Radiotherapy aims to deliver a prescribed dose to the tumor while sparing neighboring organs at risk (OARs). Increasingly complex treatment techniques such as volumetric modulated arc therapy (VMAT), stereotactic radiosurgery (SRS), stereotactic body radiotherapy (SBRT), and proton therapy have been developed to deliver doses more precisely to the target. While such technologies have improved dose delivery, the implementation of intra-fraction motion management to verify tumor position at the time of treatment has become increasingly relevant. Artificial intelligence (AI) has recently demonstrated great potential for real-time tracking of tumors during treatment. However, AI-based motion management faces several challenges, including bias in training data, poor transparency, difficult data collection, complex workflows and quality assurance, and limited sample sizes. This review presents the AI algorithms used for chest, abdomen, and pelvic tumor motion management/tracking for radiotherapy and provides a literature summary on the topic. We will also discuss the limitations of these AI-based studies and propose potential improvements.

UNASSIGNED: The purpose of this study was to assess the dosimetric improvements achieved in prostate stereotactic body radiotherapy (SBRT) treatment within the PROMETHEUS and NINJA trials using an in-house real-time position monitoring system, SeedTracker.
UNASSIGNED: This study considered a total of 127 prostate SBRT patients treated in the PROMETHEUS (ACTRN12615000223538) and NINJA (ACTRN12618001806257) clinical trials. The SeedTracker position monitoring system was utilized for real-time position monitoring with a 3-mm position tolerance. The doses delivered to the clinical target volume (CTV), rectum, and bladder were assessed by incorporating the actual target position during treatment. The dose that would have been delivered without monitoring was also assessed by incorporating the observed position deviations.
UNASSIGNED: Treatment with position corrections resulted in a mean (range) CTV D99 difference of -0.3 (-1.0 to 0.0) Gy between the planned and delivered dose. Without corrections, this difference would have been -0.6 (-3.7 to 0.0) Gy. Not correcting for position deviations resulted in a statistically significant difference between the planned and delivered CTV D99 (p < 0.05). The mean (range) dose difference between the planned and delivered D2cc of the rectum and bladder for treatment with position corrections was -0.1 (-3.7 to 4.7) Gy and -0.1 (-1.7 to 0.5) Gy, respectively. Without corrections, these differences would have been -0.6 (-6.1 to 4.7) Gy and -0.2 (-2.5 to 0.9) Gy.
UNASSIGNED: SeedTracker improved clinical dose volume compliance in prostate SBRT. Without monitoring and corrections, delivered dose would significantly differ from the planned dose.

This work investigates the use of a multi-2D cine magnetic resonance imaging-based comprehensive motion monitoring (CMM) system for the assessment of prostate intrafraction 3D drifts. The data of six healthy volunteers were analyzed and the values of a clinically-relevant registration quality factor metric exported by CMM were presented. Additionally, the CMM-derived prostate motion was compared to a 3D-based reference and the 2D-3D tracking agreement was reported. Due to the low quality of SI motion tracking (often > 2 mm tracking mismatch between anatomical planes) we conclude that further improvements are desirable prior to clinical introduction of CMM for prostate drift corrections.

OBJECTIVE: To evaluate the dosimetric impact on spatially fractionated radiation therapy (SFRT) plan quality due to intrafraction patient motion via multi-field MLC-based method for treating large and bulky (≥8 cm) unresectable tumors.
METHODS: For large tumors, a cone beam CT-guided 3D conformal MLC-based SFRT method was utilized with 15 Gy prescription. An MLC GTV-fitting algorithm provided 1 cm diameter apertures with a 2 cm center-to-center distance at the isocenter. This generated a highly heterogeneous sieve-like dose distribution within an hour, enabling same-day SFRT treatment. Fifteen previously treated SFRT patients were analyzed (5 head & neck [H&N], 5 chest and lungs, and 5 abdominal and pelvis masses). For each plan, intrafraction motion errors were simulated by incrementally shifting original isocenters of each field in different x-, y-, and z-directions from 1 to 5 mm. The dosimetric metrics analyzed were: peak-to-valley-dose-ratio (PVDR), percentage of GTV receiving 7.5 Gy, GTV mean dose, and maximum dose to organs-at-risk (OARs).
RESULTS: For ±1, ±2, ±3, ±4, and ±5 mm isocenter shifts: PVDR dropped by 3.9%, 3.8%, 4.0%, 4.1%, and 5.5% on average respectively. The GTV(V7.5) remained within 0.2%, and the GTV mean dose remained within 3.3% on average, compared to the original plans. The average PVDR drop for 5 mm shifts was 4.2% for H&N cases, 10% for chest and lung, and 2.2% for abdominal and pelvis cases. OAR doses also increased. The maximum dose to the spinal cord increased by up to 17 cGy in H&N plans, mean lung dose (MLD) changed was small for chest/lung, but the bowel dose varied up to 100 cGy for abdominal and pelvis cases.
CONCLUSIONS: Due to tumor size, location, and characteristics of MLC-based SFRT, isocenter shifts of up to ±5 mm in different directions had moderate effects on PVDR for H&N and pelvic tumors and a larger effect on chest tumors. The dosimetric impact on OAR doses depended on the treatment site. Site-specific patient masks, Vac-Lok bags, and proper immobilization devices similar to SBRT/SRT setups should be used to minimize these effects.

UNASSIGNED: We aimed to establish if stereotactic body radiotherapy to the prostate can be delivered safely using reduced clinical target volume (CTV) to planning target volume (PTV) margins on the 1.5T MR-Linac (MRL) (Elekta, Stockholm, Sweden), in the absence of gating.
UNASSIGNED: Cine images taken in 3 orthogonal planes during the delivery of prostate SBRT with 36.25 Gray (Gy) in 5 fractions on the MRL were analysed. Using the data from 20 patients, the percentage of radiotherapy (RT) delivery time where the prostate position moved beyond 1, 2, 3, 4 and 5 mm in the left-right (LR), superior-inferior (SI), anterior-posterior (AP) and any direction was calculated.
UNASSIGNED: The prostate moved less than 3 mm in any direction for 90% of the monitoring period in 95% of patients. On a per-fraction basis, 93% of fractions displayed motion in all directions within 3 mm for 90% of the fraction delivery time. Recurring motion patterns were observed showing that the prostate moved with shallow drift (most common), transient excursions and persistent excursions during treatment.
UNASSIGNED: A 3 mm CTV-PTV margin is safe to use for the treatment of 5 fraction prostate SBRT on the MRL, without gating. In the context of gating this work suggests that treatment time will not be extensively lengthened when an appropriate gating window is applied.

HERMES is a phase II trial of MRI-guided daily-adaptive radiotherapy (MRIgART) randomising men with localised prostate cancer to either 2-fractions of SBRT with a boost to the tumour or 5-fraction SBRT. In the context of this highly innovative regime the dose delivered must be carefully considered. The first ten patients recruited to HERMES were analysed in order to establish the dose received by the targets and organs at risk (OARS) in the context of intrafraction motion. A regression analysis was performed to measure how the volume of air within the rectum might further impact rectal dose secondary to the electron return effect (ERE). One hundred percent of CTV target objectives were achieved on the MRI taken prior to beam-on-time. The post-delivery MRI showed that high-dose CTV coverage was achieved in 90% of sub-fractions (each fraction is delivered in two sub-fractions) in the 2-fraction cohort and in 88% of fractions the 5-fraction cohort. Rectal D1 cm3 was the most exceeded constraint; three patients exceeded the D1 cm3 < 20.8 Gy in the 2-fraction cohort and one patient exceeded the D1 cm3 < 36 Gy in the 5-fraction cohort. The volume of rectal gas within 1 cm of the prostate was directly proportional to the increase in rectal D1 cm3, with a strong (R = 0.69) and very strong (R = 0.90) correlation in the 2-fraction and 5-fraction cohort respectively. Dose delivery specified in HERMES is feasible, although for some patients delivered doses to both target and OARs may vary from those planned.

UNASSIGNED: MRI-guided online adaptive treatments can account for interfractional variations, however intrafraction motion reduces treatment accuracy. Intrafraction plan adaptation methods, such as the Intrafraction Drift Correction (IDC) or sub-fractionation, are needed. IDC uses real-time automatic monitoring of the tumor position to initiate plan adaptations by repositioning segments. IDC is a fast adaptation method that occurs only when necessary and this method could enable margin reduction. This research provides a treatment planning evaluation and experimental validation of the IDC.
UNASSIGNED: An in silico treatment planning evaluation was performed for 13 prostate patients mid-treatment without and with intrafraction plan adaptation (IDC and sub-fractionation). The adaptation methods were evaluated using dose volume histogram (DVH) metrics. To experimentally verify IDC a treatment was mimicked whereby a motion phantom containing an EBT3 film moved mid-treatment, followed by repositioning of segments. In addition, the delivered treatment was irradiated on a diode array phantom for plan quality assurance purposes.
UNASSIGNED: The planning study showed benefits for using intrafraction adaptation methods relative to no adaptation, where the IDC and sub-fractionation showed consistently improved target coverage with median target coverages of 100.0%. The experimental results verified the IDC with high minimum gamma passing rates of 99.1% and small mean dose deviations of maximum 0.3%.
UNASSIGNED: The straightforward and fast IDC technique showed DVH metrics consistent with the sub-fractionation method using segment weight re-optimization for prostate patients. The dosimetric and geometric accuracy was shown for a full IDC workflow using film and diode array dosimetry.

UNASSIGNED: The total time of radiation treatment delivery for pancreatic cancer patients with daily online adaptive radiation therapy (ART) on an MR-Linac can range from 50 to 90 min. During this period, the target and normal tissues undergo changes due to respiration and physiologic organ motion. We evaluated the dosimetric impact of the intrafraction physiological organ changes.
UNASSIGNED: Ten locally advanced pancreatic cancer patients were treated with 50 Gy in five fractions with intensity-modulated respiratory-gated radiation therapy on a 0.35-T MR-Linac. Patients received both pre- and post-treatment volumetric MRIs for each fraction. Gastrointestinal organs at risk (GI-OARs) were delineated on the pre-treatment MRI during the online ART process and retrospectively on the post-treatment MRI. The treated dose distribution for each adaptive plan was assessed on the post-treatment anatomy. Prescribed dose volume histogram metrics for the scheduled plan on the pre-treatment anatomy, the adapted plan on the pre-treatment anatomy, and the adapted plan on post-treatment anatomy were compared to the OAR-defined criteria for adaptation: the volume of the GI-OAR receiving greater than 33 Gy (V33Gy) should be ≤1 cubic centimeter.
UNASSIGNED: Across the 50 adapted plans for the 10 patients studied, 70% were adapted to meet the duodenum constraint, 74% for the stomach, 12% for the colon, and 48% for the small bowel. Owing to intrafraction organ motion, at the time of post-treatment imaging, the adaptive criteria were exceeded for the duodenum in 62% of fractions, the stomach in 36%, the colon in 10%, and the small bowel in 48%. Compared to the scheduled plan, the post-treatment plans showed a decrease in the V33Gy, demonstrating the benefit of plan adaptation for 66% of the fractions for the duodenum, 95% for the stomach, 100% for the colon, and 79% for the small bowel.
UNASSIGNED: Post-treatment images demonstrated that over the course of the adaptive plan generation and delivery, the GI-OARs moved from their isotoxic low-dose region and nearer to the dose-escalated high-dose region, exceeding dose-volume constraints. Intrafraction motion can have a significant dosimetric impact; therefore, measures to mitigate this motion are needed. Despite consistent intrafraction motion, plan adaptation still provides a dosimetric benefit.

UNASSIGNED: Distended rectums on pre-radiotherapy scans are historically associated with poorer outcomes in patients treated with two-dimensional IGRT. Subsequently, strict rectal tolerances and preparation regimes were implemented. Contemporary IGRT, daily online registration to the prostate, corrects interfraction motion but intrafraction motion remains. We re-examine the need for rectal management strategies when using contemporary IGRT by quantifying rectal volume and its effect on intrafraction motion.
UNASSIGNED: Pre and during radiotherapy rectal volumes and intrafraction motion were retrospectively calculated for 20 patients treated in 5-fractions and 20 treated in 20-fractions. Small (rectal volume at planning-CT ≤ median), and large (volume > median) subgroups were formed, and rectal volume between timepoints and subgroups compared. Rectal volume and intrafraction motion correlation was examined using Spearman\'s rho. Intrafraction motion difference between small and large subgroups and between fractions with rectal volume < or ≥ 90 cm3 were assessed.
UNASSIGNED: Median rectal volume was 74 cm3, 64 cm3 and 65 cm3 on diagnostic-MRI, planning-CT and treatment imaging respectively (ns). No significant correlation was found between patient\'s rectal volume at planning-CT and median intrafraction motion, nor treatment rectal volume and intrafraction motion for individual fractions. No significant difference in intrafraction motion between small and large subgroups presented and for fractions where rectal volume breached 90 cm3, motion during that fraction was not significantly greater.
UNASSIGNED: Larger rectal volumes before radiotherapy and during treatment did not cause greater intrafraction motion. Findings support the relaxation of strict rectal diameter tolerances and do not support the need for rectal preparation when delivering contemporary IGRT to the prostate.

BACKGROUND: In pediatric radiotherapy treatment planning of abdominal tumors, dose constraints to the pancreatic tail/spleen are applied to reduce late toxicity. In this study, an analysis of inter- and intrafraction motion of the pancreatic tail/spleen is performed to estimate the potential benefits of online MRI-guided radiotherapy (MRgRT).
METHODS: Ten randomly selected neuroblastoma patients (median age: 3.4 years), irradiated with intensity-modulated arc therapy at our department (prescription dose: 21.6/1.8 Gy), were retrospectively evaluated for inter- and intrafraction motion of the pancreatic tail/spleen. Three follow-up MRIs (T2- and T1-weighted ± gadolinium) were rigidly registered to a planning CT (pCT), on the vertebrae around the target volume. The pancreatic tail/spleen were delineated on all MRIs and pCT. Interfraction motion was defined as a center of gravity change between pCT and T2-weighted images in left-right (LR), anterior-posterior (AP) and cranial-caudal (CC) direction. For intrafraction motion analysis, organ position on T1-weighted ± gadolinium was compared to T2-weighted. The clinical radiation plan was used to estimate the dose received by the pancreatic tail/spleen for each position.
RESULTS: The median (IQR) interfraction motion was minimal in LR/AP, and largest in CC direction; pancreatic tail 2.5 mm (8.9), and spleen 0.9 mm (3.9). Intrafraction motion was smaller, but showed a similar motion pattern (pancreatic tail, CC: 0.4 mm (1.6); spleen, CC: 0.9 mm (2.8)). The differences of Dmean associated with inter- and intrafraction motions ranged from - 3.5 to 5.8 Gy for the pancreatic tail and - 1.2 to 3.0 Gy for the spleen. In 6 out of 10 patients, movements of the pancreatic tail and spleen were highlighted as potentially clinically significant because of ≥ 1 Gy dose constraint violation.
CONCLUSIONS: Inter- and intrafraction organ motion results into unexpected constrain violations in 60% of a randomly selected neuroblastoma cohort, supporting further prospective exploration of MRgRT.