Measuring Setup Error At Multiple Neck Levels In Npc: A Case For Variable Planning Target Volume Expansion

To establish the magnitude of systematic and random setup errors (SEs) at the clivus, mid-neck, and supra-clavicular (SCF) region with image guided radiation therapy (IGRT) in NPC with recommendation for appropriate planning target volume (PTV) margins for each level. Thirty-six patients with NPC underwent radical IGRT to a dose of 70 Gy delivered over 33 fractions. Each had 9 scheduled cone beam computed tomography scans (CBCTs) done, consisting of 4 daily CBCTs in the first week followed by 5 weekly CBCTs. CBCTs are matched preferentially at the clivus in view of proximity to the nasopharynx. Each CBCT was reviewed to determine the SEs separately at the level of the clivus, C4, and C7 vertebral levels. The SEs were measured before CBCT correction (as the initial error), after CBCT correction (residual error) and recorded in the medio-lateral (ML), superior-inferior (SI) and antero-posterior (AP) axes. The 3-dimensional (3D) displacement was determined at each anatomical level as the square root of the quadratic sum of the error on each of the 3 axes. The errors were also analyzed separately to determine the systematic (Σ) and random (σ) SEs. Van Herk’s formula (2.5Σ + 0.7σ) was then used to arrive at the recommended PTV margin which would ensure 95% isodose coverage of the CTV for 90% of the patients. The mean 3D displacement at the clivus, C4, and C7 were 1.9 mm, 2.7 mm, and 3.4 mm respectively before CBCT correction. The differences in 3D displacement were statistically significant cranio-caudally between clivus and C4 (P