Esophageal cancer remains a highly lethal malignancy, with over 0.6 million new cases and 0.54 million deaths worldwide in 2020, and its global burden is projected to further increase through 2030 and 2040 (1). The disease exhibits marked geographic variation, with squamous cell carcinoma predominating in Asia, particularly in China, which accounts for nearly half of global cases, whereas adenocarcinoma is more common in Western countries, reflecting differences in smoking, alcohol consumption, obesity, and gastroesophageal reflux disease (2). Most tumors arise in the thoracic esophagus, while cervical esophageal cancer accounts for approximately 5% of all cases (3). Notably, cervical esophageal squamous cell carcinoma is often associated with poorer prognosis due to more advanced T stage and unfavorable tumor differentiation compared with tumors at other locations (4).

Optimal treatment selection is crucial in cervical esophageal cancer (5). This disease is frequently diagnosed at an advanced stage and is characterized by a high incidence of lymph node metastasis and local invasion into adjacent organs. Definitive chemoradiotherapy can achieve laryngeal preservation in selected patients; therefore, treatment decisions should carefully balance curability and quality of life (6). In contrast, given that lymphatic spread is often relatively confined to the cervical region, radical surgery remains a viable option in selected cases (7).

Therefore, the present study aimed to evaluate clinical outcomes in patients with previously untreated cervical esophageal squamous cell carcinoma treated with radiotherapy-based treatment at a single institution.

We retrospectively screened consecutive patients diagnosed with cervical esophageal cancer who underwent radiotherapy-based treatment at our Institution, Department of Radiology, The University of Tokyo Hospital. Radiotherapy was initiated between July 2012 and October 2025. Patients were eligible when they had histologically confirmed squamous cell carcinoma and received radiotherapy with curative intent. Clinical stage was assigned according to the eighth edition of the American Joint Committee on Cancer staging system (8). Patients with distant metastases were excluded, except for those with supraclavicular nodal metastases that could be encompassed within the curative-intent radiation field. All patients received concurrent platinum-based doublet chemotherapy. Patients who underwent surgery within 6 months of the start of radiotherapy were analyzed with the period up to surgery considered part of the initial treatment course. Clinical information was retrieved from institutional medical records. This study was approved by the Research Ethics Committee of the Faculty of Medicine of the University of Tokyo (No. 3372-9) and conducted in accordance with the Declaration of Helsinki; the requirement for informed consent was waived due to the retrospective study design.

For radiotherapy, patients were immobilized in the supine position using a thermoplastic mask, and planning computed tomography was performed for all patients in accordance with our institutional head and neck immobilization protocol for cervical esophageal targets (9-11). The clinical target volume encompassed the primary tumor and radiologically or clinically involved lymph nodes. A margin was added to generate the planning target volume in accordance with institutional practice. Elective nodal irradiation was not applied. All patients received external-beam radiotherapy using megavoltage photon beams delivered by linear accelerators. Treatment techniques included three-dimensional conformal radiotherapy or intensity-modulated radiotherapy depending on the treatment period and clinical indication. Patients were followed up at regular intervals with clinical examinations and imaging studies. Treatment response was assessed based on clinical and radiological findings. Disease recurrence was categorized as local, regional, or distant.

Statistical analyses were conducted using R software (version 4.1.2; R Foundation for Statistical Computing, Vienna, Austria). Overall survival (OS) was calculated from the initiation of radiotherapy to death from any cause, and progression-free survival (PFS) was calculated from the initiation of radiotherapy to disease progression or death from any cause, whichever occurred first. Survival outcomes were estimated using the Kaplan–Meier method and compared between groups using the log-rank test. A two-sided value of p<0.05 was considered statistically significant.

A total of 33 patients with previously untreated cervical esophageal squamous cell carcinoma were included in this study. The median age was 64 years (range=45-83 years), and 25 patients (75.8%) were male. Performance status was favorable in most patients, with Eastern Cooperative Oncology Group performance status of 0 for 28 patients (84.8%) and 1 for 5 (15.2%). Clinical stage was I in 3 patients (9.1%), II in 3 (9.1%), III in 7 (21.2%), IVA in 14 (42.4%), and IVB in 6 (18.2%). Seventeen patients (51.5%) had T4b disease and seven (21.2%) had T3 disease, while 27 patients (81.8%) had M0 disease and 6 (18.2%) had M1 disease.

With respect to treatment, 29 (88%) patients received definitive radiotherapy-based treatment without surgery, whereas 4 (12%) patients underwent surgery 37-108 days after the initiation of chemoradiotherapy. For these four patients, the therapeutic effect was graded as Grade 0, 1a, 1b, 2, or 3 according to the Japanese Classification of Esophageal Cancer, 12^th Edition (12) (Table I). The concurrent chemotherapy regimen was nedaplatin plus S-1 for 31 patients and cisplatin plus 5-fluorouracil for the remaining two. Induction chemotherapy was administered to five (15%) patients, all of whom received docetaxel, cisplatin, and 5-fluorouracil. The prescribed radiation dose ranged from 50.0 to 60.0 Gy, with a median of 50.4 Gy.

Follow-up ranged from 4.3 to 127.1 months, with a median of 20.28 months (interquartile range=9.73-71.24 months). During follow-up, 17 patients (51.5%) died and 21 (63.6%) experienced disease progression. The median OS was 58.2 months, while the median PFS was 12.1 months (Figure 1). The 1-, 2-, and 3-year OS rates were 86.2%, 60.1%, and 55.8%, respectively. The corresponding PFS rates were 51.3%, 37.3%, and 37.3%. After stratification by clinical stage (I-II vs. III- IVA/B), the 2-year OS rate was 100% for the stage I-II group (n=6) and 51.2% (95% confidence interval=29.2-69.5%) for the stage III-IVA/B group (n=27) (Figure 2). The median OS was not reached for those with stage I-II disease and was 26.2 months for those with stage III-IVA/B. OS differed significantly between the groups (log-rank p=0.0383). The 2-year PFS rate was 60.0% (95% confidence interval=12.6-88.2%) for those with stage I-II (n=6) disease and 32.3% (95% confidence interval=15.2-50.8%) for those with stage III-IVA/B (n=27). The median PFS was not reached for those with stage I-II and was 10.9 months for those with stage III-IVA/B. The difference in PFS did not reach statistical significance (log-rank p=0.117).

During the observation period, recurrence was observed in 21 patients (Table II). Locoregional recurrence only, defined as recurrence confined to the primary site or regional lymph nodes, was observed in 15 out of these 21 patients (71.4%). Distant-only recurrence occurred in four patients (19.0%): liver, lung, bone, and nonregional lymph nodes in one patient each. Combined locoregional and distant recurrence was identified in the remaining 2 patients (9.5%), including regional lymph nodes plus liver (n=1) and primary site plus liver (n=1).

Of these 21 patients with recurrence, best supportive care was selected for 3. Among those who received active treatment, salvage surgery was performed for four patients (partial resection of the right lower lung lobe in one, right neck dissection in one, and total pharyngolaryngectomy with esophagectomy in two). Salvage radiotherapy was delivered to two patients, consisting of either concurrent chemoradiotherapy or stereotactic body radiotherapy. Systemic therapy was administered to the remaining patients, most commonly docetaxel, cisplatin, and 5-fluorouracil in eight patients, followed by 5-fluorouracil plus cisplatin with pembrolizumab in one, 5-fluorouracil plus cisplatin with nivolumab in one, weekly paclitaxel in one, and nivolumab monotherapy in one.

This single-institution retrospective study summarizes real-world outcomes of radiotherapy-based treatment for previously untreated cervical esophageal squamous cell carcinoma, in which advanced disease was common (T4b in 51.5% and stage III-IVA/B in the majority). OS was 86.2%, 60.1%, and 55.8% at 1, 2, and 3 years, indicating that durable survival is achievable in a subset of patients. In contrast, progression was frequent (63.6%), and the median PFS was 12.1 months.

Our results are consistent with prior reports when interpreted in the context of the high proportion of patients with locally advanced disease. Huang et al. reported 2- and 5-year OS rates of 35% and 21% for their overall cohort and 46% and 28% in the curative subgroup among patients treated with concurrent high-dose cisplatin and conformal radiotherapy, without a clear survival advantage over earlier regimens (13). Tasaki et al. reported 2- and 5-year OS rates of 73.8% and 59.4%, and PFS rates of 57.8% and 48.0%, respectively, following definitive concurrent chemoradiotherapy, with complete response associated with improved survival (14). In a multicenter phase II study of definitive concurrent chemoradiotherapy (cisplatin and 5-fluorouracil plus 60 Gy), the complete response rate was 73%, with a 3-year OS of 66.5% and laryngectomy-free survival of 52.5%, supporting the feasibility of organ-preserving strategies (15). A meta-analysis by De Virgilio et al. (22 studies, n=1,222) reported 1-, 3-, and 5-year OS rates of 77.9%, 48.4%, and 35.3%, respectively, with a median OS of 33.4 months, and a 5-year PFS of 29.8% (16). Recent studies have further supported the role of definitive radiotherapy in esophageal squamous cell carcinoma. A population-based target trial emulation study by Lin et al. suggested that the addition of radiotherapy to chemotherapy may improve treatment effectiveness in patients with locally advanced esophageal squamous cell carcinoma (17). Yang et al. reported a multicenter real-world study of 290 patients with inoperable esophageal squamous cell carcinoma treated with definitive chemoradiotherapy plus anti-programmed cell death protein-1 immunotherapy, achieving 2-year OS and PFS rates of 66.9% and 47.3%, respectively (18). In addition, Shigenaga et al. reported that definitive volumetric-modulated arc therapy, administered alone or with concurrent chemotherapy, was feasible and associated with an acceptable toxicity profile (19). In their study, grade 3 neutropenia, esophagitis, and esophageal stricture occurred in 5.4%, 2.7%, and 1.4% of patients, respectively, with no grade 4 or higher toxicities. These recent findings indicate that modern radiotherapy techniques may contribute to effective disease control while limiting severe treatment-related toxicity in patients with esophageal squamous cell carcinoma.

Surgery remains another potentially effective modality in appropriately selected patients, particularly when laryngeal preservation is feasible. A meta-analysis of 18 studies (n=868) reported 1- and 5-year OS rates of 74.4% and 26.6% after primary surgery; larynx-preserving surgery yielded a higher 5-year OS (35.1%) than non-preserving procedures (14.6%) and achieved outcomes comparable to chemoradiotherapy (20). Yuan et al. similarly reported a 2-year OS of 58.9% and better survival with larynx-preserving surgery compared with total pharyngo-laryngo-esophagectomy (71.2% vs. 43.5%), noting higher perioperative mortality and worse OS for the latter group (21). However, surgical series have also highlighted substantial morbidity and recurrence; for example, Lee et al. reported 1- and 3-year OS rates of 75% and 33.3%, with a median survival of 17 months, alongside high complication and recurrence rates (22).

Immune checkpoint inhibitors have become increasingly central to the management of advanced esophageal cancer (23-25). In the phase 3 KEYNOTE-590 trial, pembrolizumab plus chemotherapy significantly improved OS and PFS compared with chemotherapy alone as first-line treatment, with a manageable safety profile (26). Their application in the salvage setting is also being explored. Dai et al. reported that induction chemotherapy combined with immunotherapy enabled response-adapted treatment selection, achieving a 1-year cancer-specific survival of 93.3% for the surgical group versus 71.4% with chemoradiotherapy among non-responders, with an overall laryngeal preservation rate of 85.0% (27). These findings support the rationale for optimizing systemic control and integrating contemporary immunotherapy into future treatment strategies.

Study limitations. The retrospective design and relatively small sample size limit the statistical power and generalizability of the findings. In addition, treatment heterogeneity, including variations in chemotherapy regimens and radiation dose, may have influenced the outcomes. Furthermore, detailed data on patterns of failure and salvage treatments were not fully analyzed, which may have provided further insights into the observed survival discrepancy.

Despite these limitations, this study provides valuable real-world evidence on the outcomes of cervical esophageal cancer treated with radiotherapy-based approaches. Given the rarity of the disease, our findings contribute to the existing literature and may help inform clinical decision-making.

In conclusion, concurrent chemoradiotherapy-based treatment resulted in encouraging long-term survival in patients with cervical esophageal squamous cell carcinoma. However, disease progression remained frequent, predominantly as locoregional failure, underscoring the need to improve locoregional control and optimize multimodal treatment strategies.

The Authors declare that they have no competing interests.

AK conceived and designed the study, collected and analyzed the data, interpreted the results, and drafted the manuscript. HY supervised the study, contributed to the interpretation of the results, and critically revised the manuscript for important intellectual content. Both Authors reviewed and approved the final version of the manuscript.

During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.