Trans Tasman Radiation Oncology Group

Automated contour quality assurance (QA) has the potential to reduce resource and cost requirements for clinical trial QA. While several studies have developed such models, few have been translated for use in prospective real-world trials. This study aimed to implement a contour QA tool using a previously developed model and evaluate its performance during deployment within the TROG 18.01 NINJA prostate cancer trial. A software tool was developed using an existing prostate clinical target volume (CTV) QA model and integrated into the trial QA workflow. A pilot study was conducted over an 18-month period, during which 56 CTVs were assessed. Reports generated by the tool flagged cases for review and were provided to radiation oncologists to support QA processes. All five protocol-violating CTVs were correctly identified during deployment, yielding a sensitivity of 1.0. However, a higher-than-expected false positive rate resulted in an accuracy of 0.46 and specificity of 0.41. Retrospective analysis showed that many cases submitted deviated from the model's training distribution, primarily due to inconsistencies in MRI acquisition and variation in submitted CTV definitions. Incorporating out-of-distribution detection based on histogram correlation and model uncertainty improved accuracy to 0.69 for in-distribution cases. Radiation oncologists reported time savings of up to 60min per case. However, preparation of data remained time intensive for QA coordinators, highlighting the need for further workflow automation. These findings support the feasibility of automated contour QA in multicentre trials and offer guidance for future implementation at scale.

Abstract BACKGROUND The prospective multi-site trial evaluating O-(2-[18F]-fluoroethyl)-L-tyrosine Positron Emission Tomography (FET-PET) in Glioblastoma (FIG) study is currently recruiting. This abstract aims to evaluate the impact of full central Nuclear Medicine physician (NMP) review of prospective FET-PET1 delineation of the biological target volume (BTV) for radiotherapy (RT) planning. MATERIAL AND METHODS Adult GBM participants across 11 Australian sites undergo up to 3 FET-PET studies post-surgery/pre-chemo-RT [CRT] (FET-PET1), one month post CRT (FET-PET2) and at suspected progression evaluation timepoint (FET-PET3). Group 1 participants enter at timepoint 1 (FET-PET1 with MRI1), with Group 2 at timepoint 2. Adjuvant RT target volumes are derived per standard contrast MRI. These are compared to hybrid post-hoc RT volumes incorporating the FET-PET1 NM-derived BTV utilising MiM version 7.0 to evaluate the impact of FET PET on potential treatment planning. All trial sites and NMP have passed credentialling which included three benchmarking cases involving FET1-BTV delineation. RESULTS Recruitment commenced in January 2021, with 253 (n=156 Group 1 and n=97 Group 2) participants enrolled to date, n=144 with evaluable FET-PET1 data. Trial credentialling demonstrated variations in FET-PET1-derived BTV in 25/72 (34.7%) -13 minor and 12 major. All n=144 prospective participant FET-PET1 with BTV delineation cases across 11 sites have undergone central NMP review, with 20/144 (13.9%) requiring resubmission. Reasons for resubmission/protocol deviation included incorrect imaging sequence selection within MiM workflow (n=3/20), static GTV overcontouring (n=12/20), dynamic volume of interest change in size/position during workflow (n=1/20) and static FET interpretation issues (n=3/20). Central radiation oncology review of hybrid BTV-derived RT volumes is underway. CONCLUSION The importance of credentialling and full central review of all prospective FET-PET-derived BTV delineation in FIG trial participants is demonstrated. The FIG study remains the largest prospective multi-site study of its kind addressing the impact of FET-PET on radiation planning, management of pseudoprogression and prognostication.

The capability to caracterize small bodies by stellar occultations has expanded enormously thanks to the Gaia mission, whose accurate astrometry has improved by order of magnitude the accuracy of star positions and proper motions, and asteroid orbits. More and more occultations by minor planets as small as 5-10~km in the Main Belt are reported by observers often equipped with mobile telescopes of modest size. Appropriate techniques to recover the absolute timing of the observed events and software tools that are now widely available permit to recover asteroid astrometry (relative to the star positions) with accuracies comparable to Gaia, especially at the smallest asteroid sizes.After some successful observations of occultations by Near Earth Asteroids such as Phaethon and Apophis, we took the challenge to push the limit at smaller sub-km NEAs. Several months before the impact by the DART probe on Dimorphos, satellite of (65803) Didymos, we established a networking activity of collaboration among amateur and professional astronomers, named ACROSS (Asteroid Collaborative Research via Occultation Systematic Survey), initially supported by the European Space Agency.One of the primary goals of ACROSS is to obtain very accurate astrometry over a few years, in order to evaluate the effect of the (September 2022) DART impact on the heliocentric orbit of Didymos. However, just a few months before the DART arrival to the target the uncertainty in the ephemeris was more than an order of magnitude larger than the size of Didymos, making predictions and telescope deployment practically not feasible.After a long work of astrometric reduction of available CCD images collected during the DART/Hera ground-based campaign (fully exploited for photometry but not for astrometry) and careful weighting of the observations, we were able to gradually converge to an exploitable orbit accuracy very close to the date of the impact. Following several failed attempts, the first successful event was observed just after the DART impact.This positive detection was followed by several others (19 in total), over 6 months, providing an astrometric accuracy comparable to the DART telemetry. At the end of the campaign astrometry by occultations was able to constrain the orbit at the same level as the use of the DART telemetry.However, the success of the observations was also enhanced by the detection in 4 events of occultations by the satellite Dimorphos. While their contribution to our knowledge about the position of Dimorphos along its orbit is not major at this stage, we found that the occultation signal, modulated by diffraction effects, can provide relevant information about its size and shape.The new ongoing campaign of stellar occultation in 2024 can thus provide an important contribution not only to constraint the dynamical properties of the Didymos system, but also to understand the post-impact properties of Dimorphos, before the arrival of Hera, including its shape. A first occultation in May 2024 proved the high quality of the orbit, resulting in the first positive of the new sequence. In this talk, we will discuss the results obtained in 2023 and present some preliminary results from the ongoing 2024 campaign.