ESTRO 2020 Abstract Book

S1098 ESTRO 2020

OC-1048 Use of Ultrasound-Compatible Models for Simulation-Based Gynecological Education L. Singer 1 , S. Damast 2 , J. Chino 3 , N. Taunk 4 , L. Lin 5 , L. Lee 1 , P. Mohindra 6 , K. Bradley 7 , C. Fisher 8 , E. Fields 9 , M. Joyner 5 1 Brigham and Women's Hospital and Dana-Farber Cancer Institute, Radiation Oncology, Boston, USA ; 2 Yale, Therapeutic Radiology, New Haven, USA ; 3 Duke Cancer Center, Radiation Oncology, Durham, USA ; 4 University of Pennsylvania, Radiation Oncology, Philadelphia, USA ; 5 The University of Texas MD Anderson Cancer Center, Radiation Oncology, Houston, USA ; 6 University of Maryland, Radiation Oncology, Baltimore, USA ; 7 University of Wisconsin, Department of Human Oncology, Madison, USA ; 8 University of Colorado, Radiation Oncology, Denver, USA ; 9 Virginia Commonwealth University, Radiation Oncology, Richmond, USA Purpose or Objective In the U.S., multiple studies have shown a decline in clinically appropriate brachytherapy (BT) use for patients with cervical cancer. National surveys have also identified a lack of adequate BT training amongst radiation oncology (RO) trainees and simulation-based education (SBE) has been utilized to address this gap in training. Clinically, ultrasound (US) provides real-time image-guidance for cervical cancer BT; however, commercially available US- compatible cervical cancer BT mannequins are currently unavailable in the U.S.. In an effort to improve the quality of BT education, we developed an SBE workshop integrating higher fidelity gynecological models compatible with real-time US-guidance. The purpose of this study is to assess the feasibility and efficacy of integrating higher fidelity procedural models into a national SBE gynecological BT workshop. Material and Methods An SBE session was held at a national RO meeting to teach gynecological BT. The session was open to attendees at all skill levels. Among 9 stations (Figure 1) led by gynecological BT experts, two stations incorporated US- guidance to improve intrauterine tandem placement. One station incorporated a high-fidelity virtual reality gynecologic skills simulator consisting of a pelvic model, with anteverted and retroverted uteri inserts, tandem-like training devices, and a laptop computer. The other station incorporated a custom-made prototype gynecologic brachytherapy pelvic model which was ultrasound- compatible. Resident attendees were encouraged to attend the US stations but all stations were available to all attendees. Optional, anonymous pre- and post-test surveys were used to assess efficacy of the workshop in meeting educational objectives.

fraction. Median age was 48 years and FIGO stage was I(39%), II(48%) and III(13%). Patients received EBRT and 2 insertions and 4 fractions of HDR BT using tandem&ovoids with/without needles. DSMs were generated at 2mm depth from delineated vagina surface, termed in this study as vagina wall surface (VWS). VWS contours were unfolded by cutting them anteriorly through the urethra, inferiorly through the posterior–inferior border of the symphysis (PIBS) and superiorly such that the surface of the cervix was excluded. DSMs of EBRT and each BT fraction, converted to Gy EQD2 (α/β=3), were added based on a system of homologous points, to generate cumulative DSMs. D 20cm2 , D 90% (Gy EQD2 3 ), S 60Gy (%), location of high/low dose regions along VWS and vagina length (VL) measured from vagina top to PIBS were investigated for correlation with St/Muc according to morbidity groups. Average/difference DSMs of upper 3cm VWS and one-way ANOVA statistical testing were used to compare between groups. Results A total of 96 DSMs (31 EBRT and 62 BT) were generated. Figure 1 shows VWS D 20cm2 and VL according to morbidity group. Patients with St G≥2 and Muc G≥2 compared to patients with St/Muc G0/1 had, respectively: mean VWS D 20cm2 higher by 16Gy (p<0.001) and 10Gy (p=0.009), mean VL shorter by 9mm (p=0.05) and 5mm (p=0.297). St G≥2 vs St/Muc G0/1 patients had larger VWS areas irradiated to intermediate doses, with S 60Gy larger by 27% (p=0.001). Moreover, average/difference DSMs in Figure 2a,c,e show higher doses throughout the entire upper 3cm circumference, with D 90% higher by 8Gy (p<0.001). In Muc G≥2 vs St/Muc G0/1 group, VWS S 60Gy was larger by 10% (p=0.196) while high dose differences were significant mainly in the latero-posterior part of upper VWS (Figure 2f).

Conclusion Vaginal dose reporting is currently based on dose points. A novel method was developed to generate vaginal DSMs that incorporate 3D spatial information and new dose descriptors for a more comprehensive view on vaginal doses and correlation with morbidity. St G≥2 patients had significantly higher vaginal doses (high and intermediate levels) and shorter VL than St/Muc G0/1 patients. Similarly, Muc G≥2 patients showed higher vaginal doses (D 20cm2 ), however the overall dose relation was less strong. Generally, ventral vagina received lower dose than the dorsal and lateral vagina, in majority of patients.

Results The session was completed in 2 hours, allowing 10 minutes to complete pre-workshop surveys, a maximum of five 20- minute stations, and 10 minutes to complete post- workshop surveys. 27 attendees completed pre-and post-