Development of a Pathway for Earlier PET-CT Scanning in Lung Cancer Patients

Objectives: To develop a new imaging algorithm for lung cancer patients by determining radiographic criteria that accurately predict potentially curable lung cancer, enabling direct referral to PET-CT scanning and streamlining the diagnostic pathway. Methods: A non-randomised retrospective analysis was conducted on chest radiographs and radiologic reports of 397 patients referred for Chest CT with suspected lung cancer between November 2012 and October 2013 at Hull and East Yorkshire Hospitals NHS Trust. Chest radiographs were categorised, and imaging pathways were suggested using predefined criteria based on tumour characteristics. The predictive accuracy was evaluated against actual imaging pathways in 137 confirmed lung cancer cases. Data analysis was performed using SPSS (IBM Version 24). Results: The predefined radiographic criteria achieved 73% accuracy in predicting patients who underwent PET-CT during diagnostic workup, with sensitivity 57%, specificity 89%, positive predictive value 77%, and negative predictive value 78%. Peripheral tumor location (p=0.019), T1-T2 stage (p=0.005), and N0 nodal status (χ²=23.44,p=0.001) were significantly associated with PET-CT requirement. Contralateral lung nodules showed negative association with PET-CT (χ²=4.34,p=0.04). Conclusions: Chest radiographic features can predict with good specificity which lung cancer patients require PET-CT scanning. Implementation of this pathway could reduce diagnostic delay by bypassing intermediate CT staging in selected patients with potentially treatable disease.