PERCIST (PET Response Criteria in Solid Tumors) [1,2] defines an objective methodology to assess the tumor burden and treatment response for FDG PET images. In summary, the following steps are performed:
1.A 3cm diameter reference sphere is (manually) placed in the right side of healthy liver, midway between the dome and inferior margin, excluding central ducts and vessels. If the liver is diseased, the reference should be placed into the blood pool of the descending aorta using the CT image. A cylinder (1cm diameter, 2cm long) is specified in , but in PSEG an ellipsoid is more practical.
2.The minimal level of tumor uptake is calculated from the average and the standard deviation of the activity in the reference region. Liver: 1.5*Average+2*Stdv; Blood: 2*Average+2*Stdv.
3.An iso-contouring segmentation is performed at the calculated minimal level, resulting in a list of lesion VOIs. A restriction can be set to filter tiny VOIs representing noise and large VOIs representing organs like the brain, bladder or heart.
4.The lesions are sorted according to their peak uptake (which corresponds to the maximum average of a 1mL sphere placed in the lesion).
5.The lesion with highest peak uptake is considered the target lesion at baseline and also after treatment. Note that a non-target lesion at baseline can become the target lesion at follow-up due to poor treatment response.
6.Although only a single target lesion is used for assessing treatment response in PERCIST, it is recommended saving additional exploratory data, particularly from the 5 hottest lesions.