2127 - The Art of Adapting: Plan Selection Rationale in CT-Guided Online Adaptive Radiation Therapy (CT-ART)

Purpose/Objective(s):

In CT-ART, treatment plans are generated de-novo (adaptive) and compared with the simulation plan (scheduled) for superiority prior to each radiation fraction (fx). We evaluated treatment variables and plan selection rationale across delivered CT-ART fxs. We hypothesized CT-ART plan selection is not modified if the prescribing physician’s fraction is delivered by a covering physician.

Materials/Methods:

All patients treated with CT-ART to the prostate, pelvis, abdomen, liver, and pancreas beginning with program initiation from 2023-2024 were evaluated (n=233; 1293 fxs). Clinical variables included covering vs prescribing physician, adaptive vs scheduled treatment, fx time < or =30 min (initial CBCT to fx completion), disease site, and plan selection rationale. Rationale was defined using standard institutional practices for each respective disease site and categorized as: superior target coverage or superior OAR avoidance alone, both superior target and OAR, or comparable coverage (i.e. similar but the selected plan was clinically preferred). Clinical variables were compared between fx time groups using Chi-square and Fisher’s Exacts tests. Associations between site and plan selection were examined with a Chi-square test. A logistic regression model was created to determine association between site and selected treatment. Regression models controlled for physician and clustered by anonymized patient ID. Analysis was conducted in R v4.4.2.

Results:

The most frequent target was the prostate (n=476 fxs, 37%). Adaptive plans were selected in 88% of delivered fxs and more often took >30 min to deliver vs scheduled plans (56% vs 27%). No individual site had a significantly different likelihood of receiving an adaptive treatment vs prostate as a reference. Most fxs were treated by a covering physician (n=738 fxs, 57%). Prescribing physicians were more likely than a covering physician to select an adaptive plan (OR 1.6; p=0.03). Rationale for adaptive plan selection was well balanced and included target coverage (29%), both target/OAR (27%), and OAR avoidance alone (26%). Among fxs treated with a scheduled plan, comparable coverage was the most frequent selection rationale (64%), albeit scheduled plans were superior in only a minority of fxs. Across all plans, the most common site where comparable coverage was observed was in prostate (32%) and least in pancreas (11%). 62% of plans selected for target coverage superiority and 55% of plans selected for comparable coverage were delivered in <30 min. In contrast, 63% of plans selected for OAR avoidance and 62% of plans selected for both target/OAR superiority took =30 min to deliver.

Conclusion:

Prescribing physicians were significantly more likely to select an adaptive plan vs a covering physician. Adaptive plans were selected for reasons including superior target coverage in >50% of fxs. Adaptive plans selected for reasons including OAR avoidance took longer to deliver than those selected for target coverage superiority.