Effect of more versus less frequent abdominopelvic computed tomography follow-up testing on overall survival in patients with stage II or III colon cancer.

Abstract

3604 Background: Patients with colon cancer undergo an abdominopelvic computed tomography (AP-CT) scan surveillance every 6-12 months for five years after surgery. However, evidence of survival benefit from more intensive AP-CT patients in stage II and III colon cancer is still elusive. Methods: A total of 2,137 patients with stage II-III colon cancer who received curative aim surgery were analyzed in the Yonsei Cancer Registry Database between Jan 1, 2005, to Dec 31, 2015. The surveillance start date was defined as 90 days after curative resection or one day after adjuvant chemotherapy. All patients had at least one AP-CT for two years after the surveillance period. The average interval of AP-CT per patient was calculated as the postoperative surveillance duration divided by the number of AP-CT examinations. Patients who underwent AP-CT with an average interval of 6 months and 12 months were assigned to the high-frequent and low-frequent surveillance groups. Association of AP-CT frequency and demographic factors with overall survival (OS) were evaluated. Univariate and multivariate analyses were conducted. Results: Among 2,137 patients who underwent curative aim surgery, the median intervals of AP-CT scan were 6.1 months in high-frequent surveillance and 10.8 months in the less-frequent surveillance group, respectively. The 5-year overall survival (OS) was not significantly different in both groups; however, OS was significantly longer in the high-frequent surveillance group of patients who harvested lymph nodes less than 12 in curative resection ( P = 0.015) or measured postoperative serum carcinoembryonic antigen (CEA) level higher than 5 ng/ml ( P = 0.023). Of note, high-frequent surveillance group detected liver metastasis earlier (6.7 months) than the low-frequent group, leading to more curative aim metastasectomy (28.0% vs 4.3%, P = 0.020), and significantly longer 5-year OS (62.2% vs 34.8%, P = 0.017). Multivariate analysis showed that harvesting lymph nodes less than 12 (HR = 4.37, P = 0.004) and postoperative CEA level higher than 5 ng/ml (HR = 7.39, P < 0.001) were prognostic factors for overall survival. Conclusions: The average interval of AP-CT in patients with stage II-III colon cancer is not associated with improved OS. However, a high-frequent AP-CT enabled early detection of liver metastases, leading to the improved OS in high-risk patients, including the number of harvested lymph nodes less than 12 or postoperative CEA level is higher than 5 ng/ml. The risk-stratified approach is warranted to guide postoperative colon cancer surveillance.