Introduction

Lugol chromoendoscopy (LCE) has been established as a tool to increase the sensitivity for the detection of esophageal squamous cell carcinoma (ESCC). However, LCE has some disadvantages such as the time consumed for the technique, patients’ discomfort, risk of aspiration and allergic reactions when compared to white light imaging (WLE) and image enhanced endoscopy (IEE).

Aims & Methods

We attempted to develop virtual lugol chromoendoscopy (V-LCE) images for ESCC using cycle-consistent generative adversarial network (CycleGAN) and to compare their diagnostic performance with that of WLE and real lugol chromoendoscopy (R-LCE). To obtain V-LCE images, we used 3425 images (1601 WLE and 1834 LCE) of superficial ESCC from 195 cases to train CycleGAN. For the test set to evaluate the usefulness of V-LCE images, other 25 WLE images from 25 patients with superficial ESCC were translated to 25 V-LCE images using trained CycleGAN. Six endoscopists (three experts and three non-experts) evaluated images obtained under WLE, V-LCE and R-LCE with respect to the detection and the margin of ESCC, using a five-point scale from 1 (poor) to 5 (excellent). We also calculated and compared the color difference (∆E) between cancerous and non-cancerous areas each in WLE, V-LCE and R-LCE image.

Results

Comparison of WLE and V-LCE images showed a trend towards higher scores for V-LCE images (detection 3.3 vs 3.8, p=0.089, margin 2.7 vs 3.0, p=0.13). A comparison between V-LCE and R-LCE images revealed the scores in R-LCE were significantly higher than those in V-LCE (detection; 4.7 vs 3.8, p<0.001, margin; 4.3 vs 3.0, p<0.001). Among the experts, the score for margin in V-LCE tended to be higher than that in WLE (3.0 vs 2.7, p=0.051). Among the non-experts, the score for detection in V-LCE tended to be higher than that in WLE (4.0 vs. 3.3, p=0.064). ∆E in V-LCE was significantly higher than that in WLE (29.6 vs. 18.3, p<0.001), while it was significantly lower than in R-LCE (29.6 vs. 39.3, p<0.001).

Conclusion

Images in V-LCE have the potential for high sensitivity for the diagnosis of superficial ESCC, while the imaging quality need to be improved.

Introduction

Endoscopic submucosal dissection (ESD) for esophageal squamous cell carcinoma (ESCC) with esophageal varices (EVs) is one of the most challenging endoscopic procedures due to the high risk of bleeding and fibrosis caused by prior treatment of varices. However, due to the small number of such cases, the clinical outcomes of ESD for ESCC with EVs have been reported in only a limited number of cases at single centers^​1-3.

Aims & Methods

We conducted a multicenter, retrospective, cohort study to clarify the safety and efficacy of ESD for ESCC with EVs.Eleven institutions in the Tohoku region of Japan participated in this multicenter, collaborative, retrospective study. We performed endoscopic resection for 3619 patients with esophageal carcinoma at the participating institutions during the period from January 2010 to April 2022. Among those, 30 patients who underwent ESD for ESCC with EVs were recruited for the present study.
Rates of en bloc resection and R0 resection, procedure time and adverse events were evaluated as indicators of the feasibility and safety of ESD. Additional treatment, recurrence and metastasis of the lesions were evaluated as indicators of the long-term efficacy of ESD.

Results

Portal hypertension was caused by cirrhosis, of which alcohol was the most common cause. En bloc resection was achieved in 93.3% and R0 resection in 80.0% of the patients. The median procedure time was 92 minutes. Adverse events included a case of uncontrolled intraoperative bleeding leading to discontinuation of ESD and a case of esophageal stricture due to extensive resection. During the follow-up period of a median for 42 months, a patient with local recurrence and another patient with liver metastasis were observed. One patient died of liver failure after receiving chemoradiotherapy as an additional treatment after ESD. No patient died of ESCC.

Conclusion

This multicenter, retrospective cohort study demonstrated the safety and efficacy of ESD for ESCC with EVs. Further studies are needed to establish appropriate treatment methods for EVs before ESD and additional treatments for patients with insufficient ESD.

References

1. Sawaguchi M, Jin M, Matsuhashi T, et al. The feasibility of endoscopic submucosal dissection for superficial esophageal cancer in patients with cirrhosis (with video). Gastrointest Endosc 2014;79:681-5.
2. Choi YK, Noh JH, Kim DH, et al. Outcomes of endoscopic submucosal dissection for superficial esophageal neoplasms in patients with liver cirrhosis. Clin Endosc 2022;55:381-9.
3. Zhang S, Liu YD, Chai NL, et al. Endoscopic submucosal dissection for early cancers or precancerous lesions of the upper gastrointestinal tract in cirrhotic patients with esophagogastric varices: Ten-year experience from a large tertiary center in China. Gastrointest Endosc 2023 Online ahead of print.

Disclosure

The authors declare that they have no conflicts of interest.

Introduction

Endoscopic submucosal dissection (ESD) for superficial esophageal squamous cell carcinoma (SESCC) involving the entire circumference carries the extremely high risk of stricture and also raises concerns about curative resection. Therefore, the current Japanese guidelines weakly recommend ESD only for circumferential SESCCs diagnosed as cEP/LPM and sized ≤ 50 mm, while surgery or chemoradiotherapy (CRT) is recommended for circumferential SESCCs other than the above. ¹ However, in clinical practice, ESD is implemented on a case-by-case basis through the discussion in a multi-disciplinary cancer board at each institution because the outcomes of each treatment for circumferential SESCCs remain unclear.²

Aims & Methods

This multicenter prospective observational study aimed to reveal the outcomes of each treatment for circumferential SESCCs. Patients with cT1a-EP/LPM or cT1a-MM/T1b-SM1 N0M0 SESCCs involving the entire circumference were enrolled in this study. At this stage, short-term outcomes, such as the proportion of curative resection after ESD, complete response (CR) after CRT/radiotherapy (RT), and adverse events including esophageal stricture, were evaluated. Curative resection in ESD was defined as pT1a without lymphovascular invasion (LVI). Esophageal stricture was defined as inability to pass the ordinary-sized endoscope thorough the treated site, and refractory stricture was defined as stricture requiring ≥ 6 endoscopic balloon dilation procedures.

Results

A total of 126 patients with circumferential SESCCs were enrolled from 16 Japanese tertiary institutions between May 2021 and November 2024. ESD was performed for 90 (71.4%) patients including 6 undergoing stepwise ESD, surgery for 18 (14.3%), CRT for 16 (12.7%), and RT alone for 2 (1.6%). Median longitudinal length of tumor was 55 (range: 30–140) mm in the ESD group, 80 (50–150) mm in the surgery group, and 70 (30–100) mm in the CRT/RT group. Tumor depth was diagnosed as cEP/LPM in 65.4% (55/84) of the ESD group, 38.9% (7/18) of the surgery group, and 50.0% (9/18) of the CRT/RT group, and the others were diagnosed as cMM/SM1. As stricture prevention method after ESD, local injection or filling of triamcinolone (TA) alone was used in 9 patients (10.7%), and both local injection of TA and oral steroid of prednisolone were used in 75 (89.3%). Among the 63 patients who did not receive additional treatment after ESD, the proportion of stricture and refractory stricture were 46.0% (29/63) and 27.0% (17/63), respectively. Curative resection after ESD was achieved in 58 patients (69.0%), including 46 (83.6%) with cEP/LPM and 12 (41.4%) with cMM/SM1. Of 26 patients with non-curative resection after ESD, 20 (76.9%) received additional treatment of either surgery, CRT, or RT, while the remaining 6 opted for follow-up. Of 18 patients in the surgery group, 7 with cEP/LPM and 4 with cMM/SM1 were diagnosed as pT1aN0 without LVI, while the remaining 7 with cMM/SM1 were diagnosed as pT1b, 3 of whom had lymph node metastasis. Anastomotic stricture was observed in 11.1% (2/18) of patients treated with surgery. Of 18 patients in the CRT/RT group, CR was confirmed in all patients without forming stricture.

Conclusion

Considering the high curative resection rate, ESD could be indicated for circumferential SESCCs diagnosed as cEP/LPM. In contrast, the careful consideration may be required in indicating ESD for circumferential SESCCs diagnosed as cMM/SM1 due to its low curative resection rate. Long-term outcome needs to be confirmed for further treatment comparisons.

References

1. Ishihara R, Arima M, Iizuka T, et al. Endoscopic submucosal dissection/endoscopic mucosal resection guidelines for esophageal cancer. Dig Endosc. 2020; 32: 452–93.
2. Kadota T, Ishihara R, Hatta W, et al. Multi-institutional questionnaire on treatment strategies for superficial entire circumferential esophageal squamous cell carcinoma. DEN Open. 2023; 3: e206.

Introduction

Despite the usefulness of white light endoscopy (WLE) and nonmagnified narrow band imaging (NBI) for screening superficial esophageal squamous cell carcinoma (SESCC) and precancerous lesions, these lesions may be missed due to their subtle features and interpretation variations among endoscopists.

Aims & Methods

Our team has developed an artificial intelligence (AI) system to detect SESCC and precancerous lesions using WLE and nonmagnified NBI. This study aims to evaluate the impact of this AI system on clinical diagnosis.
A prospective tandem study was conducted to evaluate the auxiliary diagnostic performance of the AI system in a real clinical setting. Consecutive patients undergoing sedated upper gastrointestinal endoscopy at 12 hospitals in China were recruited. Patients were randomly assigned to either the AI-assisted first group or the routine first group at a 1:1 ratio. The same endoscopist at each center performed tandem upper gastrointestinal endoscopy for each eligible patient on the same day. In the AI-assisted group, the endoscopist performed the first examination with the assistance of the AI system and the second examination without it. In the routine first group, the order of examinations was reversed. The primary outcome was the miss rate of SESCC and precancerous lesions. The secondary outcome including the detection rate of SESCC and precancerous lesions.

Results

Between October 2021 and June 2022, a total of 5,856 eligible patients were included and randomly assigned to the AI-assisted first group and 5,850 were assigned to the routine first group. The per-lesion miss rate was significantly lower in the AI-assisted first group than in the routine first group (1.7% vs 6.7%; Risk ratio (RR)=0.25; p=0.08). The detection rate in the AI-assisted first group was significantly higher than that in the routine first group (1.8% vs 1.3%; RR=1.38; p=0.03).

Conclusion

The proposed AI system can reduce the miss rate and improve the detection rate of SESCC and precancerous lesions during upper gastrointestinal endoscopy under WLE and non-magnified NBI.

Introduction

Patients with multiple Lugol-voiding lesions (mLVLs) in esophageal squamous cell carcinoma (ESCC) were known to have an increased risk of metachronous ESCC or pharyngeal and laryngeal squamous cell carcinoma (PLSCC) following endoscopic resection (ER), necessitating intensive endoscopic surveillance. The Japan Esophageal Cohort study, a multicenter prospective study evaluating the cumulative incidence of metachronous ESCC after esophageal ER, demonstrated that ER was selected as the initial treatment for most newly detected metachronous ESCC (72/74 patients). However, in clinical practice, ESCC or PLSCC outside the indications for ER are occasionally detected. Additionally, patients with mLVLs are known to have a high risk of developing other secondary primary cancers, nonetheless, the long-term clinical course in this population remains unclear.

Aims & Methods

The aim of this retrospective study was to evaluate the long-term prognosis of patients with mLVLs after endoscopic submucosal dissection (ESD) for superficial ESCC. This study included patients with mLVLs who underwent ESD for primary ESCC at our institution between April 2008 and August 2016. The exclusion criteria were patients with non-curative resection after ESD requiring additional treatment, such as submucosal invasion, positive vertical margins (pVM1), or lymphovascular invasion (LVI), and patients with a follow-up period of less than one year. After esophageal ESD, endoscopic surveillance was conducted every six months to one year, and metachronous cancers were considered for treated if detected. The cumulative incidence of metachronous ESCC and PLSCC, as well as the cumulative incidence of metachronous major ESCC and PLSCC, and overall survival (OS) were evaluated. Major ESCC were defined as lesions outside the indications for ER or those in which ER was performed resulting in a non-curative resection. Major PLSCC were defined as lesions outside the indications for ER or endoscopic laryngopharyngeal surgery (ELPS), or those in which ER or ELPS were performed resulting in pVM1 or LVI.

Results

A total of 140 patients were included, with a median age of 69 years (range: 47-86), and 30 patients (21.4%) had a history of head and neck cancer. During the median follow-up of 8.8 years (range: 1.2–15.5), 85 patients (60.7%) developed metachronous ESCC, all of whom were initially treated with ER, and 81 (95.3%) managed with ER alone, including repeated ER. Meanwhile, 35 patients developed metachronous PLSCC, 30 (85.7%) of whom were managed with ELPS alone. The cumulative incidence of metachronous ESCC and PLSCC at 5/10 years was 47.5%/67.1% for ESCC, and 13.8%/35.4% for PLSCC, respectively. The cumulative incidence of metachronous major ESCC and PLSCC at 5/10 years was 0.8%/2.8% for ESCC, and 1.5%/1.5% for PLSCC. Additionally, 30 (21.4%) patients died during follow-up, including progression of superficial metachronous ESCC (n=1) and superficial metachronous PLSCC (n=2) under observation after detection due to poor performance status, recurrences of previously treated PLSCC (n=7), aspiration pneumonia (n=4), hematological malignancies (n=3), tongue cancer (n=2), and others (n=11). The 5- and 10-year OS rates were 92.4%/ 79.5%.

Conclusion

In patients with mLVLs, endoscopic surveillance including pharyngo-laryngeal area following ESD for ESCC enabled early detection of metachronous lesions, the majority of which were successfully treated with ER. However, deaths from other causes were not negligible, highlighting the importance of careful long-term systemic monitoring.

Introduction

Endoscopic submucosal dissection (ESD) is now widely used to treat superficial esophageal squamous cell carcinoma (SESCC) that is preoperatively diagnosed as having a negligible risk of lymph node metastasis. SESCC sometimes experiences delays in ESD due to the patient's social background and the treatment of other synchronous cancers, but the effect of these delays has not been well established. In this study, we aimed to elucidate the relationship between the time from initial diagnosis to ESD and the curability in SESCC indicated for ESD.

Aims & Methods

The subjects were lesions diagnosed as SESCC with indication for ESD from January 2017 to December 2021 at our institution; the indication for ESD included lesions diagnosed endoscopically with a depth of cEP/LPM (cancer invading up to the mucosal epithelium and the lamina propia mucosa) or cMM/SM1 (cancer invading up to .2 mm into the submucosa) and confirmed via computed tomography to be absence of lymph nodes and distant metastasis. We compared the curability between lesions treated within three months after initial diagnosis (group A) and those treated after seven months from initial diagnosis (group B) in a retrospective analysis. Curative resection was defined as lesions that histologically confined within the muscularis mucosae, negative vertical margin (VM), with no lymphovascular invasion, while non-curative resection was defined as anything else. High-risk non-curative resection was defined as lesions that histologically reached SM2 (invasion depth into the submucosal layer >.2 mm) or with lymphovascular invasion or VM positive or unclear.

Results

The analysis included 177 patients with 249 lesions in group A and 61 patients with 74 lesions in group B, with no significant differences in gender, age and prior cancer. The median time from initial diagnosis to ESD in group A and B were 1 (interquartile range [IQR], 0-1) and 11 (IQR, 8-14) months, respectively. The rate of lesions treated with chemotherapy, radiation therapy or surgery for concurrent cancer within the period from initial diagnosis to ESD was significantly higher in group B (A vs. B: 4.4% vs 24.3%; p = 0.000). Median lesion size at initial diagnosis was significantly larger in group A (15mm (IQR, 10-25) vs. 10mm (IQR,8-20), p=0.03). The rate of lesions with cEP/LPM at initial diagnosis in group A and B were 83.1% and 93.2%, respectively.
The change in endoscopic tumor size at initial diagnosis and pathologic tumor size after ESD was significantly larger in group B (A vs. B: 1 mm (IQR,0-5) vs. 4 mm (IQR, 0-8), p=0.000), whereas there was no significant difference in the rate of non-curative and high-risk non-curative resection between the two groups for all lesions (A vs. B:16.9% vs. 12.2%; 6.8% vs. 8.5%). In cEP/LPM lesions at initial diagnosis (group A: 207, group B: 69), there was also no significant difference in the rate of non-curative and high-risk non-curative resection ( A vs. B: 8.2% vs. 5.8%; 6.8% vs. 5.8%). In cMM/SM1 lesions at initial diagnosis　(group A: 42, group B: 5), the rate of non-curative resection tended to be higher in group B (A vs. B: 59.5% vs 100%, p=0.143), and high-risk non-curative resection was significantly higher (A vs. B: 50% vs 100%, p=0.0429).

Conclusion

Our study suggested that delayed ESD did not significantly impact curability for cEP/LPM SESCC, whereas it reduced curability for cMM/SM1 lesions. ESD for SESCC can be delayed due to social factors and the management of other synchronous cancers; however, prompt treatment is important for cMM/SM1 lesions, while a delay might be acceptable for cEP/LPM lesions.

Introduction

Esophageal squamous cell carcinoma (ESCC) is an aggressive malignancy often detected through surveillance esophagogastroduodenoscopy (EGD). Despite regular screening, some cases still progress to advanced stages, including invasion into the muscularis mucosa (pT1a-MM) or deeper layers. The detection of interval ESCC during surveillance presents a clinical challenge, as these lesions can impact long-term prognosis.

Aims & Methods

This study aimed to investigate the characteristic endoscopic features and prognosis of post-EGD ESCC (PEESCC), defined as ESCC that progressed to pT1a-MM or deeper within two years of a prior surveillance EGD that did not identify neoplasia or carcinoma. We retrospectively analyzed 225 superficial ESCC lesions from 215 patients treated with endoscopic submucosal dissection (ESD) at the Hiroshima University Hospital between April 2010 and March 2023. Among them, 29 lesions in 28 patients were classified as the post-EGD ESCC (PEESCC) group, defined as cases where EGD performed 24 months before diagnosis did not detect neoplasia or carcinoma. The remaining 196 lesions in 188 patients were the control group. Subsequently, endoscopic findings [the presence of multiple Lugol-voiding lesions (LVLs), cancer location, circumferential involvement, direction, tumor size, macroscopic type, submucosal tumor (SMT)-like elevation, tumor color, intrapapillary capillary loop (IPCL), and clinical depth of invasion] and prognosis were compared. For accurate endoscopic findings and prognostic evaluation, patients with concurrent pT2 or deeper ESCC, those who had undergone treatment for pT2 or deeper ESCC, and those with a history of esophagectomy were excluded.

Results

In the PEESCC group, the proportions of cases with the presence of multiple-LVLs (69.0% vs. 39.3%, p < 0.01), cervical esophageal lesions (17.2% vs. 3.1%, p < 0.01), lesions localized between the 11 o’clock and 1 o’clock positions (20.7% vs. 7.7%, p = 0.037), and SMT-like marginal elevations (20.7% vs. 7.7%, p = 0.037) were significantly higher, whereas the proportion of red-colored lesions (24.5% vs. 44.9%, p = 0.027) was significantly lower compared to the control group. Additionally, the tumor size in the PEESCC group was significantly smaller than that in the control group (21.6±12.3 mm vs. 34.0±18.3 mm, p<0.01). From the multivariate analysis using the stepwise method, the presence of multiple-LVLs (OR=3.2, 95%CI:1.3-8.5, p<0.01), cervical esophageal lesions (OR=9.9, 95%CI:2.2-45.6, p<0.01), small tumor diameter(≦30mm)(OR=3.3, 95%CI:1.2-10.6, p=0.017), and SMT-like elevation (OR=3.8, 95%CI:1.1-12.0, p=0.034) were significanty identified as characteristic endoscopic findings of PEESCC. No statistically significant findings were detected in the multivariate analysis of patients without ESD history. But in the multivariate analysis limited to non-CRT cases to exclude the influence of CRT, the presence of multiple-LVLs (OR=6.5, 95%CI:2.1-23.4, p<0.01), cervical esophageal lesions (OR=15.7, 95%CI:3.1-86.8, p<0.01), tumor circumferential involvement: <1/3(OR=8.5, 95%CI:1.6-46.1, p=0.013), and SMT-like marginal elevation(OR=6.5, 95%CI:0.8-9.8, p=0.015) were identified as independent risk factors for PEESCC.The PEESCC group had a lower 5-year disease-specific survival (93.1% vs. 98.5%, p=0.039) and recurrence-free survival (69.0% vs. 83.7%, p=0.0245).

Conclusion

PEESCC lesions are associated with distinct endoscopic features and poor prognosis than non-PEESCC lesions.