Abstract

7220 Background: The spread of computed tomography (CT) screening for lung cancer (LC) leads to the increased detection of small sized LC. Performance and accuracy of radiological diagnosis had not been evaluated and indication of limited surgery for CT detected LC remains unknown. Methods: Between Dec/2002 and May/2004, a prospective study was performed. Patients (pts) having peripheral lung adenocarcinoma (AD) of T1N0M0 were eligible. Radiologic findings of LC were evaluated as to ground-glass opacity (GGO) with thin-section CT (TSCT). A radiological non-invasive cancer (NIC) was defined as a tumor having consolidation less than half of the maximum tumor dimension. A pathological NIC, which is future candidate for limited surgery, was defined as a tumor with no lymph node mets, lymphatic, nor vascular invasion. A validity of radiological diagnosis by above criteria was investigated. Primary endpoint is specificity as the proportion of pts with radiologically diagnosed invasive LC in pts with pathologically diagnosed invasive LC. Planned sample size was 450 with precision-base. We expected that lower limit of 95% CI of specificity is over 97%. Results: Totally 811 pts were enrolled onto the study at 31 institutions, including 357 men. Age ranged 27 to 75 (median 61). Primary endpoint was evaluated on central-reviewed 545 AD resected by lobectomy. Specificity and sensitivity for the diagnosis of pathological NIC were 96.4% (161/167 95% CI 92.3% to 98.7%) and 30.4% (115/378 95% CI 25.8% to 35.3%), respectively. As to relationship between radiologic NIC and pathological nodal involvement, specificity and sensitivity for the diagnosis were 98.3% (59/60 95% CI 91.1% to 100%) and 24.7% (120/485 95% CI 21.0% to 28.8%), respectively. Conclusions: Our pre-determined criterion for the specificity was not confirmed, and candidates for limited surgical resection can not be selected based on our criteria. However, radiological diagnosis of node-negative LC with TSCT was well coincide, and the criterion might be used for selecting node negative peripheral LC. A phase II study of limited surgical resection for peripheral LC 2.0 cm or less in size is now planned based on the results of exploratory analysis of this study. No significant financial relationships to disclose.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call