Clinical Utility of Dual Phase FDG PET for Distinguishing Tumor Progression from Radionecrosis in the Modern Era

Abstract

Dual-phase PET CT, a technique by which two PET/CT scans are taken hours apart, has previously shown utility in differentiating radionecrosis (RN) from tumor progression (TP) after radiation for brain metastases. We sought to validate the utility of this technique in an independent, contemporary patient cohort. Understanding the difficulty of validation without the gold standard of pathologic confirmation in a majority of cases, we hypothesized that a dual-phase PET/CT indicating likely RN would correlate with improved overall survival (OS). We performed a retrospective cohort study of all patients who received dual-phase FDG-PET between April 2015 and January 2023 at a single center for the purpose of distinguishing RN from TP in brain metastases previously treated with radiation. Scans were classified as indicating likely RN (dpPET-RN), likely TP (dpPET-TP), or unclear (dpPET-Unc) based on final radiology report. A Kaplan Meier (K-M) analysis was performed to evaluate differences in survival among the different dual-phase PET finding categories. An alpha level of 0.05 in the log-rank test was used to determine statistical significance. We identified 36 patients who met inclusion criteria. All patients had received prior radiation - stereotactic radiosurgery (SRS), whole brain radiotherapy (WBRT) or both- to the lesion, and 21 (58%) had received surgery at some point prior to the dual phase PET CT. One patient received brachytherapy. Median time from the most recent course of radiation until dual-phase PET was 294 days, while median time from most recent surgery to dual-phase PET was 379 days. 14/36 (39%) of scans were called as dpPET-TP,15/36 (42%) were dpPET-RN, and 7/36 (19%) were dpPET-Unc. There were 10 cases where pathology was available after subsequent resection of the lesion. Five of the 10 path reports were concordant with the dual phase PET read, 5 were discordant. On K-M, median survival was 11 months (95% CI: 7.8 to 14.2 months) for dpPET-TP patients, 18 months (95% CI: 7.9 to 33.5 months) for dpPET-RN, and 12 months (95% CI: 0.0 to 51.2 months) for dpPET-Unc. On log rank testing, differences in the survival distribution for the different groups of dual-phase FDG-PET results were not statistically significant, χ2(2) = 4.085, p = .130. Dual-phase PET CT has been proposed as a useful tool for distinguishing TP from RN after prior radiation. In the small subset of cases from our study with pathologic confirmation of outcomes, dual-phase PET failed to show high concordance with pathology. Although the dpPET-RN cohort had numerically superior OS compared to the dpPET-TP and dpPET-unc groups, this result was not statistically significant. Further independent validation of this imaging technique is warranted before it can be relied upon for routine clinical management.