Standardized added metabolic activity (SAM): a partial volume independent marker of total lesion glycolysis in liver metastases

Abstract

Purpose: The standardized added metabolic activity (SAM) is a new marker of total lesion glycolysis that avoids partial volume effect (PVE) and thresholding. SAM is calculated by drawing a volume of interest (VOI(1)) around the tumour and a larger VOI (VOI(2)) around VOI(1). Subtracting the background activity in VOI(2)-VOI(1) from VOI(1) yields SAM. If VOI(1) is set at a reasonable distance from the tumour, PVE are avoided. Phantom and initial clinical validation data are presented.

Methods: Spheres of a Jaszczak phantom were filled with a 5.4, 3.64 and 2.0 times higher concentration relative to background activity and positron emission tomography (PET) data were acquired during 10 min. SAM of all spheres was expressed as a percentage of the expected value (the actual activity ratio minus 1). In 15 patients a 10-min list-mode acquisition PET study centred on their primary squamous cell carcinoma (PSCC) was performed and images of 1-10 min reconstructed. SAM1-9min values of PSCC were expressed as a percentage of SAM10min. Nineteen patients suffering from liver metastases treated with chemotherapy underwent PET/CT prior to (scan 1) and after 3-6 cycles of chemotherapy (scan 2). SAM and maximum standardized uptake values (SUV(max)) of the liver lesions on scan 1 (SAM1 and SUV(max)1) and the percentage reduction between both ΔSAM and ΔSUV(max) were related to Response Evaluation Criteria in Solid Tumors (RECIST) response.

Results: For the phantom acquisitions, the mean normalized SAM/sphere volume calculated was 94.9 % (SD 5.9 %) of the expected value. In the PSCC patients, the mean difference between SAM1min and SAM10min was only 4 % (SD 5 %). SUV(max)1min and SUV(max)10min proved to be not significantly different, but the variability was slightly larger than that of SAM (SD 6.4 %). SAM1 and ΔSAM values for responders versus non-responders were, respectively, 57 (SD 119) versus 297 (SD 625) for SAM1 (p = 0.2) and 99 % (SD 3 %) versus 32 % (SD 44 %) for ΔSAM (p = 0.001). SUV(max)1 and ΔSUV(max) values in responders versus non-responders were, respectively, 3.9 (SD 2.4) versus 6.3 (SD 3.1) for SUV(max)1 (p = 0.08) and 94 % (SD 17) versus 7 % (SD 40 %) for ΔSUV(max) (p = 0.0001). The AUC of ΔSAM and ΔSUV(max) were not significantly different on receiver-operating characteristic (ROC) analysis (AUC 1.0 and 0.99, respectively, p = 0.6).

Conclusion: SAM is a promising parameter for tumour response assessment of liver metastases by means of (18)F-fluorodeoxyglucose PET.