Esophageal Cancer Radiation Management

Scope and treatment selection

Radiation management should be determined by clinical TNM stage, resectability, tumor location, histology, performance status, comorbidity, and the feasibility of surgery; the evidence supplied here is heterogeneous and includes randomized trials, meta-analyses, prospective single-arm data, retrospective cohorts, and diagnostic studies ^{[1]A1a[4]A1b[6]A1b[9]B2b[12]B2b[18]B3b}. The recommendations below therefore distinguish curative endoscopic, trimodality, definitive, postoperative, salvage, and investigational approaches rather than assigning one radiation strategy to every patient within a stage group ^{[1]A1a[4]A1b[7]B2a[12]B2b[14]C4}.

Stage I and superficial disease

For T1aN0M0 squamous cell carcinoma, endoscopic resection (ER) can provide definitive local treatment when the lesion is completely removed with a negative margin and lacks lymphovascular invasion; in the prospective selective-treatment study, patients meeting these criteria received no additional treatment ^[12]B2b. After ER, prophylactic chemoradiotherapy (CRT) using 41.4 Gy to locoregional lymph nodes was used for selected patients with T1b (SM1–2) N0M0 disease with a negative margin or T1a disease with lymphovascular invasion, although the supplied abstract does not provide the complete outcome data or all treatment-group results ^[12]B2b.

Esophagectomy remains a standard option for stage I esophageal squamous cell carcinoma, while selective CRT after ER is a prospective alternative for appropriately selected superficial tumors ^[12]B2b. Definitive CRT is also a curative option for cT1bN0M0 esophageal squamous cell carcinoma, but residual local disease or recurrence after an apparent complete response can occur; retrospective endoscopic findings may help identify patients at increased risk of non-radical cure ^[16]C4.

The role of intraluminal brachytherapy as a curative boost in stage I squamous cell carcinoma remains uncertain. A systematic review found that the optimal radiation method for early disease has not been established and that the effects of brachytherapy on local control and survival remain controversial ^[7]B2a. Thus, brachytherapy should not be considered routine stage I treatment on the basis of the supplied evidence ^[7]B2a.

For superficial disease after noncurative endoscopic submucosal dissection (ESD), esophagectomy is described as the standard adjuvant treatment, but retrospective data also support adjuvant CRT as a potentially effective alternative in selected patients; the comparison included 60 patients, of whom 34 received CRT and 26 underwent surgery ^[17]B3b.

Stage II–III and resectable locally advanced disease

For resectable, nonmetastatic locally advanced disease, neoadjuvant CRT followed by esophagectomy is an established curative strategy in the cited clinical literature ^{[18]B3b[19]C4}. In a meta-analysis of three randomized trials involving 375 patients, neoadjuvant CRT significantly increased R0 resection and pathologic complete response rates compared with neoadjuvant chemotherapy alone, but did not significantly improve 3- or 5-year progression-free survival or overall survival ^[1]A1a.

The CROSS regimen described in routine practice consisted of five weekly courses of carboplatin and paclitaxel with concurrent external-beam radiation delivered as 23 fractions of 1.8 Gy, followed by surgery ^[18]B3b. The supplied evidence does not establish that one neoadjuvant regimen is universally preferred across histologies, tumor sites, or healthcare systems ^{[1]A1a[18]B3b}. Patients should be reassessed after neoadjuvant therapy before proceeding to esophagectomy, because a retrospective cohort found that 16.8% of patients scheduled for neoadjuvant treatment and surgery did not ultimately undergo resection ^[19]C4.

The optimal neoadjuvant radiation dose remains unsettled. A large retrospective National Cancer Database analysis compared 41.4, 45, and 50.4 Gy in patients receiving neoadjuvant CRT for nonmetastatic disease and evaluated overall survival, downstaging, and perioperative mortality; because this was not a randomized comparison, it does not define a universally superior dose ^[84]C.

Routine adjuvant chemotherapy after neoadjuvant CRT and esophagectomy should not be assumed to improve outcomes. In a retrospective ESCC cohort of 382 patients, only 69 received adjuvant chemotherapy; reported 5-year overall and disease-free survival were 47.1% and 42.6%, respectively, and adjuvant chemotherapy did not improve overall survival in the reported analysis ^[15]B3b.

Unresectable locally advanced disease

Definitive CRT is described as the current standard for locally advanced unresectable esophageal carcinoma, particularly thoracic squamous cell carcinoma ^[4]A1b. JCOG1510/TRIANgLE was designed to test induction docetaxel, cisplatin, and 5-fluorouracil followed by conversion surgery or definitive CRT versus definitive CRT alone; the planned phase III study enrolled 230 patients, with overall survival as the primary endpoint, so the supplied trial description is not evidence of superiority ^[4]A1b.

For unresectable disease treated with definitive CRT, overall survival remains the definitive clinical endpoint. A later correlation analysis was specifically designed to evaluate whether event-free survival can serve as a surrogate for overall survival in trials of definitive CRT, but the supplied abstract does not report its results ^[3]A1a.

Cervical esophageal squamous cell carcinoma may be treated definitively when surgery is unsuitable or not selected. In a retrospective series of 30 patients treated with three-dimensional conformal radiotherapy, the median gross-tumor and involved-node dose was 60 Gy; 26 patients received concurrent chemotherapy and 26 received elective nodal irradiation to cervical and upper mediastinal regions at a median 40 Gy ^[56]C4.

Postoperative, recurrent, and anatomically borderline disease

Adjuvant concurrent CRT after radical resection has been evaluated in patients with stage IIB–IVA disease. In a retrospective series of 125 patients, 122 received more than 50 Gy, but only 52 completed more than 5 weeks of combined treatment, indicating substantial treatment-intensity variation and limiting interpretation of efficacy ^[6]A1b.

For locoregional recurrence after surgery, salvage RT or concurrent CRT may provide a potentially curative or disease-controlling option in selected patients without prohibitive toxicity. A retrospective study of 147 patients used 60 Gy in 20 fractions for RT alone and 60–70 Gy in 30–35 fractions for concurrent CRT; the study evaluated clinical outcomes but does not establish a randomized standard ^[14]C4.

Concurrent cervical-node metastasis in thoracic esophageal or gastroesophageal-junction cancer represents a border zone between locoregional and distant disease. A Dutch nationwide cohort evaluated definitive CRT, neoadjuvant therapy followed by surgery, and chemotherapy with or without limited RT of ≤30 Gy, but the supplied abstract does not provide comparative survival results; treatment should therefore be individualized in a multidisciplinary setting ^[9]B2b.

Response assessment and treatment adaptation

Contrast-enhanced CT is used for diagnosis and initial staging of esophageal squamous cell carcinoma, but conventional CT alone is insufficient to predict early recurrence reliably; CT radiomics and clinical-feature models have been investigated for recurrence prediction after trimodality therapy ^[2]A1b. MRI-based methods have also been studied for early response assessment during neoadjuvant therapy or concurrent CRT, but a 2024 meta-analysis concluded that the clinical role remains under exploration ^[5]B2a. Interim 18F-FDG PET metabolic response during neoadjuvant CRT has been systematically evaluated for predicting pathologic response, progression-free survival, and overall survival ^[8]B2a. Dynamic contrast-enhanced MRI response criteria have been retrospectively studied for identifying pathologic complete response after neoadjuvant CRT, but these data are not sufficient to replace pathological assessment or establish omission of surgery ^[13]B3b.

Role and treatment intent

EBRT is used across the definitive, postoperative, and neoadjuvant management settings for esophageal cancer. The supplied evidence includes definitive radiotherapy or chemoradiotherapy, postoperative radiotherapy after resection, and neoadjuvant chemoradiation followed by surgery. [21-25] Evidence quality is heterogeneous: the involved-field versus elective-nodal irradiation systematic review is graded 2a, whereas most comparative or prognostic studies are retrospective and graded 3b–5. [21-31]

Definitive EBRT and target volumes

The principal target-volume question is involved-field irradiation (IFI) versus elective nodal irradiation (ENI). The 2022 systematic review and meta-analysis specifically compared these approaches in patients receiving definitive radiotherapy or chemoradiotherapy, assessing overall survival and radiotherapy-related adverse events. ^[21]B2a A large retrospective series evaluated 924 patients treated with definitive IMRT; 272 received ENI and 652 received IFI, with propensity-score matching used to compare groups. ^[28]B3b These studies address whether limiting the irradiated nodal volume can preserve disease control while reducing toxicity, but the supplied abstract information does not provide the pooled effect estimates or the matched-study numerical results. ^{[21]B2a[28]B3b}

Definitive IMRT has also been evaluated without chemotherapy. In a retrospective cohort of 103 patients with pathologically confirmed squamous cell carcinoma, patients with stage I–IVA disease received IMRT alone and were analyzed for overall- and progression-free-survival prognostic factors. ^[24]B3b For early disease, a multicenter retrospective study included 196 patients with T1–2N0M0 esophageal squamous cell carcinoma treated with definitive radiotherapy; 78 received radiochemotherapy and 118 radiotherapy alone. ^[31]B3b In that study, 96 patients received three-dimensional conformal radiotherapy and 100 IMRT, while the median doses to both the planning target volume and gross tumor volume were 60 Gy; median follow-up was 59.2 months. ^[31]B3b

A large retrospective IMRT analysis included 1,637 patients treated between 2005 and 2017. Reported 1-, 3-, and 5-year overall-survival rates were 65.9%, 34.2%, and 27.0%, respectively, while corresponding progression-free-survival rates were 45.8%, 25.0%, and 18.5%. ^[30]B3b These outcomes describe an institutional IMRT experience and should not be interpreted as randomized comparisons of dose, target volume, or chemotherapy. ^[30]B3b

Dose and dose-escalation strategies

Standard-dose IMRT has been compared retrospectively with simultaneous integrated boost IMRT (SIB-IMRT) in locally advanced squamous cell carcinoma treated with definitive chemoradiotherapy. ^[23]C4 The standard-dose regimen was 60 Gy in 30 fractions, at 2 Gy per fraction, delivered once daily, five days per week. ^[23]C4 From a single-center population of 1,748 patients, 109 eligible SIB-IMRT patients and 266 standard-dose IMRT patients were selected, and propensity-score matching was used. ^[23]C4

Another retrospective study assessed selective local-tumor dose escalation using SIB-IMRT in 302 patients. ^[27]D5 The overall cohort had a median overall survival of 30.0 months and median disease-free survival of 21.3 months; chemotherapy, clinical TNM stage, and dose per fraction were examined as prognostic variables in multivariable analyses. ^[27]D5 Because these dose-escalation studies are retrospective, their results support further evaluation rather than establishing a universally preferred escalated-dose regimen. ^[23]C4[27]D5

Postoperative and neoadjuvant EBRT

Postoperative IMRT was retrospectively evaluated in 228 patients with resected thoracic esophageal squamous cell carcinoma. ^[22]B3b The reported median postoperative radiation dose was 60 Gy; one-, three-, and five-year overall-survival rates were 89.9%, 56.7%, and 45.1%, respectively. ^[22]B3b These results represent a selected postoperative cohort and do not establish the benefit of postoperative IMRT over surgery alone or another adjuvant approach. ^[22]B3b

In the neoadjuvant setting, a study of 77 patients with locally advanced esophageal cancer treated with IMRT and image-guided radiotherapy followed by surgery examined whether sarcopenia, estimated using total psoas area on a cross-sectional CT image at the L4 level, predicted grade ≥3 acute toxicity, pathological response, and overall survival. ^[25]B3b The study therefore supports assessment of nutritional and skeletal-muscle status when counseling patients for trimodality therapy, although the supplied abstract does not provide the final effect estimates. ^[25]B3b

Advanced planning and charged-particle therapy

Charged-particle therapy, including proton-beam and carbon-ion radiotherapy, has been reviewed as an approach intended to reduce radiation exposure to surrounding normal tissues because of the Bragg-peak dose distribution. ^[26]D5 The review reports that proton therapy offers a more favorable irradiation-dose distribution than X-ray techniques, including IMRT, but it is a clinical-results review rather than evidence of superiority from randomized comparative trials. ^[26]D5

Patient selection and practical interpretation

In elderly patients, definitive concurrent chemoradiotherapy with S-1 has been retrospectively studied in 80 patients older than 70 years; radiation was delivered using IMRT or VMAT, and toxicity was evaluated using radiation-therapy toxicity criteria. ^[29]B3b This experience may inform treatment of selected older adults, but it does not define a standard regimen for all elderly patients. ^[29]B3b Overall, the available evidence supports individualized EBRT planning using IMRT or VMAT, careful consideration of IFI versus ENI, and attention to dose, comorbidity, age, nutritional status, and treatment intent. [21-31] Randomized evidence remains limited for target-volume reduction, postoperative IMRT, selective dose escalation, and particle therapy. [21-27]

Scope and evidence base

Brachytherapy has been investigated as exclusive treatment, an external-beam radiotherapy (EBRT) boost, reirradiation, and palliation in esophageal cancer. The available literature is heterogeneous and includes a systematic review of curative-intent treatment in stage I squamous cell carcinoma, a retrospective single-institution series, a retrospective EBRT-based chemoradiotherapy cohort without intracavitary brachytherapy, and a dosimetric study of iodine-125 seed stents. ^{[7]B2a[32]C4[33]B3b[35]D5}

The highest-level source identified here is a systematic review that searched three bibliographic databases for studies published from 1950 through January 2019. It specifically evaluated brachytherapy as boost therapy in stage I esophageal squamous cell carcinoma, with overall survival and disease-free survival as primary outcomes and locoregional outcomes as secondary outcomes. Palliative treatment was excluded. ^[7]B2a The review was designed to address whether intraluminal brachytherapy improves local control or survival in early disease, a question the authors characterized as controversial. ^[7]B2a

Stage I and superficial disease

The stage I evidence should be interpreted as disease- and selection-specific: the systematic review focused on curative-intent brachytherapy, particularly brachytherapy used as a boost in stage I squamous cell carcinoma, rather than palliative stenting or treatment of all histologies. ^[7]B2a The supplied evidence does not provide the review’s pooled survival estimates, local-control estimates, toxicity results, or definitive comparative conclusion; therefore, these outcomes should not be inferred from the citation alone. ^[7]B2a

A separate retrospective study evaluated 123 patients with superficial esophageal cancer treated from 1998–2015 using definitive EBRT plus systemic chemotherapy without intracavitary brachytherapy. ^[33]B3b The prescription was 60 Gy in 30 fractions to the planning treatment volume, and elective nodal irradiation was not performed. ^[33]B3b Most patients had squamous cell carcinoma (122/123), and the cohort included 27 cT1a and 96 cT1b tumors; the median age was 66 years, with a male-to-female ratio of 106:17. ^[33]B3b Because this study did not administer intracavitary brachytherapy, it provides contextual evidence for definitive EBRT-based chemoradiotherapy but does not establish the efficacy or necessity of a brachytherapy boost. ^[33]B3b

Institutional clinical experience

The Institut Gustave Roussy retrospective series included 90 consecutive patients treated with high-dose-rate esophageal brachytherapy between 1992 and 2018. ^[32]C4 Treatment indications were exclusive therapy in 7 patients, boost after EBRT in 41, reirradiation in 36, and palliation in 6. ^[32]C4 The most frequently prescribed regimens were 3 × 5 Gy for boost treatment and 6 × 5 Gy for exclusive treatment or reirradiation, prescribed at the applicator surface or at 5 mm. ^[32]C4

This series demonstrates clinical use across multiple treatment settings but is retrospective, unicentric, and subject to substantial selection and treatment-era effects. ^[32]C4 The supplied abstract indicates that follow-up included reporting of patients who developed complications, but the excerpt does not provide the complete complication profile, survival outcomes, local-control results, or comparative effectiveness data; these endpoints should therefore not be quantified here. ^[32]C4

Dosimetry and iodine-125 seed stents

A dosimetric study evaluated self-expanding coated esophageal stents carrying iodine-125 seeds. ^[35]D5 The investigators examined dose-distribution characteristics according to seed spacing, stent diameter, stent length, and iodine-125 activity, with the aim of generating a dosimetry table for clinical application. ^[35]D5 The experimental model used a 0.8-cm-thick, 20-cm-diameter, 20-cm-high cylindrical organic-glass setup containing solid paraffin; CT imaging was used to transfer the model information to a treatment-planning system. ^[35]D5 This study is dosimetric and does not establish clinical tumor control, survival, dysphagia relief, migration risk, or toxicity outcomes. ^[35]D5

Practical interpretation

Based on these references, brachytherapy may be considered an institutionally specialized technique for selected curative-intent, boost, reirradiation, exclusive, or palliative strategies, but the evidence provided does not support a universal indication or a definitive survival advantage. ^[7]B2a[32]C4 Decisions should account for disease stage, histology, prior EBRT, treatment intent, anatomy, and the availability of experienced applicator placement and treatment planning; the cited studies demonstrate these different applications but do not provide a validated selection algorithm. ^[7]B2a[32]C4

The esophageal schwannoma imaging study is not evidence for brachytherapy: it concerns 21 surgically resected, pathologically confirmed esophageal schwannomas and evaluates imaging and preoperative diagnosis rather than radiation treatment. ^[34]D5 It should not be used to guide brachytherapy indications for esophageal carcinoma. ^[34]D5

Clinical role

Concurrent chemoradiation (CRT) combines external-beam radiotherapy with radiosensitizing systemic therapy and is used in neoadjuvant, definitive, selective early-stage, postoperative, and salvage settings. The strength of evidence differs substantially by setting, and the available references include randomized trials, meta-analyses, prospective cohorts, and retrospective studies.^{[1]A1a[4]A1b[6]A1b[12]B2b[14]C4[16]C4[17]B3b[18]B3b}

For resectable, locally advanced esophageal cancer, neoadjuvant CRT followed by esophagectomy increases tumor eradication at surgery compared with neoadjuvant chemotherapy alone. A meta-analysis of three randomized trials involving 375 patients found significantly higher R0-resection and pathologic complete-response rates with neoadjuvant CRT than with chemotherapy alone; however, it found no significant improvement in 3- or 5-year progression-free survival or overall survival.^[1]A1a In routine practice, the CROSS regimen consists of five weekly cycles of carboplatin (area under the curve 2) and paclitaxel (50 mg/m²) given concurrently with external-beam radiotherapy delivered as 23 fractions of 1.8 Gy—a total of approximately 41.4 Gy—followed by surgery.^[18]B3b The cited daily-practice cohort evaluated this approach in primary resectable, nonmetastatic esophageal cancer, but its retrospective design limits causal comparison with alternative neoadjuvant strategies.^[18]B3b

Weight and nutritional status require active management during preoperative therapy. In the OGSG1003 exploratory analysis, weight loss during neoadjuvant chemotherapy was investigated in relation to postoperative infectious complications and prognosis in locally advanced squamous-cell carcinoma; the supplied abstract identifies the clinical question and study population but does not provide the final effect estimates.^[37]B2b Patients unable to receive cisplatin because of renal or cardiac dysfunction or advanced age may receive preoperative FOLFOX, although the supporting evidence is retrospective and concerns chemotherapy rather than concurrent CRT.^[81]C

Definitive CRT

Definitive CRT is an established curative-intent option for locally advanced unresectable thoracic squamous-cell carcinoma. The JCOG1510/TRIANgLE phase III protocol describes definitive CRT with concurrent cisplatin plus 5-fluorouracil as the current standard comparator and evaluates whether induction docetaxel, cisplatin, and 5-fluorouracil followed by conversion surgery or definitive CRT improves overall survival; the reference is a trial-design report rather than efficacy evidence.^[4]A1b Definitive CRT is also described as a curative option for cT1bN0M0 squamous-cell carcinoma, although residual or recurrent local disease can occur after an apparent complete remission.^[16]C4

For selected stage I disease, endoscopic resection can guide treatment intensity. In a prospective study of 176 patients with T1b (SM1–2) N0M0 thoracic squamous-cell carcinoma, patients with favorable endoscopic-resection pathology could receive no additional treatment, whereas prophylactic CRT using 41.4 Gy to locoregional lymph nodes was used for selected higher-risk findings, including pT1b disease with a negative margin or pT1a disease with lymphovascular invasion.^[12]B2b After noncurative endoscopic submucosal dissection, retrospective comparisons have evaluated adjuvant CRT against esophagectomy; the cited study included 60 patients—34 treated with CRT and 26 with surgery—but does not establish equivalence because of nonrandom treatment allocation.^[17]B3b

Response assessment and surveillance

Early response assessment may help characterize treatment effectiveness but should not be used alone to replace standard restaging or pathologic assessment. A systematic review and meta-analysis evaluated interim 18F-FDG PET changes during neoadjuvant CRT for prediction of pathologic response, progression-free survival, and overall survival; pooled sensitivity, specificity, and hazard-ratio analyses were the prespecified methods.^[8]B2a MRI-based response assessment remains investigational. A 2024 meta-analysis examined MRI sequences and parameters for early response prediction during neoadjuvant therapy or concurrent CRT, reflecting continuing uncertainty about the optimal imaging approach.^[5]B2a In a prospective cohort of 38 squamous-cell carcinoma patients receiving concurrent CRT, serial weekly diffusion-weighted MRI was obtained; at treatment completion, 20 patients (52.6%) were assessed as having complete response and 18 (47.4%) as partial response.^[39]C4 A retrospective study evaluated dynamic contrast-enhanced MRI criteria for identifying pathologic complete response after neoadjuvant CRT, but its diagnostic performance should be interpreted as early observational evidence.^[13]B3b A 2026 correlation analysis is specifically examining whether event-free survival can serve as a surrogate for overall survival in randomized trials of definitive CRT for unresectable locally advanced disease; the supplied abstract states the objective and methods but does not provide the correlation results.^[3]A1a

Postoperative, recurrent, and special situations

Adjuvant concurrent CRT after radical resection has been assessed retrospectively in stage IIB–IVA disease. In one series, 122 of 125 patients (97.6%) received more than 50 Gy, but only 52 patients (41.6%) received more than 5 weeks of chemotherapy with radiotherapy, indicating substantial treatment-intensity variation.^[6]A1b After neoadjuvant CRT and surgery for squamous-cell carcinoma, retrospective data do not support routine adjuvant chemotherapy: among 382 patients, only 69 (18.1%) received postoperative chemotherapy, and the study reported no overall-survival improvement in the overall cohort.^[15]B3b

Salvage CRT is used for selected locoregional recurrence after surgery. A retrospective study of 147 patients used 60–70 Gy in 30–35 fractions with concurrent chemotherapy, whereas radiotherapy alone was delivered as 60 Gy in 20 fractions; outcomes were reported after a median follow-up of 18.8 months, but selection bias limits treatment comparisons.^[14]C4 For resectable disease with concurrent cervical-node metastasis, a Dutch nationwide cohort compared definitive CRT with neoadjuvant therapy followed by surgery and other strategies, reflecting an evidence-limited border zone between locoregional and distant disease.^[9]B2b Pembrolizumab plus cisplatin/5-fluorouracil improves outcomes in advanced esophageal cancer in the KEYNOTE-590 Japanese subgroup follow-up, but this is systemic first-line therapy evidence and does not establish a role for combining pembrolizumab with concurrent radiotherapy.^[36]A1b

Scope and treatment selection

Adjuvant and palliative radiotherapy (RT) should be selected according to resection status, recurrence pattern, symptom burden, prior treatment, expected survival, and the feasibility of systemic therapy. The available evidence supplied for this section is predominantly retrospective, single-institution, or population-based observational evidence; therefore, treatment associations should not be interpreted as proof of benefit over alternative strategies.^{[6]A1b[9]B2b[32]C4[41]B3b[46]C4[50]D5[52]D5}

Cervical lymph-node metastasis in otherwise resectable thoracic esophageal or gastroesophageal-junction cancer represents a clinical boundary between locoregional and distant disease. A Dutch nationwide cohort specifically compared definitive chemoradiotherapy, neoadjuvant treatment followed by surgery, and chemotherapy with or without limited RT (≤30 Gy) in this population and evaluated overall survival, but the supplied abstract does not provide the comparative survival estimates.^[9]B2b Treatment should therefore be individualized through multidisciplinary review rather than determined solely by the cervical-node location.^[9]B2b

Postoperative and post-endoscopic RT

Evidence for postoperative concurrent chemoradiotherapy remains nonrandomized. In a retrospective series of 125 patients with stage IIB–IVA disease treated after radical resection, 122 patients received >50 Gy (97.6%). Only 52 patients completed more than 5 weeks of concurrent chemotherapy and RT, whereas 73 received 1–4 weeks, indicating substantial difficulty completing combined treatment; the study assessed tolerability, short-term efficacy, and survival.^[6]A1b

A separate retrospective analysis of 195 patients with curatively resected thoracic esophageal squamous-cell carcinoma compared no adjuvant treatment, chemotherapy, RT, and chemoradiotherapy. The median RT dose was 45.0 Gy (range, 34.8–59.4 Gy), and outcomes examined included overall survival, disease-free survival, and locoregional recurrence.^[50]D5 Because treatment assignment was retrospective and the supplied evidence does not report the adjusted comparative results, these data support postoperative RT or chemoradiotherapy as consideration in selected high-risk patients rather than as a universal standard.^[50]D5

For patients undergoing non-radical resection of locally advanced esophageal squamous-cell carcinoma, a retrospective Chinese study evaluated planned neoadjuvant RT in 33 patients and concurrent chemoradiotherapy in 119 patients. Median follow-up was 29.8 months, and 101 patients survived longer than 3 years; the study analyzed survival and prognostic factors, but the supplied abstract does not establish a randomized comparison or identify a preferred regimen.^[46]C4

After endoscopic submucosal dissection (ESD) for superficial clinical T1N0M0 cancer, prophylactic chemoradiation using 40–41.4 Gy has been reported to produce favorable outcomes but does not completely prevent regional lymph-node metastasis. A retrospective series of 44 patients instead examined dose-escalated, definitive-dose RT of 50–61.2 Gy, with or without chemotherapy, after ESD in patients with pathological muscularis-mucosa involvement with lymphovascular invasion or other high-risk superficial disease. This approach should be regarded as selected adjuvant treatment rather than routine therapy because the evidence is retrospective.^[40]B3b

Recurrent and metastatic disease

Recurrence after potentially curative surgery commonly requires individualized salvage planning. A retrospective series of patients recurring after neoadjuvant chemo(radio)therapy and radical esophagectomy, treated between 1994 and 2015, emphasized that potentially curative salvage strategies are available only in limited circumstances and evaluated survival according to dissemination pattern and prognostic factors.^[41]B3b RT may therefore be considered for localized or symptomatic recurrence when anatomy, prior dose, and performance status permit, while disseminated recurrence generally requires integration with systemic therapy.^{[41]B3b[51]D5}

For metastatic esophagogastric cancer, palliative RT to the primary tumor has been studied after initial chemotherapy in patients with well-controlled metastatic disease. Among 132 patients receiving palliative chemotherapy, 97 had responding or stable disease after 3 months; 53 subsequently received primary-tumor palliative RT and 44 continued chemotherapy alone. The study assessed time to local progression and other treatment-related outcomes, but its retrospective design limits causal interpretation.^[52]D5 Real-world systemic-treatment data from 2,204 patients with synchronous metastatic esophagogastric cancer demonstrate substantial heterogeneity in first-line monotherapy, doublet, triplet, and trastuzumab-containing strategies; this study was not designed to determine the value of RT.^[51]D5

Brachytherapy and symptom-directed palliation

High-dose-rate esophageal brachytherapy has been used as exclusive treatment, an external-beam boost, reirradiation, or palliation. In a retrospective Institut Gustave Roussy series of 90 patients treated from 1992–2018, the most frequent schedules were 3 × 5 Gy for boost treatment and 6 × 5 Gy for exclusive treatment or reirradiation, prescribed at the applicator surface or 5 mm depth. The evidence supports feasibility in carefully selected patients but does not define a universally preferred dose or demonstrate superiority over external-beam RT.^[32]C4

Stents, RT, and endoscopic palliation

Self-expanding metallic stents (SEMS) remain an established means of relieving malignant dysphagia or fistula when rapid mechanical palliation is required, but the interaction between SEMS and RT requires caution. In a retrospective cohort of 323 patients, predominantly with squamous-cell carcinoma (79.6%), investigators evaluated adverse events in patients who had received or would receive RT; the study was specifically motivated by uncertainty regarding SEMS safety with RT.^[42]C4 Another retrospective series of 192 patients undergoing 236 SEMS procedures found significant dysphagia-score improvement within 1 week and 1 month, including patients receiving additional palliative treatment.^[45]B3b

Stenting in the cervical esophagus or hypopharynx is technically challenging and may be poorly tolerated. A retrospective review of 69 patients evaluated cervical-esophageal stenting according to whether the hypopharynx was involved.^[43]C4 In patients with prior RT, a low-radial-force Niti-S stent produced major adverse-event rates of 6.3% versus 5.9% without prior RT (p=0.95) in a retrospective comparison of 83 patients, although these findings should not be generalized to every stent design or clinical setting.^[53]D5 A prospective pilot study also evaluated a multisegmented, fully covered SEMS with antimigration features against historical controls, focusing on technical success, clinical efficacy, migration, restenosis, and adverse events.^[54]D5

Photodynamic therapy can be combined with EBRT or SEMS for recurrent obstruction. In 20 patients, dysphagia improved in 90% at 4 weeks; recurrent dysphagia led to stent insertion after a mean of 63 days, and major complications occurred in 10%.^[47]B2b Rare skeletal-muscle metastases were identified in 4 of 205 PET/CT-evaluated patients, with one apparent case subsequently attributed to a second primary pancreatic cancer; PET/CT findings should therefore be confirmed when they would alter RT intent.^[44]C4

Scope and clinical framing

Toxicity after esophageal-cancer radiotherapy (RT) is influenced by treatment intent, concurrent systemic therapy, prior irradiation, tumor location, airway involvement, surgery, and the use of an esophageal stent. The available evidence supplied here is heterogeneous and is dominated by retrospective series, technical reviews, and selected prospective studies; therefore, reported rates should not be generalized across treatment settings. ^{[10]A1b[23]C4[32]C4[42]C4[53]D5[56]C4}

Acute toxicity during and shortly after RT

Acute esophageal mucosal injury commonly presents clinically as worsening dysphagia and impaired oral intake, although the supplied studies do not provide a uniform grading system or pooled incidence. Risk is particularly relevant when RT is combined with chemotherapy: a retrospective target-trial emulation of 432 adults aged ≥70 years compared definitive RT alone with concurrent chemoradiotherapy (CCRT), evaluating treatment completion, grade ≥3 toxicity, and unplanned readmission within 90 days. The study was designed to address the balance between disease control and tolerability in older adults, but the supplied abstract does not report the comparative outcome estimates. ^[10]A1b

In a large retrospective comparison of simultaneous integrated-boost intensity-modulated RT and standard-dose intensity-modulated RT for locally advanced esophageal squamous cell carcinoma, standard treatment was 60 Gy in 30 fractions; the investigators specifically assessed safety alongside efficacy. The evidence base consisted of 109 SIB-IMRT patients and 266 standard-dose IMRT patients selected from a larger cohort of 1,748 definitive-CCRT patients, with propensity-score matching used to reduce baseline imbalance. ^[23]C4

Definitive three-dimensional conformal RT for cervical esophageal squamous cell carcinoma was evaluated in 30 patients, with a median gross-tumor and involved-node dose of 60 Gy; 26 patients received concurrent chemotherapy and elective nodal irradiation. This anatomic site requires particular attention to swallowing, laryngeal, and airway symptoms, although the supplied abstract does not provide detailed acute-toxicity frequencies. ^[56]C4

Airway-involved tumors represent a high-risk toxicity setting. In a retrospective series of 83 patients, 97.6% had squamous histology and all received taxane–platinum induction chemotherapy before reassessment for possible radical therapy. The rationale was to reduce the morbidity and treatment-related mortality associated with immediate radical treatment, particularly the risk of treatment-induced tracheoesophageal fistula (TEF). ^[61]C4

Airway, fistula, and postoperative complications

Esophageal tumors abutting or invading the tracheobronchial tree can cause TEF or bronchoesophageal fistula, aspiration, respiratory infection, and potentially fatal airway compromise. In a retrospective study of 335 patients receiving self-expandable metallic stents (SEMS), 37 patients (11.0%) developed SEMS-induced esophagorespiratory fistula, with a median interval of 129 days after stent placement. ^[57]B3b A separate 20-year case-series review examined 15 patients with TEF or bronchoesophageal fistula and evaluated open surgery, airway or esophageal stenting, and gastrostomy/jejunostomy; the supplied abstract indicates that these fistulas are uncommon but potentially life-threatening, but does not provide treatment-specific outcome estimates. ^[58]C4

Primary tracheobronchial necrosis after esophagectomy is a distinct postoperative complication defined as necrosis without anastomotic leakage or another cervical or mediastinal abscess. A nationwide Japanese questionnaire study collected data from 6,370 esophagectomy patients treated at 67 institutions between 2010 and 2019 and classified severity as grade 1 mucosal necrosis, grade 2 transmural bronchial-wall necrosis without fistula or perforation, and higher grades involving structural complications. ^[55]B3b Postoperative imaging is essential when complications are suspected after Ivor-Lewis esophagectomy because altered anatomy complicates interpretation; anastomotic leak, gastric-conduit necrosis, and pleuropulmonary complications are among the most serious complications. ^[63]D5 Minimally invasive surgery may reduce blood loss and postoperative pain and accelerate recovery, but the reviewed experience did not demonstrate consistent reductions in early postoperative complications or improvements in long-term outcomes. ^[64]D5

Stents, re-irradiation, and late toxicity

SEMS placement in an irradiated esophagus requires caution because RT may alter tissue integrity and healing. In a retrospective study of 323 patients with malignant stenosis or fistula, investigators evaluated whether prior or subsequent RT affected SEMS adverse events; squamous carcinoma accounted for 79.6% of tumors and partially covered stents predominated. ^[42]C4 Conversely, a low-radial-force Niti-S SEMS series included 83 patients, including 32 with prior RT and 51 without prior RT. Major adverse events occurred in 6.3% of the prior-RT group versus 5.9% of the non-RT group, a nonsignificant difference (P=0.95), supporting—but not proving—the feasibility of carefully selected low-radial-force stenting after RT. ^[53]D5 In another retrospective series, SEMS insertion improved dysphagia within 1 week and 1 month; outcomes were compared in patients with and without additional palliative treatment. ^[45]B3b

Late local toxicity includes persistent or recurrent dysphagia, benign stricture, ulceration, bleeding, perforation, and fistula. High-dose-rate esophageal brachytherapy was retrospectively assessed in 90 patients treated for exclusive therapy, external-beam RT boost, re-irradiation, or palliation. Common schedules included 3 × 5 Gy as a boost and 6 × 5 Gy for exclusive treatment or re-irradiation, prescribed at the applicator surface or 5 mm depth; at follow-up, 50% of patients had experienced the reported late clinical event in the supplied abstract, although the event and denominator details are incomplete. ^[32]C4 For refractory benign strictures, a prospective, nonrandomized, single-arm, eight-institution trial evaluated a biodegradable stent after at least 5 dilations or at least 1 radial incision and cutting procedure in patients with dysphagia score ≥2 and an endoscope unable to pass; the primary endpoint was safety and efficacy. ^[60]B2b

Non-esophageal radiation injury and diagnostic pitfalls

Focal hepatic FDG uptake after neoadjuvant chemoradiotherapy may represent radiation-induced liver injury rather than metastasis. A retrospective study and literature review emphasized that this finding can mimic liver metastases on post-treatment FDG-PET/CT and lead to overstaging; interval metastases are detected in almost 10% of patients undergoing restaging, making correlation with treatment fields and imaging morphology important. ^[59]C4

A rare but clinically important toxicity concern is marked radiosensitivity in Fanconi anemia. A case report described a patient with synchronous esophageal and tongue cancers who received 30 Gy to the esophageal mass and later 27 Gy to the tongue surgical bed, with no additional treatment and no evidence of disease at 7 months; the case illustrates the need for individualized risk assessment rather than establishing a standard dose or safety threshold. ^[65]C4

Outcome framework

Overall survival (OS) remains the definitive endpoint for evaluating esophageal-cancer treatment, although its long follow-up requirements can delay assessment of therapeutic benefit.^[3]A1a In unresectable locally advanced disease treated with definitive chemoradiotherapy (dCRT), event-free survival (EFS) is being evaluated as a potential surrogate for OS; however, the validity of EFS in this setting requires trial-level correlation evidence and should not be assumed from resectable-disease studies.^[3]A1a In resectable disease, recurrence-free survival (RFS) or disease-free survival (DFS) may be useful surrogate endpoints, but a 2025 meta-analysis found only a modest pooled association between treatment effects on OS and RFS/DFS, with pooled hazard ratios of 0.90 for OS and 0.87 for RFS/DFS.^[76]B2a

Definitive and neoadjuvant chemoradiotherapy

Treatment response after neoadjuvant therapy or concurrent chemoradiotherapy remains difficult to predict reliably before treatment completion.^[5]B2a A systematic review and meta-analysis evaluated MRI-based response assessment across different sequences and parameters, reflecting ongoing heterogeneity in imaging methods and the need for further validation before routine response-adaptive treatment decisions.^[5]B2a Diffusion-weighted MRI apparent diffusion coefficient (ADC) has also been studied as a predictor of pathologic response to neoadjuvant therapy; evidence was synthesized according to baseline ADC, percentage change, and other response scenarios, but methodological and clinical heterogeneity limit universal thresholds.^[69]B2a

Interim 18F-FDG PET has prognostic and predictive potential during neoadjuvant chemoradiotherapy.^[8]B2a The evidence base assessed early change in standardized uptake value for prediction of pathologic response, progression-free survival (PFS), and OS, including pooled diagnostic performance for pathologic response and pooled hazard ratios for survival outcomes.^[8]B2a Interim PET should therefore be regarded as a risk-stratification tool rather than an established stand-alone indication to escalate, de-escalate, or terminate radiation treatment.^[8]B2a

For patients with resectable esophageal cancer and concurrent cervical lymph-node metastasis, treatment selection remains uncertain because this nodal distribution lies at the boundary between locoregional and distant disease.^[9]B2b A Dutch nationwide cohort compared dCRT, neoadjuvant treatment followed by surgery, chemotherapy with or without limited radiotherapy of ≤30 Gy, and palliation, with OS as a principal outcome.^[9]B2b These population-based findings are relevant when radiation fields and intent are individualized, but they do not replace randomized comparisons of treatment strategies.^[9]B2b

Systemic therapy outcomes relevant to radiation pathways

In advanced esophageal cancer, the Japanese subgroup of the phase 3 KEYNOTE-590 study provided 5-year follow-up for first-line pembrolizumab plus cisplatin/5-fluorouracil versus chemotherapy alone.^[36]A1b OS and PFS were primary endpoints, while objective response rate and safety were secondary endpoints; the earlier analysis had shown numerically favorable OS and PFS with pembrolizumab-containing therapy, and the extended follow-up was intended to characterize durability.^[36]A1b These results inform multimodality treatment planning but do not establish pembrolizumab as a radiosensitizer or define a radiation dose or field.^[36]A1b

For patients receiving second-line nivolumab monotherapy for unresectable or metastatic disease, a multicenter real-world cohort reported a response rate of 23% among patients with measurable lesions and a disease-control rate of 45% in the full cohort.^[70]B2b Grade ≥3 adverse events occurred in 14%, with no treatment-related deaths reported in that cohort.^[70]B2b A separate retrospective study examined whether immune-related adverse events during immune-checkpoint-inhibitor combination regimens were associated with CR/PR rate, PFS, and OS, but retrospective association should not be interpreted as proof that immune toxicity improves survival.^[78]C4

Endoscopic and surgical outcomes that affect radiation selection

Endoscopic eradication therapy is an effective treatment approach for Barrett’s esophagus and related neoplasia, although it carries greater risk of harms and resource use than surveillance endoscopy.^[68]A1c Liquid-nitrogen spray cryotherapy can ablate malignant esophageal tissue and has been used for esophageal disease for more than 20 years, but technical complexity can limit widespread adoption and procedural success.^[82] These approaches are most relevant to selected superficial or residual lesions and should not be conflated with definitive radiation for locally advanced cancer.^[68]A1c[82]

Endoscopic submucosal dissection (ESD) is described as the preferred treatment for early esophageal neoplasms.^[71]A1b When mucosal defects involve >75% of the esophageal circumference, post-ESD stricture formation is a major outcome concern; a randomized trial compared intralesional steroid injection with oral prednisolone using stricture rate, resolution, and number of dilations as endpoints.^[71]A1b Traction-assisted techniques, including clip-line traction and submucosal tunneling, can reduce procedural time during esophageal ESD compared with conventional approaches.^[66]A1a Meta-analysis of endoscopic submucosal tunnel dissection versus ESD evaluated en-bloc and complete resection, procedural time, dissection speed, complications, and recurrence, but the evidence included mixed superficial upper-gastrointestinal lesions.^[77]B2a In a randomized-trial dataset of large superficial esophageal neoplasms, knife type was associated with procedural outcomes and operator handover because of ESD difficulty, emphasizing that local expertise can influence treatment success.^[74]A1b

Surgical morbidity also affects the comparative value of radiation-based strategies. A meta-analysis of transthoracic hybrid minimally invasive esophagectomy versus open esophagectomy included 29 studies and 3,994 patients and found lower overall postoperative morbidity with hybrid minimally invasive surgery, with an odds ratio of 0.66 versus open surgery.^[67]A1a A randomized comparison of mediastinoscope-assisted and conventional transhiatal esophagectomy evaluated perioperative results in 62 patients, although the abstract does not establish superiority for long-term cancer outcomes.^[83] Prolonged thromboprophylaxis after esophagectomy was tested in a randomized study comparing one-month low-molecular-weight heparin with standard in-hospital prophylaxis; biochemical coagulation markers, venous thromboembolism, and mortality were assessed.^[72]A1b Multimodal prehabilitation during neoadjuvant chemotherapy was tested in a randomized feasibility trial using exercise, nutrition, and psychosocial support; recruitment, retention, dropout, and functional outcomes were assessed, but feasibility evidence is insufficient to prove an oncologic survival benefit.^[73]C4

Prognostic modeling and interpretation

MRI-based radiomics combined with clinical factors has been developed to predict DFS and OS in ESCC using training and validation cohorts, with model discrimination assessed by concordance indices and calibration curves.^[75]A1b Such models may support individualized risk estimation but require external validation before guiding radiation dose, target volume, or treatment omission.^[75]A1b Across all strategies, response, local control, recurrence, PFS, DFS/RFS, OS, treatment-related toxicity, postoperative morbidity, and patient function should be reported separately because improvement in an intermediate endpoint does not necessarily demonstrate an OS benefit.^{[3]A1a[76]B2a}

Part of the Esophageal Cancer family. Cross-cutting management is split across dedicated child pages:

• Esophageal Cancer , diagnostic page (definition, epidemiology, staging, biomarkers, prognosis)
• Esophageal Cancer Surgical Management , operations by stage, fertility-sparing options, sentinel node mapping, adjuvant triggers (Sedlis / Peters)
• Esophageal Cancer Systemic Therapy , concurrent / adjuvant / metastatic chemotherapy, targeted therapy, immune checkpoint inhibitors
• Esophageal Cancer Palliative Care , early integration, symptom management, palliative procedures, end-of-life care
• Esophageal Cancer Surveillance and Follow-up , post-treatment surveillance schedule, late toxicity, survivorship, patient counselling
• Esophageal Cancer Recurrent and Metastatic Disease , local-regional salvage, distant metastatic systemic therapy, oligometastatic disease

Pearl: Use these links to hop between management modalities; the parent Esophageal Cancer page carries diagnosis + staging that informs every decision here.