PET/CT can direct diagnostic and therapeutic decisions in pancreatic lesions

Abstract

14048 Background: This study aimed to determine the role of PET/CT in differentiating pancreatic lesions and changing therapeutic approach. Methods: A retrospective analysis of prospectively collected data of 33 PET/CT scans (Philips, Cleveland, USA) was performed. Scans were made between October 2004 and July 2005 for differentiating primary or recurrent pancreatic disease found on CT, MRI or/and EUS. All PET/CT scans were assessed by a radiologist in co-operation with a nuclearist. Pancreatic lesions were interpreted based on visual assessment and by use of a semi-quantitative evaluation (SUV). These results were correlated with pathological diagnoses or clinical follow-up, dividing patients into 2 groups: ‘positive’ (malignant) versus ‘negative’ (benign disease/no lesion). In each patient therapeutic change after PET/CT was assessed. Results: The patient group consisted of 33 consecutive patients, of whom 14 were female. Median age was 61 years (range 33–76 years). SUV values in malignant lesions were significantly higher than those found in benign lesions [5.77 (95%CI 4.08–7.46) versus 1.76 (95%CI 0.79–2.7), P<0.01]. Eleven patients with benign disease or no lesion, had a negative PET/CT. Twenty-two patients had malignant disease of the pancreas. Twenty of 22 patients had a positive PET/CT. Sensitivity and specificity of PET/CT for diagnosing benign or malignant disease visually was 90.9% and 100% respectively. In the benign/no lesion group, a change in therapeutic approach was noted for 9/11 patients (81,8%): no surgery (n=6), no chemotherapy (recurrence excluded, n=3). No change was noted in 2/11 patients (18,2%): no differentiation pancreatitis/tumor. In the malignant group, a change in therapeutic approach was noted for 14/22 patients (63,6%): recurrence confirmed (n=8), up-staged (n=5), down-staged (n=1). No change was seen in 8/22 patients (36,4%): incorrectly sent to surgery (n=4), no differentiation pancreatitis/tumor (n=2), recurrence missed (n=1), false positive diagnosis of liver metastasis (n=1). Conclusions: PET/CT has 90.9% sensitivity and 100% specificity for diagnosing benign or malignant disease visually. PET/CT is a useful tool for directing therapeutic strategy. Overall, therapeutic approach was changed in 23/33 (69,7%) of patients. No significant financial relationships to disclose.