Slobodan Devic PhD, Presenter: Consultant, International Specialty Products, Inc

Huriyyah Mohammed, Abstract Co-Author: Nothing to Disclose

Saad Aldelaijan, Abstract Co-Author: Nothing to Disclose

Nada Tomic MS, Abstract Co-Author: Nothing to Disclose

Jan Seuntjens PhD, Abstract Co-Author: Nothing to Disclose

Francois Deblois PhD, Abstract Co-Author: Nothing to Disclose

Sergio Faria, Abstract Co-Author: Nothing to Disclose

Most of contemporary efforts in implementing functional information into radiotherapy treatment planning process are based on an attempt to replace CT/MRI based GTV with PET based GTV. However, the fact is that such approach did not lead to widespread clinical implementations of PET based GTV in patient treatments. Exploring a new venue of complementing the CT/MRI based anatomical information with FDG-based PET/CT functional data we developed a novel Uptake Volume Histogram approach for biological target volumes (BTV).

A cohort of 31 NSCLC patients was used. Background uptake in PET scan was defined as a weighted mean over the FDG uptake values within the contra-lateral healthy lung tissue through slices containing the tumor, and then scaled by factor of 3 to account for difference in tissue density between healthy lung and solid tumor. To decrease inter-patient variability we divided each PET slice raw data by the background uptake to obtain local signal-to-background ratio (S/B) images. By sampling a region of interest containing the tumor on S/B images, an uptake volume histogram was constructed for each patient.

Distinct biological volumes were observed in the uptake volume histograms and fitted using convolution of four Gaussians. We areonly making a hypothesis that they may represent three different regions of active tumor, proliferation and glycolysis. The thresholding values for these volumes, as found by the cross-section of the corresponding fitted Gaussian curves, vary significantly over the patient population. On the other hand, we found that corresponding partial volume ratios are on average 0.5, 0.3, and 0.2 for the active tumor, proliferating and glycolytic BTV, respectively.

We hypothesized that FDG-based PET data could be used to isolate at least three distinct biological target volumes in NSCLC patients by constructing uptake volume histograms over the functional FDG-based PET/CT data. The large variability observed in the threshold values throughout the patient cohort can be explained as a consequence of the large variability in the physiological status of the tumor volume at the time of the PET/CT scan.

Proposed method suggests that multiple BTVs should be determined on a patient-to-patient basis.The concept is in synergy with a contemporary custom-made patient-specific oncologic treatment planning.

Devic, S, Mohammed, H, Aldelaijan, S, Tomic, N, Seuntjens, J, Deblois, F, Faria, S, FDG-based Uptake Volume Histograms: A Novel Venue towards Biological Target Volumes for Radiotherapy Treatment Planning.  Radiological Society of North America 2011 Scientific Assembly and Annual Meeting, November 26 - December 2, 2011 ,Chicago IL. http://archive.rsna.org/2011/11034411.html