Abstract
Background/Aim: Currently, there are no standard guidelines for the waiting time from the diagnosis of gastric neoplasms to endoscopic submucosal dissection (ESD). Patients and Methods: A total of 1,605 patients who had undergone ESD for early gastric cancer (EGC) or high-grade dysplasia (HGD) were enrolled. Waiting time for ESD was defined as the time from the first diagnosis to ESD. Multivariable logistic regression analysis was conducted. Results: The curative resection rate was 86.8% and the mean waiting time was 36.8 days. In the multivariable model, longer waiting time did not significantly affect non-curative resection, whereas age >70 years, submucosal fibrosis, and initial cancer diagnosis were significantly associated with non-curative resection. Waiting time was still not identified as a risk factor for non-curative resection in EGC and HGD groups. Conclusion: A longer waiting time from diagnosis to ESD was not associated with non-curative resection.
Endoscopic submucosal dissection (ESD) is the most popular resection method and is widely used for the treatment of early gastric cancer (EGC) and gastric adenoma. This is because ESD enables a single piece resection, facilitating complete histological evaluation and curative resection at the same time (1). The long-term outcome of ESD for EGC has been reported to be similar to that of surgery, and ESD involves local resection with preservation of the stomach (2). ESD cannot be performed immediately after identifying the lesion endoscopically; biopsy and pathologic confirmation must precede the procedure. In EGC, computed tomography scan should be performed to evaluate the lymph node involvement and distant metastasis before ESD (3). This may lead to some delay from EGC diagnosis to ESD, which can be defined as the waiting time. Several studies have documented that prolonged waiting time for treatment adversely affects the outcome in various cancers (4). However, while it has been reported that waiting time affects the psychosocial status of patients scheduled for ESD (5), data on such clinical outcomes are lacking.
EGC and gastric adenoma with high-grade dysplasia (HGD) can progress over time. If the lesion does progress, the procedure may only result in non-curative resection, although the lesion may have been identified as being treatable by curative resection during the initial work-up. Therefore, clinicians aim to perform ESD as soon as a lesion is diagnosed. There is scant evidence regarding the influence of a longer waiting time for ESD on the outcome. Therefore, we hypothesized that a longer waiting time affects the success of curative resection as a result of disease progression.
To our knowledge, this is the first study to evaluate whether longer waiting time for ESD affects curative resection. We conducted this study with an aim to evaluate the association between the time from diagnosis to ESD and curative resection rates in patients with EGC and HGD.
Patients and Methods
This was a retrospective, single-center study conducted at the Ajou University Medical Center (Suwon, Republic of Korea). The study protocol was approved by our institutional review board (approval no. AJIRB-MED-MDB-19-109), and the requirement of obtaining informed consent from each participant was waived owing to the retrospective nature of the study. All co-authors had access to the study data, and each author reviewed and approved the final manuscript.
Patients. Between January 1, 2009, and December 31, 2019, a total of 4,476 patients underwent ESD for gastric neoplasms, including gastric adenoma with low-grade dysplasia, HGD, and EGC. The inclusion criteria were histopathologic confirmation of HGD after endoscopic forceps biopsy before ESD and histopathologic confirmation of EGC that was included as an expanded indication (3, 6) for ESD. Exclusion criteria were cases of gastric adenoma with low-grade dysplasia after endoscopic forceps biopsy and complicated cases, including micro- and macro-perforations. We also excluded patients with a waiting time of >90 days because there may be special reasons for the delay in ESD.
Pre-ESD protocol. There are two pathways to ESD after histologic confirmation of EGC and HGD, as follows (Figure 1): 1) patients with histologically confirmed EGC or HGD at a local hospital are referred to our outpatient services and then hospitalized for ESD, and 2) patients whose lesions are confirmed via upper endoscopy at our center are immediately hospitalized. For patients diagnosed at a local hospital, the tissue sample slides are reviewed by our pathologist prior to their admission. The agreement between the existing results and opinions of the expert pathologists in our center was 96.3%. When the histologic diagnoses differed between our center and the referral center, all slides from the specimens were reviewed again.
Illustration showing the process from diagnosis to endoscopic submucosal dissection (ESD) The following two processes were followed: patients were referred to our center for admission after diagnosis at other centers or patients diagnosed at our center were admitted directly for ESD.
ESD protocol. All ESD procedures were performed by three expert endoscopists using single-channel (GIF-Q260J; Olympus, Tokyo, Japan) or two-channel (GIF-2TQ260M; Olympus, Tokyo, Japan) endoscopy. After identifying the lesion by narrow band imaging and chromoendoscopy using indigo carmine, circumferential marking was done 5 mm outside the tumor margin using a needle knife (Dual knife, Olympus, Tokyo, Japan) or argon plasma coagulation (Erbe Elektromedizin, Tübingen, Germany). Epinephrine mixed fluid (1:10,000) was injected for submucosal lifting, and dissection was performed using an insulated-tip knife (IT knife, Olympus, Tokyo, Japan). The resected specimen was retrieved using a Swirl Net (Olympus, Tokyo, Japan), and all samples were fixed in 10% buffered formalin solution and embedded in paraffin. A standard histopathological process including hematoxylin and eosin staining was conducted. The histologic type, depth of invasion, lymphatic and vascular invasion, perineural invasion, and presence of tumor cells in the resection margin were assessed. Pathological diagnoses were made according to the Japanese Classification of Gastric Cancer (3). Status of H. pylori infection was evaluated by conducting a rapid urease test and histologic confirmation with hematoxylin and eosin stain and Wright-Giemsa stain before ESD. H. pylori infection status was determined as positive when one or two of these tests were positive. If the results of both tests were negative, H. pylori infection status was determined as negative. If a patient was infected with H. pylori, its eradication was performed after ESD, and a follow-up test was conducted using a rapid urease test or urea breath test.
Definition of the waiting time and ESD outcomes. Waiting time is defined as the time from histological diagnosis following the discovery of lesion and biopsy sampling to the point of ESD. In this study, we divided the waiting time into two categories: one for patients diagnosed at a local hospital and underwent ESD at our center and another for patients both diagnosed with and underwent ESD at our center (Figure 1). Curative resection was defined as all of the following conditions being met (3): en bloc resection (defined as complete removal of the lesion in one piece without fragmentation), negative horizontal and vertical margins on histologic examination, no lymphovascular infiltration, no perineural invasion, and (a) irrespective of tumor size, histologically of differentiated type, pT1a without ulcerative findings [UL (–)]; (b) tumor size ≤3 cm, histologically of differentiated type, pT1a UL (+); (c) tumor size ≤2 cm, histologically of undifferentiated type, pT1a UL (–); or (d) tumor size ≤3 cm, histologically of differentiated type, pT1b [submucosa 1 (SM1) <0.5 mm from the muscularis mucosa]. Upgrade discrepancy was defined as the diagnoses of subsequent ESD specimens with a histology of more malignant potential than that at initial forceps biopsy (from HGD to adenocarcinoma).
Study outcomes. The primary outcome of this study was the association between the waiting time (from diagnosis to ESD) and curative resection. The secondary outcomes were risk factors associated with non-curative resection in all patients, with EGC or HGD, and the association between the waiting time and upgrade discrepancy (between the endoscopic forceps biopsy and final ESD pathology) in patients with HGD.
Statistical analysis. According to the initial diagnosis, we divided the patients into two groups, EGC and HGD. Continuous variables are expressed as mean±standard deviation and categorical variables as total number and percentages. For baseline characteristics, continuous variables were compared using the t-test and categorical variables using the chi-squared test. Univariate and multivariate logistic regression analyses were conducted to identify significantly associated factors for curative resection and upgrade discrepancy after ESD. All reported p-values were two-sided, and p<0.05 was considered significant. All statistical analyses were performed using SAS version 9.4 (SAS Institute, Cary, NC, USA).
Results
Study population and baseline characteristics. A total of 1,605 patients were finally included in this study (Figure 2); their baseline characteristics depending on curative resection are summarized in Table I. In the initial forceps biopsy (pre-ESD pathology), there were 1,094 cases (68.2%) of EGC and 511 cases (31.8%) of HGD. The mean lesion diameter was 18.5±8.5 mm, and the mean waiting time to ESD was 36.8±15.5 days for the entire study population. A total of 45.8% of the study population had H. pylori infection, and there were more cases that had been diagnosed and referred from another center (1,052; 65.5%) than those diagnosed at our center (553; 34.5%). The overall curative resection rate was 86.8% (1,393 patients). We divided the study population into two groups according to curative resection. The HGD group showed a higher curative resection rate (98.2%) than the EGC group (81.4%). There were significant differences in age, pre-ESD pathology, waiting time to ESD, post-ESD pathology, and presence of submucosal fibrosis during ESD between the two groups (p<0.05 for all). However, the two groups did not significantly differ in their outcomes according to the referral center (p=0.187).
Flow diagram showing patient enrolment. A total of 1,605 patients, including those with early gastric cancer and gastric adenoma with high-grade dysplasia, were retrospectively reviewed. ESD: Endoscopic submucosal dissection; APC: argon plasma coagulation.
Clinicopathological characteristics of the included patients who underwent endoscopic submucosal dissection.
Association between waiting time and non-curative resection. We analyzed the association factors for non-curative resection after ESD in 1,605 patients (Table II). In the univariate analyses, age >70 years, presence of submucosal fibrosis during ESD, cancer identified during initial diagnosis, and waiting time were significantly associated with non-curative resection (p<0.05 for all). In the multivariable analysis, older age [odds ratio (OR)=1.664, 95% confidence interval (CI)=1.217-2.274, p=0.001], presence of submucosal fibrosis during ESD (OR=1.865, 95%CI=1.345-2.587, p<0.001), and cancer detection at initial evaluation (OR=11.899, 95%CI=6.014-23.542, p<0.001) were significantly associated with non-curative resection. Therefore, the risk for non-curative resection did not increase with the increase in waiting time.
Logistic regression analysis for the identification of the independent risk factors for non-curative resection in all included patients.
As cancer detection at initial diagnosis was the strongest risk factor (OR=11.899), we identified the respective factors that affected non-curative resection in the EGC and HGD groups. In multivariable analyses in the EGC group, age >70 years (OR=1.655, 95%CI=1.199-2.283, p=0.002) and presence of submucosal fibrosis (OR=1.866, 95%CI=1.332-2.612, p<0.001) were significantly associated with non-curative resection (Table III), and in multivariable analyses in the HGD group, only tumors located in the middle one-third portion of the stomach (OR=5.737, 95%CI=1.145-28.752, p=0.034) were associated with non-curative resection (Table III). Waiting time still did not affect non-curative resection in both groups.
Logistic regression analysis for the identification of the independent risk factors for non-curative resection in patients with early gastric cancer and high-grade dysplasia.
Association between waiting time and upgrade discrepancy in patients with HGD. In our study, 274 patients (53.6%) in the HGD group (pre-ESD pathology) were confirmed to have adenocarcinoma as the final pathology (Table I). Logistic regression analysis was performed to confirm whether waiting time affected this upgrade discrepancy (Table IV). However, surface depression (OR=3.632, 95%CI=1.773-7.438, p<0.001) was the only identified significant risk factor in the multivariable regression model. Waiting time was not observed as a risk factor for the discrepancy.
Logistic regression analysis for the identification of the risk factors for upgrade discrepancy between initial diagnosis (endoscopic forceps biopsy) and final endoscopic submucosal dissection pathology in patients with high-grade dysplasia.
Discussion
One of the important goals of ESD in gastric neoplasm is curative resection. Although an increased waiting time raises concerns of disease progression in EGC and HGD, in our study, the waiting time did not influence curative resection. Old age (age >70 years) and submucosal fibrosis in the EGC group and tumor in the middle-third portion of the stomach in the HGD group were significantly associated with non-curative resection. Additionally, in the HGD group, waiting time did not affect upgrade discrepancy.
Upper endoscopy is the most effective examination for the early diagnosis of gastric cancer (7, 8), and screening endoscopy is positively associated with gastric cancer-related mortality (9). Owing to these benefits, the number of people undergoing this examination is continuously rising. Consequently, the likelihood of early detection of gastric cancer has increased along with the growing significance of ESD as a treatment modality for gastric neoplasm. In Korea, the National Cancer Screening Program for gastric cancer provides biannual screening with either upper gastrointestinal series or upper endoscopy for men and women aged ≥40 years every other year. Thus, endoscopy is frequently performed at the screening institutes, which contributes to a prolonged waiting time. In our study, diagnosis at another center was associated with 1.4 times longer waiting time than that at our center (40.4 vs. 29.4 days, p<0.001). Moreover, the number of patients who are diagnosed at another center and referred to our center continues to increase. This trend is predicted to continue in the coming years, and patients may become psychologically anxious as their waiting time before the procedure is prolonged. Therefore, our study would present a guideline for patients and physicians in such situations.
Studies have consistently reported that a longer waiting time is a predictor of worse overall survival in various cancers (4, 10-14). However, they failed to prove this relationship in certain cancers (4, 11, 15-18). For gastric cancer, a longer waiting time was not associated with the overall survival in patients treated with neoadjuvant chemotherapy (19) and primary gastrectomy (19-22). The primary goal of ESD is a curative resection, as a failure to achieve that would require additional treatment. This study was performed with the hypothesis that longer waiting time might lead to non-curative resection due to disease progression. This is because assessing the possibility of a curative resection at the time of diagnosis and assessing it later at the time of ESD may be different. However, in our study, waiting time did not predict curative resection. In our data, we excluded cases involving a waiting time ≥90 days, and most patients (1,463 patients, 91.2%) underwent ESD within 60 days of diagnosis. Additional studies are needed to examine whether longer waiting time may affect curative resection.
In our study, the waiting time did not contribute to upgrade discrepancy. Our previous study (23) revealed that 50.2% of the patients diagnosed with HGD as the pre-ESD pathology showed a change in the final pathological diagnosis (post-ESD pathology). We called this change as upgrade discrepancy. However, with respect to low-grade adenoma, only 16.3% of the patients showed upgrade discrepancy (23). Therefore, we enrolled patients with HGD in the present study to evaluate the association between upgrade discrepancy and waiting time. Upgrade discrepancy cannot be explained solely based on disease progression due to a longer waiting time, as cancer cells exist focally in the lesion and thus, forceps biopsy may induce sampling errors (24). If the waiting time affects disease progression, the outcomes of ESD may also be affected. This may be because it may not be possible to identify lesions suitable for ESD in the pre-ESD evaluation owing to disease progression. Therefore, further studies are needed to examine whether the waiting time for ESD may cause a recurrence due to disease progression.
This study has several limitations. First, this study was retrospective. Therefore, we could not completely control for confounding factors, such as patients’ comorbidities and performance status. Randomized trials could address this limitation; however, these are unlikely to be approved owing to ethical issues. Second, data from a single center were used in this study; thus, the possibility of selection bias cannot be disregarded. Many studies that have reported a relationship between waiting time to treatment and overall survival were multicenter studies utilizing national databases (10, 12-16, 18). Nevertheless, the strengths of single-center studies include consistent ESD protocol, quality control, access to accurate medical data, and minimal variance in the technical expertise levels among operators. The absence of heterogeneity of data across centers may lead to more accurate outcomes. Third, the waiting time may be altered by the progression status of cancer and a patient’s performance status, but these factors were not taken into consideration in our analyses. If a patient’s status is stable and the tumor is in an early stage, the waiting time may be prolonged to prioritize cases involving unstable patient status and more progressed cancers. However, as our study was conducted on cases with EGC and HGD, we suspect that the influence would have been weaker than that of studies that examined the waiting time until surgery in advanced cancer. Fourth, we did not analyze the long-term outcomes, such as recurrence. Post-ESD recurrence is multifactorial. As these factors cannot be controlled for, we did not analyze them in our study. Finally, ESD outcome differs according to the operator’s expertise; however, our results were obtained by three highly skilled expert operators at a single center; therefore, the data obtained are more homogeneous than those obtained in multicenter studies.
In conclusion, waiting time to ESD did not significantly affect the rate of non-curative resection after ESD in patients with EGC and HGD in our study. Endoscopists should strive to shorten the time from diagnosis to procedure as much as possible; however, both patients and physicians should not be concerned with the inevitable wait time caused by the clinical process. These results may aid in reducing the anxiety of patients in clinical practice and establish a guideline regarding the waiting time for ESD. More retrospective multiple- and single-center studies assessing this association between the waiting time to ESD and non-curative resection for EGC and HGD are needed to support our results.
Acknowledgements
The Authors would like to thank Woohyun Cho (Medical information & media center, Ajou university school of medicine) for providing editing services for images and illustrations. She did not receive compensation.
Footnotes
↵* These Authors contributed equally to this work.
Authors’ Contributions
Gil Ho Lee and Jin Woong Park contributed equally to this work. Conceptualization: CK.N.; Data curation: G.H.L., J.W.P., CK.N.; Formal analysis: G.H.L., E.L., CK.N.; Investigation: G.H.L., J.W.P.; Methodology: G.H.L., E.L., CK.N.; Resources: E.L., J.R., Y.B.K., S.G.L., S.J.S., K.M.L.; Pathology review: J.R., Y.B.K.; Software: E.L., G.H.L.; Supervision: CK.N.; Validation: G.H.L., E.L., CK.N.; Visualization: G.H.L., J.W.P.; Writing – original draft: G.H.L., J.W.P., CK.N.; Writing – review & editing: G.H.L., CK.N
Conflicts of Interest
The Authors declare that they have no competing interests in regard to this study.
- Received November 23, 2020.
- Revision received December 3, 2020.
- Accepted December 4, 2020.
- Copyright© 2021, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.