Positron Emission Tomography in Staging Early Lung Cancer A Randomized Trial

Background: Among patients with early-stage non-small cell lung cancer (NSCLC), preoperative imaging tests are important in defining surgical candidates. Objective: To assess whether whole-body positron emission tomography and computed tomography (PET-CT) plus cranial imaging correctly upstages cancer in more patients with NSCLC than does conventional staging plus cranial imaging. Design: Randomized clinical trial with recruitment from June 2004 to August 2007. The centralized, computer-generated, variable block size randomization scheme was stratified by treatment center and cancer stage. Participants, health care providers, and outcome assessors were not blinded to imaging modality assignment. Setting: 8 hospitals and 5 PET-CT centers in academic institutions. Patients: Eligible patients were older than 18 years; had histologic or cytologic proof of stage I, II, or IIIA NSCLC on the basis of chest radiography and thoracic CT; and had a tumor considered to be resectable. Intervention: PET-CT or conventional staging (abdominal CT and bone scan). All patients also had cranial imaging using CT or magnetic resonance imaging. Measurements: The primary outcome was correct upstaging, thereby avoiding stage-inappropriate surgery. Secondary outcomes were incorrect upstaging and incorrect understaging. Results: 170 patients were assigned to PET-CT and 167 to conventional staging. Eight patients (3 who had PET-CT and 5 who had conventional staging) did not have planned surgery. Disease was correctly upstaged in 23 of 167 PET-CT recipients and 11 of 162 conventional staging recipients (13.8% vs. 6.8%; difference, 7.0 percentage points [95% CI, 0.3 to 13.7 percentage points]), thereby sparing these patients from surgery. Disease was incorrectly upstaged in 8 PET-CT recipients and 1 conventional staging recipient (4.8% vs. 0.6%; difference, 4.2 percentage points [CI, 0.5 to 8.6 percentage points]), and it was incorrectly understaged in 25 and 48 patients, respectively (14.9% vs. 29.6%; difference, 14.7 percentage points [CI, 5.7 to 23.4 percentage points]). At 3 years, 52 patients who had PET-CT and 57 patients who had conventional staging had died. Limitation: The relatively small sample and the fact that some patients did not have planned surgery limited the ability to determine precise differences in clinical outcomes that were attributable to testing strategies. Conclusion: Preoperative staging with PET-CT and cranial imaging identifies more patients with mediastinal and extrathoracic disease than conventional staging, thereby sparing more patients from stage-inappropriate surgery, but the strategy also incorrectly upstaged disease in more patients.