3095 - Evaluation of Breath-Hold Techniques for Pancreatic Cancer Stereotactic Ablative Radiotherapy via Intra-Fractional Kilo-Voltage Imaging and Fiducial Tracking

S. Wang¹, X. Zhong¹, S. J. Domal¹, Y. Gonzalez², L. Chen¹, Z. Iqbal², S. N. Badiyan², Y. K. Park¹, D. J. Sher¹, A. R. Godley¹, T. A. Aguilera¹, and D. D. M. Parsons¹; ¹Department of Radiation Oncology, University of Texas Southwestern Medical Center, Dallas, TX, ²Department of Radiation Oncology, UT Southwestern Medical Center, Dallas, TX

Purpose/Objective(s): This work investigates the accuracy and consistency of three different breath-hold (BH) techniques for pancreatic cancer stereotactic ablative radiotherapy (SAbR). We hypothesize that surface-guided radiotherapy (SGRT)-based BH techniques can improve internal positional target accuracy and treatment efficiency compared to conventional spirometry-based deep inhalation breath hold (DIBH) for pancreatic SAbR. Materials/

Methods: Treatment accuracy and efficiency were analyzed for 36 pancreatic cancer patients that underwent 5-fraction SAbR, 12 with active breathing coordinator (ABC)-enabled DIBH, 13 with SGRT-based DIBH, and 11 with SGRT-based expiration BH (ExBH). In the ABC DIBH workflow, respiratory motion was monitored via spirometry by comparing the inhalation volume to a set threshold. In SGRT-based BH, the live patient surface was compared to a reference surface with 3-mm and 3-degree limits for translational and rotational variations. For each patient, 1 to 6 implanted fiducials were used as the surrogates of the true target position for beam gating during treatment delivery. For each beam-on and every 30 degrees of gantry rotation thereafter, a triggered kilo-voltage (kV) image was acquired, and the fiducials were automatically detected and compared to their expected positions. Beam-on only occurred when triggered imaging verified that the fiducials were within a 3-mm margin, thus excessive BH attempts would translate to longer treatment time. To compare the targeting accuracy and precision between different breath hold techniques, the superior-inferior (SI) deviations (the only resolvable direction from 2D kV imaging) of the fiducials from their reference locations were extracted from over 1000 triggered kV images. Treatment efficiency was quantified via patient in-vault time tracked by an in-house Bluetooth-based real-time location system.
Results: SGRT ExBH and DIBH were found to be more accurate and consistent with 81.9% and 82.5% of the fiducial deviations within 3 mm (the typical PTV margin for pancreatic SAbR) and average fiducial locations of -0.07 ± 2.36 mm and 0.12 ± 2.46 mm, compared favorably with 68.1% and -1.11 ± 3.28 mm for ABC DIBH. Patients treated with SGRT-based BHs on average spent 8 minutes less in the vault with more consistent treatment durations (average in-vault time of 49 ± 11 minutes, 48 ± 11 minutes, and 56 ± 19 minutes for SGRT ExBH, SGRT DIBH, and ABC DIBH, respectively).
Conclusion: In addition to facilitating faster and more uniform treatments, SGRT-based breath-hold techniques improved targeting accuracy and precision for pancreatic cancer SAbR as validated by intra-fractional fiducial tracking, proving its capacity as a viable replacement for spirometry breath hold techniques.

Abstract 3095 – Table 1

Breath Hold Technique	Patient In-Vault Time (min)	Fiducial Deviation = 1 mm	Fiducial Deviation = 3 mm	Fiducial Deviation = 5 mm
SGRT ExBH	49 ± 11 [27, 83]	32.8%	81.9%	96.9%
SGRT DIBH	48 ± 11 [28, 80]	38.9%	80.5%	93.6%
ABC DIBH	56 ± 19 [28, 106]	25.5%	68.1%	88.6%