Systematic review and meta-analysis of endoscopic ultrasonography in staging diagnosis of esophageal cancer after neoadjuvant radiotherapy and chemotherapy

Introduction

Esophageal cancer, as a common malignant tumor of digestive tract, has a high morbidity and mortality rate (1,2). Because of its insidious onset and high malignancy, patients are often in the advanced stage at the time of seeing a doctor, and sometimes other parts are found to be invaded during surgery, so radical resection cannot be performed (3-5). Therefore, it is very important to accurately evaluate the preoperative staging of patients. For esophageal cancer patients, accurate preoperative staging is helpful to provide the most appropriate individualized treatment plan for these patients (6-10). For example, some patients with preoperative stage T4 can increase the chance of radical resection through neoadjuvant therapy (11,12). For patients with T1, N0 and M0 stages, especially those with high surgical risk, endoscopic mucosal resection or endoscopic submucosal dissection can avoid the possible complications of surgical treatment and strive for radical treatment with minimal trauma (13-15).

At present, the most commonly used methods for preoperative staging of esophageal cancer mainly include computed tomography (CT), endoscopic ultrasonography (EUS) and positron emission tomography (PET) (16-21). Among them, CT, especially enhanced CT, is the most commonly used examination method to exclude distant metastasis of esophageal cancer patients, which can accurately find the most common liver, brain and lung metastasis (21,22). However, it has a limited role in judging the accurate depth of tumor invasion into esophageal wall, and the accuracy of evaluating T staging is poor (23). PET is more sensitive than CT in the diagnosis of primary tumors, but its limited information on the depth of tumor invasion limits its role in T staging (24,25). EUS is a gastrointestinal tract examination technology that combines an endoscope and ultrasound, and it plays a vital role in the staging of esophageal cancer and determination of the origin and depth of tumors (26,27). In terms of the structure of EUS, a miniature high-frequency ultrasonic probe is installed on the top of the endoscope, and an ultrasonic scan can be performed by the endoscope entering the body to obtain the ultrasonic images of the features of gastrointestinal tissue. The obtained images can be of some help to more intuitive and accurate diagnosis of disease and the implementation of follow-up targeted treatment plans (28,29). EUS can measure the diameter and cross-sectional area of the esophageal wall and the gastric wall by ultrasound. A study (30) indicates that the percentage of the reduction in the maximum cross-sectional area of EUS is closely related to the efficacy of neoadjuvant therapy. If the maximum cross-sectional area of EUS is reduced by over 50%, neoadjuvant chemotherapy is effective for tumors.

This study was innovatively incorporated into the current literature research on EUS in evaluating the staging of esophageal cancer after neoadjuvant chemotherapy at home and abroad, and evaluated the diagnostic ability of EUS in staging of esophageal cancer after neoadjuvant chemotherapy through meta-analysis system, so as to evaluate the reference value of EUS in the analysis and diagnosis of esophageal cancer, and provide theoretical reference for clinical diagnosis of preoperative staging of esophageal cancer. We present the following article in accordance with the PRISMA-DTA reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-437/rc).

Methods

Article retrieval

PubMed, this paper uses computer to search PubMed, MEDLINE, EBSCO, Science Direct, Cochrane Library and CNKI. With “ethical ultrasound, esophageal cancer, neo adjuvant chemotherapy, diagnosis, tumor node metastasis” as the detection words, the relevant literatures published from the database establishment to January 2022 were searched manually, and professional journals were searched to avoid omissions. In addition, the research object of the literature search was human beings.

In the retrieval process, subject words and free words were combined to carry out multiple retrievals to obtain the references that could be included. After that, a search engine was used to trace each article. Rev Man 5.3 software provided by Cochrane collaboration network was used to evaluate the risk of inclusion in the literature.

Article inclusion and exclusion criteria

The articles were included based on the following criteria:

• Patients were engaged in the study on neoadjuvant chemotherapy before surgery;
• Histopathological diagnostic result was the gold standard;
• Patients were adults over 18 years old;
• True positive, false positive, false negative, and true negative values of staging diagnosis could be obtained directly or indirectly;
• Articles were published in English language.

The articles were excluded based on the following criteria:

• Articles were case reports, overviews, conferences, letters, and reviews;
• Histopathological features were not the gold standard;
• The research objects were animals or in vitro study was involved;
• There was no sufficient data determining diagnostic indexes in articles;
• TNM classification system was not used to conduct the study on esophageal cancer staging.

Data extraction

Two professionals used uniform Microsoft Excel (Microsoft, the United States) to screen articles and extract data independently. The main data extracted include the final results. If there was a disagreement, they resolved it by discussion. The main extracted data were as follows:

• General data included in articles, such as title, first author, and publication year;
• The basic features of research objects, including sample size and detection methods;
• Diagnostic index test results;
• The detection rate of esophageal cancer patients at each stage after neoadjuvant chemotherapy in each article and the data that determined the accuracy of tests (sensitivity and specificity).

Evaluation criteria of articles

Quality assessment of diagnostic accuracy studies (QUADAS) recommended by Cochrane (the United States) criteria was adopted to evaluate the quality of the included articles. According to each evaluation index, the quality of the original included articles was evaluated. Each article was evaluated as “yes”, “no”, or “uncertain”.

Statistical methods

Rev Man 5.3 software (Cochrane, the United States) and Stata software (Stata Corp, the United States) were used to draw the bias risk assessment map. Furthermore, Q-test and heterogeneity (I²) were used to evaluate the heterogeneity among each article. The sensitivity and specificity of the assessment of each esophageal cancer stage using EUS were calculated, compared, and expressed by 95% confidence interval (CI). In addition, forest plots and summary receiver operating characteristic (SROC) curves were drawn. Funnel plots of different diagnostic indexes were adopted to test potential publication bias and carry out sensitivity analysis.

Results

Retrieval results and basic information about articles

A total of 211 articles were obtained by database retrieval [173] and manual journal retrieval [38]. In 173 articles, 4 articles were duplicates and 68 disqualified articles were excluded. Furthermore, 11 articles were also excluded for other reasons (There is a problem with the statistical method, the sample size is too small). The remaining 90 articles were initially selected. After that, 32 articles (there is content in the abstract and title that is not relevant to this article) were excluded by reading abstracts and titles, and there were 58 remaining articles. In addition, 29 research reports and review articles were excluded, and there were 29 remaining articles. By reading the full-text of all remaining articles, 12 articles with incorrect research types were excluded, and 4 articles were also excluded because the required diagnostic results were incomplete or unavailable. TNM staging criteria were not used in 1 article, so this article was excluded Finally, a total of 12 articles were included in meta-analysis. In 38 articles, 25 research reports and review articles were excluded, and there were 13 remaining articles. By reading the full-text of all remaining articles, 4 articles with incorrect research types were excluded, 7 articles with incorrect research types were excluded, and 2 articles were also excluded because the required diagnostic results were incomplete or unavailable. Figure 1 shows the process of article retrieval.

Figure 1 Article retrieval process.

By reading the content of the included articles, the basic information about the articles was extracted. Among the 12 included articles, there were 824 patients receiving esophageal cancer neoadjuvant chemotherapy, and the sample size ranged from 17 to 143. In the 12 included articles, each index of the preoperative diagnostic staging for esophageal cancer patients using EUS after neoadjuvant chemotherapy was described in detail. Furthermore, vital histopathological staging diagnostic results were the gold standard in all 12 articles. In 11 articles, T1, T2, and T3 patients were involved, T4 esophageal cancer patients were included in 7 articles, and N0 and N1 esophageal cancer patients were involved in 9 articles. The results of the evaluation of the quality of 12 included articles demonstrated that 8 articles were rated level A (66.66%), 2 articles were rated level B (16.67%), and 2 articles were rated level C (16.67%). Table 1 shows the basic features of the included articles. Figure 2 displays the evaluation of the risk bias of the references drawn with Rev Man 5.3. Figure 3 illustrates the summary of the risk bias of references.

Figure 2 Risk bias evaluation diagram of included articles.

Figure 3 Risk bias evaluation and summary diagram of included articles. “+” referred to low risk, “−” represented high risk, and “?” denoted unclear.

Evaluation results of heterogeneity

The heterogeneity of the diagnosis of each stage using EUS in the included articles was evaluated. There was heterogeneity in the sensitivity and specificity among each article (I²=88.55%, 75.44%). In terms of the heterogeneity results of the diagnosis of T2 stage, there was heterogeneity in the sensitivity and specificity among each article (I²=68.39%, 70.07%). The heterogeneity results of the diagnosis of T3 stage showed that there was heterogeneity in the sensitivity and specificity among each article (I²=76.79%, 90.24%). The heterogeneity results of the diagnosis of T4 stage indicated that there was little heterogeneity in the sensitivity and specificity among each article (I²=35.15%, 49.78%). The heterogeneity results of the diagnosis of N0 stage revealed that there was little heterogeneity in the sensitivity and specificity among each article (I²=59.40%, 45.87%). The heterogeneity results of the diagnosis of N1 stage demonstrated that there was heterogeneity in the sensitivity and specificity among each article (I²=47.54%, 62.07%). There was heterogeneity among the diagnostic data of EUS for each stage, and it needed to be summarized and analyzed by random effect model and fitted with summary receiver operating characteristic (SROC) curve.

Meta-analysis of diagnosis of T1 stage using EUS

In the 12 included articles (31-42), the diagnostic results of T1 stage using EUS in diagnostic experiments were analyzed. Figure 4 is a forest plot showing the sensitivity and specificity of individual studies and summary studies at T1 stage. Furthermore, heterogeneity test was carried out for the sensitivity of the diagnosis of T1 stage in the 12 included articles, and the results showed that Q=96.06, degree of freedom (df) =11.00, I²=88.55%, and P=0.00, which indicated that there was heterogeneity among each research group. Furthermore, the combined sensitivity was 0.16 with a 95% CI was 0.05–0.39. The lowest sensitivity was 0.00 and 95% CI was 0.00–0.37. The highest sensitivity reached 1.00 and 95% CI was 0.03–1.00. In addition, heterogeneity test was conducted for the specificity of the diagnosis of T1 stage in the 12 included articles. The results revealed that Q=44.79, df =11.00, I²=75.44%, and P=0.00, which suggested that there was heterogeneity among each research group. The combined specificity was 0.99 and 95% CI was 0.94–1.00. The lowest specificity was 0.82 and 95% CI was 0.60–0.95. The highest specificity was 1.00 and 95% CI was 0.96–1.00. Figure 5 was SROC curve of T1 staging diagnosis. If the SROC was closer to the upper left corner of the image, the area under the SROC curve became larger with higher diagnostic accuracy. The results of T1 staging diagnosis showed that the proportion of false negatives and false positives was low, and the diagnostic accuracy was high.

Figure 4 Forest plot of sensitivity and specificity of individual study and summary study on the diagnosis of T1 stage using EUS. CI, confidence interval; df, degree of freedom; EUS, endoscopic ultrasound.

Figure 5 SROC curve of the diagnosis of T1 stage using EUS. SROC, summary receiver operating characteristic; AUC, area under curve; SENS, sensitivity; SPEC, specificity; EUS, endoscopic ultrasound.

Meta-analysis of diagnosis of T2 stage using EUS

In the 12 included articles, the diagnostic results of T2 stage using EUS in diagnostic experiments were analyzed. Figure 6 is a forest plot showing the sensitivity and specificity of individual studies and summary studies at T2 stage. A heterogeneity test was conducted for the sensitivity of T2 staging diagnosis in the 12 included articles. The results demonstrated that Q=34.80, df =11.00, I²=68.39%, and P=0.00, which showed that there was heterogeneity among each research group. The combined sensitivity was 0.34 and 95% CI was 0.20–0.52. The lowest sensitivity was 0.07 and 95% CI was 0.01–0.24. The highest sensitivity was 1.00 and 95% CI was 0.63–1.00. In addition, heterogeneity test was performed on the specificity of T2 staging diagnosis in the 12 included articles. The results showed that Q=36.75, df =11.00, I²=70.07%, and P=0.00, which demonstrated that there was heterogeneity among each research group. The combined specificity was 0.80 and 95% CI was 0.74–0.85. The lowest specificity was 0.63 and 95% CI was 0.50–0.74. The highest specificity was 1.00 and 95% CI was 0.81–1.00. Figure 7 displays the SROC curve of T2 staging diagnosis. If the SROC curve was closer to the upper left corner of the image, the area under the SROC curve became larger with higher diagnostic accuracy. The results of T2 staging diagnosis indicated that the proportion of false negatives and false positives was high with low diagnostic accuracy.

Figure 6 Forest plot of sensitivity and specificity of individual study and summary study on the diagnosis of T2 stage using EUS. CI, confidence interval; df, degree of freedom; EUS, endoscopic ultrasound.

Figure 7 SROC curve of the diagnosis of T2 stage using EUS. SROC, summary receiver operating characteristic; AUC, area under curve; SENS, sensitivity; SPEC, specificity; EUS, endoscopic ultrasound.

Meta-analysis of diagnosis of T3 stage using EUS

In the 12 included articles, the diagnostic results of T3 stage using EUS in diagnostic experiments were analyzed. Figure 8 is a forest plot showing the sensitivity and specificity of individual studies and summary studies at T3 stage. A heterogeneity test was conducted for the sensitivity of T3 staging diagnosis in the 12 included articles. The results showed that Q=47.39, df =11.00, I²=76.79%, and P=0.00, which indicated that there was heterogeneity among each research group. The combined sensitivity was 0.78 and 95% CI was 0.63–0.88. The lowest sensitivity was 0.38 and 95% CI was 0.14–0.68. The highest sensitivity was 1.00 and 95% CI was 0.85–1.00. Furthermore, heterogeneity test was conducted for the specificity of T3 staging diagnosis in the 12 included articles. The results revealed that Q=112.69, df =11.00, I²=90.24%, and P=0.00, which demonstrated that there was high heterogeneity among each research group. The combined specificity was 0.52 and 95% CI was 0.36–0.68. The lowest specificity was 0.17 and 95% CI was 0.09–0.28. The highest specificity was 1.00 and 95% CI was 0.59–1.00. Figure 9 presents the SROC curve of T3 staging diagnosis. If the SROC curve was closer to the upper left corner of the image, the area under the SROC curve became larger with higher diagnostic accuracy. The results of T3 staging diagnosis showed that the proportion of false negatives and false positives was high with low diagnostic accuracy.

Figure 8 Forest plot of sensitivity and specificity of individual study and summary study on the diagnosis of T3 stage using EUS. CI, confidence interval; df, degree of freedom; EUS, endoscopic ultrasound.

Figure 9 SROC curve of the diagnosis of T3 stage using EUS. SROC, summary receiver operating characteristic; AUC, area under curve; SENS, sensitivity; SPEC, specificity; EUS, endoscopic ultrasound.

Meta-analysis of diagnosis of T4 stage using EUS

In 7 included articles (33-39), the diagnostic results of T4 stage using EUS in diagnostic experiments were analyzed. Figure 10 is a forest plot showing the sensitivity and specificity of individual studies and summary studies at T4 stage. A heterogeneity test was conducted for the sensitivity of T4 staging diagnosis in 7 included articles. The results revealed that Q=9.25, df =6.00, I²=35.15%, and P=0.16, which demonstrated that there was low heterogeneity among each research group. The combined sensitivity was 0.16 and 95% CI was 0.04–0.50. The lowest sensitivity was 0.00 and 95% CI was 0.00–0.52. The highest sensitivity was 1.00 and 95% CI was 0.03–1.00. In addition, heterogeneity test was conducted for the specificity of T4 staging diagnosis in 7 included articles. The results indicated that Q=11.95, df =6.00, I²=49.78%, and P=0.06, which demonstrated that there was heterogeneity among each research group. The combined specificity was 0.98 and 95% CI was 0.95–0.99. The lowest specificity was 0.91 and 95% CI was 0.78–0.97. The highest specificity was 1.00 and 95% CI was 0.97–1.00. Figure 11 displays the SROC curve of T4 staging diagnosis. If the SROC curve was closer to the upper left corner of the images, the area under the SROC curve became larger with higher diagnostic accuracy. The results of T4 staging diagnosis showed that the proportion of false negatives and false positives was low with high diagnostic accuracy.

Figure 10 Forest plot of sensitivity and specificity of individual study and summary study on the diagnosis of T4 stage using EUS. CI, confidence interval; df, degree of freedom; EUS, endoscopic ultrasound.

Figure 11 SROC curve of the diagnosis of T4 stage using EUS. SROC, summary receiver operating characteristic; AUC, area under curve; SENS, sensitivity; SPEC, specificity; EUS, endoscopic ultrasound.

Meta-analysis of diagnosis of N0 stage using EUS

In 9 included articles (32-35,37-39,41,42), the diagnostic results of N0 stage using EUS in diagnostic experiments were analyzed. Figure 12 is a forest plot showing the sensitivity and specificity of individual studies and summary studies at N0 stage. A heterogeneity test was conducted for the sensitivity of N0 staging diagnosis in 9 included articles. The results demonstrated that Q=19.70, df =8.00, I²=59.40%, and P=0.01, which showed that there was heterogeneity among each research group. The combined sensitivity was 0.62 and 95% CI was 0.53–0.71. The lowest sensitivity was 0.45 and 95% CI was 0.17–0.77. The highest sensitivity was 1.00 and 95% CI was 0.03–1.00. In addition, heterogeneity test was implemented for the specificity of N0 staging diagnosis in 9 included articles. The results showed that Q=14.78, df =8.00, I²=45.87%, and P=0.06, which revealed that there was heterogeneity among each research group. The combined specificity was 0.65 and 95% CI was 0.58–0.71. The lowest specificity was 0.38 and 95% CI was 0.18–0.62. The highest specificity was 0.83 and 95% CI was 0.36–1.00. Figure 13 presents the SROC curve of N0 staging diagnosis. If the SROC curve was closer to the upper left corner of the images, the area under the SROC curve became larger with higher diagnostic accuracy. The results of N0 staging diagnosis showed that the proportion of false negatives and false positives was high with low diagnostic accuracy.

Figure 12 Forest plot of sensitivity and specificity of individual study and summary study on the diagnosis of N0 stage using EUS. CI, confidence interval; df, degree of freedom; EUS, endoscopic ultrasound.

Figure 13 SROC curve of the diagnosis of N0 stage using EUS. SROC, summary receiver operating characteristic; AUC, area under curve; SENS, sensitivity; SPEC, specificity; EUS, endoscopic ultrasound.

Meta-analysis of diagnosis of N1 stage using EUS

In 9 included articles, the diagnostic results of N1 stage using EUS in diagnostic experiments were analyzed. Figure 14 is a forest plot showing the sensitivity and specificity of individual studies and summary studies at N1 stage. Heterogeneity test was carried out for the sensitivity of N1 staging diagnosis in 9 included articles. The results showed that Q=15.25, df =8.00, I²=47.54%, and P=0.05, which demonstrated that there was heterogeneity among each research group. The combined sensitivity was 0.65 and 95% CI was 0.58–0.72. The lowest sensitivity was 0.38 and 95% CI was 0.18–0.62. The highest sensitivity was 0.83 and 95% CI was 0.36–1.00. In addition, heterogeneity test was implemented for the specificity of N1 staging diagnosis in 9 included articles. The results indicated that Q=21.09, df =8.00, I²=62.07%, and P=0.01, which showed that there was heterogeneity among each research group. The combined specificity was 0.63 and 95% CI was 0.54–0.72. The lowest specificity was 0.45 and 95% CI was 0.17–0.77. The highest specificity was 1.00 and 95% CI was 0.03–1.00. Figure 15 displays the SROC curve of N1 staging diagnosis. If the SROC curve was closer to the upper left corner of the image, the area under the SROC curve became larger with higher diagnostic accuracy. The results of N1 staging diagnosis demonstrated that the proportion of false negatives and false positives was high with low diagnostic accuracy.

Figure 14 Forest plot of sensitivity and specificity of individual study and summary study on the diagnosis of N1 stage using EUS. CI, confidence interval; df, degree of freedom; EUS, endoscopic ultrasound.

Figure 15 SROC curve of the diagnosis of N1 stage using EUS. SROC, summary receiver operating characteristic; AUC, area under curve; SENS, sensitivity; SPEC, specificity; EUS, endoscopic ultrasound.

Sensitivity analysis

Sensitivity analysis was carried out by changing the analysis model. The results of the meta-analysis and the summary results of the application of different analysis models showed no obvious changes, which indicated that the included articles showed good stability.

Discussion

Esophageal cancer is a common malignant tumor characterized by the invasion of peripheral tissues and metastasis. In recent years, surgical operational and medical technical levels have constantly improved, but the postoperative 5-year survival rate of patients with esophageal cancer is still extremely low. Simple surgical treatment cannot meet the clinical needs of patients with esophageal cancer (43,44). To further enhance the long-term survival rate of patients with esophageal cancer, the implementation of neoadjuvant chemotherapy before surgery attracts extensive attention from medical staff and it shows good therapeutic effects (45). Multiple studies show that neoadjuvant chemotherapy can effectively improve the long-term survival rate of patients with esophageal cancer and alleviate patients’ clinical symptoms (46,47). Nonetheless, neoadjuvant chemotherapy shows some limitations at present. Clinically, there no effective assessment method for the therapeutic effects of neoadjuvant chemotherapy. Furthermore, patients cannot be staged accurately, and the misjudgment of the staging for patients with esophageal cancer after neoadjuvant chemotherapy will affect the direction of subsequent treatment for patients (48).

EUS is mainly to observe and evaluate the depth of cancer invasion into esophageal wall, the extent of invasion into mediastinum and swollen lymph nodes in mediastinum to complete the diagnosis and staging of patients (49). Furthermore, EUS can measure the diameter and cross-sectional area of the esophageal wall and gastric wall by ultrasound. Relevant study (50) demonstrates that the percentage of the reduction in the maximum cross-sectional area of EUS is closely related to the efficacy of neoadjuvant therapy. If the maximum cross-sectional area of EUS is reduced by over 50%, neoadjuvant chemotherapy is effective for tumors. Clinically, the studies on the adoption of EUS to assess TNM staging after neoadjuvant chemotherapy for patients with esophageal cancer flourish. Among each study, the accuracy of determining the depth of esophageal cancer infiltration using EUS was different. According to a study (51), the accurate judgment of clinical staging using EUS can improve the clinical therapeutic effects on about three-quarters of patients. Furthermore, over half of all patients offer valid information to subsequent treatment plans after EUS examination, which reduces the risks that patients face and reduces treatment costs. Current domestic and foreign articles on the evaluation of the efficacy of neoadjuvant chemotherapy on esophageal cancer using EUS were included. A meta-analysis system was used to assess the diagnostic capacity of esophageal cancer staging after neoadjuvant chemotherapy using EUS to evaluate the references values of EUS in analyzing and diagnosing esophageal cancer, which provided a theoretical reference and basis for the assessment of therapeutic effects of neoadjuvant chemotherapy on esophageal cancer.

In the meta-analysis, the diagnostic accuracy of the clinical staging after neoadjuvant chemotherapy for patients with esophageal cancer using EUS was estimated and evaluated. A total of 12 articles were included and the diagnosis of T1-T4 sensitivity was assessed with a heterogeneity test. The results were as follows—T1: Q=96.06, df =11.00, I²=88.55%, and P=0.00; T2: Q=34.80, df =11.00, I²=68.39%, and P=0.00; T3: Q=47.39, df =11.00, I²=76.79%, and P=0.00; T4: Q=9.25, df =6.00, I²=35.15%, and P=0.16. The results demonstrated that there was heterogeneity among each research group. The combined sensitivity was 0.16 with a 95% CI of 0.05–0.39, 0.34 with a 95% CI of 0.20–0.52, 0.78 with a 95% CI of 0.63–0.88, and 0.16 with a 95% CI of 0.04–0.50. Furthermore, the specificity of T staging diagnosis using EUS in the 12 included articles was assessed with a heterogeneity test. T1 to T4 were Q=44.79, df =11.00, I²=75.44%, and P=0.00; Q=36.75, df =11.00, I²=70.07%, and P=0.00; Q=111.69, df =11.00, I²=90.24%, and P=0.00; and Q=9.25, df =6.00, I²=35.15%, and P=0.16. The results showed that there was heterogeneity among each research group. The combined specificities were 0.99 with a 95% CI of 0.94–1.00, 0.80 with a 95% CI of 0.74–0.85, 0.52 with a 95% CI of 0.36–0.68, and 0.98 with a 95% CI of 0.95–0.99. In addition, the sensitivity of N0 and N1 diagnosis using EUS was assessed with a heterogeneity test. The results showed that Q=19.70, df =8.00, I²=59.40%, P=0.01 and Q=15.25, df =8.00, I²=47.54%, P=0.05, which indicated that there was no heterogeneity among each research group. The combined sensitivities were 0.62 with a 95% CI of 0.53–0.71 and 0.65 with a 95% CI of 0.58–0.72. Furthermore, the specificity of N0 and N1 diagnosis using EUS in 9 included articles was assessed with a heterogeneity test, which showed Q=14.78, df =8.00, I²=45.87%, and P=0.06; and Q=21.09, df =8.00, I²=62.07%, and P=0.01. The results demonstrated that there was heterogeneity among each research group. The combined specificity was 0.65 with a 95% CI of 0.58–0.71 and 0.63 with a 95% CI of 0.54–0.72. In addition, the area under the SROC curve reflected the diagnostic values diagnostic methods possessed. A larger area under curve meant higher diagnostic values. The areas under curves of T1 and T4 diagnosed using EUS were 0.85 and 0.93, respectively, which indicated that they showed high diagnostic values. In addition, the sensitivity of TNM staging was generally low and the specificity was high, which was consistent with the study conducted by Sun et al. (52). The low sensitivity might be related to local fibrosis and inflammation in patients’ esophageal tissues after neoadjuvant chemotherapy, which could reduce tumor size. However, patients’ esophageal structure cannot return to normal. As a result, it was difficult to assess the depth of tumors accurately.

The diagnostic capacity of the clinical staging for patients with esophageal cancer after neoadjuvant chemotherapy using EUS was synthesized and assessed in the meta-analysis. The meta-analysis provided evidence-based suggestions on clinical practical guidance. In clinical studies, EUS could be used to accurately assess the clinical TNM staging of patients with esophageal cancer, which can offer more accurate reference for subsequent treatment.

Conclusions

The articles related to the diagnosis and assessment of the clinical staging for patients with esophageal cancer after neoadjuvant chemotherapy using EUS were selected and included in the meta-analysis. The meta-analysis investigated the accuracy of the evaluation of the clinical effects of neoadjuvant chemotherapy on patients with esophageal cancer using EUS. The results of the meta-analysis confirmed that the sensitivity of determining TNM clinical staging of patients with esophageal cancer using EUS was poor, while its specificity was good. Furthermore, EUS showed high accuracy in diagnosing patients at T1 and T4 stages. The significant heterogeneity among articles might be related to the fact that there was no uniform time criterion between neoadjuvant chemotherapy and EUS examination. Furthermore, the differences in sample size and design in articles also caused heterogeneity. A uniform criterion needs to be formulated. In addition, more samples and high-quality articles should be analyzed to provide a more accurate and effective basis for clinical practice.

Acknowledgments

Funding: The study was supported by Taizhou City Science and Technology Plan Class A Project (No. 1801ky24).

Footnote

Reporting Checklist: The authors have completed the PRISMA-DTA reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-437/rc

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-22-437/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Didamson OC, Abrahamse H. Targeted Photodynamic Diagnosis and Therapy for Esophageal Cancer: Potential Role of Functionalized Nanomedicine. Pharmaceutics 2021;13:1943. [Crossref] [PubMed]
2. Ghosh NR, Jones LA. Dietary risk factors for esophageal cancer based on World Health Organization regions. Nutrition 2022;95:111552. [Crossref] [PubMed]
3. Huang Z, Jin Y, Cai X, et al. Association of the programmed death ligand-1 combined positive score in tumors and clinicopathological features in esophageal cancer. Thorac Cancer 2022;13:523-32. [Crossref] [PubMed]
4. Ishaq HM, Mohammad IS, Sher Muhammad K, et al. Gut microbial dysbiosis and its association with esophageal cancer. J Appl Biomed 2021;19:1-13. [Crossref] [PubMed]
5. Eyck BM, Onstenk BD, Noordman BJ, et al. Accuracy of Detecting Residual Disease After Neoadjuvant Chemoradiotherapy for Esophageal Cancer: A Systematic Review and Meta-analysis. Ann Surg 2020;271:245-56. [Crossref] [PubMed]
6. Radlinski M, Martin LW, Walters DM, et al. Use of endoscopic ultrasound in pre-treatment staging of esophageal cancer did not alter management plan. J Thorac Dis 2020;12:5850-6. [Crossref] [PubMed]
7. Vazquez-Sequeiros E. Endosonography-guided fine-needle aspiration for re-evaluation of lymph node status after neoadjuvant therapy in patients with esophageal cancer: is there any role for it? Endoscopy 2020;52:164-5. [Crossref] [PubMed]
8. Vollenbrock SE, van Dieren JM, Voncken FEM, et al. Added value of MRI to endoscopic and endosonographic response assessment after neoadjuvant chemoradiotherapy in oesophageal cancer. Eur Radiol 2020;30:2425-34. [Crossref] [PubMed]
9. Ji Y, Du X, Chen M. Definitive Chemoradiotherapy for Older Patients With Esophageal Cancer-Reply. JAMA Oncol 2022;8:305-6. [Crossref] [PubMed]
10. Jin Z, Zhang J, Chen D, et al. Neoadjuvant chemoradiotherapy, chemotherapy, and radiotherapy do not significantly increase the incidence of anastomotic leakage after esophageal cancer surgery: a meta-analysis. Dis Esophagus 2021;doab089. [Crossref] [PubMed]
11. Knippen S, Duma MN. Esophageal cancer: improved prognosis with neoadjuvant chemoradiotherapy. Strahlenther Onkol 2022;198:86-8. [Crossref] [PubMed]
12. Li J, Xu J, Zheng Y, et al. Esophageal cancer: Epidemiology, risk factors and screening. Chin J Cancer Res 2021;33:535-47. [Crossref] [PubMed]
13. Li S, Qiu R, Yuan G, et al. Body composition in relation to postoperative anastomotic leakage and overall survival in patients with esophageal cancer. Nutrition 2022;94:111534. [Crossref] [PubMed]
14. Lin SH, Lee JM, Wu IH. Comparison of Clinical Outcomes between Salvage and Elective Thoracic Endovascular Aortic Repair in Patients with Advanced Esophageal Cancer with Aortic Invasion: A Retrospective Cohort Study. Biomedicines 2021;9:1889. [Crossref] [PubMed]
15. Liu X, Zhang YF, Zhan YL, et al. Study on the operation process and construction standard of an esophageal cancer screening cohort study. Zhonghua Liu Xing Bing Xue Za Zhi 2021;42:1504-8. [PubMed]
16. Mai Z, Liu Q, Wang X, et al. Integration of Tumor Heterogeneity for Recurrence Prediction in Patients with Esophageal Squamous Cell Cancer. Cancers (Basel) 2021;13:6084. [Crossref] [PubMed]
17. Nuytens F, Lenne X, Clément G, et al. Effect of Phased Implementation of Totally Minimally Invasive Ivor Lewis Esophagectomy for Esophageal Cancer after Previous Adoption of the Hybrid Minimally Invasive Technique: Results from a French Nationwide Population-Based Cohort Study. Ann Surg Oncol 2022;29:2791-801. [Crossref] [PubMed]
18. Okereke IC, Westra J, Tyler D, et al. Disparities in esophageal cancer care based on race: a National Cancer Database analysis. Dis Esophagus 2022;35:doab083. [Crossref] [PubMed]
19. Paydary K, Reizine N, Catenacci DVT. Immune-Checkpoint Inhibition in the Treatment of Gastro-Esophageal Cancer: A Closer Look at the Emerging Evidence. Cancers (Basel) 2021;13:5929. [Crossref] [PubMed]
20. Qi Y, Wu S, Tao L, et al. A Population-Based Study: How to Identify High-Risk T1-2 Esophageal Cancer Patients? Front Oncol 2021;11:766181. [Crossref] [PubMed]
21. Rho J, Quan YH, Choi BH, et al. Near-infrared fluorescent imaging with indocyanine green in rabbit and patient specimens of esophageal cancer. J Thorac Dis 2021;13:6314-22. [Crossref] [PubMed]
22. Shin CM. Treatment of Superficial Esophageal Cancer: An Update. Korean J Gastroenterol 2021;78:313-9. [Crossref] [PubMed]
23. Takeda K, Matsushita H, Umezawa R, et al. Hyperfractionated radiotherapy for re-irradiation of recurrent esophageal cancer. Radiat Oncol J 2021;39:265-9. [Crossref] [PubMed]
24. Tan L, Cheng D, Wen J, et al. Identification of prognostic hypoxia-related genes signature on the tumor microenvironment in esophageal cancer. Math Biosci Eng 2021;18:7743-58. [Crossref] [PubMed]
25. Tanaka I, Hirasawa D, Matsuda T. Extremely small but invasive esophageal cancer: endoscopists should not miss it. Gastrointest Endosc 2022;95:799-800. [Crossref] [PubMed]
26. Tanaka I, Hirasawa D, Togo D. A submucosal tumor developing 15 years after endoscopic resection with additional chemoradiotherapy for esophageal cancer. Gastroenterology 2022;163:e3-4. [Crossref] [PubMed]
27. Trindade AJ, Zhang J, Hauschild J, et al. Impact of Coronavirus Disease 2019 on the Diagnosis and Therapy for Barrett's Esophagus and Esophageal Cancer in the United States. Gastroenterology 2022;162:978-980.e6. [Crossref] [PubMed]
28. Wang XY, Maswikiti EP, Zhu JY, et al. Photodynamic therapy combined with immunotherapy for an advanced esophageal cancer with an obstruction post metal stent implantation: A case report and literature review. Photodiagnosis Photodyn Ther 2022;37:102671. [Crossref] [PubMed]
29. Xing W, Zhao L, Zheng Y, et al. The Sequence of Chemotherapy and Toripalimab Might Influence the Efficacy of Neoadjuvant Chemoimmunotherapy in Locally Advanced Esophageal Squamous Cell Cancer-A Phase II Study. Front Immunol 2021;12:772450. [Crossref] [PubMed]
30. Xu Y, Cui H, Dong T, et al. Integrating Clinical Data and Attentional CT Imaging Features for Esophageal Fistula Prediction in Esophageal Cancer. Front Oncol 2021;11:688706. [Crossref] [PubMed]
31. Bohle W, Kasper M, Zoller WG. Different accuracy of endosonographic tumor staging after neoadjuvant chemotherapy and chemoradiotherapy in esophageal cancer. Surg Endosc 2016;30:2922-8. [Crossref] [PubMed]
32. Bowrey DJ, Clark GW, Roberts SA, et al. Serial endoscopic ultrasound in the assessment of response to chemoradiotherapy for carcinoma of the esophagus. J Gastrointest Surg 1999;3:462-7. [Crossref] [PubMed]
33. DeWitt J, Kesler K, Brooks JA, et al. Endoscopic ultrasound for esophageal and gastroesophageal junction cancer: Impact of increased use of primary neoadjuvant therapy on preoperative locoregional staging accuracy. Dis Esophagus 2005;18:21-7. [Crossref] [PubMed]
34. Griffin JM, Reed CE, Denlinger CE. Utility of restaging endoscopic ultrasound after neoadjuvant therapy for esophageal cancer. Ann Thorac Surg 2012;93:1855-9; discussion 1860. [Crossref] [PubMed]
35. Heinzow HS, Seifert H, Tsepetonidis S, et al. Endoscopic ultrasound in staging esophageal cancer after neoadjuvant chemotherapy--results of a multicenter cohort analysis. J Gastrointest Surg 2013;17:1050-7. [Crossref] [PubMed]
36. Isenberg G, Chak A, Canto MI, et al. Endoscopic ultrasound in restaging of esophageal cancer after neoadjuvant chemoradiation. Gastrointest Endosc 1998;48:158-63. [Crossref] [PubMed]
37. Kalha I, Kaw M, Fukami N, et al. The accuracy of endoscopic ultrasound for restaging esophageal carcinoma after chemoradiation therapy. Cancer 2004;101:940-7. [Crossref] [PubMed]
38. Meister T, Heinzow HS, Osterkamp R, et al. Miniprobe endoscopic ultrasound accurately stages esophageal cancer and guides therapeutic decisions in the era of neoadjuvant therapy: results of a multicenter cohort analysis. Surg Endosc 2013;27:2813-9. [Crossref] [PubMed]
39. Misra S, Choi M, Livingstone AS, et al. The role of endoscopic ultrasound in assessing tumor response and staging after neoadjuvant chemotherapy for esophageal cancer. Surg Endosc 2012;26:518-22. [Crossref] [PubMed]
40. Schneider PM, Metzger R, Schaefer H, et al. Response evaluation by endoscopy, rebiopsy, and endoscopic ultrasound does not accurately predict histopathologic regression after neoadjuvant chemoradiation for esophageal cancer. Ann Surg 2008;248:902-8. [Crossref] [PubMed]
41. Willis J, Cooper GS, Isenberg G, et al. Correlation of EUS measurement with pathologic assessment of neoadjuvant therapy response in esophageal carcinoma. Gastrointest Endosc 2002;55:655-61. [Crossref] [PubMed]
42. Zuccaro G Jr, Rice TW, Goldblum J, et al. Endoscopic ultrasound cannot determine suitability for esophagectomy after aggressive chemoradiotherapy for esophageal cancer. Am J Gastroenterol 1999;94:906-12. [Crossref] [PubMed]
43. Zhang C, Luo CL, Shang GS, et al. Galangin Enhances Anticancer Efficacy of 5-Fluorouracil in Esophageal Cancer Cells and Xenografts Through NLR Family Pyrin Domain Containing 3 (NLRP3) Downregulation. Med Sci Monit 2021;27:e931630. [Crossref] [PubMed]
44. Zhang Y, Zhu M, Mo J, et al. Tumor microenvironment characterization in esophageal cancer identifies prognostic relevant immune cell subtypes and gene signatures. Aging (Albany NY) 2021;13:26118-36. [Crossref] [PubMed]
45. Zhao Y, Xu J, Chen Q. Analysis of Curative Effect and Prognostic Factors of Radiotherapy for Esophageal Cancer Based on the CNN. J Healthc Eng 2021;2021:9350677. [Crossref] [PubMed]
46. Zhong Y, Yang C, Wang N, et al. Hot Tea Drinking and the Risk of Esophageal Cancer: A Systematic Review and Meta-Analysis. Nutr Cancer 2022;74:2384-91. [Crossref] [PubMed]
47. Zhu YG, Xiao BF, Zhang JT, et al. Genetically Modified T Cells for Esophageal Cancer Therapy: A Promising Clinical Application. Front Oncol 2021;11:763806. [Crossref] [PubMed]
48. Yang X, You Z, Gong W, et al. Increased 68Ga-FAPI Uptake in Facet Joint Osteoarthritis in a Patient With Esophageal Cancer. Clin Nucl Med 2022;47:342-3. [Crossref] [PubMed]
49. Yoshida T, Nishino T, Goto M, et al. ypN0 in Patients With Definitive cN-positive Status After Preoperative Treatment Is a Prognostic Factor in Esophageal Cancer. Anticancer Res 2022;42:195-203. [Crossref] [PubMed]
50. Zhou J, Sun K, Wang S, et al. Associations between cancer family history and esophageal cancer and precancerous lesions in high-risk areas of China. Chin Med J (Engl) 2022;135:813-9. [Crossref] [PubMed]
51. Zhuo QF, Liu MQ, Li Z, et al. Effect of laparoscopic surgery for pancreatic cancer after neoadjuvant chemotherapy. Zhonghua Wai Ke Za Zhi 2022;60:134-9. [PubMed]
52. Sun F, Chen T, Han J, et al. Staging accuracy of endoscopic ultrasound for esophageal cancer after neoadjuvant chemotherapy: a meta-analysis and systematic review. Dis Esophagus 2015;28:757-71. [Crossref] [PubMed]

(English Language Editor: C. Mullens)