Efficacy comparison of first-line treatment regimens for stage IV esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype Authors: Fengyi Wang, Rong Yu, Wei Zhang, Dan Han, Wei Deng, Chengxin Liu, Shixuan Zhou, Yifei Wang, Chao Bai, Mingxia Shao, Chen Wang, Yifei Wang, Wei Huang Published: 2026 ABSTRACT AbstractBackgroundSupraclavicular lymph node metastasis (SLNM) is a common subtype of oligometastatic stage IV esophageal squamous cell carcinoma (ESCC), with a better prognosis than multifocal metastatic disease. However, the optimal first-line treatment strategy for this specific subgroup remains unclear.ObjectivesThis study evaluated the efficacy and safety of three first-line treatment modalities—chemoimmunotherapy (CI), concurrent chemoradiotherapy (CCRT), and chemoimmunotherapy followed by radiotherapy (CIRT)—in patients with ESCC with SLNM.DesignThis retrospective, multicenter cohort study analyzed data from 484 ESCC patients with SLNM treated at three major cancer centers in China.MethodsData from 484 patients between January 2019 and December 2024 were analyzed. Inverse probability of treatment weighting (IPTW) was applied to reduce selection bias. PFS and OS were estimated using the Kaplan–Meier method. Intergroup differences were compared using the log-rank test. The Fine–Gray competing risks model was applied to both unweighted and weighted cohorts to calculate the cumulative incidence. Cumulative incidence functions (CIF) were estimated using the Aalen–Johansen method. A two-sided P value < 0.05 was considered statistically significant.ResultsThe CI, CCRT, and CIRT groups enrolled 145, 137, and 202 patients. In weighted Cox regression analysis, CIRT demonstrated superior PFS (adjusted HR (aHR)=0.73, 95% CI: 0.53–0.99, P=0.043) and OS (aHR=0.48, 95% CI: 0.31–0.75, P=0.001) compared to CI. CIRT demonstrated superior PFS compared with CCRT (HR = 0.67, 95% CI: 0.52–0.87, P = 0.003). The weighted median PFS in each group was as follows: CI group 11.0 months (95% CI: 8.0–15.0), CCRT group 11.0 months (95% CI: 10.0–14.0), CIRT group 16.0 months (95% CI: 13.0–20.0) (P = 0.002).ConclusionCompared with CI (systemic therapy alone), CIRT demonstrated potential advantages in PFS, OS, and locoregional control while maintaining acceptable toxicity. Compared with CCRT, CIRT demonstrated improved PFS with no significant differences in OS or locoregional recurrence. These findings suggest that CIRT may represent a promising first-line treatment option for patients with ESCC and SLNM.Keywords: Esophageal squamous cell carcinoma, supraclavicular lymph node metastasis, first-line treatment, immunotherapy, chemotherapy, radiotherapy Introduction Esophageal cancer is one of the most common malignant tumors and the fifth leading cause of cancer-related deaths. Cancer metastasis is the primary cause of cancer-related deaths globally, 1 with 90% of cases being esophageal squamous cell carcinoma (ESCC) 2 in China. Its onset is insidious, and most patients are diagnosed at locally advanced stages or with distant metastasis. Studies indicate that the most common sites of distant metastasis are the liver (33.4%), non-regional lymph nodes (26.6%), and lung (20.5%). 3 According to the definitions of the American Joint Committee on Cancer (AJCC) 8th Edition 4 and the Japanese Esophageal Cancer Staging System (JES), 5 non-regional lymph node metastasis includes supraclavicular lymph node metastasis (SLNM), cervical lymph nodes above the supraclavicular region, para-aortic lymph node station 3A, retroperitoneal lymph nodes, and other distant lymph node stations. According to the latest ESMO guidelines, 6 non-regional lymph nodes are explicitly listed as one of the common treatable sites of oligometastatic disease (OMD). The concept of OMD was first proposed by Hellman and Weichselbaum in 1995, initially representing a transitional stage between limited metastatic disease and extensive metastatic disease. 7 Recent consensus statements and clinical guidelines have further refined the definition of OMD by incorporating factors such as the number of metastatic lesions, involved organs, and feasibility of local ablative treatment. Therefore, ESCC with SLNM is classified as non-regional lymph node metastasis and falls under the oligometastatic condition, defined as distant oligometastatic disease (M1). Furthermore, ESMO guidelines consider this subtype eligible for locally intensive therapy. 8 Systemic therapy serves as the foundation, and local treatment may be considered for patients with potential long-term benefit following MDT evaluation. CSCO 9 and NCCN 10 guidelines currently recommend chemoimmunotherapy (CI) as first-line treatment. However, in East Asia, concurrent chemoradiotherapy (CCRT) is a frequently adopted treatment regimen. The Chinese Radiotherapy Guidelines for Esophageal Cancer 11 also suggest considering CCRT. Given these differing recommendations, the relative merits of each approach remain inconclusive. From a biological mechanism perspective, radiotherapy (RT) can improve the immune microenvironment by inducing tumor antigen release, promoting dendritic cell maturation, and enhancing CD8+ T cell infiltration. Simultaneously, it achieves more effective control over both primary lesions and metastatic lymph nodes; thereby synergizing with PD-1/PD-L1 inhibitors. 12 This combination enhances radiosensitivity in ESCC by activating T cells and inducing G2/M phase arrest,13,14 while amplifying local and regional antitumor responses.15,16 The immune system also plays a crucial role in the therapeutic effects of RT by promoting tumor cell death within the irradiated field. Therefore, combining RT with immune checkpoint blockade enhances radiosensitization and suppresses local tumor recurrence and progression. 17 Thus, administering approximately 50–60 Gy of consolidation RT to controllable lymph node lesions after systemic therapy may be a key factor in achieving high local control rates and prolonged survival. A single-arm prospective phase II trial demonstrated that chemoimmunotherapy followed by radiotherapy (CIRT) was well tolerated in patients with oligometastatic ESCC and the median PFS and OS showed trends toward clinically meaningful benefits. 18 Therefore, a large-scale, multicenter cohort study was conducted to evaluate the efficacy and safety of three treatment modalities—CI, CCRT, and CIRT—as first-line treatment options for ESCC patients with SLNM. Materials and methods Enrolled patients This retrospective, multicenter cohort study analyzed data from 484 ESCC patients with SLNM treated at Shandong Cancer Hospital and Institute, Beijing Cancer Hospital, and Yantai Affiliated Hospital of Binzhou Medical University between January 2019 and December 2024. All eligible patients were grouped based on their first-line treatment regimen: Group 1 received CI as first-line therapy; Group 2 received CCRT; Group 3 received CIRT. Treatment strategies were determined through multidisciplinary team (MDT) discussions and individualized according to patient characteristics, including performance status, tumor burden, primary tumor location, anticipated treatment tolerance, patient preference, and physician judgment, while considering contemporary guideline recommendations. Patients were typically followed up every 3 months during the first year, then every 3 to 6 months thereafter. Tumor location was categorized as upper, middle, lower thoracic esophagus, or multiple-primary. Multiple-primary referred to tumors involving more than one thoracic esophageal segment and therefore not assignable to a single anatomical location. Inclusion criteria were as follows: (1) Based on pre-treatment endoscopic biopsy specimens, pathologists at each center made a histopathological diagnosis of ESCC in accordance with the World Health Organization (WHO) tumor classification criteria; (2) Initial clinical staging was cTanyNanyM1 according to the 8th edition of the AJCC; (3) Distant metastasis was confirmed by PET-CT or other imaging examinations to be limited to supraclavicular lymph node involvement; (4) No prior anticancer treatment. Exclusion criteria were as follows: (1) Presence of organ metastasis; (2) First-line treatment other than CI, CCRT, or CIRT; (3) Subsequent receipt of other treatments (e.g., surgical intervention). (4) Patients in the CIRT group were required to have non-progressive disease before initiation of radiotherapy. Patients who experienced disease progression before planned radiotherapy were not classified as receiving CIRT and were not reassigned to the CI group. This multicenter retrospective study was approved by the ethics committee of Shandong Cancer Hospital and Institute, Shandong First Medical University (Approval No. SDTHEC 202601055). All participating centers followed a consistent approach, and the diagnosis, treatment, and follow-up of all patients were conducted in accordance with standardized institutional protocols and the latest national clinical practice guidelines. The requirement for informed consent was waived due to the retrospective nature of the study. Patient information was kept confidential, and the study was conducted in accordance with the principles of the Declaration of Helsinki. The reporting of this study conforms to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement guidelines. 19 The completed STROBE checklist is provided as Supplementary File 1. Treatment process Chemotherapy consisted of either a fluoropyrimidine plus platinum regimen (PF) or a taxane (paclitaxel or nab-paclitaxel) plus platinum regimen (TP). Among all patients receiving immunotherapy, treatment regimens consisted of PD-1 inhibitors, including intravenous administration of 200 mg Camrelizumab (n = 212), 200 mg Sintilimab (n = 92), 200 mg Tislelizumab (n = 19), 200 mg Pembrolizumab (n = 16), or 240 mg Toripalimab (n = 8), on Day 1 of each 21-day treatment cycle. RT was delivered using intensity-modulated radiotherapy (IMRT). Before treatment, all patients underwent CT simulation using a large-bore CT scanner with intravenous contrast enhancement, and images were acquired at 3-mm slice thickness for treatment planning. CT images were imported into the treatment planning system for target delineation and dose optimization. Involved-field irradiation was adopted for all patients. The gross tumor volume (GTV) included the primary esophageal tumor (GTVp) and radiologically confirmed metastatic lymph nodes (GTVn). The clinical target volume of the primary tumor (CTVp) was generated by expanding the GTVp by 0–0.5 cm radially and 3 cm cranio-caudally to encompass potential microscopic disease extension, whereas the CTV of the metastatic lymph nodes (CTVn) was generated by expanding the GTVn by 0.5–1.0 cm in all directions and included the involved supraclavicular lymphatic drainage region. The planning target volume (PTV) was generated by adding a 0.5–1.0 cm margin to the CTV to account for setup uncertainty and organ motion. The prescribed radiation dose ranged from 50 to 60 Gy, delivered in daily fractions of 1.8–2.2 Gy. Organs at risk (OARs), including the spinal cord, lungs, and heart, were routinely contoured. Dose constraints included a maximum spinal cord dose ≤50 Gy, bilateral lung V20 ≤30%, heart dose (Dmean) <26Gy, V30 <46%. Patients in the CI group received chemotherapy combined with a PD-1 inhibitor. Patients in the CCRT group received platinum-based chemotherapy combined with paclitaxel/albumin-bound paclitaxel or platinum-based chemotherapy combined with tegio capsules, concurrently with local radiotherapy. Patients in the CIRT group received the same chemoimmunotherapy regimen as those in the CI group, followed by radiotherapy after a median of 4 cycles of chemoimmunotherapy (range, 2–6). As this was a retrospective real-world study, the timing of radiotherapy was individualized according to treatment response, patient performance status, treatment tolerance, and multidisciplinary team recommendations. Target volumes were delineated by experienced radiation oncologists according to the RTOG contouring recommendations and the latest national guidelines for esophageal cancer radiotherapy. During treatment, patients were monitored for treatment-related adverse reactions, including but not limited to: radiation esophagitis, radiation pneumonitis, myelosuppression, digestive reaction, and allergic reactions. Adverse reactions were graded according to the Common Terminology Criteria for Adverse Events (CTCAE) version 5.0. Outcomes The primary endpoints of this study were progression-free survival (PFS) and overall survival (OS). Secondary endpoints included adverse events and locoregional recurrence (LRR). PFS was defined as the time from initial treatment to disease progression, death or last follow-up, with follow-up duration being the interval from treatment initiation to death or last follow-up. OS was defined as the time from initial treatment to death or last follow-up. Disease progression was determined based on imaging and clinical evidence, specifically including: (1) Endoscopy, CT, or other imaging studies reveal progression of the primary tumor; (2) Emergence of new tumor lesions, lymph node metastases, or distant organ metastases; (3) Treatment response assessment indicating disease progression (PD). LRR was defined as recurrence within the field or regional boundaries, including recurrence in the primary tumor, regional lymph nodes, and supraclavicular lymph nodes. Statistical analyses To reduce selection bias, this study employed inverse probability of treatment weighting (IPTW) based on a generalized boosting model. The propensity score model incorporated covariates including age, sex, history of alcohol consumption, history of smoking, T stage, N stage, Karnofsky Performance Status (KPS), primary tumor site. PFS and OS were estimated using the Kaplan–Meier method. Intergroup differences were compared using the log-rank test. Weighted PFS and OS were analyzed using Cox proportional hazards models for univariate and multivariate analyses. Median follow-up duration was calculated using the inverse Kaplan–Meier method. The cumulative incidence of LRR was analyzed using the Fine–Gray subdistribution risk model, where distant metastasis (DM) and non-tumor-related death occurring during follow-up were treated as competing events. Based on this model, the cumulative incidence of 3-year LRR was estimated, and a nomogram was constructed accordingly. Model predictive performance was assessed using the C-index from 10-fold cross-validation, while calibration was evaluated using 5 and 10-group calibration curves. Additionally, the Fine–Gray competing risks model was applied to both unweighted and IPTW-weighted cohorts to calculate the cumulative incidence of LRR and DM at 60 months. Cumulative incidence functions (CIF) were estimated using the Aalen–Johansen method. Intergroup comparisons were performed using Wald tests based on subdistribution hazard models, with two-sided P values < 0.05 considered statistically significant. All statistical analyses were conducted using R software (version 4.4.3). Results Patient characteristics As shown in Figure 1 3,181 patients diagnosed with ESCC at three cancer centers between January 2019 and December 2024 were initially included in the study. After screening according to inclusion criteria and excluding patients with incomplete clinical data, 484 eligible patients were ultimately identified. Among these, 145 (30.0%) received CI as first-line treatment, 137 (28.3%) received CCRT, and 202 (41.7%) received CIRT. As shown in Table 1, the three groups exhibited significant heterogeneity in pre-treatment demographic and clinical-pathological characteristics. Specifically, significant differences were observed in the distribution of primary tumor sites (p < 0.001) across the three groups. Additionally, other prognostic factors, including age (p = 0.154), sex (p = 0.603), smoking history (p = 0.884), alcohol consumption history (p = 0.261), and T stage (p = 0.241) were relatively balanced across the three groups at baseline. IPTW was applied to adjust for baseline characteristic differences across treatment groups, resulting in substantial balance among the weighted cohorts. This provided a sound basis for subsequent survival analyses. All weighted SMDs were below 0.1, validating the effectiveness of the weighting process and confirming the well-balanced state of covariates. Flow chart of patient enrollment. A total of 3181 patients were screened from three cancer centers, and finally, 484 eligible patients were included. Table 1.Demographic and baseline clinical characteristics of patients.Unweighted, number (%)Weighted, mean (%)CharacteristicCCRT (n=137)CI (n=145)CIRT (n=202)PSMDCCRT (n=137)CI (n=145)CIRT (n=202)SMDAge (mean (SD))63.7 (8.4)63.4 (8.6)62.0 (8.2)0.1540.19563.262.962.70.061Sex(%)Male118(86.1)130(89.7)180 (89.1)0.6030.03588.090.188.50.020Female19(13.9)15(10.3)22 (10.9)​​12.09.911.5​Drink(%)Yes71 (51.8)79 (54.5)122 (60.4)0.2610.08657.557.858.10.006No66 (48.2)66 (45.5)80 (39.6)​​42.542.241.9​Smoke(%)Yes72 (52.6)80 (55.2)111 (55.0)0.8840.02657.456.954.10.032No65 (47.4)65 (44.8)91 (45.0)​​42.643.145.9​KPS(%)>8070 (51.1)67 (46.2)106 (52.5)0.5000.13450.452.449.30.073≤8067 (48.9)78 (53.8)96 (47.5)​​49.647.650.7​T(%)T1-222 (16.1)15 (10.3)33 (16.3)0.2410.05913.210.713.40.027T3-4115 (83.9)130 (89.7)169 (83.7)​​86.889.386.6​N(%)N0-153 (38.7)37 (25.5)66 (32.7)0.0600.13231.028.131.20.030N2-384 (61.3)108 (74.5)136 (67.3)​​69.071.968.8​Tumor(%)Upper-thoracic48 (35.0)26 (17.9)36 (17.8)<0.0010.17224.819.122.50.058Middle-thoracic34 (24.8)25 (17.2)45 (22.3)​0.07622.321.121.50.012Lower-thoracic28 (20.4)59 (40.7)69 (34.2)​0.20330.934.832.60.039Multiple-primary27 (19.7)35 (24.1)52 (25.7)​0.06022.025.123.40.031 Demographic and baseline clinical characteristics of patients. Progression and survival outcomes The median follow-up duration for all patients was 22.0 months [interquartile range (IQR): 21.0–25.0]. In the overall cohort before IPTW adjustment, the 3-year OS rate for the CI group was 52.1% (95% CI: 41.4%–65.7%); for the CCRT group, it was 70.8% (95% CI: 62.2%–80.5%); while the CIRT group achieved 69.8% (95% CI: 61.8%–78.8%). The estimated 5-year OS rate was higher in the CIRT group 69.8% (95% CI: 61.8%–78.8%) than in the CCRT group 52.9% (95% CI: 40.4%–69.2%); however, due to the limited follow-up period and the small number of patients at risk in the later stages, this long-term efficacy trend requires further investigation. The median PFS and median OS were 12.0 months (95% CI: 11.0–15.0) and NA (95% CI: 53.0–NA) for all patients, respectively. The median PFS in each group was as follows: CI group 11.0 months (95% CI: 8.0–14.0), CCRT group 11.0 months (95% CI: 10.0–15.0), CIRT group 16.0 months (95% CI: 13.0–22.0) (P = 0.016). Median OS had not been reached in any group (P < 0.001) (Figure 2(a) and (b)). After IPTW, the median PFS was 11.0 months (95% CI: 8.0–15.0), 11.0 months (95% CI: 10.0–14.0), and 16.0 months (95% CI: 13.0–20.0) (P = 0.002). CI achieved a median OS of 34.0 months (28.0–NA), while the other two groups did not reach median OS (P < 0.001) (Figure 2(c) and (d)). As summarized in Table 2, after IPTW, CIRT demonstrated superior PFS compared to CI (adjusted HR (aHR)=0.73, 95% CI: 0.53–0.99, P=0.043), with significant improvement in OS (aHR=0.48, 95% CI: 0.31–0.75, P=0.001). Compared with CCRT, CIRT also demonstrated superior PFS (HR = 0.67, 95% CI: 0.52–0.87, P = 0.003). After IPTW, baseline differences between groups were balanced, yet the trends in PFS and OS remained consistent. Multivariate Cox regression analysis yielded similar results (Supplementary Table 1 and Supplementary Table 2). Progression-free survival (PFS) and overall survival (OS) evaluated using the Kaplan–Meier method before (A, B) and after (C, D) inverse-probability of treatment weighting (IPTW) analysis for overall cohorts. Abbreviation: CI, chemoimmunotherapy; CCRT, concurrent chemoradiotherapy; CIRT, chemoimmunotherapy followed by radiotherapy. Numbers at risk in the IPTW-weighted panels represent effective sample size rather than actual patient counts. Table 2.Summary of unadjusted and propensity score-weighted analysis results.TreatmentHR(95%CI)PPFSUnadjustedCIRT vs CI0.72 (0.54-0.95)0.020CI vs CCRT1.00 (0.75-1.33)0.986CIRT vs CCRT0.71 (0.55-0.93)0.012AdjustedCIRT vs CI0.73 (0.53-0.99)0.043CI vs CCRT0.92(0.67-1.26)0.610CIRT vs CCRT0.67 (0.52-0.87)0.003OSUnadjustedCIRT vs CI0.50 (0.33-0.76)0.001CI vs CCRT2.27(1.46-3.52)<0.001CIRT vs CCRT1.14 (0.72-1.80)0.575AdjustedCIRT vs CI0.48 (0.31-0.75)0.001CI vs CCRT2.10(1.32-3.35)0.002CIRT vs CCRT1.01 (0.63-1.62)0.971 Summary of unadjusted and propensity score-weighted analysis results. In this cohort, LRR occurred in 65 (44.8%), 68 (49.6%), and 71 (35.1%) patients in the CI, CCRT, and CIRT groups, respectively; DM occurred in 20 (13.8%), 46 (33.6%), and 44 (21.8%) cases, respectively; and the number of death events was 37 (25.5%), 37 (27.7%), and 48 (23.7%), respectively. In the IPTW-weighted Fine–Gray competing risks model, treatment modality was confirmed as an independent predictor of 3-year LRR. The model demonstrated good fit in both 5- and 10-group calibration plots, with predicted cumulative LRR incidence rates highly consistent with observed values (Supplementary Figure 1). Compared with CIRT, CI significantly increased the risk of LRR (aHR = 1.595, 95% CI 1.118–2.275, P = 0.009). No other clinical variables showed significant association with LRR (P > 0.050) (Supplementary Table 3). In the competing risks analysis, the unweighted CIF showed no significant difference between CCRT and CIRT for LRR (P = 0.981) (Supplementary Table 4), with similar results observed in the multivariate analysis (P = 0.763) (Supplementary Table 5). The cumulative incidence of LRR was significantly higher in the CI group (aHR = 1.544, 95% CI 1.078–2.211, P = 0.018), and this trend persisted after IPTW (aHR = 1.595, 95% CI 1.118–2.275, P = 0.009) (Figure 3(a) and (c); Supplementary Table 5). For DM, no significant difference was observed between CCRT and CIRT (unweighted P = 0.632; weighted P = 0.851). CI showed borderline differences compared with CIRT in the unweighted analysis (P = 0.050) and reached statistical significance after weighting (aHR = 0.567, 95% CI: 0.326–0.987, P = 0.045) (Figure 3(b) and (d); Supplementary Table 5). In summary, regardless of IPTW adjustment, CI consistently demonstrated higher cumulative LRR rates and a consistent adverse trend in DM; overall local and distant control rates were comparable between CCRT and CIRT. The incidence of locoregional failure and distant progression determined by competing risk analysis before (A, B) and after (C, D) inverse-probability of treatment weighting analysis. In subgroup analysis, CIRT showed a trend toward reduced risk in PFS compared to CCRT, although this trend did not reach statistical significance (P=0.42), multivariate analysis confirmed this trend. Furthermore, after IPTW adjustment, a significant interaction with radiation dose was observed (P for interaction = 0.02): among patients who received >50.4 Gy of radiation therapy, CIRT significantly reduced the risk (aHR = 0.65, 95% CI: 0.42–1.00, P = 0.048) (Supplementary Figure 2; Supplementary Table 6 and 7). Treatment-related toxicity As shown in Table 3, in the unweighted population, the most common treatment-related adverse events (TRAEs) across these three treatment strategies included myelosuppression, digestive reaction, infection, radiation esophagitis, and radiation pneumonitis, among others. Among these, 50 patients (14.70%) reported radiation esophagitis, with 17 cases (12.4%) in the CCRT group and 33 cases (16.3%) in the CIRT group (P<0.001), (Grade ≥3) myelosuppression (CI 11.00% vs. CCRT 21.90% vs. CIRT 14.40%, P=0.035), and radiation esophagitis (CI 0.00% vs. CCRT 6.60% vs. CIRT 7.40%, P<0.001). As summarized in Supplementary Table 8, after IPTW, (Grade ≥3) myelosuppression occurred most frequently in the CCRT group (CI 10.40% vs. CCRT 21.50% vs. CIRT 14.60%, P = 0.035); and radiation esophagitis occurred more frequently in the CIRT group (CCRT 5.10% vs. CIRT 7.40%, P<0.001). Table 3.Treatment-emergent adverse events in the entire cohort before IPTW.Adverse eventsAny gradeP.valueGrade≥3P.valueCI(n=145)CCRT(n=137)CIRT(n=202)CI(n=145)CCRT(n=137)CIRT(n=202)Myelosuppression103 (71.00%)96 (70.10%)149 (73.80%)0.73116 (11.00%)30 (21.90%)29 (14.40%)0.035Hyponatremia2 (1.40%)1 (0.70%)1 (0.50%)0.8211 (0.70%)0 (0.00%)0 (0.00%)0.583Hypokalemia3 (2.10%)1 (0.70%)0 (0.00%)0.1012 (1.40%)1 (0.70%)0 (0.00%)0.269Hypoproteinemia3 (2.10%)4 (2.90%)2 (1.00%)0.4032 (1.40%)2 (1.50%)2 (1.00%)1.000Hypothyroidism1 (0.70%)2 (1.50%)1 (0.50%)0.6941 (0.70%)2 (1.50%)0 (0.00%)0.121Digestive reaction29 (20.00%)38 (27.70%)23 (11.40%)<0.00111 (7.60%)9 (6.60%)1 (0.50%)<0.001Infection1 (0.70%)2 (1.50%)1 (0.50%)0.6940 (0.00%)1 (0.70%)0 (0.00%)0.283Radiation esophagitis0 (0.00%)17 (12.40%)33 (16.30%)<0.0010 (0.00%)9 (6.60%)15 (7.40%)<0.001Radiation pneumonitis0 (0.00%)2 (1.50%)3 (1.50%)0.3960 (0.00%)2 (1.50%)2 (1.00%)0.392Radiation dermatitis0 (0.00%)0 (0.00%)2 (1.00%)0.5130 (0.00%)0 (0.00%)1 (0.50%)1.000Radiation pharyngitis0 (0.00%)1 (0.70%)2 (1.00%)0.6300 (0.00%)0 (0.00%)1 (0.50%)1.000Immune myocarditis2 (1.40%)0 (0.00%)1 (0.50%)0.4822 (1.40%)0 (0.00%)1 (0.50%)0.482Immune dermatitis0 (0.00%)0 (0.00%)1 (0.50%)1.0000 (0.00%)0 (0.00%)1 (0.50%)1.000Immune pituitary inflammation1 (0.70%)0 (0.00%)1 (0.50%)1.0001 (0.70%)0 (0.00%)1 (0.50%)1.000Immune hemangioma0 (0.00%)0 (0.00%)1 (0.50%)1.0000 (0.00%)0 (0.00%)1 (0.50%)1.000Allergic reaction2 (1.40%)1 (0.70%)3 (1.50%)0.8861 (0.70%)1 (0.70%)0 (0.00%)0.339Cutaneous disorder3 (2.10%)10 (7.30%)5 (2.50%)0.0502 (1.40%)4 (2.90%)1 (0.50%)0.214Abnormal liver function6 (4.10%)4 (2.90%)10 (5.00%)0.6793 (2.10%)3 (2.20%)5 (2.50%)1.000Abnormal renal function1 (0.70%)1 (0.70%)0 (0.00%)0.3391 (0.70%)1 (0.70%)0 (0.00%)0.339Peripheral neurotoxicity1 (0.70%)1 (0.70%)1 (0.50%)1.0001 (0.70%)0 (0.00%)0 (0.00%)0.583Other1 (0.70%)2 (1.50%)2 (1.00%)0.8661 (0.70%)2 (1.50%)2 (1.00%)0.866 Treatment-emergent adverse events in the entire cohort before IPTW. Discussion This study enrolled 484 untreated stage IV esophageal carcinoma patients with SLNM and employed IPTW to mitigate selection bias among three treatment strategies (CI, CCRT, CIRT). The study evaluated their impact on survival outcomes, safety, and locoregional recurrence-free survival (LRR-free survival) in this ESCC subtype. Results demonstrated that compared with systemic therapy alone, CIRT demonstrated superior outcomes in terms of PFS, OS, and local control; compared with CCRT, it offered an advantage in PFS while associated with a relatively lower incidence of high-grade toxicity. Fine-Gray competing risks modeling and multivariate Cox regression analysis further confirmed that RT was significantly associated with improved OS, consistent with previous studies.20,21 These findings suggested that combining aggressive local RT with systemic therapy plays a crucial role in reducing progression and recurrence. For example, Eric J Lehrer. 22 et al. provided supporting evidence. Given radiotherapy’s excellent local control rates and, perhaps more importantly, its low incidence of severe acute and late toxicities, RT is particularly important for patients with metastatic disease and limited life expectancy. Regarding survival outcomes, this study demonstrated that the median PFS in the CIRT group reached 16.0 months, significantly outperforming both the CI and CCRT groups. For instance, in Chen et al.'s study, the median PFS for the radical chemoradiotherapy group and the systemic therapy-only group was only 8.7 months vs. 7.3 months (p=0.002) 23 ; In two large Phase III trials of first-line chemotherapy combined with immunotherapy (ESCORT-1st 24 and ORIENT-15 25 ), the median PFS for this combination therapy was 6.9 months and 7.2 months, respectively, significantly lower than the 16.0 months observed in CIRT. The median OS in the CI group was 34.0 months, while neither the CCRT nor CIRT groups reached median survival. However, the Kaplan–Meier curves showed that the curves for these two groups diverged in the later stages, suggesting a potential late separation favoring CIRT. Specifically, Lindsay L. Puckett et al.'s retrospective analysis of radical chemoradiotherapy showed a 5-year survival rate of 50.5%, 26 lower than the 69.8% observed in CIRT. Given differences in patient selection criteria, histological composition, and systemic treatment regimens across studies, the statistical divergence in survival data between Chen et al.'s research and this study is reasonable. Nevertheless, the cross-study trend indicates that our findings are highly consistent with existing evidence: aggressive local radiotherapy in combination with systemic therapy significantly improves disease control and long-term survival outcomes in patients with ESCC and SLNM. Regarding the risk of local recurrence, the impact of different radiotherapy strategies on LRR and DM was systematically evaluated using an IPTW-weighted competing risks model. CI exhibited a significantly higher LRR risk than CIRT. No significant difference was observed between CCRT and CIRT. Additionally, DM analysis in this study revealed a statistically significant difference between treatment modalities after IPTW, with CIRT associated with a lower risk of DM than CI (P=0.045). However, this finding should be interpreted with caution, as DM is influenced by multiple factors, including tumor molecular characteristics, hematogenous dissemination capacity, and host immune status, rather than solely depending on local treatment modality.1,27 In summary, localized radiation therapy offers significant advantages in reducing primary tumor progression and regional recurrence risk, but its impact on the risk of distant metastasis remains uncertain. Additionally, the impact of different radiotherapy doses on treatment efficacy and safety in this patient subtype was investigated. Subgroup analysis revealed that patients in the CIRT group receiving >50.4 Gy demonstrated more significant survival improvement, consistent with the conclusion from Guttmann DM et al. that “curative doses are superior to palliative doses”. 28 Other studies also indicate that ESCC patients with non-regional lymph nodes metastasis significantly benefit from high-dose radiotherapy. 24 This suggests that dose escalation may contribute to improved local control. In terms of safety, all three groups demonstrated good tolerability, though differences existed in the toxicity profiles. Weighted analysis revealed that Grade ≥3 myelosuppression was most common in CCRT and least frequent in CI. The thoracic spine, a critical irradiation region in esophageal radiotherapy, is also a primary site of hematopoiesis. Moderate-to-high radiation doses increase the risk of myelosuppression. Concurrently, radiation exposure to immune cells—particularly lymphocytes—can cause significant lymphopenia, impairing immune function and compromising survival prognosis. 29 In the two radiotherapy groups, this difference may be partly attributable to treatment sequencing. Grade ≥3 myelosuppression was most frequent in the CCRT group, which may reflect the overlapping toxic effects of concurrent chemotherapy and thoracic radiotherapy on hematopoietic function. In contrast, patients in the CIRT group received radiotherapy sequentially after chemoimmunotherapy, this may have caused the lesions to shrink prior to radiotherapy, resulting in a correspondingly smaller treatment field, while also allowing for partial recovery of bone marrow function, thereby reducing the risk of severe hematologic toxicity. It is noteworthy that despite the satisfactory efficacy of CIRT in our study, treatment-related toxicities cannot be overlooked. Grade ≥3 radiation esophagitis was not uncommon (7.4%). Although such toxicities are generally manageable, this finding suggests that future clinical practice should place greater emphasis on optimizing dose distribution, immunotherapy, radiotherapy sequencing design, and mucosal protection strategies. This approach aims to enhance therapeutic efficacy while reducing the risk of severe adverse reactions. This study also has limitations. First, owing to the retrospective and non-randomized nature of this study, treatment allocation was determined according to real-world clinical practice rather than random assignment. Although IPTW was applied to balance measured baseline characteristics across treatment groups, residual confounding from unmeasured variables, such as patient comorbidities, socioeconomic factors, and other clinical considerations, cannot be completely excluded. Second, the follow-up duration remains insufficient to fully reflect OS outcomes, particularly as median OS has not yet been reached in CIRT and CCRT. Therefore, the observed survival advantage associated with CIRT should be interpreted with caution and requires further validation in prospective randomized studies with longer follow-up periods. Previous studies have predominantly focused on the broad concept of OMD with limited attention directed toward the more common and distinct subtype characterized by SLNM. This subtype exhibits clear biological features, and relevant literature has confirmed its superior prognosis compared to patients with multifocal metastatic disease.30–32 Given these characteristics, targeted guidelines for this specific subtype remain lacking. Current guidelines generally recommend systemic therapy as the primary approach for distant metastasis, supplemented by local treatment. However, our rigorous analysis of real-world clinical data suggests that CIRT offers a PFS benefit, with potential advantages in OS, toxicity management, and local disease control. These findings provide additional evidence supporting precision-targeted treatment for this specific subtype and enhance our understanding of the characteristics, treatment sensitivity, and prognosis of this subtype, which may inform the development of more targeted clinical pathways and guidelines in the future. Conclusions In summary, the findings of this study suggest that CIRT may represent a rational and biologically grounded therapeutic strategy for ESCC patients with SLNM. By integrating the distant control capabilities of systemic therapy with the regional curative potential of radiotherapy, CIRT may represent a promising treatment strategy for this specific subtype of oligometastatic ESCC patients. Supplemental material Supplemental material - Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype Supplemental material for Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype by Fengyi Wang, Rong Yu, Wei Zhang, Dan Han, Wei Deng, Chengxin Liu, Shixuan Zhou, Yifei Wang, Chao Bai, Mingxia Shao, Chen Wang, Yifei Wang, Wei Huang in Therapeutic Advances in Medical Oncology Supplemental material - Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype Supplemental material for Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype by Fengyi Wang, Rong Yu, Wei Zhang, Dan Han, Wei Deng, Chengxin Liu, Shixuan Zhou, Yifei Wang, Chao Bai, Mingxia Shao, Chen Wang, Yifei Wang, Wei Huang in Therapeutic Advances in Medical Oncology Acknowledgements The authors thank all patients and staff who participated in this study. Author contributions: F.Y.W., R.Y., and W.Z. contributed equally to this work and are designated as co-first authors. F.Y.W.: Investigation, Methodology, Statistical analysis, Data curation, Writing – original draft. R.Y. and W.Z.: Investigation, Data curation, Supervision, Resource. D.H., W.D., and C.L.: Supervision. S.Z., Y.W., C.B., M.S., C.W., and Y.W.: Investigation, Data curation. W.H.: Conceptualization, Methodology, Writing – review & editing, Supervision, Funding acquisition, Resource. All authors read and approved the final manuscript. Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (U24A20731), the Shandong Provincial Natural Science Foundation Project (ZR2023MH296), the Noncommunicable Chronic Diseases—National Science and Technology Major Project (2023ZD0502300), and the Collaborative Academic Innovation Project of Shandong Cancer Hospital (ZF005). The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Supplemental material: Supplemental material for this article is available online. ORCID iDs Fengyi Wang https://orcid.org/0009-0006-4165-5179 Wei Deng https://orcid.org/0000-0003-1778-2013 Chengxin Liu https://orcid.org/0000-0003-2962-0820 Chao Bai https://orcid.org/0009-0006-7631-1923 Wei Huang https://orcid.org/0000-0002-1667-657X Ethical considerations This study was approved by the Ethics Committee of Shandong Cancer Hospital and Institute, Shandong First Medical University (approval number: SDTHEC 202601055), the Ethics Committee of Yantai Affiliated Hospital of Binzhou Medical University (Approval No. LW-20260722043), and the Ethics Committee of Beijing Cancer Hospital (Approval No. 2026YJZ24). Consent to participate The requirement for informed consent was waived due to the retrospective nature of the study. Data Availability Statement The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.* References Associated Data Supplementary Materials Supplemental material - Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype Supplemental material for Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype by Fengyi Wang, Rong Yu, Wei Zhang, Dan Han, Wei Deng, Chengxin Liu, Shixuan Zhou, Yifei Wang, Chao Bai, Mingxia Shao, Chen Wang, Yifei Wang, Wei Huang in Therapeutic Advances in Medical Oncology Supplemental material - Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype Supplemental material for Efficacy comparison of first-line treatment regimens for stage IV Esophageal squamous cell carcinoma with supraclavicular lymph node metastasis: A study of the oligometastatic subtype by Fengyi Wang, Rong Yu, Wei Zhang, Dan Han, Wei Deng, Chengxin Liu, Shixuan Zhou, Yifei Wang, Chao Bai, Mingxia Shao, Chen Wang, Yifei Wang, Wei Huang in Therapeutic Advances in Medical Oncology Data Availability Statement The datasets generated and/or analysed during the current study are available from the corresponding author on reasonable request.*