Background: Previous studies demonstrated that a decrease of the standardized uptake value between pretreatment and posttreatment positron emission tomography (PET) scans can predict histopathologic treatment response in patients with esophageal cancer.

Methods: Forty-seven patients who underwent PET-computed tomography (CT) scans before (scan 1) and after (scan 2) neoadjuvant chemoradiotherapy and during the follow-up period after surgery (scan 3) were included in this study. It was evaluated whether decrease of metabolic tumor length between scan 1 and scan 2 can predict histopathologic response to treatment. Moreover, the value of PET-CT was compared with PET in the assessment of tumor recurrence based on a visual analysis of scan 3. Reference standards for treatment response and recurrence were histopathology results.

Results: The reduction of tumor length between before and after chemoradiotherapy scans (between scan 1 and scan 2) was a better predictor of histopathologic response and of time to recurrence than the decrease in standardized uptake value. The most accurate differentiation was achieved when using a cut-off value of 33% reduction of the initial tumor length. Using this threshold to define metabolic response, the sensitivity was 91% (19 of 21) and the specificity was 92% (24 of 26) for predicting histopathologic treatment response. Based on a visual analysis, PET-CT was more accurate than PET in the differentiation of tumor recurrence from posttreatment tissue changes. Integrated PET-CT achieved a sensitivity of 91% (48 of 53) and a specificity of 81% (30 of 37) in identifying sites of tumor recurrence, compared with 83% (44 of 53) and 65% (24 of 37) with PET.

Conclusions: Decrease of tumor length was shown to be a better predictor of treatment response and disease-free survival than decrease of standardized uptake value. Furthermore, PET-CT is more accurate in the evaluation of recurrence than PET.