The effectiveness of Halifax-produced 18FDG-PET/CT in the evaluation of patients with solitary pulmonary nodule or suspected lung cancer and the impact on patient management

Abstract

Integrated Positron Emission Tomography and Computerized Tomography (PET/CT) is a powerful imaging technique that combines functional and anatomical information for early and accurate detection of cancer. Limited availability and cost necessitate appropriate case selection based on its impact on patient management. The Halifax-based Lung Cancer Site Team (LCST) oncologists request PET/CT scans using online requisitions that prospectively capture pre-PET/CT case information including cancer indication, stage, management intent (curative or palliative) and treatment modality (surgery, chemotherapy, radiotherapy, multimodality therapy, or observation) in a database. Using the database we identified 77 scans completed on patients with suspected lung cancer (SLC, including solitary pulmonary nodule, SPN) during July 3rd, 2010 to November 30th, 2010. After considering PET/CT results and subsequent follow up medical records, with the aid of oncologists we determined similar post-PET/CT information to assess for changes in these parameters as well as confirmation of the nature of the suspected lung lesion. When used to diagnose a SLC or SPN, the PET/CT showed 88.6% sensitivity, 83.3% specificity, 86.5% accuracy, 88.6% positive predictive value, and 88.3% negative predictive value. PET/CT established a different diagnosis (usually benign) in 41.6%, and changed stage in 58.4% of cases. PET/CT changed management intent in 7.8% of cases (usually to palliative) and altered treatment modality in 59.7%. If the lung nodule size measured above 30mm, then 61% of the cases were confirmed as malignant. The Halifax PET/CT protocol is effective in correctly diagnosing malignancy in cases of suspected lung cancer at rates similar to that reported in literature. As utilized by the LCST oncologists, the impact of PET/CT on expected management (59.7%) is higher than that reported in literature (26%) suggesting a widening of the case selection criteria. In those cases where nodule size was >30mm, where an unexpectedly low number of malignant cases were confirmed, further analysis is planned to investigate pre-PET imaging criteria of malignancy.