Back

Surface changes of clinical target volume and its sublocations in nasopharyngeal cancer with image-guided radiation therapy

Tan, W.; Yang, M.; Hu, D.

2022-05-19 oncology
10.1101/2022.05.15.22275068 medRxiv
Show abstract

PURPOSEWe aimed to quantify the sublocal geometric uncertainties of the neck prophylactic clinical target volume (CTVprophy) during image-guided radiotherapy for nasopharyngeal cancer (NPC). MATERIALS AND METHODSTwenty patients with locally advanced NPC underwent one planning computed tomography (CTplan) followed by six weekly CT (CTrepeat) scans during chemotherapy and intensity-modulated radiation therapy. The sternocleidomastoid muscle (SCM) and its anterior, middle, and posterior parts, as well as the body contours at the 1st (C1) and 2nd (C2) cervical vertebrae, hyoid bone (HB), and cricoid cartilage (CC) in transverse CT sections, were manually delineated in the CTplan and each CTrepeat. The residual error and 2D or 3D vector displacements of each sublocation were calculated, and the planning target volume (PTV) margins were estimated using the PTV margin formula. RESULTSThe left- and right-sided SCM volume decreased by 3.7 {+/-} 9.6% (1.9-5.4%) and 5.1 {+/-} 6.7% (3.9-6.3%), respectively, and the center of mass shifted medially 0.8-0.9 mm. An anisotropic PTV margin of 2-4 and 1-5 mm was needed in the left-right and anterior-posterior directions, respectively. The geometric changes in the upper neck at the C1 and C2 sections were smaller than those in the middle-lower neck at the HB and CC levels. At the same sublocation, the margin needed in the anterior-middle part was smaller than that needed in the posterior part of the neck. The rigid imaging registration-induced anatomical errors in the upper neck were < 1.9%, and those in the middle and lower neck level were 0.6-3.8%. CONCLUSIONSThe surface geometrical changes of the neck prophylactic CTV in the sublocations are substantial and an anisotropic PTV margin of 1-5 mm is needed in the context of image-guided radiotherapy for NPC.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.