2976 - A Prospective Trial to Measure Contralateral Breast Dose during Breast Radiotherapy Using Cherenkov Imaging and Scintillation Dosimetry

03:00pm - 04:00pm PT

Hall F

Screen: 16

POSTER

Presenter(s)

Allison Matous, MD - Dartmouth-Hitchcock Medical Center, Lebanon, NH

A. L. Matous¹, M. Clark², R. Vasyltsiv², S. M. McVorran³, A. Fariss⁴, P. Bruza², D. J. Gladstone⁵, and L. A. Jarvis⁶; ¹Dartmouth-Hitchcock Medical Center, Lebanon, NH, ²Thayer School of Engineering, Dartmouth College, Hanover, NH, ³Dartmouth, Geisel School of Medicine, Hanover, NH, ⁴Department of Medicine, Radiation Oncology, Dartmouth-Hitchcock Medical Center / Norris Cotton Cancer Center, Lebanon, NH, ⁵Department of Medicine, Geisel School of Medicine, Dartmouth College, Hanover, NH, ⁶Dartmouth Health, Lebanon, NH

Purpose/Objective(s):

Minimizing radiation to the untreated, contralateral breast is a critical treatment goal in breast radiotherapy, as this extraneous dose can contribute to subsequent malignancies. Prior studies assessing contralateral breast dose (CBD) lack on-patient dosimetry data, and treatment techniques have evolved since this prior work was completed. The baseline incidence, causes, and clinical implications of CBD therefore remain unclear. We hypothesize that Cherenkov imaging, which enables direct visualization of radiotherapy beams on patients and can detect dose to off-target tissues, can be used to identify the incidence and root causes of CBD in routine clinical practice.

Materials/Methods:

Cherenkov images were prospectively reviewed for 62 patients receiving supine breast radiotherapy across 1,115 fractions at 2 institutions. Treatment techniques included standard tangents (38), wide tangents (11), tangents with medial electrons (1), non-coplanar beams (1), accelerated partial breast irradiation (aPBI) with volumetric modulated arc therapy (VMAT) (9), and chest wall and nodal irradiation using VMAT (2). When CBD was identified on imaging, patients were consented to an IRB-approved trial for dose measurements at the site of CBD, and a root cause analysis (RCA) was conducted. Dose measurements were made with scintillators and thermoluminescent dosimeters. Dose visualized on Cherenkov images was compared to treatment planning system (TPS) predictions of dose distribution, and a 3-point scale was used to score the extent that visualized CBD was predicted in the TPS. Measured doses were compared to TPS dose estimates at the site of interest.

Results:

CBD was observed during treatments for 40% (25) of patients. Techniques associated with CBD included standard tangents (3), wide tangents (10), tangents with medial electrons (1), aPBI VMAT (9), and chest wall and nodal irradiation using VMAT (2). RCA showed CBD was consistent with TPS predictions in 56% (14/25), unplanned and resulting from setup inaccuracies in 8% (2/25), and planned plus additional CBD due to setup in 36% (9/25) of instances. On-patient CBD measurements for a single fraction ranged from 5.7 - 62.9 cGy for aPBI treatments and 27.3 - 215.8 cGy for tangent-based treatments. TPS predictions underestimated CBD compared to on-patient measurements in 83% of cases.

Conclusion:

Cherenkov imaging combined with scintillation dosimetry is an efficient and accurate method for identifying and quantifying CBD. In this cohort, CBD was common and most often attributed to treatment planning. However, unplanned CBD caused by patient setup inaccuracies was also identified. Furthermore, on-patient dose measurements were routinely higher than TPS predictions, underscoring the importance of studying CBD through actual patient treatments. Future work will expand this study to additional institutions and incorporate measured CBD data into models of second malignancy risk.