increasing survival is controversial. This study aimed to determine whether surgical resection increases the survival of patients with GC in a Mexican population.

Methods:

A systematic review of literature searches (Evidence-based MEDLINE/PubMed, Web of Science, Cochrane Library, and SciELO) and meta-analysis were performed based on the Preferred Reporting Items for Systematic Reviews and Meta-Analysis criteria. The published articles from 2000 to the current time were divided into cross-sectional and randomized studies. The inclusion criteria were survival, surgical resections, patients treated in Mexico, and primary GC. The effect estimation was calculated using the risk ratio (RR). The random-effects model and a confidence interval (CI) of 95% were used.

Results:

The RR of the pooled studies was 1.09 (95% CI, 0.71–1.67). RR of 0.82 (95% CI, 0.63–1.07) was obtained in cross-sectional studies, and randomized studies showed a RR of 2.08 (95% CI, 0.25–17.07).

Conclusion:

This work is the first systematic study that assesses the role of surgery on the survival of patients with GC in the Mexican population, the results showed that surgical resection did not improve survival in patients with GC.

1. Introduction

Gastric cancer (GC) is one of the most frequent types of cancer worldwide.^[1] In Mexico, an incidence rate of 6.2% and a mortality rate of 4.7 per 100 000 people are reported.^[2] Early detection of GC is still a challenge, resulting in high rates of diagnosis at advanced stages and poorer survival.^[1]

Despite recent medical progress, surgical treatment remains the procedure with a curative intent among GC patients.^[3] However, not all patients and stages of GC are candidates to undergo surgery. Moreover, the use of adjuvant and neoadjuvant chemotherapy can increase the survival of the patient.^[1]

The Mexican guidelines describe that open surgery or laparoscopic surgical resection are indicated in most cases where endoscopy does not achieve curative resection due to the high risk of lymph node metastases.^[4] Although tumor resection aims to cure gastric adenocarcinoma, some cancers cannot be totally removed.^[5] Currently, there are no records regarding the number of surgical resections or the impact of surgical procedures on the survival of GC patients in Mexico. However, it seems that the increase in surgical resections has decreased the mortality rate.^[6] A study in the Netherlands mentioned that treatment decisions in hospitals have an impact on overall survival; those patients with a lower probability of surgical resection had worse outcomes.^[7] A most recent study in Sweden concluded that a higher resection rate seems to be beneficial with regard to overall survival among the Swedish population.^[8]

The physicians consider different factors such as age, sex, comorbidities, and socioeconomic status to perform a surgical procedure on GC patients.^[7] Mexico is one of the top countries with elevated frequencies of comorbidities such as diabetes, obesity, and metabolic syndrome,^[9] conditions that impact the survival of patients with GC. Besides, the Mexican health system still faces many challenges, such as health inequity and the implementation of an adequate reporting system for cancer patients.^[10] There is scarce nationwide data regarding the number of surgical resections performed and the overall outcomes in GC patients in Mexico. Therefore, we hypothesize that surgical resection is associated with longer survival. This present work aims to assess the impact of surgical resection on the survival of CG patients in Mexico.

2. Materials and Methods

2.1. Study design and search strategy/data source

A systematic review was performed independently by 4 researchers (O.D.P.-C., H.S.L.V.-S., C.D.F.-G., A.G.P.-A.) using published articles on the survival of GC patients treated in Mexico. Studies were identified by searching terms in English and Spanish using online sources and databases of MEDLINE/PubMed, Web of Science, Cochrane Library, and SciELO. Grey literature was searched using Google Scholar and the Library of the National Autonomous University of Mexico (UNAM in the Spanish acronym). The combination of the keywords that yielded the best results in the databases was: ((Gastric Cancer) AND (Mexico) AND (Treatment) AND (Survival)). The search was done for titles and/or abstracts and was carried out between December 2021 and January 2022.

2.2. Selection of studies

A total of 114 studies were identified with the previous keywords and exported to Excel. After removing duplicates, 99 studies were considered as candidates for this review. Then, the following inclusion criteria were applied: studies that include patients with a diagnosis of primary GC; over 18 years of age; Mexican population; surgical procedures as corrective treatment of cancer; reported data on the survival of the population studied. The exclusion criteria for our study were: manuscripts written in other languages than Spanish or English; studies with molecular pathways; studies using cell lines or animal models; letters to the editor, systematic reviews, only abstracts, files not available or incomplete.

After preliminary screening of studies, all abstracts were evaluated for eligibility (in some cases, studies were evaluated in full-length) based on the established criteria. There was no restriction on the recruitment period or sample size. Only articles published from 2000 to the current time were considered.

2.3. Data extraction

The present study was carried out based on Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) criteria^[11] (PRISMA checklist).

Data from all the selected articles were independently extracted by 3 researchers (A.L.G.-S., A.A.-N., R.L.) into a predefined database, including first author, year of publication, sample type, and population size. Also, data on the patient’s follow-up after surgical treatment and control group was reported (Table 1). Survival was measured from the first day of treatment until death from any cause. The quality of the methodology used in each study was evaluated using the Critical Appraisal tools in JBI Systematic Reviews. In case of discrepancy among researchers (O.D.P.-C., A.G.P.-A.), the final consensus was made by another investigator (R.L.) (Supplementary Figures S1 and S2, Supplemental Digital Content, http://links.lww.com/MD/J56).

Table 1.

Characteristics of the cross-sectional studies included in this systematic review.

Num	Author/Study	With treatment (surgical), n (%)		Follow-up (mo)	Without treatment (control), n (%)		Follow-up (mo)	Gastric cancer stages, n (%)
		Survival	Death		Survival	Death
1	Oñate-Ocaña, et al (2007). Treatment of the adenocarcinoma of the esophagogastric junction at a single institution in Mexico	29 (42.65)	39 (57.35)	16	10 (6.02)	156 (93.98)	6	I: 23 (9.8), II: 84 (35.9); III: 109 (46.6); UND: 18 (7.7)
2	Green, et al (2002). Adenocarcinoma of the stomach: univariate and multivariate analysis of factors associated with survival	38 (33.33)	76 (66.67)	12	10 (27.77)	26 (72.23)	6	I: 8 (5.3), II: 9 (6), III: 33 (22.0); IV: 96 (64.0); UND: 4 (2.4)
3	Ramos de la Medina, et al (2004). Clinicopathologic characteristics of gastric cancer in a young patient population	31 (42.46)	42 (57.54)	10	6 (60.0)	4 (40.0)	6	I–II: 8 (9.6), III: 22 (26.5); IV: 53 (63.9)
4	Medina-Franco, et al (2007). Prognostic factors in primary gastric lymphoma	16 (80.00)	4 (20.00)	65	11 (52.38)	10 (47.62)	18	I: 11 (26.8); II: 8 (19.5); III: 7 (17.1); IV: 15 (36.6)
5	Oñate-Ocaña, et al (2008). Selection of best candidates for multiorgan resection among patients with T4 gastric carcinoma	27 (15.97)	142 (84.03)	25	52 (9.47)	497 (90.53)	8	III: 86 (12.0); IV: 632 (88.0)

UND = undefined stage.

2.4. Ethical clearance

The study was approved by the Research Committee (DG/2599/2018) and the Ethics Committee of the Regional High Specialty Hospital of the Yucatan Peninsula (No. CEI/030/2018) in connection with the research project (identification code: 2018-006).

2.5. Statistical analysis

The risk ratio (RR) was assessed from the total number of patients at different stages of GC that did not survive in the following study periods.

When comparing the surgical and nonsurgical treatments, the effect estimation was pooled using a random-effects model and presented in a forest plot with a confidence interval (CI) of 95%. The heterogeneity was assessed using the I² index. Publication bias was assessed by the visual inspection of the funnel plot. All statistical analyses were undertaken using the Meta package version 4.18 for R Studio software version 4.1.^[12]

3. Results

3.1. Study selection

Through PRISMA criteria, the conducted literature search displayed a total of 114 articles (PRISMA Flow Diagram, Supplementary Figure S5, Supplemental Digital Content, http://links.lww.com/MD/J160). Altogether, 15 articles were eliminated due to duplications, 99 relevant records were identified, and their titles and abstracts were reviewed. After applying the inclusion and exclusion criteria, a total of 8 articles were included in this systematic review and meta-analysis.

3.2. Characteristics of the studies

The years of the studies' publications ranged from 2000 to 2008. Nevertheless, the recruitment period of the participants was different across the selected studies, the period of data collection was between 1985 to 2001.

Gallardo-Rincón et al (2000),^[13] Green et al (2002),^[14] Oñate-Ocaña et al (2007),^[15] and Oñate-Ocaña et al (2008)^[16] conducted studies that included patients treated at the Instituto Nacional de Oncología in Mexico; Avilés et al (2004)^[17] and Avilés et al, (2006)^[18] carried out studies with patients from the Centro Médico Nacional Siglo XXI; Ramos-De la Medina et al, (2004)^[19] and Medina-Franco et al, (2007)^[20] conducted studies in patients from the Instituto Nacional de Ciencias Médicas y Nutrición Salvador Zubirán. All studies used biopsy as the diagnostic method for GC. In summary, patients included in this work were treated in hospitals located in Mexico City.

Tumor classification was reported using different criteria, 3 of the studies that included adenocarcinoma samples used the TNM Staging Criteria, and 2 studies used the American Joint Committee on Cancer Cancer Staging Criteria. Regarding the studies that included samples of lymphoma, 2 studies used Mushoff Criteria, and only one used the Ann Arbor scale.

The results were organized into 2 categories according to the study type: 5 studies were cross-sectional studies, and 3 were randomized (Tables 1 and 2).

Table 2.

Characteristics of the randomized studies included in this systematic review.

Num	Author/Study	With treatment (surgical), n (%)		Follow-up (mo)	Without treatment (control), n (%)		Follow-up (mo)	Gastric cancer stages, n (%)
		Survival	Death		Survival	Death
1	Aviles, et al (2006). Surgery and chemotherapy versus chemotherapy as treatment of high-grade MALT gastric lymphoma	38 (77.55)	11 (22.45)	60	36 (75.00)	12 (25.00)	60	I: 21 (21.6), II: 76 (78.4)
2	Aviles, et al (2004). The role of surgery in primary gastric lymphoma results of a controlled clinical trial	211 (48.06)	228 (54.94)	120	144 (96.00)	6 (4.00)	120	I: 404 (68.6); II: 185 (31.4)
3	Gallardo-Rincon, et al (2000). Neoadjuvant chemotherapy with P-ELF (cisplatin, etoposide, leucovorin, 5-fluorouracil) followed by radical resection in patients with initially unresectable gastric adenocarcinoma: a phase II study	5 (50.00)	5 (50.00)	13	17 (36.18)	30 (63.82)	8	III: 20 (33.3); IV: 40 (66.7)

A total number of 1932 patients were included in this analysis. Grouped by the study type, 1226 patients were included in cross-sectional studies, and 743 patients belonged to randomized studies. In addition, 475 patients (24.09%) were classified at Stage I, 362 (18.35%) at Stage II, 277 (14.05%) at Stage III, and 836 (42.40%) at Stage IV; only 22 patients were undefined (1.11%). Among all the studies, the age range of participants was 18 to 89 years, and the participants’ follow-up range was 6 to 65 months for the cross-sectional studies and 8 to 120 months for the randomized studies.

3.3. Characteristics of the studies

A RR of 1.09 (95% CI, 0.71–1.67) was obtained when all studies were pooled (Fig. 1), meaning that those patients who underwent a surgical resection did not show an improvement in survival compared with those patients that were not treated by surgery. Also, we showed the funnel plot (Fig. 2) for these studies.

Figure 1.

Forest plot of the risk ratio for mortality in patients with gastric cancer.

Figure 2.

Funnel plot of the risk ratio for mortality in patients with gastric cancer.

Following the general analysis, study subgroups were analyzed. Cross-sectional studies were grouped, and a RR of 0.82 (95% CI, 0.63–1.07) was obtained (Fig. 3). In the same way, randomized studies showed a RR of 2.08 (95% CI, 0.25–17.07) (Fig. 4). In summary, the RR was not significant in establishing whether surgery improves survival in patients with GC.

Figure 3.

Forest plot of the risk ratio for cross-sectional studies in patients with gastric cancer.

Figure 4.

Forest plot of the risk ratio for randomized studies in patients with gastric cancer.

4. Discussion

This study aimed to determine whether surgical resection increases the survival of patients with GC in the Mexican population. The results analyzed from 8 studies showed that the overall effect size of the RR did not improve the survival in patients with CG undergoing surgery.

Although minimally invasive techniques have been implemented in the treatment of GC or a personalized approach depending on the cancer stages, the primary method used as a curative treatment for GC is still surgical resection followed by adjuvant therapies.^[21]

Many risk factors are related to GC development, such as dietary habits and mutations due to chemical exposition.^[22,23] In Mexico, the recognized risk factors related to GC development are smoking, alcoholism, processed meat consumption, nitrosamines, Epstein-Barr virus infections, and high salt intake.^[24,25] Moreover, the population of Mexico has severe comorbidities and unhealthy dietary habits, and our health system still has limited social security for attending to all patients with GC. In addition, the prognosis worsens because many patients with comorbidities arrive at the hospitals with a late diagnosis and advanced stages of the disease^[26]; Likewise, comorbidities show considerable challenges to cancer management and contribute to the treatment decision, and as a consequence impact the overall survival of patients. Unfortunately, in the included studies for this meta-analysis, the comorbidities of patients, such as health habits such as dietary habits, alcohol or tobacco intake, or any other information were not provided.

Many studies have described the clinicopathological characteristics of the GC, the effect of the palliatives in advanced stages, and the histopathological characteristics of GC, among other features, but few studies have described the impact of surgical resection on survival.

Among the studies selected in this work, we identify 2 methodological designs, cross-sectional and randomized studies. Also, the studies analyzed in this work were performed in the 80s, 90s, and 00s, and the vast majority of the sample represents only the central part of Mexico. This analysis does not include populations from the north and south of Mexico.

Although there is new evidence in cancer prevention (methods of diagnosis, control, and therapies), there is a lack of information about the effect of surgical procedures and survival in patients with GC. In our analysis, only 1 study conducted by Oñate-Ocaña et al^[15] had shown that surgical resections in patients with GC were adequate to extend the survival of patients. The study showed that the primary tumor was in the esophagogastric junction, most of the cases presented did not show metastases, hence, being the initial cancer stages (I–II ~46%). The small diameter of the esophageal lumen generates dysphagia, resulting in hypoxia and excessive weight loss among patients. The endoscopic study allows a timely diagnosis and, consequently, removal of the localized primary tumor, contributing to the complete recovery of the patients.^[4,27] Medina-Franco et al^[20] obtained similar results that Oñate-Ocaña et al^[15] when they studied patients with primary gastric non-Hodgkin lymphoma at earlier stages of cancer (I–II ~ 46%). The authors observed that surgical resection produced remission of cancer. Unfortunately, the RR was not significant to establish that surgery improves survival in patients with GC.

Studies performed by Green et al,^[14] Ramos-de la Medina et al,^[19] and Oñate-Ocaña et al^[16] reported that surgical resection did not show an improvement in the survival in patients with GC. All studies included patients with advanced cancer stages and an overall survival time of 12, 10, and 25 months respectively. Green et al determined that advanced-stage patients can receive surgery as a palliative procedure, with a subsequent follow-up of chemotherapy sessions. Ramos-de la Medina et al included patients with a family history of GC. The results showed no significant impact on survival in patients with or without a family history of GC. Finally, Oñate-Ocaña et al determined that surgical resection in patients with advanced GC should only be performed if there is a minimum chance of survival for the patient. Unfortunately, 84% of the patients with aggressive surgical resection died.

In a randomized study by Gallardo-Rincon et al,^[13] the authors reported that patients with unresectable gastric adenocarcinoma showed a mean survival time of 13 months after partial gastrectomy resection and observed that patients required adjuvant chemotherapy to achieve such survival. The partial gastrostomies showed a few postoperative complications, such as fistulas and wound dehiscence, and provided benefits to the patient such as better nutritional status, better functional capacity, and better hospital stay.^[28,29] In 2004, Aviles et al^[17] studied the role of surgery in primary gastric lymphoma through controlled clinical trials. They observed that almost half of the participants survived after the surgical procedure, and 96% without surgery (only chemotherapy and/or radiotherapy) survived before the follow-up of 120 months. Gastric non-Hodgkin lymphoma has good survival even without oncological management, and its presence is usually associated with Helicobacter pylori infection. Therefore, surgery is not the best option for patients with early stages diagnoses, but the pharmacological treatment of bacteria has been shown to be the best approach together with radiotherapy.^[30,31] In conjunction, they have shown an overall survival of >10 years. Another randomized study performed in 2006 by Aviles et al^[18] determined that the combination of surgery and chemotherapy in patients with gastric lymphoma increases survival time when they are identified and treated at early stages. The RRs of 3 previous studies were not concluding in establishing whether surgery improves survival in patients with GC.

As mentioned above, when all studies were pooled, the RR was not significant for improvement in the survival in patients between surgical and nonsurgical treatments (Fig. 1), suggesting that surgical resection as a curative treatment does not increase the time survival in the patients (RR 1.09; 95% CI, 0.71–1.67). Nonetheless, the obtained pooled RR showed high heterogeneity among studies (I² = 93%). The heterogeneity confirms that the variability in the survival reported in each study was perhaps caused by the different cancer stages included in the studies.

The high survival in patients with and without surgery reported by Aviles et al (2004) introduced bias in the forest plot analysis. Hence, the estimation of the effect when the study was removed and presented in a new forest plot was RR of 0.82 (95% CI, 0.66–1.03) and a heterogeneity of I² = 70%, which shows that regardless of the Aviles et al study, the surgery had not to impact on the overall survival in GC patients (Supplementary Figures S3 and S4, Supplemental Digital Content, http://links.lww.com/MD/J57).

The role of surgery in the survival of patients with GC has been controversial, so it has recently been reported that surgical procedures not only have cancer remission purposes but the surgeries can also help to deal with the symptoms of GC.^[32] Despite surgical resection being the primary method used as a curative treatment for GC, there is no agreement on using other procedures such as robotic or laparoscopic surgeries. Open surgery was the most used in the studies included in the present meta-analysis; only Oñate-Ocaña et al (2007) reported that laparoscopy procedures were performed in some of the surgical procedures without concluding remarks about it. In our study, the low survival reported could have been due to the advanced stages of cancer in the patients and the postoperative complications. In this regard, postoperative complications can reduce overall survival in patients with GC by 47.6% in the next 5 years.^[33] Among postoperative complications, we identified postoperative infections, digestive bleeding, surgical dehiscence, and thrombosis. However, the information provided was insufficient to conclude their impact on survival in patients with GC.

This meta-analysis demonstrates for the first time that surgical procedures have not contributed to the survival of Mexican GC patients. Similar results were obtained by Alkhaffaf et al reporting the inconsistencies in the outcome measures in trials examining therapeutic, and surgical interventions for GC.^[34] Their study populations included patients from South Korea, Japan, and China (>87% of studies population), Poland, and Italy. Regarding overall survival, their results were inconsistent. In fact, 41% of clinical control trials analyzed studies showed 84% of mortality.

Several studies agree that an early diagnosis through endoscopic procedures could contribute to a better disease prognosis. Regions such as Korea and Japan with a high incidence of GC have campaigns promoting a universal scheme of screening; therefore, in Japan, upper digestive tract endoscopies are performed annually for adults over 50 years old.^[35] Meanwhile, in Korea, the screening is carried out on people between 40 and 75 years with a biannual periodicity.^[36]

In our country, GC should be considered a priority because is one of the main causes of hospital morbidity in the Mexican population over 65 years old.^[24] Hence, health personnel together with government authorities, should implement specific strategies focused on: a national screening of persons with suspected risk factors for developing GC; a complete record of the clinic history of patients, which includes the diagnosis, treatment, complications, and outcome of the detected cases; and finally, strategies for the surgical treatment that increase the survival of the patients with GC.

In this meta-analysis and systematic review, we showed that surgical procedures, as a curative treatment, do not increase the overall survival in patients with GC, at least in patients affected by advanced stages of cancer; therefore, preventive measures are urgent. Also, these results show us that new studies that provide more information about the disease in the Mexican population are required. Moreover, the implementation of new therapeutic alternatives for the treatment of GC is needed.

A timely diagnosis of GC is possible by identifying initial symptoms and performing endoscopy in people with suspected GC or with a family history of the pathology. These actions can help to take decisions in performing surgical procedures at the early stages of the disease, and the implementation of treatments with neoadjuvant therapies could increase patient survival.

4.1. Strengths and limitations

This study is the first meta-analysis to determine the role of surgical resection on the survival of patients with GC in the Mexican population. Among the limitations of the study, the population analyzed in this meta-analysis was only from the center of Mexico, excluding patients with GC from other regions of Mexico. The underdiagnosis of GC in our country did not allow us to identify other strategies for the management of the disease. The misinformation and the aggressiveness of GC favor tumor proliferation and metastases, as well as the lack of self-care and prevention in the patient. The lack of information on the risk factors in the studies and the higher indices of diabetes, hypertension, and obesity in the Mexican population all together contribute to reducing the survival of the population. In addition, there is no mandatory national screening of GC through endoscopy procedures in people >65 years of age or with a clinical or familial history of cancer. And finally, the studies did not describe the specific type of surgery (open surgery, partial or total gastrectomy) and type of adjuvant therapy (neoadjuvant, concomitant), and there is no record of this information in a nationwide public database.

5. Conclusions

This work is the first systematic study to assess the role of surgery on the survival of patients with GC in the Mexican population. The results showed that surgical resection did not improve survival in patients with GC. Future survival studies need to be performed considering appropriate cancer stratification, as well as controlled clinical trials, are needed to objectively establish the role of surgery as treatment in GC.

Acknowledgments

We are grateful to Julio Vega MSc, for helping with the statistical analysis.

Author contributions

Conceptualization: Ana Ligia Gutiérrez-Solis, Azalia Avila-Nava, Roberto Lugo.

Data curation: Oscar Daniel Pacheco-Can, Hayde Sara Leticia Vázquez-Segura, Alfredo Geovanny Pech-Aguilar, Carlos David Franco-González.

Formal analysis: Ana Ligia Gutiérrez-Solis, Roberto Lugo.

Funding acquisition: Roberto Lugo.

Investigation: Oscar Daniel Pacheco-Can, Hayde Sara Leticia Vázquez-Segura, Alfredo Geovanny Pech-Aguilar, Carlos David Franco-González.

Methodology: Ana Ligia Gutiérrez-Solis, Azalia Avila-Nava, Roberto Lugo.

Writing – original draft: Ana Ligia Gutiérrez-Solis, Oscar Daniel Pacheco-Can, Hayde Sara Leticia Vázquez-Segura, Alfredo Geovanny Pech-Aguilar, Carlos David Franco-González, Roberto Lugo.

Writing – review & editing: Ana Ligia Gutiérrez-Solis, Azalia Avila-Nava, Roberto Lugo.