Assuming that this could be helpful in tumor response assessment to chemotherapy (CT), we decided to study the scan-to-scan intra-patient variations of blood-pool and liver FDG uptake.
We retrospectively analyzed the SUVs at blood-pool (at the aortic arch level) and liver level, of 627 PET obtained in 209 pts (79 with HL and 130 with NHL), who had at least 3 repeated PET scans (for staging, after the 2nd – 4th and after the 6th round of CT). Patients preparation, acquisition and reconstruction protocols were held stable, uniform and reproducible.
In the whole series, mean SUV did not change significantly among the 3 PET studies, both in blood-pool (2.47, SD 0.36; 2.38, SD 0.47; 2.37, SD 0.43) and in the liver (3.37, SD 0.48; 3.30, SD 0.6; 3.34, SD 0.76), having Gaussian distributions.
Neither significant differences could be found in scan-to-scan intra-patient liver SUVs, both in HL and NHL pts nor in blood-pool among the 2nd and 3rd PET.
Conversely, a significant scan-to-scan intra-patient blood-pool SUV variation between staging and 2nd-4th CT round, was detected in case of complete response in NHL pts (mean ⊗ -0.11, SD 0.41, p <0.05) and in all HL responding pts (mean ⊗ -0.20, SD 0.41, p<0.05), both in case of complete (mean ⊗ -0.24, SD 0.43, p <0.05) and partial response (mean ⊗ -0.15, SD 0.38, p <0.05).
The significant differences of SUVs were always due to a decrease from staging to 2nd-4th CT round in NHL, with complete response (from 2.57 to 2.43) and in HL with, partial (from 2.4 to 2.21), and complete response (from 2.39 to 2.24,).
In our whole series the mean SUVs did not changed significantly after CT, as compared to staging, while a significant SUV decrease at the only blood-pool level was found when analyzing the intra-pts variations in case of response to CT (in all responding HL and in complete responding NHL pts).
We do not confirm any scan-to-scan intra-patient liver SUV variation, as reported by AA, who assumed an SUV increase after CT was related to a successful complete response.