Superiority of trimodality therapy to surgery alone in esophageal cancer: Results of CALGB 9781

Reviewer: Charles Wood, MD
Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 6, 2006

Presenter: Tepper JE
Presenter's Affiliation: UNC Lineberger Comprehensive Cancer Center, Chapel Hill, NC
Type of Session: Scientific


  • There is disagreement regarding the proper treatment of resectable esophageal carcinoma
  • The purpose of this study was to compare surgical resection alone versus neoadjuvant chemoradiation followed by surgical resection for patients with resectable esophageal cancer

Materials and Methods

  • The study design called for accrual of 500 patients with 3 years of planned follow-up
  • The primary endpoint was overall survival (OS), and secondary endpoints included local control (LC), distant control (DC), and toxicity
  • Eligibility requirements included patients with Stage I-III adenocarcinoma or squamous cell carcinoma tumors of the esophagus ≥ 2 cm from the gastroesophageal junction, no metastases beyond the regional draining nodes (<1.5 cm celiac nodal involvement was allowed for distal tumors), and adequate organ function
  • Staging requirements included EGD with biopsy, barium esophagogram, and CT; additional studies including EUS and/or thoracoscopy/laparoscopy were recommended but not required
  • Patients were randomized to either surgery alone or neoadjuvant cisplatin (100 mg/m 2 ) and 5-FU (1000 mg/m 2 x 4d) on weeks 1 and 5 concurrent with radiotherapy (RT) to a dose of 50.4 Gy in 1.8 Gy fractions over 5.6 weeks followed by esophagectomy with lymph node dissection
  • Preferred surgical resection included left or right thoracotomy with maximal nodal dissection and was performed in the trimodality group 3-8 weeks following the completion of neoadjuvant treatment; transhiatal esophagectomies were discouraged but allowed
  • Patients were stratified according to nodal status, histology, and invasive versus non-invasive staging


  • This was a phase III prospective, multicenter, randomized trial with analysis via intent-to-treat
  • A total of 56 patients were enrolled onto the study between October 1997 and March 2000 at which time the study was closed due to poor accrual; 30 patients were randomized to the trimodality arm and 26 to the surgery alone arm
  • Median follow-up was 6 years
  • Patients were primarily male, Caucasian, and with adenocarcinoma histology (~75%)
  • The treatment arms were well-balanced according to major prognostic factors
  • Response rate data was available for only 25 patients; complete response (CR) was defined as absence of tumor on surgical pathologic examination; partial response (PR) was defined by shrinkage of tumor on surgical pathologic examination
  • CR rates were 40%, and CR and PR were 80% in the trimodality group
  • The primary Grade 3-4 toxicities in the trimodality group were hematopoietic (54%) and esophagitis/dysphagia (40%)
  • 14 and 17 patients on the trimodality and surgery arms, respectively, had surgical complications, with 2 post-surgical deaths within 30 days on the surgery alone arm
  • There were 2 anastomotic leaks in the trimodality group
  • Postoperative hospital stays were 11.5 days (range 0-24 days) in the trimodality arm and 10 days (range 0-56 days) in the surgery alone arm
  • There was a statistically significant OS advantage favoring the trimodality group (4.5 years vs. 1.8 years, p=0.02), and a log rank test stratified by N stage, staging approach, and histology yielded a p value of 0.005
  • 5-year OS was 39% vs. 16 % favoring the trimodality arm
  • There was a statistically significant progression-free survival (PFS) benefit favoring the trimodality group (p=0.01)
  • There were 9 relapses in the trimodality group (3 local and distant, 1 local only, and 5 distant only)
  • There were 14 relapses in the surgery alone group (4 local only and 10 distant only)

Author's Conclusions

  • This study demonstrated a long-term survival benefit for patients receiving trimodality treatment
  • Trimodality treatment represents the standard of care in patients with resectable esophageal adenocarcinoma

Clinical/Scientific Implications

This study, though small, demonstrates a clear OS and PFS advantage in favor of trimodality treatment versus surgery alone for patients with resectable esophageal cancer. Though it possesses little power, this is essentially irrelevant, as statistical significance was demonstrated for the survival endpoints.

It should be noted that the pattern of failure data is incomplete, and those results are invalid from which to draw conclusions. Ultimately, trimodality therapy should be considered the clear standard of care in treatment of patients with resectable esophageal cancer.