Although the use of PET/CT in radiation therapy planning has some advantages, there are still several caveats with utilization of the imaging technology, according to a presentation Monday from Katherine Mah, MD, at the 50th annual meeting American Society for Therapeutic Radiology and Oncology (ASTRO).
Mah, a physicist at the Odette Cancer Center, Sunnybrook Health Sciences Center in Toronto, said that PET/CT has random and scatter coincidences, which increase the background noise. She noted that there are number of ways to correct for the scatter coincidences—the biggest contributor of the background.
While the 2D hardware has fewer coincidences, she said that the 3D systems have increased sensitivity and thus more true coincidences in their field of view (FOV).
Mah said that there are pros and cons to the 2D and 3D systems. For 2D systems, there is better scatter rejection and better uniformity across the axial FOV. However, the 2D drawback is longer scan times, and requires higher doses. For 3D systems, she said that their increased scatter coincidences requires more software reconstruction algorithms, but requires shorter scan times and less doses.
Unlike CT, Mah noted that there is no standard algorithms, and as a result, all the vendors offer varied algorithms.
She said that attenuated photons result in “lost” events which cannot contribute to the image, adding that use of a 511 kVp transmission image provide probability of attenuation in a particular voxel. She stressed the importance of viewing the attenuation and the uncorrected images at the same time.
For PET/CT image quality, Mah said its spatial resolution is limited to a detector size 4 to 6 mm; its temporal resolution is limited to acquisition times; and its standard aperture size is 70-cm and 85-cm for big bore.
About the most common oncologic imaging—FDG-PET—Mah said it is highly adaptable, but its images are maps of glucose metabolism, acquiring “rather non-specific images.” She also noted that due to the common physiological uptake, FDG-PET images tend to display artifacts.
In comparing CT and PET/CT, Mah said that respiratory motion is more often present in PET because of the prolonged studies. She said that the difference in breathing for both CT and PET creates artifacts when CT attenuation correction is used.
She added that PET commonly underestimates activity in regions of mismatch, typically above the diaphragm, and erroneously locates lesions in mismatched regions. She also said that “if a patient extends beyond the CT FOV, PET reconstructions assumes no attenuation correction is required in the cut-off region, underestimating PET activity.”
Despite its drawback, Mah said that the main contribution of PET is its superior targeting in radiation treatment planning, which reduces observability.