2991 - Timing of CT Simulation after Surgery in Online Adaptive Partial Breast Radiotherapy

12:30pm - 01:45pm ET

Poster Hall - Exhibit Hall A

Screen: 2

POSTER

Presenter(s)

Madeline Evans, BS - UAB Heersink School of Medicine, Birmingham, AL

M. Evans¹, J. A. Pogue², J. A. Soldner¹, J. Hoyle³, C. G. Reid⁴, H. Boggs³, C. Cardenas², N. Viscariello², C. Stanley², D. N. Stanley², and M. Soike³; ¹The University of Alabama at Birmingham Heersink School of Medicine, Birmingham, AL, ²University of Alabama at Birmingham, Birmingham, AL, ³University of Alabama at Birmingham Department of Radiation Oncology, Birmingham, AL, ⁴Brookwood Baptist Medical Center, Birmingham, AL

Purpose/Objective(s):

Online adaptive radiotherapy (oART) allows for accelerated partial breast (APBI) plans to be generated for each fraction by recontouring the lumpectomy cavity and replanning in real time. To maintain expected plan quality during adaption, ideally the lumpectomy cavity will be similar in size between CT simulation and treatment delivery. In the setting of oART, the ideal windows of times 1) from surgery to sim and 2) from sim to first fraction delivery are unknown. The purpose of this study was to identify the effect of timing intervals between surgery, sim, and first fraction for oART with endpoints of PTV volume change, time spent during each oART session, and dosimetric plan quality.

Materials/Methods:

We evaluated oART using Ethos (Varian Medical Systems) in 31 patients with 33 lumpectomy cavities treated between 2021 and 2023. APBI consisted of five fraction oART sessions using the Ethos workflow. The lumpectomy cavity was contoured with a 10 mm expansion for clinical target volume + 3 mm to create PTV. The cavity and organs at risk were contoured at time of sim and at each oART fraction. The treated PTV volume (cc) was determined at each session and averaged over the treatment course. This averaged PTV volume in cc was subtracted from the original PTV at time of sim to create ?PTV (cc). We correlated days from surgery to sim and sim to first fraction with ?PTV, time per session, and plan quality metrics with Spearman’s rank correlation coefficient.

Results:

The median PTV size at time of sim was 104 cc (Interquartile range [IQR] 52cc) and median PTV size during oART was 83 cc (IQR 50cc). Median number days from surgery to sim was 44 (IQR 25) and sim to first fraction was 23 days (range 9-40). As a continuous variable, more days from surgery to sim was moderately associated with a smaller ?PTV (?= -0.456, p=0.0077). Time from sim to first fraction was not associated with a change ?PTV (?=0.196, p=0.275).

The median treatment session length was 42 minutes (IQR 6). Session length was not correlated with days from surgery to sim or sim to first fraction intervals. PTV coverage, volume of breast receiving 30 Gy, lung V9, skin max dose, and heart V1.5 Gy dosimetry did not significantly change compared to the original plan based on days from surgery to sim or sim to first fraction (P>0.05 for all variables).

Conclusion:

Longer intervals from surgery to simulation were associated with smaller differences between simulation and on-treatment PTV volumes. However, across the timing ranges observed, oART consistently maintained PTV coverage and organ-at-risk sparing without prolonging session duration. These findings suggest that Ethos-based online adaptive APBI is robust to common variations in postoperative timing, supporting flexible scheduling once patients are beyond the early postoperative period. Larger studies are warranted to define whether narrower timing recommendations provide incremental benefit.