Targeting Pancreatic Cancer: The Dosimetric Impact Of Image Guided Radiation Therapy By Intratumoral Fiducial Markers

High local recurrence rates and significant toxicity have prompted several recent advances in pancreatic cancer radiation therapy including Image Guided Radiation Therapy (IGRT), on-board imaging by cone beam CT (CBCT), and motion control by 4D-CT. Although each technique shows clear dosimetric benefit at planning, the quality of daily treatment delivery is highly dependent on precise target localization. Pancreatic fiducials have proven to be superior in isocenter localization when compared to other markers including stents and various anatomic landmarks. Yet to our knowledge, the more clinically relevant dosimetric impact of IGRT delivered by fiducial markers has not yet been published. To this end, the current study applies positional and anatomic data to VMAT pancreatic cancer treatment plans to quantify the dosimetric effects of image guided radiation therapy (IGRT) using intra-tumoral fiducial markers versus bony anatomy. Using daily orthogonal kV images, positional and dosimetric data were analyzed for 12 patients with locally advanced inoperable disease, who were treated using fiducial-based IGRT and volumetric-modulated arc therapy (VMAT) to the intact pancreas. The shifts from fiducial to bone (Fid-Bone) that were required to co-register the daily pre-treatment images to the planning CT were recorded as Left-Right (LR), Anterior-Posterior (AP) and Superior-Inferior (SI). The iso-center was then shifted accordingly for a total of 60 kV based bone-matched plans. Each was recalculated to determine the dosimetric impact of the Fid-Bone positional shifts on the planning target volume coverage (PTV50.4), liver, kidney, and stomach/ duodenum. The absolute Fid-Bone positional shifts were greatest in the SI direction, (LR=2.8 ± 2.8, AP=2.7 ± 3.0, and SI 6.3 ± 7.9 mm, mean ± SD, p=0.03). The PTV50.4 was significantly reduced by 13.5%, (Fid-Bone=95.3 ± 2.0 vs. 82.3 ± 11.8%, p=0.02). This varied widely among patients (Fid-Bone PTV50.4 Range=2-60%), where 33% of patients had a reduction in PTV50.4 of more than 10%. The impact on OARs was greatest to the liver, (Fid-Bone=14.6 vs. 16.1 Gy, 10%), and stomach, (Fid-Bone=23.9 vs. 25.5 Gy, 7%), approaching statistical significance (p=0.10 both). Positional shifts between kV and CBCT based plans were nearly identical (Fid-Bone LR, AP, and SI, each <1%). Compared to image co-registration using fiducial markers, IGRT delivered by bone-matched orthogonal kV images has a highly variable and significant negative impact on PTV coverage, with limited effect on OARs. Together, these data support the use of implanted intra-tumoral pancreatic markers to improve the accuracy of PTV localization for IGRT in pancreatic cancer.