P14.93 The utility of the brain 18-FDG-PET and perfusion magnetic resonance imaging in the radionecrosis differential diagnosis

Abstract BACKGROUND Radiation-induced necrosis (RDN) is a side effect observed in patients who underwent stereotactic radiosurgery (SRS) alone or combined with whole brain radiotherapy (WBRT) as treatment for their brain tumors. Nowadays, RDN diagnosis and differentiation from tumor progression using the standard imaging techniques represents a challenge, and histological diagnosis still is the gold standard.Our aim is to assess the positive and negative predictive values (PPV, NPV) of FDG-PET and perfusion magnetic resonance imaging (MRI) in RDN diagnosis. MATERIAL AND METHODS From our Pathology department database, all patients with RDN or mixed (tumor plus RDN) brain lesions diagnosis during last 5 years were reviewed. Demographic, clinical and oncologic treatment characteristics were recorded. MRI and FDG-PET images at the suspicion (clinical or radiological) of progression or RDN were registered and compared with the definite diagnosis provided by the tissue sample analysis. Sensitivity, specificity, as such as PPV and NPV for perfusion MRI sequences and FDG-PET image analysis were calculated. RESULTS 162 patients underwent SRS+/- WBRT in a 5 year period. During follow-up, 11 patients had surgery-confirmed RDN. There are 11 patients (3 women) with 12 lesions, 3 from a breast cancer, 6 from a lung tumor, 1 from a melanoma, 1 atypical meningioma and 1 glioblastoma. 9 of them were treated with both SRS and WBRT, and the 3 others with WBRT alone. The mean age was 65.36 (range: 44–77) years. The median time between the completion of radiation therapy and the suspicion of RDN was 19.7 (range: 3–34) months. With the evolution, it was observed an evident increase in the size of surronding oedema (2–6 times) by FLAIR RMI. We estimate a PPV 0.40 and NPV 0.80 for perfusion MRI, and PPV 0.25 and NPV 0.75 for FDG-PET, respectively. CONCLUSION The diagnostic performance of both techniques in our series is low and similar to published data; therefore its results must be carefully interpreted in each case. It is peremptory to implement new diagnostic tools in the standard of care with better diagnostic outcomes.