Colorectal cancer (CRC) patients surveyed by 18FDGPET-CT (PET-CT): An open-label multicenter randomized trial (NCT 00624260)

3520 Background: Curative surgery is the best therapy of CRC and recurrences. We assessed whether adding semi-annual PET-CT to the usual surveillance would be cost-effective in high risk recurrent CRC patients. Methods: CRC patients (stage II tumor perforated, stages III and IV) in remission after curative surgery were randomly assigned (1:1) to trimester usual surveillance (control) or usual surveillance + semi-annual course PET-CT (intervention) for a 3-yr follow up period. Every 3 months, multidisciplinary committee decided about recurrence by yes/no/doubtful. If yes, curative surgery alone (when relevant), or chemotherapy alone (unresecable recurrence) were conducted; additional exams could be performed if doubtful. Primary composite endpoint (failure) comprised unresectable recurrence & death. The economic assessments according to standards (CHEERS) were performed and costs were compared between groups. Statistical tests for calculation of the relative risk (RR) were used and survival was analyzed using Kaplan-Meier method, Log-Rang test and Cox models. Results: Baseline characteristics of 239 patients (120/119) enrolled in 12 centers were balanced. The failure rate was 29.2% (31 unresectable recurrences & 4 deaths) and 23.5% (27 & 1) in Interventional vs Control, respectively with no significant difference (RR = 1.24, 95% CI: 0.81-1.90; P = .32). Similar results were observed in multivariate analysis (Cox models) adjusted for stage and tumor differentiation (HR = 1.33, 95% CI: 0.8-2.19, P = .27). Period until the unresectable recurrence was significantly shorter in Interventional (median = 7; IQR: 3-20 months) than in Control group (14.3; 7.3-27; P= 0.016). This was consistent with lower elevation (median; IQR) of tumour marker in interventional (3.8; 2.8-19) than in control group (10; 5.2-28.6) at the first recurrence time as compared to the baseline (p = 0.007). Overall (mean; SD) cost (euros)/patient was higher in the PET-Scan (9385; 11658) than in the control group (7027; 7656). Conclusions: Although recurrences were detected earlier in PET-CT group, the strategy was less effective, more expensive. This exam should not be advised routinely. Clinical trial information: NCT 00624260.