BACKGROUND:Patients with esophageal cancer often receive neoadjuvant chemoradiotherapy and their clinical re-staging often influences their care. We evaluated the accuracy of these tests.
METHODS: A prospective trial on a series of consecutive patients was performed. Entry criteria required an initial PET-CT using flouro-2-deoxy-D-glucose (FDG) an EUS-FNA, neoadjuvant chemoradiotherapy, a repeat PET-CT and EUS-FNA and complete surgical resection with lymphadenectomy within one month of re-staging. The objective was to measure the accuracy of repeat PET-CT and EUS-FNA.
RESULTS:There were 32 patients. For EUS-FNA the accuracy for the T status was 50% (16/32) and for the N status was 72% (23/32). When EUS was wrong for the T status, it never understaged a patient and was never off by more than 1 level except once. For PET-CT the accuracy for the N status was 94% (30/32) (p<0.04) and for the M status was 97% (31/32). There were 11 patients with a complete pathologic response and EUS predicted only 6/11 (55%) whereas PET-CT predicted 10/11 (91%).
CONCLUSIONS: PET-CT is more accurate then EUS-FNA for the restaging of patients with esophageal cancer who have undergone neoadjuvant therapy for the N status. PET may also be better for predicting complete responders. However, it cannot assess the T status. Although EUS-FNA is wrong half the time for the T status, it rarely understages the patient's tumor. Thus patients offered surgery can be resected. These findings may help guide treatment.

Back to Final Program