Evaluation of routine image follow-up in nonmetastatic colorectal cancer after curative surgical resection.

Abstract

524 Background: Follow-up surveillance is performed after primary treatment in colorectal cancer (CRC), but it is controversial in the literature the real benefit of an intensive examination in terms of outcomes and resources. Intensive follow-up after surgery for colorectal cancer has been challenged by some new published data (CEA watch trial and FACS trial). These new data suggest that a less intensive follow-up program based on carcinoembryonic antigen (CEA) measurements or CEA-Triggered imaging would be enough to detected most of the recurrences. We believe that there is a high percentage of patients with recurrence disease and normal CEA value, for whom image screening would be necessary to detect early disease recurrence suitable to metastasectomy with curative intention. Methods: This is a retrospective study that included 372 patients from a tertiary cancer center in São-Paulo (Brazil) diagnosed with colorectal adenocarcinoma stages I to III. We observed, after primary treatment, a pattern of recurrence detected either by CT (computed tomography) imaging only or CEA elevation and CT image in combination. Results: Out of the 372 patients analyzed, 110 (29,5%) had recurrent disease with a median follow-up time of 34 months. Of the 110 recurrences detected, 75 (68,18%) were detected by CEA elevation in combination with CT image, 33 (30%) were detected only by CT image and 2 (1,81%) neither by CT nor by CEA alteration. There was no clinic feature that would predict pattern of recurrence when analyzed by qui square test. Metastasectomy rate from this analysis 53,6% and it was similar among both groups. Recurrence rate after metastasectomy was 59,3%. There is a 5-year overall survival difference between patients that underwent or not metastasectomy (79,4% vs. 54%, p 0,01). Conclusions: CEA-based follow-up program and CEA-triggered imaging failed to detect early recurrence in almost 30% of cases. We believe that this number is high enough to allow us to continue to perform image test during CRC follow-up.