D Branco*, J Mayadev, K Moore, X Ray, UC San Diego, La Jolla, CA

• 
  D Branco

• 
  
• 
  
• 
  
• 
  
• 
  
• 
  
• 
  

TU-IePD-TRACK 3-7 (Tuesday, 7/27/2021) 3:00 PM - 3:30 PM [Eastern Time (GMT-4)]

Purpose: To evaluate a physicist-driven, CBCT-based daily adaptive protocol in locally advanced cervical cancer for multiple endpoints: on-couch time, physicist contouring accuracy of daily CTVs, and target coverage with reduced PTV margins versus non-adapted standard-of-care.

Methods: CT scans of 13 retrospective cervical cancer patients with original CTV_SIM were re-planned on the Varian Ethos platform using 12-field IMRT and 3mm symmetric CTV-to-PTV margins. Each plan was adapted on five CBCT scans to simulate Ethos’ auto-segmentation and auto-planning workflow. Required contouring edits were made by one of three physicists, mimicking daily adaptation without real-time physician review. Time spent editing OARs and CTVs was recorded, as was total treatment time (CBCT acquisition to treatment start, inclusive of auto-segmentation, contour review/edits, replanning, and evaluation). All physicist-adapted targets, CTV_PHYS, were reviewed by an experienced gynecologic radiation oncologist for accuracy and, when necessary, corrected to establish ground truth, CTV_GT. The adaptive coverage frequency was compared to non-adaptive standard-of-care treatment by assessing whether daily CTV_GT was contained inside PTV_PHYS=CTV_PHYS+3mm and/or PTV_SOC=CTV_SIM+(5mm-nodal, 10mm-vagina/parametrium, 15mm-uterus/cervix).

Results: The average time spent editing OAR/CTV contours was 3.8(range:1.0-15) and 7.1(range:2.5-12.5) minutes, respectively. The average total time was 16.7(range:7.0-29.0) minutes. 54/65 sessions were completed <20 min; only 1/65 required >23 min. Physician reviews of CTV_PHYS resulted in: 47/65 requiring no further editing (CTV_GT=CTV_PHYS), 14/65 requiring target size reduction (CTV_GT CTV_PHYS). CTV_GT extended outside the non-adapted PTV_SOC in 53/65(81.5%) fractions, while the daily CTV_GT extended outside of PTV_PHYS in only 4/65(6.2%) fractions.

Conclusion: Our results demonstrate that CBCT-driven, daily adaptation of cervical cancer without real-time physician review is feasible in a typical time slot (<20 minutes). With daily adaptation, even substantially reduced PTV margins resulted in markedly increased CTV coverage frequency. These margin reductions would also improve normal tissue sparing, providing further motivation for daily adaptation.

Funding Support, Disclosures, and Conflict of Interest: KM acknowledges funding support from AHRQ R01HS025440; consulting/honoraria from Varian Medical Systems. XR acknowledges a lab services agreement with Varian Medical Systems. JM acknowledges funding support from GOG Foundation, AstraZeneca, NRG Oncology; consulting/honoraria from Varian Medical Systems.

ePosters

Keywords

Cone-beam CT, Image-guided Therapy, Radiation Therapy

Taxonomy

TH- External Beam- Photons: adaptive therapy

Contact Email