Main Session
Sep 28
PQA 01 - Radiation and Cancer Physics, Sarcoma and Cutaneous Tumors

2246 - Investigation of Knowledge-Based Planning for Spine Stereotactic Body Radiation Therapy with Dose Concentration on the Gross Tumor

02:30pm - 04:00pm PT
Hall F
Screen: 28
POSTER

Presenter(s)

Fumiya Tsurumaki, PhD - Tokyo Metropolitan Komagome Hospital, Bunkyo-ku, Tokyo

F. Tsurumaki1, Y. Nakajima1,2, K. Ito1, S. Ide1,2, Y. Suda1, S. Kito1, Y. Fujita2, K. Murofushi1, and N. Tohyama2; 1Department of Radiation Oncology, Tokyo Metropolitan Cancer and Infectious Diseases Center, Komagome Hospital, Tokyo, Japan, 2Department of Radiological Sciences, Komazawa University, Tokyo, Japan

Purpose/Objective(s):

The minimum dose to the gross tumor volume (GTV) in spine stereotactic body radiation therapy (SBRT) has been associated with local tumor control. Recently, spine SBRT with increased GTV dose has been implemented, but this dose escalation complicates the planning process, resulting in extended planning time and increased inter-planner variability. Knowledge-based planning (KBP) improves planning efficiency and standardizes plan quality in spine SBRT. However, no reports have applied it to spine SBRT with dose concentration on the gross tumor. This study aims to evaluate the feasibility of KBP to spine SBRT with dose concentration on the gross tumor.

Materials/Methods:

A retrospective analysis of 92 spinal metastases cases was conducted. Among them, 72 cases were manually planned for this study and used as training datasets for the KBP model. The remaining 20 cases had manually created plans (MP) used for actual clinical treatment and served as testing datasets. All plans aimed to deliver 24 Gy in 2 fractions to 95% of the planning target volume (PTV). Furthermore, these plans followed institutional dose constraints and incorporated dose escalation to the GTV, ensuring the minimum dose was as high as possible, with the PTV D2% reaching 160% of the prescribed dose. The KBP model was applied to the testing datasets to generate KBP plans in a single optimization process. We evaluated whether KBP plans satisfied organ-at-risk (OAR) dose constraints. Three dosimetric metrics (PTV D95%, PTV D2%, and GTV D98%) and two conformity metrics (Paddick conformity index and D2 cm) of KBP plans were compared to MP using the Wilcoxon signed-rank test (significance level: p < 0.05).

Results:

KBP plans satisfied OAR dose constraints in all 20 testing cases. No statistically significant differences were found between MP and KBP plans in PTV D95%, PTV D2%, or the conformity index. In 90% of cases, PTV D2% was within approximately 1.5% of agreement between MP and KBP plans. The mean values of GTV D98%, and D2 cm for MP vs. KBP plans were 26.4 vs. 30.7 Gy (p < 0.001), and 17.5 vs. 15.7 Gy (p < 0.001), respectively.

Conclusion:

KBP plans, generated through a single optimization process, achieved target coverage and conformity equivalent to those of MP. KBP successfully increased the minimum dose to the GTV while maintaining PTV D2%, suggesting its feasibility to spine SBRT with dose concentration on the gross tumor.

Abstract 2246 - Table 1: Mean ± SD of evaluation metrics between manual and KBP plans

Metric

 Manual plan

 KBP plan

 p value
PTV D95% (Gy)

 23.5 ± 0.75

 23.6 ± 0.76

 0.94
PTV D2% (Gy)

 38.2 ± 0.23

 38.2 ± 0.58

 0.84
GTV D98% (Gy)

 26.4 ± 2.6

 30.7 ± 2.6

 <0.001*
Paddick conformity index

 0.87 ± 0.03

 0.87 ± 0.03

 0.45
D2 cm (Gy)

 17.5 ± 1.5

 15.7 ± 1.3

 <0.001*
PRVcorda D0.035 cc (Gy)

 16.9 ± 0.14

 15.8 ± 0.81

 <0.001*
Esophagus D0.035 cc (Gy)

 21.3 ± 4.2

 19.7 ± 4.1

 0.01*