This study was approved by the Medical Ethics Committee of Sun Yat-sen University Cancer Center. Informed written consent was obtained from all participants. Patients diagnosed with OSCC who underwent transthoracic esophagectomy and lymphadenectomy at the thoracic surgery department of Sun Yat-sen University Cancer Center from December 1996 to December 2009 were screened for study recruitment. Our institutional electronic medical record system was initiated in December 1996; therefore, this was considered the date after which highly reliable data could be obtained.

All patients with pathologically confirmed OSCC who fit the following inclusion criteria were included for analysis: (1) pathologic T status of T1, T2, T3, or T4a; (2) pathologic stage of I, II, or III; (3) microscopically complete resection (R0); and (4) patients who underwent surgery or surgery followed by cytotoxic chemotherapy/radiotherapy. Patients with T1N0 were excluded because most of them were treated with endoscopic resection. We did not include patients who received neoadjuvant treatment due to the limited power in predicting the number of LNMs prior to surgery that was observed during the study period.

Preoperative examination included physical examination, blood test, chest X-ray, barium esophagography, endoscopy, and CT scans of the chest, abdomen and cervical region. If tracheal invasion was suspected, a bronchoscope test was recommended. Endoscopic ultrasonography was widely applied beginning in 2003. Positron emission tomography/CT was initiated in 2005, but was used for few cases due to its high cost. Pathologic staging was performed based on the 7^th American Joint Committee on Cancer staging system[26].

Surgical approaches applied in this study included left transthoracic, Ivor-Lewis, and tri-incisional procedures, which were mainly selected based on tumor location and preoperative inspection results. For patients with middle/lower thoracic OSCC with no evidence of superior mediastinum metastasis, we generally performed left transthoracic esophagectomies. For patients with middle/lower thoracic OSCC with evidence of superior mediastinum metastasis, we performed Ivor-Lewis/tri-incisional esophagectomies. In patients with upper thoracic OSCC, we only carried out tri-incisional procedures.

Lymphadenectomy extent was influenced by the surgical approach. For left transthoracic esophagectomy, we could not achieve upper mediastinal lymphadenectomy in most cases, and instead performed standard two-field lymphadenectomy, including subcarinal, lower mediastinal, and upper abdominal dissections. However, for Ivor-Lewis or tri-incisional esophagectomies, an additional upper mediastinal dissection was completed in most cases by so-called extended two-field lymphadenectomy. Cervical lymphadenectomy was not routinely carried out.

Because there were no standard guidelines for the adjuvant treatment of OSCC, treatment options were selected based on tumor stage, the doctor’s opinion, and the patient’s desires. In our hospital, adjuvant treatment was started at 4-6 wk after operation, mostly in patients with LNM. Chemotherapy was typically applied as platinum-based two-drug regimen for 4-6 cycles. Radiotherapy was mainly delivered to the position of the anastomosis, mediastinum, pericardium, and left epiploic lymphatic vessels, at a total dose of 50-60 Gy.

After completion of primary treatment, the patients were asked to participate in outpatient follow-up every 3 mo for the first 2 years, every 6 mo for years 3-5, and every 12 mo thereafter. Regular assessment included physical examination, blood test, endoscopy, chest X-ray, and ultrasound testing. CT scans of the chest, abdomen and cervical region were performed at least once a year. For those could not afford regular follow-up visits, a telephone follow-up was performed. Survival status was reclassified using the best available methods in March 2014. The median time from surgery to the last censoring date for the entire cohort was 72.2 mo.

Statistical analysis was performed using SPSS 19.0 software package (IBM Corp., Armonk, NY, United States). We defined death as the event and overall survival (OS) as the time from the date of surgery to the date of death or final follow-up. Survival analysis was performed using the Kaplan-Meier method and differences between curves were assessed by the log-rank test. Univariate and multivariate Cox regression analyses were used to identify factors associated with prognosis. Statistical significance was assumed at a P < 0.05. The statistical methods of this study were reviewed by Yin Guo from Sun Yat-sen University Cancer Center.

A total of 1490 patients were enrolled as the target population. There were 1150 male and 340 female patients with a median age of 58 years (range: 30-88 years). In this study, 4.1% (61/1490) of the patients were pathologically diagnosed as stage I, 48.6% (724/1490) as stage II, and 47.3% (705/1490) as stage III. The most frequently applied approach was left transthoracic esophagectomy (917/1490; 61.5%), and the median number of resected lymph nodes (LNs) was 14 (range: 1-91). Combined therapy was used for 319 patients, including 194 who received adjuvant chemotherapy, 91 who received adjuvant chemoradiotherapy, and 34 who received adjuvant radiotherapy. The patient characteristics are listed in Table 1.

A total of 631 patients were still alive at the last follow-up, and the median survival time was 35.5 mo. To investigate the influence of nodal status on therapeutic response, we divided the entire cohort based on their nodal status (N0 vs N1 vs N2-3).

The influence of nodal status on surgery is shown in Table 2. The patients with a status of N0 or N1 who underwent the Ivor-Lewis/tri-incisional procedure presented with significantly better survival compared with those who underwent left transthoracic esophagectomy, with adjusted hazard ratios (HRs) of 0.632 (95%CI: 0.499-0.801) and 0.661 (95%CI: 0.516-0.847), respectively (P < 0.05) (Figure 1). Similar results were also observed for the number of resected LNs (Figure 2). For the patients with an N2-3 status, neither the surgical approach nor the number of resected LNs reached significance by univariate analysis, with unadjusted HRs of 0.826 (95%CI: 0.644-1.058) and 0.849 (95%CI: 0.668-1.078), respectively. The results indicate the poor curative effect of surgery in the patients with an N2-3 status.

Open in New Tab   Full Size Figure   Download Figure

Figure 1 Overall survival curves stratified by surgical approaches for esophageal squamous cell carcinoma patients. A: Patients with N0 status; B: Patients with N1 status; C: Patients with N2-3 status (Log-rank test).

Open in New Tab   Full Size Figure   Download Figure

Figure 2 Overall survival curves stratified by resected lymph nodes for esophageal squamous cell carcinoma patients. A: Patients with N0 status; B: Patients with N1 status; C: Patients with N2-3 status (Log-rank test).

The influence of nodal status on combined therapy is shown in Table 3. Compared with surgery alone, the combined therapy did not generate better outcomes in the patients with N0 status (unadjusted HR = 0.878, 95%CI: 0.617-1.249). For those with an N1 or N2-3 status, survival benefits were observed in those who received the combined therapy, with adjusted HRs of 0.712 (95%CI: 0.537-0.944) and 0.672 (95%CI: 0.521-0.867), respectively (Figure 3). These data supported the use of combined therapy to treat OSCC with LNM.

Open in New Tab   Full Size Figure   Download Figure

Figure 3 Overall survival curves stratified by treatment for oesophageal squamous cell carcinoma patients. A: Patients with N0 status; B: Patients with N1 status; C: Patients with N2-3 status (Log-rank test).

To search for an optimal treatment, extensive surgery was defined as Ivor-Lewis/tri-incisional procedure with resection of ≥ 15 LNs, and limited surgery as the other procedures assessed. Therefore, patients were classified into four groups based on their therapeutic regimens, including Group A (limited surgery alone), Group B (extensive surgery alone), Group C (limited surgery followed by adjuvant treatment), and Group D (extensive surgery followed by adjuvant treatment). The prognosis of patients with N0 status who underwent surgery alone was comparable to that of the combined therapy, and the longest survival times were observed in Group B and Group D (Figure 4A). The optimal treatment for the patients with N1 status was extensive surgery followed by adjuvant treatment (Group D) (Figure 4B). The outcomes of limited surgery for the patients with an N2-3 status were similar to those of extensive surgery, and the best curative effects were observed in Group C and Group D (Figure 4C).

Open in New Tab   Full Size Figure   Download Figure

Figure 4 Overall survival curves stratified by therapeutic regimens for esophageal squamous cell carcinoma patients. A: Patients with N0 status; B: Patients with N1 status; C: Patients with N2-3 status (Log-rank test). Group A, limited surgery alone; Group B, extensive surgery alone; Group C, limited surgery followed by adjuvant treatment; Group D, extensive surgery followed by adjuvant treatment.

Esophageal cancer is a frequent cause of death worldwide, and the traditional management has always been surgical resection. In previous decades, great improvements in preoperative examination and surgical technology have led to significantly enhanced long-term survival. However, survival is still unsatisfactory for locally advanced esophageal cancer, which has prompted an evaluation of neoadjuvant (preoperative) and adjuvant (postoperative) combined-modality therapy. Until now, the optimal multimodality therapy has not yet been established.

Lymph node metastasis (LNM) is one of the worst prognostic factors for localized esophageal cancer. Furthermore, LNM decreases the curative effects of surgery and enhances the necessity of multimodality treatment. Therefore, we speculated that the assessment of nodal status might facilitate the selection of the most effective treatment.

With a median follow-up time of 72.2 mo, a total of 1490 patients were analyzed for survival and the median survival time was 35.5 mo. The surgical approach (left transthoracic vs Ivor-Lewis/tri-incisional) was verified as independent prognostic significance in patients with an N0 or N1 status, but not in those with an N2-3 status. Similar results were also observed with the number of resected LNs (≤ 14 vs≥ 15). Compared with surgery alone, combined therapy achieved better outcomes in patients with N1 or N2-3 status, but not in those with an N0 status. For those with N2-3 status, neither the surgical approaches nor the number of resected LNs was significant, and aggressiveness of surgery did not influence the outcome; the longest survival was observed in those patients who received the combined therapy.

Our results demonstrate the positive role of combined therapy in esophageal squamous cell carcinoma with LNM, and that aggressive surgical resection does not improve survival in patients with an N2-3 status.

This retrospective study with a large cohort of patients with esophageal squamous cell carcinoma (n = 1490) contributes to the existing literature regarding surgical therapy of esophageal cancer. It adds relevant data and concludes that aggressive surgery is not justified at an advanced tumor stage. The limitations of this study are clear and declared by the authors. The aspect of a neoadjuvant therapy is not included but discussed and answered in other studies.