Surveillance frequency in resected esophageal cancer: Towards personalization of follow-up

INTRODUCTION: In spite of advances in curative management of esophageal cancer, a significant proportion of patients have early recurrence following resection. The role of CT-guided surveillance remains undefined. This study aims to determine if follow-up can be personalised, to allow detection of clinically relevant recurrence, while reducing low-yield surveillance for patients. METHODS: A retrospective review was conducted encompassing patients undergoing esophagectomy with curative intent from 1st March 2018-31st May 2022. Routine 3-monthly CT scanning was conducted for 2 years, followed by 6-monthly surveillance for 1 year, and annual surveillance to 5 years. Disease characteristics, time to recurrence and time to death were recorded and interrogated to determine their impact on recurrence and personalization of surveillance. RESULTS: In total, 190 patients underwent surveillance. Seventy-one (37 %) developed recurrence, with most in the first two years. Those who recurred were younger (63 vs 67, p < 0.001), had higher pathologic staging (p < 0.001), higher tumour regression grade (p = 0.005), higher lymph node ratio (p < 0.001) and high-risk histology (p < 0.001). Most recurrences detected were asymptomatic (94 %). A personalised surveillance score was devised. With strict criteria, 12 % of patients could be excluded from surveillance without compromising detection of asymptomatic recurrence. By broadening criteria, a larger portion of patients could avoid imaging, with a small number of asymptomatic recurrences missed. This would require significant balancing of the risk-benefit ratio for individuals. CONCLUSION: Intensive surveillance post-resection of esophageal cancer will detect most recurrences while asymptomatic, potentially facilitating intervention. In select patients, routine surveillance could be excluded without compromising oncologic or patient outcomes.