The therapeutic effect of novel endoscopic submucosal dissection with lymph node dissection surgery on T1 ESCC: A single center, prospective, and nested case-control study Authors: Yiqing Mei, Jianning Zhou, Weiwei Jiang, Suya Pang, Quanfu Huang, Peiyuan Mei, Yu Huang, Kuo Li, Yilin Dong, Yongde Liao, Rong Lin Published: 2026 ABSTRACT AbstractBackgroundEndoscopic submucosal dissection is effective for treating T1 early esophageal squamous cell carcinoma. However, there is a risk of lymph node metastasis . Standard radical surgery may be overtreatment for T1 ESCC. This trial was designed to clarify the efficacy of novel dual-scopy combined surgery in the treatment of T1 ESCC.MethodsBetween December 2021 and May 2025, 21 patients underwent dual-scopy combined surgery, 197 underwent ESD, and 213 underwent radical surgery for early (T1 stage) ESCC. This study included 63 patients who underwent ESD, dual-scopy combined surgery, and standard radical surgery (1:1:1) in the efficacy analysis. The patients’ overall survival, recurrence-free survival, disease-specific survival, complications, and clinical outcomes were evaluated.ResultsThere was no significant difference among the three groups in OS, RFS, and DSS. Complication rates were similar between the ESD and dual-scopy groups, both lower than in the radical surgery group. Compared to the radical surgery group, the ESD and dual-scopy groups demonstrated significantly shorter hospital stays and lower costs. Postoperative quality of life and nutritional levels were better in both the ESD and dual-scopy groups compared to radical surgery.ConclusionsFor T1-stage ESCC patients without clear metastasis, the novel dual-scopy combined surgery may achieve lesion resection and lymph node dissection without affecting patient survival. Moreover, it may reduce postoperative complications, hospitalization time, and costs, and potentially improve quality of life and nutritional status compared to standard radical surgery.Trial registrationThis study was registered as a clinical trial with the China Clinical Trial Registration Center (ChiCTR2100053603)Keywords: Esophageal squamous cell carcinoma, Endoscopic submucosal dissection, Lymph node dissection, Radical surgery Background Esophageal cancer ranks among the most prevalent and lethal malignancies worldwide [1, 2], with treatment selection critically dependent on clinical staging. Current therapeutic modalities encompass endoscopic interventions, standard radical surgery, and multimodal approaches including chemoradiotherapy and immunotherapy [3]. The depth of tumor infiltration and the risk of related lymph node metastasis (LNM) fundamentally guide the treatment decision-making and prognosis evaluation of esophageal squamous cell carcinoma (ESCC) [4, 5]. For Tis, T1a-EP, and T1a-LPM ESCC demonstrating <5% LNM risk, endoscopic submucosal dissection (ESD) achieves comparable oncologic outcomes to surgery with superior organ preservation, establishing it as the preferred treatment [6, 7]. Conversely, T1b-SM3 ESCC carries up to 46% LNM risk, mandating radical esophagectomy with lymphadenectomy as the current standard. The treatment of ESCC currently faces the challenge of balancing tumor cure and functional preservation [8]. For lesions limited to the mucosal layer, ESD can achieve a 5-year survival rate (>90%) comparable to surgical procedures, also having advantages such as minimal trauma, fast recovery, and low cost [9]. However, ESD cannot evaluate and treat regional lymph nodes, and LNM status is a key factor affecting the prognosis of ESCC [10]. Multiple studies have shown that T1 esophageal cancer revealed LNM rates of 16% for T1b-SM1, 25–35% for T1b-SM2, and 46–62% for T1b-SM3. However, the optimal surgical strategy for intermediate-risk lesions (T1b-SM1 and T1b-SM2) remains poorly defined, presenting a significant clinical dilemma. Current guidelines only weakly recommend ESD as a relative indication for clinically diagnosed T1b-SM1 (non-circumferential) lesions, whereas T1b-SM2 and deeper lesions fall beyond standard ESD indications, with surgical resection or chemoradiotherapy recommended initially [6]. Radical surgery remains the standard for T1b-SM3 ESCC [11]. Although radical surgery can achieve complete tumor resection and systematic lymph node dissection (LND) of ESCC, the surgical trauma, postoperative complications, and decreased quality of life caused by it cannot be ignored. More noteworthy is that existing preoperative evaluation methods (such as endoscopic ultrasound, computed tomography (CT), etc.) have limited accuracy in determining the depth of tumor infiltration, with a misjudgment rate of about 15%–20%, which may lead to some T1 ESCC patients receiving overtreatment [12, 13]. Therefore, exploring new treatment strategies for ESCC that can preserve organ function while ensuring curative effects is particularly important. The novel dual-scopy combined surgery is a promising and innovative procedure that integrates the advantages of ESD with surgical LND, with treatment decisions guided in real-time by intraoperative pathological findings [14]. The procedure begins with thoracoscopic LND based on tumor location, and the harvested nodes are immediately sent for rapid frozen section examination. If no metastasis is detected, the procedure proceeds with ESD. If LNM is confirmed, patients undergo standard radical esophagectomy. For patients who refuse or cannot tolerate the radical surgery, ESD is performed first to remove the primary lesion, followed by postoperative chemoradiotherapy. This strategy differs from standard treatment approaches: compared with ESD, this approach enables systematic pathological evaluation of lymph nodes while improving patient survival through LND; compared with conventional radical surgery, it spares node-negative patients from unnecessary esophagectomy and its associated complications, thereby maximizing the preservation of esophageal function while ensuring tumor cure. Nevertheless, outstanding questions regarding patient selection criteria, technical standardization, and long-term outcomes demand rigorous clinical validation through prospective studies. This study aims to explore the application prospects of the novel dual-scopy combined surgery in T1 stage ESCC and to investigate the patient selection criteria, technical standardization, and long-term outcomes of this strategy. This approach is intended to improve therapeutic efficacy, reduce complications and postoperative recurrence, thereby offering a potential treatment option for early-stage esophageal cancer. Methods Trial design and ethics This is a single-center prospective study and a nested case-control study registered at the Chinese Clinical Trial Registry (ChiCTR2100053603). The study was conducted at Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, in accordance with the Revised Helsinki Declaration, and was approved by the hospital’s Ethics Committee (Approval ID: 2021-0887-01). Written informed consent was obtained from all patients for the study. Patients The patients were recruited according to the following inclusion criteria: 1) age between 18 and 80 years; 2) can tolerate routine digestive endoscopy, imaging, biochemistry, and endoscopic or surgical treatment; 3) T1 stage ESCC patients diagnosed by endoscopy and pathology; 4) declined curative surgery for esophageal cancer; 5) no preoperative radiotherapy or chemotherapy performed; 6) good compliance and ability to cooperate throughout the entire clinical trial; 7) the subjects voluntarily participate in the trial, and the subjects or their legal representatives sign the notice. Exclusion criteria for enrolment were 1) Preoperative workup demonstrated confirmed lymph node or distant metastasis. After multidisciplinary team (MDT) discussion, the consensus was that a combined endoscopic approach would not be beneficial; 2) Other conditions that may affect the patient’s survival and prognosis include severe cardiac, renal, and hepatic diseases, as well as malignancies; 3) Patients who actively withdraw halfway; 4) Patients who lack the capacity to provide informed consent, or for whom continuation in the study is contraindicated on medical or ethical grounds. All patients underwent pathological diagnosis through magnifying dye endoscopy, endoscopic ultrasonography, CT, and endoscopic biopsy before treatment. Patients at risk of LNM could choose to undergo separate ESD, dual-scopy combined surgery, or radical surgery with sufficient informed consent. Interventions The dual-scopy combined surgery first involved lymph node dissection and intraoperative rapid frozen section examination to determine whether the lymph nodes had metastasized. The scope of LND included the right upper tracheal lymph nodes, upper thoracic esophageal lymph nodes, lower protuberance lymph nodes, and upper diaphragmatic lymph nodes [15–18]. If the rapid frozen section examination during surgery was negative, ESD was performed. If it was positive, radical surgery for ESCC was performed; if the patient refused to undergo radical surgery, ESD followed by radiotherapy and chemotherapy treatment was performed. ESD was performed by an experienced endoscopist, using an insulated-tip knife (IT knife, KD-611L; Olympus, Tokyo, Japan) or dual knife (KD-650Q; Olympus) with a VIO 300D (Erbe, Tübingen, Germany) electrosurgical generator and a GIF-Q260 J endoscope (Olympus). At least two experienced pathologists performed pathological diagnosis on the specimens diagnosed the specimens (Fig. 1). Novel dual-scopy combined surgery procedure Follow-up The follow-up protocol consisted of: endoscopic examinations at 3, 6, and 12 months post-treatment and annually thereafter; chest CT scans every 6 months in the first year and annually thereafter to assess for lymph node enlargement and distant metastasis; and telephone follow-ups at 3, 6, and 12 months and then annually to evaluate swallowing function and general health status. All examinations and follow-ups were performed by the hospital staff. Follow-up started on the treatment date and ended at the date of death or July 10, 2025, with a minimum follow-up period of 1 month for all patients. Outcomes The primary outcomes were overall survival (OS), recurrence-free survival (RFS), and disease-specific survival (DSS). Recurrences included local recurrence and distant metastasis. Local recurrence was defined as both high grade dysplasia or ESCC within 1 cm of the primary excision site (scar) at least 6 months after ESD and local LNM. Distant metastasis was defined as secondary squamous cell carcinoma of distant organs or supraclavicular lymph nodes discovered during follow-up after ESD. OS was defined as the time from the date of surgery to the date of death from any cause. RFS was defined as the time from the date of surgery to either record of recurrence or death from ESCC or surgery-related complications or adverse events. DSS was defined as the time from the date of surgery to the date of death by ESCC or surgery-related complications or adverse events. The secondary outcomes comprised complication and adverse reaction rates, and short-term clinical outcomes evaluations. Complications were categorized as intraoperative (major bleeding ≥ 400 ml, perforation, visceral organ damage, and anesthesia-related complications) and postoperative (chest discomfort, postoperative bleeding, respiratory failure, hoarseness, pleural effusion, perforation, etc.). Main complications were defined as esophageal stenosis (graded both by Stooler’s dysphagia score and endoscopy) [10], esophageal fistula, gastroesophageal reflux. Serious adverse events were defined as multiple complications or death within 30 days after surgery. The short-term clinical outcome evaluations included the highest body temperature within the first 3 days after surgery and the time to the first soft diet after surgery. Nested case-control study This study adopted a matched nested case-control study to better control selection bias. Twenty-one patients were selected from each ESD group and the radical surgery group through propensity score matching (PSM) for comparative analysis, with a 1:1 ratio comparison with the dual-scopy combined surgery group. Baseline characteristics before PSM are shown (Additional file 1: Table S1). Statistical analysis The t-test or Mann–Whitney test was performed for continuous variables. A chi-squared test, or Fisher’s exact test, was performed for categorical variables. OS, RFS, and DSS were illustrated using Kaplan–Meier curves and were compared using the log-rank test. Risk factors for postoperative stenosis were first screened by univariable logistic regression. Variables with a p < 0.2 in the univariable analysis were subsequently entered into a multivariable logistic regression model. To identify factors associated with postoperative recurrence, univariable Cox regression analyses were performed. A two-sided p < 0.05 was considered statistically significant. Data analysis was performed using SPSS software (SPSS Statistics, version 26.0; IBM Corp., Armonk, New York, USA). Results Baseline characteristics From December 2021 to May 2025, 21 T1 ESCC patients underwent dual-scopy combined surgery. At the same stage, 197 T1 ESCC patients underwent ESD, and 213 T1 ESCC patients underwent radical surgery. Of these,68 ESD patients were excluded from the study because they had multiple lesions treated in a single ESD session or required repeated ESD procedures, while 105 patients in the radical surgery group were excluded for having undergone surgery after neoadjuvant chemotherapy or having multiple lesions. Among the remaining ESD and radical surgery group patients, 21 patients were matched from each group through PSM. Finally, a total of 63 patients from the three groups were included in the efficacy analysis (Fig. 2). Trial profile According to the 11th edition of the Japanese Esophageal Cancer Classification [19], all patients had T1 ESCC. Baseline characteristics were similar in the three groups (Table 1). No significant differences were observed among the three groups in terms of in sex, age, BMI, history of smoking and drinking, medical history, tumor location, depth of invasion, lesion circumference, or longitudinal diameter (all p > 0.05). Regarding age, only a pairwise comparison revealed a difference between the ESD and radical surgery groups (p = 0.034) Table 1Baseline patient characteristicsESDDual-scopy combined surgeryRadical surgeryPP*P†P‡PTotal, n212121Sex, n (%)1111Female2 (9.5)2 (9.5)3 (14.3)Male19 (90.5)19 (90.5)18 (85.7)Age, median (range), years67 (51–75)63 (49–74)61 (51–73)0.1040.2930.0340.272BMI, n (%)<203 (14.3)3 (14.3)1 (4.8)0.68210.6060.606≥20 to <2513 (61.9)12 (57.1)17 (81.0)0.22310.3060.182≥255 (23.8)6 (28.6)3 (14.3)0.64510.6970.454Smoking history, n (%)0.8260.75711Yes10 (47.6)12 (57.1)11 (52.4)No11 (52.4)9 (42.9)10 (47.6)Drinking history, n (%)0.63210.7510.530Yes9 (42.9)10 (47.6)7 (33.3)No12 (57.1)11 (52.4)14 (66.7)Past history, n (%)Heart disease3 (14.3)1 (4.8)2 (9.5)0.8640.60611Hypertension9 (42.9)7 (33.3)6 (28.6)0.6130.7510.5201Diabetes1 (4.8)3 (14.3)00.3120.60610.232Others5 (23.8)3 (14.3)5 (23.8)0.7910.70010.700Location of tumor, n (%)Upper thoracic2 (9.5)3 (14.3)00.35410.4880.232Middle thoracic9 (42.9)9 (42.9)5 (23.8)0.33410.3260.326Lower thoracic10 (47.6)9 (42.9)16 (76.2)0.06310.1120.059Depth of invasion, n (%)T1a12 (57.1)12 (57.1)12 (57.1)1111T1a-EP2 (9.5)2 (9.5)T1a-LPM5 (23.8)5 (23.8)T1a-MM5 (23.8)5 (23.8)T1b9 (42.9)9 (42.9)9 (42.9)1111T1b-SM17 (33.3)7 (33.3)T1b-SM22 (9.5)2 (9.5)T1b-SM300Circumference, n (%)≤ 1/47 (33.3)6 (28.6)3 (14.3)0.33710.2770.454> 1/4 – ≤ 1/26 (28.6)8 (38.1)10 (47.6)0.4460.7430.3410.755> 1/2 – ≤ 3/44 (19.0)3 (14.3)5 (23.8)0.920110.697> 3/4 – ≤ 14 (19.0)4 (19.0)3 (14.3)1111Longitudinal diameter, median (IQR), cm2.1 (1.8)2.3 (2.8)2.1 (2.3)0.834111Duration of operation, median (IQR), minutes92 (32)170 (36)365 (118)<0.001<0.001<0.001<0.001ASA scoreI12 (57.1)8 (38.1)0<0.0010.354<0.0010.004II9 (42.9)12 (57.1)18 (85.7)0.0140.5370.0100.088III01 (4.8)3 (14.3)0.31210.2320.606En bloc resection, n (%)2121211111R0 resection, n (%)18 (85.7)19 (90.5)210.35410.2320.488Lymph node metastasis, n (%)Yes03 (14.3)0.232No2118 (85.7)Hospital stay, median (IQR), day8 (4)13 (4)25 (25)<0.0010.015<0.001<0.001Hospitalization expenses, median (IQR), thousand19.60 (4.90)51.52 (21.34)108.31 (95.76)<0.001<0.001<0.001<0.001ESD, endoscopic submucosal dissection*P represents the ESD group compared to the dual-scopy combined surgery group, †P represents the ESD group compared to the radical surgery group, and ‡P represents the dual-scopy combined surgery group compared to the radical surgery group Baseline patient characteristics ESD, endoscopic submucosal dissection *P represents the ESD group compared to the dual-scopy combined surgery group, †P represents the ESD group compared to the radical surgery group, and ‡P represents the dual-scopy combined surgery group compared to the radical surgery group Survival outcomes The median follow-up duration was 23 months (3–43 months). For the ESD group, dual-scopy combined surgery group, and radical surgery group, the OS rates were 95.2%, 90.5%, and 85.7% respectively at the end of follow-up; and RFS rates were 90.5%, 95.2%, and 80.95%, respectively; and DSS rates were 100%, 100%, and 90.5%, respectively (all p > 0.05, Fig. 3 and Fig. 4). When the data were stratified into T1a and T1b, there were also no significant differences detected in the OS, RFS, and DSS curves between the ESD, dual-scopy combined surgery, and radical surgery groups (all p > 0.05, Fig. 3 and Fig. 4). At the final follow-up, one patient in the ESD group died due to underlying diseases. Two patients in the dual-scopy combined surgery group died. One patient died from newly diagnosed malignant melanoma of the nasal sinuses after surgery, and the other died from a primary gastric antral tumor with LNM discovered postoperatively. Three patients in the radical surgery group died, with two of them dying due to severe respiratory failure caused by postoperative anastomotic leakage, and one died due to refractory upper gastrointestinal bleeding. Kaplan–Meier recurrence curves of intramucosal tumor (T1), and subgroups of T1a and T1b for the dual-scopy combined surgery group and radical surgery group. A-C OS curves for T1 (A), T1a (B), and T1b (C) stages. The 3-year estimated OS rates for the dual-scopy combined surgery group and radical surgery group were 90.5% and 85.7%, respectively. D-F RFS curves for T1 (D), T1a (E), and T1b (F) stages. The 3-year estimated RFS rates for the dual-scopy combined surgery group and radical surgery group were 95.2% and 80.95%, respectively. G-I DSS curves for T1 (G), T1a (H), and T1b (I) stages. The 3-year estimated DSS rates for the dual-scopy combined surgery group and radical surgery group were 100% and 90.5%, respectively Kaplan–Meier recurrence curves of intramucosal tumor (T1), and subgroups of T1a and T1b for the ESD group and the dual-scopy combined surgery group. A-C OS curves for T1 (A), T1a (B), and T1b (C) stages. The 3-year estimated OS rates for the ESD group and the dual-scopy combined surgery group were 95.2% and 90.5%, respectively. D-F RFS curves for T1 (D), T1a (E), and T1b (F) stages. The 3-year estimated RFS rates for the ESD group and the dual-scopy combined surgery group were 90.5% and 95.2%, respectively. G-I DSS curves for T1 (G), T1a (H), and T1b (I) stages. The 3-year estimated DSS rates for the ESD group and the dual-scopy combined surgery group were 100% and 100%, respectively Procedures and complications All patients successfully completed the surgery and en bloc resection. The R0 resection rates of the ESD group and the dual-scopy combined surgery group were 85.7% and 90.5%, with no significant difference (p = 1). The median operation time of surgery in the dual-scopy combined surgery group was significantly shorter than that in the radical surgery group (p < 0.001), and longer than in the ESD group (p < 0.001, 92 min vs 170 min vs 365 min) (Table 1). No significant differences were observed in complications or adverse events between the ESD and the dual-scopy combined surgery groups. The overall rates in both the ESD (38.1%) and dual-scopy combined surgery (61.9%) groups were significantly lower than in the radical surgery group (95.2%). Furthermore, the incidence of main complications differed significantly across the three groups (p = 0.035), with the ESD group having a lower rate compared to the radical surgery group (28.6% vs 66.7%, p = 0.031). Regarding esophageal stenosis, it occurred in five patients in the ESD group, five in the dual-scopy combined surgery group, and four in the radical surgery group, with no significant difference observed among the three groups (p = 1). Regarding gastroesophageal reflux, four patients were affected in the ESD group, eight in the dual-scopy combined surgery group, and thirteen in the radical surgery group, showing a significant difference across the groups (p = 0.018). The incidence in the ESD group was significantly lower than that in the radical surgery group (19.0% vs 61.9%, p = 0.012). For esophageal fistula, no cases occurred in either the ESD group or the dual-scopy combined surgery group, while five cases were reported in the radical surgery group. The difference among the three groups was significant (p = 0.009). Both the ESD group and the dual-scopy combined surgery group had significantly lower incidences compared to the radical surgery group (p = 0.048) (Table 2). No significant differences were observed in the degree of stenosis assessed by Stooler’s dysphagia score and endoscopy and the outcomes among the three groups (Additional file 1: Table S2). Multivariate analysis indicated that a circumferential range > 3/4 (p = 0.002) and tumor location in the mid-thoracic (p = 0.021) or lower-thoracic esophagus (p = 0.036), compared to the upper-thoracic region, were significantly associated with postoperative esophageal stenosis. In contrast, the surgical approach was not identified as a significant contributing factor (Table 3). Table 2Characteristics of complications and adverse eventsESDDual-scopy combined surgeryRadical surgeryPP*P†P‡PTotal, n8 (38.1)13 (61.9)20 (95.2)<0.0010.217<0.0010.021Intraoperative Complications, n (%)major bleeding (≥ 400 ml)001 (4.8)1—11hypercapnia002 (9.5)0.323—0.4880.488others000Postoperative complications, n (%)chest discomfort3 (14.3)6 (28.6)11 (52.4)0.0280.4540.0220.209postoperative bleeding001 (4.8)1—11respiratory failure005 (23.8)0.009—0.0480.048esophageal stenosis5 (23.8)5 (23.8)4 (19.0)1111hoarseness002 (9.5)0.323—0.4880.488pleural effusion005 (23.8)0.009—0.0480.048others000Main complications, n(%)6 (28.6)8 (38.1)14 (66.7)0.0350.7430.0310.122esophageal stenosis5 (23.8)5 (23.8)4 (19.0)1111esophageal fistula005 (23.8)0.009—0.0480.048gastroesophageal reflux4 (19.0)8 (38.1)13 (61.9)0.0180.3060.0120.217Serious adverse events002 (9.5)0.323—0.4880.488ESD, endoscopic submucosal dissection*P represents the ESD group compared to the dual-scopy combined surgery group, †P represents the ESD group compared to the radical surgery group, and ‡P represents the dual-scopy combined surgery group compared to the radical surgery group Characteristics of complications and adverse events ESD, endoscopic submucosal dissection *P represents the ESD group compared to the dual-scopy combined surgery group, †P represents the ESD group compared to the radical surgery group, and ‡P represents the dual-scopy combined surgery group compared to the radical surgery group Table 3Univariate model for stenosisFactorsStenosis (n = 14)No stenosis (n = 49)UnivariateMultivariateP valueOR (CI)P valueOR (CI)Sex, n (%)Female2 (14.3)5 (10.2)Male12 (85.7)44 (89.8)0.6700.682 (0.117,3.962)Age, n (%)< 65 years8 (57.1)32 (65.3)≥ 65 years6 (42.9)17 (34.7)0.5771.412 (0.421,4.739)BMI, n (%)≥20 to <2510 (71.4)32 (65.3)<203 (21.4)4 (8.2)0.3002.400 (0.458,12.583)≥251 (7.1)13 (26.5)0.2020.246 (0.029,2.122)Smoking history, n (%)No5 (35.7)25 (51.0)Yes9 (64.3)24 (49.0)0.3161.875 (0.549,6.404)Drinking history, n (%)No7 (50)30 (61.2)Yes7 (50)19 (38.8)0.4541.579 (0.478,5.216)Heart disease, n (%)No1443 (87.8)Yes06 (12.2)1Hypertension, n (%)No10 (71.4)31 (63.3)Yes4 (28.6)18 (36.7)0.5730.689 (0.188,2.520)Diabetes, n (%)No1445 (91.8)Yes04 (8.2)1Location of tumor, n (%)Upper thoracic3 (21.4)2 (4.1)Middle thoracic6 (42.9)17 (34.7)0.1600.235 (0.031,1.768)0.0210.032 (0.002,0.601)Lower thoracic5 (35.7)30 (61.2)0.0330.111 (0.015,0.841)0.0150.036 (0.003,0.517)Invasion depth, n (%)T1a7 (50)29 (59.2)T1b7 (50)20 (40.8)0.5411.450 (0.440,4.778)Longitudinal diameter, n (%) cm< 56 (42.9)46 (93.9)≥ 58 (57.1)3 (6.1)<0.00120.444 (4.227,98.883)0.2245.439 (0.355, 83.238)Circumference, n (%)≤ 3 /44 (28.6)48 (98.0)> 3 /410 (71.4)1 (2.0)<0.001120.00 (12.094,1190.716)0.00287.018 (5.016,1509.558)Treatment methods, n (%)Dual-scopy combined surgery5 (35.7)16 (32.7)ESD5 (35.7)16 (32.7)11 (0.242,4.138)Radical surgery4 (28.6)17 (34.7)0.7070.753 (0.171,3.313)ESD, endoscopic submucosal dissection; OR, odds ratio; CI, confidence interval Univariate model for stenosis ESD, endoscopic submucosal dissection; OR, odds ratio; CI, confidence interval Clinical outcomes In terms of hospitalization duration and costs, the ESD group was slightly lower than the dual-scopy combined surgery group, and both groups were significantly lower than the radical surgery group (p < 0.001) (Table 1). Moreover, short- and long-term clinical outcomes did not differ significantly between the ESD and dual-scopy combined surgery groups, and both showed significantly better soft-diet recovery and weight recovery than the radical surgery group (p < 0.001, p = 0.002). Furthermore, there was no significant difference in recurrence among the three groups (p > 0.05) (Table 4). The recurrence rate in the ESD group and the radical surgery group was higher than that in the dual-scopy combined surgery group (9.5% and 9.5% vs 4.8%, p = 1). Two patients in the ESD group experienced recurrence, including one with local recurrence and one with distant LNM. One patient in the dual-scopy combined surgery group and two patients in the radical surgery group had distant LNM. Based on the univariate analysis in Supplementary Table 5, invasion depth (HR = 14.316, p = 0.012) and longitudinal diameter ≥ 5 cm (HR = 6.117, p = 0.048) were significantly associated with tumor recurrence. A possible explanation is that extensive mucosal lesions suggest prolonged tumor evolution and a higher risk of occult submucosal invasion [20] (Additional file 1: Table S5). Table 4Clinical outcomes of patientsESDDual-scopy combined surgeryRadical surgeryPP*P†P‡PTotal, n212121En bloc resection, n (%)2121211111R0 resection, n (%)18 (85.7)19 (90.5)210.35410.2320.488Lymph node dissection, median (IQR)Group—5 (3)6 (4)0.731Total number—10 (8)13 (8)0.010Lymph node metastasis, n (%)Yes03 (14.3)0.232No2118 (85.7)Short-term clinical outcomesHighest body temperature within the first 3 days after treatment, mean (IQR), °C36.6 (0.3)37.1 (0.8)37.6 (1.5)0.0030.1080.0020.581First soft diet time, median (IQR), day3 (2)6 (4)12 (10)<0.0010.046<0.001<0.001Long-term follow-up clinical outcomesWeight recovery, n (%)16 (76.2)13 (61.9)5 (23.8)0.0020.5040.0020.029Follow-up outcomes, n (%)Recurrence, n (%)2 (9.5)1 (4.8)2 (9.5)1111Local recurrence, n (%)1 (4.8)00111—Distant metastasis, n (%)1 (4.8)1 (4.8)2 (9.5)1111ESD, endoscopic submucosal dissection*P represents the ESD group compared to the dual-scopy combined surgery group, †P represents the ESD group compared to the radical surgery group, and ‡P represents the dual-scopy combined surgery group compared to the radical surgery group Clinical outcomes of patients ESD, endoscopic submucosal dissection *P represents the ESD group compared to the dual-scopy combined surgery group, †P represents the ESD group compared to the radical surgery group, and ‡P represents the dual-scopy combined surgery group compared to the radical surgery group Nutrition and quality of life The EORTC QLQ-C30 was used to assess the quality of life of patients in the three groups, and the Nutritional Risk Screening 2002 (NRS2002) was used to evaluate the nutritional status of the patients. Postoperative quality of life and nutritional status were significantly better in the ESD and dual-scopy combined surgery groups than in the radical surgery group. The two groups showed no significant differences in functional outcomes (such as general health, social functioning) or symptom scores (such as fatigue, nausea, pain, appetite loss), but both were superior to the radical surgery group (Additional file 1: Table S3). Furthermore, the nutritional status of patients following ESD or dual-scopy combined surgery was significantly better than that of the radical surgery group (Additional file 1: Table S4). Discussion Our study is the first research on the treatment of ESCC using the novel dual-scopy combined surgery. This study indicated that the OS, RFS, and DSS of patients in the dual-scopy combined surgery group were not inferior to those in the radical surgery group. Moreover, compared with the ESD group, there was no significant difference in adverse events, hospital stay duration, and hospitalization costs among patients in the dual-scopy combined surgery group, while these outcomes were significantly better than those in the radical surgery group. In summary, preliminary findings suggest that the novel dual-scopy combined surgery is a potentially safe and effective approach for treating stage T1 ESCC. This technique achieves complete lesion resection and LND without compromising survival rates, while preserving organ function. Its advantages include fewer complications and adverse events, shorter hospital stays, lower treatment costs, and better patient-reported quality of life and nutritional status. For T1a ESCC, it is currently believed that the best treatment is endoscopic therapy, such as endoscopic mucosal resection or endoscopic submucosal dissection [21]. However, T1a ESCC still has a 1.44%–8.96% probability of LNM [11], so the dual-scopy combined surgery may also be considered for patients who need a safer therapeutic approach. In addition, the current evidence and NCCN guidelines favor esophagectomy for most T1b ESCC, as nodal involvement is observed in 16.6%–46% of T1b ESCC patients [22]. However, most T1b patients were confirmed to have no LNM after surgery. Due to inaccuracies in preoperative evaluation, some patients were found to have cancerous lesions limited to the mucosal layer upon histological assessment. In this study, 41.46% of patients had inconsistent preoperative evaluation and postoperative pathological results in determining the depth of lesions, with 31.71% of patients having excessive preoperative depth assessment. In some studies, approximately half of clinical T1b (SM1-2) ESCC patients were diagnosed with pT1a stage ESCC after surgery [23]. For these patients, standard radical surgery may be an overly aggressive curative treatment, while the novel dual-scopy combined surgery may be the preferred treatment for organ preservation. While esophagectomy carries a reported complication risk of 17–74% [24], the incidence in our radical surgery group was higher at 95.2%, while that in the dual-scopy combined surgery group was lower (71.4%). The higher complication rate in the radical surgery group, compared to those reported in previous literature, stems from our broader definition of complications, which included endoscopic procedure-related events (e.g., chest discomfort) [25, 26]. This approach allows for a more equitable comparison across the three groups. The study indicates that dual-scopy combined surgery therapy provides survival outcomes comparable to standard radical surgery, while also causing less trauma to patients with superior health-related quality of life. Therefore, the novel dual-scopy combined surgery may serve as a potential alternative to standard radical resection for patients with T1 ESCC who have no clinically evident LNM on preoperative evaluation, or for those who refuse or are unable to tolerate the standard radical surgery. The efficacy of ESD is superior to endoscopic mucosal resection, with a higher complete resection rate (92.3% compared to 52.7%) and a lower local recurrence rate (0.3% compared to 11.5%) [27, 28]. Compared with ESD, the novel dual-scopy combined surgery can completely remove the lesions and clear lymph nodes in ESCC patients, thereby reducing the risk of LNM [15–18]. The lack of a significant survival difference between the ESD and dual-scopy groups may be attributed to limited follow-up duration and sample size. Notably, the RFS was lower in the ESD group than in the dual-scopy group (90.5% vs 95.2%), suggesting that the dual-scopy combined surgery could potentially improve survival outcomes with longer follow-up and larger cohorts. This study still has some shortcomings. First, it is a single-center analysis with a limited sample size and relatively short follow-up duration. Second, due to the limited number of events for postoperative recurrence, multivariate Cox regression analyses for the outcome were not feasible. In the future, we plan to conduct standardized, multi-center studies to enroll more patients and extend the follow-up period, thereby providing more accurate and reliable findings. Conclusions In conclusion, our preliminary findings suggest that the novel dual-scopy combined surgery may offer potential safety and feasibility in selected patients with T1 ESCC, and could represent a more suitable therapeutic strategy for some patients at risk of LNM or unable to tolerate radical surgery. Supplementary information Additional File 1: Table S1-S5. Table S1 Baseline characteristics before PSM. Table S2 The degree of postoperative stenosis occurrence. Table S3 Comparison of EORTC QLQ-C30 Scale Scores. Table S4 Comparison of NRS2002 Scores. Table S5 Univariate analysis for recurrence. Acknowledgements The authors thank all study participants, care teams, investigators, research staff, and administrative personnel from the participating institutions for their contributions. Abbreviations Lymph node metastasis Endoscopic submucosal dissection Esophageal squamous cell carcinoma Lymph node dissection Computed tomography Multidisciplinary team Overall survival Recurrence-free survival Disease-specific survival Propensity score matching Author contributions RL and YDL designed and supervised the study including all data collection and analysis. YQM, JNZ, WWJ and SYP are co–first authors. Concept and design: RL and YDL, JNZ and YQM. Acquisition, analysis, or interpretation of data: YQM, JNZ, YLD and SYP. Drafting of the manuscript: YQM, JNZ and WWJ. Critical review of the manuscript for important intellectual content: RL, YDL, YQM, JNZ and WWJ. Statistical analysis: SYP, YQM and JNZ. Obtained funding: RL and WWJ. Supervision: QFH, PYM, YH and KL. Funding This study was supported in part by the National Key Research and Development Program (No.2023YFC2307001), National Natural Science Foundation of China (No. U25A20141, 82100563) and the Natural Science Foundation of Hubei Province (Nos. 2022CFA009). Data Availability Due to the sensitive nature of the data collected for this study, the raw data are not publicly available to protect participant confidentiality. The author confirms that all data generated or analysed during this study are included in this article. The data supporting the findings of this study can be obtained from the corresponding author upon reasonable request. Declarations Ethics approval and consent to participate The study was conducted at Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, in accordance with the Revised Helsinki Declaration, and was approved by the hospital’s Ethics Committee (Approval ID: 2021-0887-01). Written informed consent was obtained from all patients for the study. Competing interests The authors declare no competing interests. Footnotes Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. These authors contributed equally: Yiqing Mei, Jianning Zhou, Weiwei Jiang, Suya Pang. Contributor Information Yongde Liao, Email: 2019xh0249@hust.edu.cn. Rong Lin, Email: linrong@hust.edu.cn. Supplementary information The online version contains supplementary material available at 10.1186/s12916-026-04899-5. References Associated Data Supplementary Materials Additional File 1: Table S1-S5. Table S1 Baseline characteristics before PSM. Table S2 The degree of postoperative stenosis occurrence. Table S3 Comparison of EORTC QLQ-C30 Scale Scores. Table S4 Comparison of NRS2002 Scores. Table S5 Univariate analysis for recurrence. Data Availability Statement Due to the sensitive nature of the data collected for this study, the raw data are not publicly available to protect participant confidentiality. The author confirms that all data generated or analysed during this study are included in this article. The data supporting the findings of this study can be obtained from the corresponding author upon reasonable request.