International Journal of Radiation Oncology, Biology, Physics vol:78 issue:3 pages:S531-S532
Annual Meeting of the American-Society-For-Radiation-Oncology edition:52 location:San Diego, CA date:31 October - 04 November 2010
Purpose/Objective(s): To evaluate the dosimetric quality of the treatment plan early during radiotherapy (RT) by assessing the impact of volumetric changes and
baseline shifts of the primary tumor and the involved mediastinal lymph nodes. FDG-PET/CT imaging was chosen because of the optimal image quality and
allowing 1) accurate target delineation of tumor and lymph nodes and 2) dose recalculation.
Materials/Methods: For 35 patients repeated 4D FDG-PET/CT imaging with a contrast-enhanced CT during the second week of RT was made using the same
procedure as the planning scan. Target volumes and normal tissues were delineated on both scans. Dose prescription followed a dose-escalation protocol based on
normal tissue constraints. Patients received sequential RT (N=19), chemo RT (N=14) or RT only (N=2). The gross tumor volume (GTV) and clinical target volume
(CTV) was compared between planned and repeated imaging. The treatment plan was copied to the repeated scan and dose-volume histograms were calculated.
Investigated parameters were the minimum dose coverage (D99) and the volume receiving 90% of the prescribed dose (V90) of the primary target volume (pr) and
nodal volume (ln). Also investigated were the mean lung dose (MLD) and the maximum (D0.1) dose to the spinal cord.
Results: For one patient no primary tumor was visible on the repeated scan, and one patient had two primary tumors which were analyzed separately. The average
volume reduction of the GTVpr was 5.7±19.0% (p=0.04) and was not correlated with changes in lymph node volume (N=26), 2.0±22.0% (p=0.78). The average
displacement of the primary tumor between planning and repeated scan was 5.2±3.7 mm (range 0.1 to 15.8 mm). The effect of volume changes and baseline shifts
on the coverage of CTVpr was small: for 1 patient the V90 was smaller than 99% due to a growth of GTVpr. Differences in D99 were small -2.2±11.8% (range -
69% to +5%, p=0.27). No correlation was observed between baseline shift of the GTVpr and D99 or V90. D99 of the CTVln showed a reduction larger than 5% for
3 patients (12%) of which 2 patients had large changes (>10%) in GTVpr. MLD was approx. equal for planned and repeated scan: 16.1±3.9 Gy vs. 16.3±4.1 Gy,
resp. For the spinal cord (D0.1) these numbers were: 45.1±9.6 Gy vs. 45.7±9.9 Gy.
Conclusions: Repeated FDG-PET/CT imaging is a suitable method for quantification of volumetric and dosimetric changes in primary tumor, mediastinal lymph
nodes and normal tissues. For the entire patient population no correlation with dosimetric parameters were observed but volume and baseline changes of the
primary tumor may already occur early during treatment. For individual patients, (large) differences occur due to volume changes and baseline shifts of primary
tumor and lymph nodes and may justify more frequent imaging.