Multimodal imaging of esophagus and cardia cancer before and after treatment.

PURPOSE

Prognosis and treatment of esophagus and cardia cancer (ECC) depend on the precision with which the disease is staged according to the American Joint Committee of Cancer (AJCC) criteria. Imaging modalities normally used in clinical staging are esophagography, esophagoscopy, endoscopic ultrasound (EUS), computed tomography (CT) and positron emission tomography- CT fusion (CT-PET). The combination of these methods is crucial in determining not only the right diagnosis but also the stage and follow-up after multimodal treatment. The purpose of our investigation was to define the role of each imaging modality in determining the most appropriate treatment options in patients with ECC.

MATERIALS AND METHODS

Fifty-six patients with ECC diagnosed by X-ray of the upper digestive tract, endoscopy and biopsy were staged using EUS, chest and abdomen CT scan, and CT-PET. Thirty-four patients in stage II and 18 patients in stage III underwent surgery after neoadjuvant chemotherapy; four patients in stage IV were treated with the positioning of an endoprosthesis after chemoradiotherapy. In the 52 patients who had surgery, follow-up included digestive tract X-ray, endoscopy and CT of the chest and abdomen every 6-8 months for the first 3 years. CT-PET was only performed in patients with a clinical suspicion of recurrence and/or CT findings suspicious of persistent disease (12 cases).

RESULTS

In all 56 patients, endoscopy, EUS, CT and CT-PET in combination were crucial in determining the site of disease, locoregional extent and depth of esophageal wall penetration (T), and any involvement of the mediastinal lymph nodes (N1), extrathoracic lymph nodes (M1) or hepatic metastases. In the locoregional staging of ECC before chemotherapy, we were able to differentiate T2-T3 from T4 in 40 patients; T4 disease was found in 12 potentially resectable cases. We were able to distinguish N0 from N1 in 12 patients. In four cases, the presence of small lymph node and/or liver metastases prompted positioning of an endoprosthesis. The specificity of CT in detecting small lymph nodes in the mediastinum was less than 50% while for CT-PET, it was more than 80%; EUS revealed sensitivity higher than 90% but a low specificity in seven cases. Only CT-PET revealed metastatic subdiaphragmatic lymph nodes (diameter <15 mm) in three cases. Presurgical restaging of the 18 patients (stage III) who had chemotherapy was based on endoscopy, EUS, CT of the chest and abdomen and CT-PET (only in suspected cases) and was compatible with surgery. Anastomotic recurrence was diagnosed in 16 patients by endoscopy with associated biopsy; any intramediastinal spread from anastomotic recurrences was evaluated by chest CT, and CT-PET in suspected cases.

CONCLUSIONS

X-ray of the upper digestive tract and chest and abdomen CT scan are useful in preliminary evaluation of ECC. Endoscopy is particularly indicated for evaluating tumour morphology, taking biopsies for a histological diagnosis and the early diagnosis of anastomotic recurrences. EUS is indicated mainly for evaluating T stage before and after chemotherapy or chemoradiotherapy. CT-PET is extremely useful in identifying small mediastinal metastatic lymph nodes (N1) or extrathoracic lymph nodes (M1) and hepatic metastases (</=1 cm), which may escape multislice CT. PET alone is useful for identifying residual or recurrent tumour in the esophageal wall when an endoprosthesis is in place.