Involved-field irradiation or elective-nodal irradiation in neoadjuvant chemo-radiotherapy for locally-advanced esophageal cancer: comprehensive analysis for dosimetry, treatment-related complications, impact on lymphocyte, patterns of failure and survival

Purpose To compare the differences between involved-field irradiation (IFI) and elective nodal irradiation (ENI) in selecting the optimal target area for neoadjuvant chemoradiotherapy (nCRT) in patients with locally advanced esophageal squamous cell carcinoma (LA-ESCC). Materials and methods We retrospectively analyzed 267 patients with LA-ESCC, of whom 165 underwent ENI and 102 underwent IFI. Dosimetry, treatment-related complications, pathological responses, recurrence/metastasis patterns, and survival were compared between the two groups. Results The median follow-up duration was 27.9 months. The R0 resection rates in the IFI and ENI groups were 95.1% and 92.7%, respectively (p=0.441), while the pathological complete response (pCR) rates were 42.2% and 34.5%, respectively (p=0.12). The ENI group received higher radiation doses to the heart (HV30:23.9% vs. 18%, p=0.033) and lungs (LV30:7.7% vs. 4.9%, p<0.001) than the IFI group. Consequently, the ENI group showed a higher incidence of grade 2 or higher radiation pneumonitis (30.3% vs. 17.6%, p=0.004) and pericardial effusion (26.7% vs. 11.8%, p=0.021) than the IFI group. Post-operation fistulas were observed in 3 (2.9%) and 17 cases (10.3%) in the IFI and ENI groups, respectively (p=0.026). In the multivariate analysis, smoking, positive lymph node involvement (pN+), and anastomotic fistula were independent predictors of overall survival (OS). The pN+ patients exhibited a greater propensity for recurrence compared to pN- patients, especially in the first year of follow-up (6.67% vs. 0.56%, p=0.003). Conclusion The ENI group had a higher incidence of radiation-induced adverse events compared to the IFI group, likely due to the higher radiation doses to normal tissues. Considering the similar disease-free survival (DFS) and OS rates in the two groups, IFI may be suitable for nCRT in patients with LA-ESCC, although further prospective studies are warranted.


Introduction
Esophageal cancer is the seventh most prevalent malignancy and the sixth leading cause of cancer-related fatalities (1).Despite advances in treatment, the disease is characterized by a high incidence of local and distant recurrence following surgical resection, leading to a dismal 5-year OS rate that rarely exceeds 30% (2).The long-term findings of the landmark CROSS trial established the survival benefits of neoadjuvant radiotherapy combined with surgery for resectable esophageal cancer, setting the standard of care for locally advanced cases (3,4).However, the optimal radiation field for neoadjuvant radiotherapy remains controversial.Although the CROSS trial assessed the efficacy and recurrence patterns in the involved-field irradiation (IFI) group, it did not compare these parameters with those in the elective nodal irradiation (ENI) group (3).A retrospective study including 118 patients with esophageal squamous cell carcinoma (ESCC) receiving neoadjuvant chemoradiotherapy (nCRT) compared the efficacy and failure patterns between the ENI and non-ENI groups, as both were applicable to most of the population because of the similar prognoses of the two groups.However, considering the higher risk of complications in older patients (>70 years old), the authors recommended IFI only for this subgroup (5).
This study compared the application of ENI and IFI to evaluate the optimal radiation fields in terms of the dosimetric parameters, chemoradiotherapy-related and operational complications, impact on lymphocytes, pathological response, disease-progression patterns, and clinical outcomes.

Patients
Patients with locally advanced esophageal cancer (AJCC Ver. 8, stages II-IVA) were reviewed at the West China Hospital between March 2017 and October 2022.All patients had pathologically confirmed squamous cell carcinoma and underwent nCRT followed by radical esophagectomy.Patients with performance status (PS) ≥ 2, distant metastases, prior chest radiation therapy, or coexisting other malignant tumors were excluded.The clinical stage was assessed by the following examinations, including esophagography, endoscopy, and computed tomography (CT), with some patients undergoing positron emission tomography.

Surgery
Surgery was scheduled 4 to 8 weeks after completion of nCRT.Most of the patients underwent McKeown esophagectomy (240/ 267, 89.9%).R0 resection was defined as complete resection of the tumor, with no tumor visible under the naked eye or microscope.A microscopic residual tumor (R1) was defined as a tumor located < 1 mm from the proximal, distal, or circumferential resection margins.

Pathological analysis
Pathological complete response (pCR) was defined as the absence of cancer cells in the primary lesion and regional lymph nodes after neoadjuvant therapy.Based on the degree of response of the primary tumor to treatment, the evaluation protocol for esophageal cancer (modified Ryan scheme for tumor regression score) from the College of American Pathologists classified the tumor regression grade (TRG) into four stages as follows: grade 0, no surviving cancer cells; grade 1, single cells or rare small groups of cancer cells; grade 2, residual cancer with evident tumor regression but more than single cells or rare small groups of cancer cells; and grade 3, extensive residual cancer with no evident tumor regression (11,12).

Treatment-related complications
Complications of nCRT included hematological toxicity (anemia, thrombocytopenia, leukopenia, neutropenia, and lymphopenia), radiation pneumonia, radiation esophagitis, and radiation heart disease.Postoperative complications included fistula, esophageal stenosis, pleural effusion, pneumothorax, pulmonary atelectasis, acute respiratory distress syndrome (ARDS), mortality, and readmission within 30 days.We scored the severity of treatment-related complications using the Common Terminology Criteria for Adverse Events (CTCAE) 5.0.

Definition of endpoint and patterns of failure
The main endpoints were OS, disease-free survival (DFS), complications, pathological response, and failure modes.In-field failure (IFF) was defined as the presence of recurrence or metastasis within the irradiated field area.Out-of-field failure (OFF) was defined as the presence of recurrence or metastasis outside the irradiated field area.Patients in both groups had local recurrences.

Statistical analysis
The x2 test or Fisher's exact test were used to compare the differences in the patient and tumor characteristics, toxicity, and first failure between the ENI and IFI groups.Spearman's correlation analysis was performed to analyze the correlation between two continuous variables.The time between the start of nCRT or surgery and date of death, recurrence, or last follow-up was used to compute the survival outcomes.OS and DFS were calculated using the Kaplan-Meier method and compared using the log-rank test.Cox proportional hazard models were used to conduct both single-factor and multi-factor analyses as well as to compute the hazard ratios (HR) and 95% confidence intervals (CI).Variables with p <0.1 in the univariate analysis were entered into the multivariate model.SPSS (version 26.0), R Studio (version 4.2.3), and GraphPad Prism 9 software were used for all analyses.Statistical significance was set at p < 0.05.

Patient characteristics
A total of 267 patients with esophageal cancer who completed preoperative nCRT followed by radical surgery were enrolled in the study.Among these, 102 (38%) were in the IFI group and 165 (62%) were in the ENI group.Table 1 shows the characteristics of the 267 patients.The median numbers of lymph node dissections were 25 (range:15-46) and 24 (range:16-43) in the IFI and ENI groups, respectively.The R0 resection rates in the IFI and ENI groups were 95.1% and 92.7%, respectively.Age sex, PS, smoking, tumor location, length, and stage, radiation dose, and R0 resection were not significantly different between the IFI and ENI groups (all p > 0.05).

Dosimetric parameters
The radiation doses in the IFI and ENI groups were 40.0-41.4Gy and 39.6-41.4Gy in 1.8-2.0Gy fractions, respectively.Table 2 presents the cardiac and pulmonary dosimetric parameters of the IFI and ENI groups.Patients who underwent ENI had significantly higher heart V 30 and lung V 5 , V 10 , V 20 , and V 30 values than those in the IFI group (p < 0.05).

Treatment-related complications
All serious adverse events that occurred during treatment are summarized in Table 3. Hematological toxicity has emerged as the most prevalent complication of radiation therapy.In the IFI and ENI groups, grade 3 or higher leukopenia was observed in 35 (34.3%) and 36 (21.8%) patients, respectively, and neutropenia was observed in 30 (29.4%) and 23(13.9%)(p < 0.05) patients, respectively.Grade 2 pericardial effusion (p=0.021) and radiation pneumonitis (p=0.004)occurred at considerably higher rates in the ENI group than in the IFI group (30.3% vs. 17.6% and 26.7% vs. 11.8%,respectively).No significant differences in arrhythmia or radiation esophagitis were observed between the two groups.There were 20 postoperative fistulas: three (2.9%) in the IFI group and 17 (10.3%) in the ENI group (p=0.026).Within 30 days of surgery, four patients in the IFI group and six in the ENI group were readmitted for anastomotic fistula (seven cases), respiratory failure (two cases), and wound infection (one case).There were two deaths in each group within 30 days after surgery: two from hemorrhage and two from severe pneumonia.Intraoperative bleeding, surgery duration, ARDS, pleural effusion, pneumothorax, atelectasis, and esophageal stenosis were not significantly different between the two groups (all p > 0.05).
The differences in the three disease progression patterns of ALLIFF, ALLOFF, and ALLDM between the pN-/pN+ groups are summarized in the follow-up years (Figure 3B).The risk of ALLIFF was greater in the pN+ group than in the pN-group (but only in the first year, p=0.003).This difference gradually diminished over the next two/three years.Although there was no significant difference in the overall pattern of recurrence between the pN-/pN+ groups, the pN+ group showed a higher tendency of recurrence (25.6% vs. 16.4%,p=0.074).

Discussion
ESCC is one of the most prevalent cancers in Asian countries and is typically locally advanced or advanced when first diagnosed and has a high fatality rate (13).In patients with LA-ESCC, the combined use of nCRT and surgery has a considerable survival advantage over surgery alone.The CROSS trial and NEOCRTEC 5010 study laid the foundation for nCRT plus surgery as the    standard of care for this patient population (3,(14)(15)(16).A large sample study based on the National Cancer Database showed that in neoadjuvant radiotherapy, the pCR and OS did not differ between the three higher radiation doses [39.6-44.9 vs. 45-49.9Gy vs. 50 Gy; pCR (p = 0.1) vs. OS (p = 0.097)] ( 17).While higher radiation doses could increase toxicity, all patients in this study had radiation doses of 39.6-41.4Gy.However, the current guidelines do not provide explicit recommendations on the scope of target outlining for neoadjuvant RT.Based on the similar values of OS and DFS obtained in our study, the IFI technique might effectively decrease the radiation dose to normal tissues and consequently reduce the treatment-related adverse effects compared to ENI.ESCC is more likely to metastasize through the esophageal axial lymphatics to multilevel lymph nodes or lymph nodes far from the primary site because of extensive longitudinal lymphatic connections within the esophageal wall (18,19).The theoretical rationale for ENI is its ability to control lymph node micrometastases and potentially enhance the treatment efficacy by irradiating larger anatomical areas (19).However, the comparative efficacies of IFI and ENI have been found to be inconsistent in many studies on definitive chemoradiotherapy for LA-ESCC.A study conducted at the University of Tokyo Hospital involving 239 cases of esophageal cancer revealed that IFI did not increase the risk of lymph node failure in clinically unaffected nodal stations and demonstrated superior progression-free survival (PFS) and OS compared to the ENI group (20).Similarly, in a study of definitive radiotherapy for locally advanced non-small cell lung cancer (LA-NSCLC), IFI did not increase the incidence of lymph node failure in uninvolved nodal sites but significantly reduced esophageal toxicity (21).Conversely, a retrospective analysis of a larger sample of patients with ESCC favored ENI over IFI in terms of improved OS, with comparable toxicity profiles between the two groups (19).Furthermore, a meta-analysis indicated comparable rates of local control and OS in the ENI and IFI groups; however, the latter exhibited significantly lower incidences of esophageal and pulmonary toxicity (22).Several studies have shown that ENI can  did not receive adjuvant chemotherapy, those who received adjuvant chemotherapy benefited in terms of OS in all stages of lymph node (37).Therefore, patients with pN+ ESCC may benefit from adjuvant chemotherapy.Recently, the results of the CheckMate-577 study confirmed that adjuvant immunotherapy may improve tumor-free survival in patients with high-risk esophageal cancer who did not achieve pCR after nCRT and R0 surgery, and that adjuvant nivolumab reduced the risk of distant metastases after surgery compared to placebo treatment (29% vs. 39%), with a median distant metastasis-free survival of 28.3 months and 17.6 months, respectively (38).Thus, patients with pN+ may benefit from postoperative adjuvant immunotherapy.Interestingly, our data indicated an increased risk of anastomotic leakage in the ENI group compared to the IFI group according to the surgical procedure, whereas no significant differences were observed in the occurrence of pleural effusion, pneumothorax, esophageal stricture, and other postoperative complications.The larger radiotherapy target area of ENI may cover the anastomotic site in the currently used thoracoscopic approach for esophageal cancer surgery.Radiation exerts negative effects on wound repair through various mechanisms, including vascular system alterations, inflammatory response changes, and cellular function disruption (39,40).Collagen, a vital matrix protein responsible for the strength and integrity of intestinal wall anastomosis, can be significantly hindered by high radiation doses, thereby affecting anastomotic healing (41,42).Consequently, considering the range of target areas outlined for nCRT, the IFI may be a more suitable approach in the nCRT settings.
This study has some limitations.First, this study has the inherent limitations associated with both retrospective and observational studies.Future prospective studies with randomized controlled designs will provide stronger evidence in this field.Second, our study focused exclusively on patients with ESCC, which limits the generalizability of our findings to other histological types of esophageal cancer, although almost 95% of them are squamous cell carcinomas in Asian countries.Furthermore, as this was a single-center study, the external validity of our results may be limited.Findings from a single institution may not fully represent diverse patient populations encountered in broader clinical settings.Therefore, we anticipate the emergence of large-scale multicenter prospective studies in the future.

Conclusion
In conclusion, our study demonstrates that IFI is not inferior to ENI in terms of the pathological response and survival outcomes.The smaller target area of IFI has the potential to reduce cardiopulmonary irradiation, leading to a decrease in treatmentrelated adverse effects, which theoretically supports the utilization of IFI in nCRT for esophageal cancer.Patients with pN+ disease after nCRT are more likely to experience recurrence and metastasis, which are associated with a poorer prognosis, thus requiring more comprehensive treatment options.

2 (
FIGURE 2 (A) Disease-free survival (DFS) analysis and (B) overall survival (OS) analysis of elective-nodal irradiation (ENI) and involved-field irradiation (IFI).(C) Heatmap of covariance test for variables with p<0.1 in univariate analysis.(D) Forest plot of Cox multivariable regression analysis for OS.

TABLE 2
Comparison of dosimetric parameters between ENI and IFI groups.

TABLE 1
Demographic and baseline variables and treatment characteristics of the study population.

TABLE 3
Adverse events analysis based on neoadjuvant chemoradiotherapy and surgery.Events of grade ≥3 according to CTCAE 5.0.§, Events of grade ≥3 with fever or grade 4 according to CTCAE 5.0.*, Events of grade ≥2 according to CTCAE 5.0; ¶, Events of any grade according to CTCAE 5.0.

TABLE 5
Distribution of pathologic response and stage after surgery.

TABLE 6
Sites of the first treatment failure between IFI and ENI groups.
FF, in-field failure; OFF, out-of-field failure; DM, distant metastasis; ALL IFF, all in-field failure; ALL OFF, all out-of-field failure; ALL DM, all distant metastasis.