Detection of Recurrence after Thoracic Stereotactic Ablative Radiotherapy Using FDG-PET-CT.

To evaluate the performance of FDG-PET-CT in distinguishing between local recurrence vs. post radiation changes in patients with early-stage non-small cell lung cancer treated with stereotactic ablative radiotherapy, we performed a retrospective analysis of patients with concerning post-treatment FDG-PET-CT scans. Under these circumstances, the positive predictive value of a concerning FDG-PET-CT is relatively low especially at 3 months. Introduction/Background: Differentiating local recurrence (LR) from post-treatment changes following stereotactic ablative radiotherapy (SABR) for thoracic tumors is challenging. We sought to evaluate the performance of FDG-PET-CT in distinguishing recurrence from post-radiation changes in patients with stage I-II non???small cell lung cancer (NSCLC) treated with SABR. Materials and Methods: We performed a retrospective review of patients with stage I-II NSCLC treated with SABR and subsequently followed with surveillance FDG-PET-CT scans from 2004 to 2014. The radiology reports were coded as 0 or 1 if minimally or substantially concerning for LR, respectively, and correlated with outcome. Prognostic factors for false-positive FDG-PET-CT were assessed using logistic regression models. Results: We identified 145 patients meeting inclusion cr iter ia for the retrospective analysis. Amongst the 39 (26.9%) patients with FDG-PET-CT scans concerning for LR 3 to 24 months after treatment, 14 were confirmed to have LR. Thus, the positive predictive value (PPV) of FDG-PET-CT in identifying LR was 36% (14/39). Factors associated with a false-positive scan included concerning FDG-PET-CT at the earliest post-treatment time point (3 months) (odds ratio 0.67, P = .04) and older age (odds ratio 2.3, P = .02). Conclusion: Our analysis indicates that the PPV of a concerning FDG-PET-CT after SABR for early-stage NSCLC is relatively low, especially at early post-treatment timepoints, but accuracy is improving over time with institutional experience.