Development of a national deep learning-based auto-segmentation model for the heart on clinical delineations from the DBCG RT nation cohort.

BACKGROUND: This study aimed at investigating the feasibility of developing a deep learning-based auto-segmentation model for the heart trained on clinical delineations. MATERIAL AND METHODS: This study included two different datasets. The first dataset contained clinical heart delineations from the DBCG RT Nation study (1,561 patients). The second dataset was smaller (114 patients), but with corrected heart delineations. Before training the model on the clinical delineations an outlier-detection was performed, to remove cases with gross deviations from the delineation guideline. No outlier detection was performed for the dataset with corrected heart delineations. Both models were trained with a 3D full resolution nnUNet. The models were evaluated with the dice similarity coefficient (DSC), 95% Hausdorff distance (HD95) and Mean Surface Distance (MSD). The difference between the models were tested with the Mann-Whitney U-test. The balance of dataset quantity versus quality was investigated, by stepwise reducing the cohort size for the model trained on clinical delineations. RESULTS: During the outlier-detection 137 patients were excluded from the clinical cohort due to non-compliance with delineation guidelines. The model trained on the curated clinical cohort performed with a median DSC of 0.96 (IQR 0.94-0.96), median HD95 of 4.00 mm (IQR 3.00 mm-6.00 mm) and a median MSD of 1.49 mm (IQR 1.12 mm-2.02 mm). The model trained on the dedicated and corrected cohort performed with a median DSC of 0.95 (IQR 0.93-0.96), median HD95 of 5.65 mm (IQR 3.37 mm-8.62 mm) and median MSD of 1.63 mm (IQR 1.35 mm-2.11 mm). The difference between the two models were found non-significant for all metrics (p > 0.05). Reduction of cohort size showed no significant difference for all metrics (p > 0.05). However, with the smallest cohort size, a few outlier structures were found. CONCLUSIONS: This study demonstrated a deep learning-based auto-segmentation model trained on curated clinical delineations which performs on par with a model trained on dedicated delineations, making it easier to develop multi-institutional auto-segmentation models.

Personal innovative approach in radiation therapy of lung cancer- functional lung avoidance SPECT-guided (ASPECT) radiation therapy: a study protocol for phase II randomised double-blind clinical trial.

BACKGROUND: Radiation therapy (RT) plays a key role in curative-intent treatment for locally advanced lung cancer. Radiation induced pulmonary toxicity can be significant for some patients and becomes a limiting factor for radiation dose, suitability for treatment, as well as post treatment quality of life and suitability for the newly introduced adjuvant immunotherapy. Modern RT techniques aim to minimise the radiation dose to the lungs, without accounting for regional distribution of lung function. Many lung cancer patients have significant regional differences in pulmonary function due to smoking and chronic lung co-morbidity. Even though reduction of dose to functional lung has shown to be feasible, the method of preferential functional lung avoidance has not been investigated in a randomised clinical trial. METHODS: In this study, single photon emission computed tomography (SPECT/CT) imaging technique is used for functional lung definition, in conjunction with advanced radiation dose delivery method in randomised, double-blind trial. The study aims to assess the impact of functional lung avoidance technique on pulmonary toxicity and quality of life in patients receiving chemo-RT for lung cancer. Eligibility criteria are biopsy verified lung cancer, scheduled to receive (chemo)-RT with curative intent. Every patient will undergo a pre-treatment perfusion SPECT/CT to identify functional lung. At radiation dose planning, two plans will be produced for all patients on trial. Standard reference plan, without the use of SPECT imaging data, and functional avoidance plan, will be optimised to reduce the dose to functional lung within the predefined constraints. Both plans will be clinically approved. Patients will then be randomised in a 2:1 ratio to be treated according to either the functional avoidance or the standard plan. This study aims to accrue a total of 200 patients within 3 years. The primary endpoint is symptomatic radiation-induced lung toxicity, measured serially 1-12 months after RT. Secondary endpoints include: a quality of life and patient reported lung symptoms assessment, overall survival, progression-free survival, and loco-regional disease control. DISCUSSION: ASPECT trial will investigate functional avoidance method of radiation delivery in clinical practice, and will establish toxicity outcomes for patients with lung cancer undergoing curative chemo-RT. TRIAL REGISTRATION: Clinicaltrials.gov Identifier: NCT04676828 . Registered 1 December 2020.

Strategies for Motion Robust Proton Therapy With Pencil Beam Scanning for Esophageal Cancer.

PURPOSE: Proton therapy of esophageal cancer is superior to photon radiation therapy in terms of normal tissue sparing. However, respiratory motion and anatomical changes may compromise target dose coverage owing to density changes, geometric misses, and interplay effects. Here we investigate the combined effect on clinical target volume (CTV) coverage and compare proton therapy with intensity modulated radiation therapy (IMRT). METHODS AND MATERIALS: This study includes 26 patients with esophageal cancer previously treated with IMRT planned on 4-dimensional computed tomography (4D-CT). For each patient, 7 proton pencil beam scanning (PBS) plans were created with different field configurations and optimization strategies. The effect of respiration was investigated by calculating the phase doses, 4D dose, and 4D dynamic dose (including interplay effects). The effect of anatomical changes was investigated by recalculating all plans on all phases of a 4D-CT surveillance scan. RESULTS: The most robust PBS plans were achieved using 2 posterior beams requiring coverage of planning target volume (PTV) and simultaneously using robust optimization (RO) of CTV (2PAPTVRO), resulting in only 1 patient showing V95%CTV <97% in 1 or more phases of the planning CT. For the least robust PBS plans obtained using lateral + posterior beams and CTV-RO, but not requiring PTV coverage (2LPRO), 10 patients showed underdosage. For IMRT, 2 patients showed underdosage. Interplay effects reduced V95%CTV significantly when delivering only 1 fraction, but the effects generally averaged out after 10 fractions. The effect of interplay was significantly larger for RO-only plans compared with plans optimized with RO combined with PTV coverage. Combining the effect of anatomical changes and respiration on the 4D-CT surveillance scan resulted in V95%CTV <97% for 3 2PAPTVRO, 16 2LPRO, and 8 IMRT patients. CONCLUSIONS: PBS using posterior beam angles was more robust to anatomical changes and respiration than IMRT. The effect of respiration was enhanced when anatomical changes were present. Single fraction interplay effects deteriorated the dose distribution but were averaged out after 10 fractions.

Variation in current prescription practice of stereotactic body radiotherapy for peripherally located early stage non-small cell lung cancer: Recommendations for prescribing and recording according to the ACROP guideline and ICRU report 91.

BACKGROUND AND PURPOSE: In 2017 the ACROP guideline on SBRT for peripherally located early stage NSCLC was published. Later that year ICRU-91 about prescribing, recording and reporting was published. The purpose of this study is to quantify the current variation in prescription practice in the institutions that contributed to the ACROP guideline and to establish the link between the ACROP and ICRU-91 recommendations. MATERIAL AND METHODS: From each of the eight participating centres, 15 SBRT plans for stage I NSCLC were analyzed. Plans were generated following the institutional protocol, centres prescribed 3 × 13.5 Gy, 3 × 15 Gy, 3 × 17 Gy or 3 × 18 Gy. Dose parameters of the target volumes were reported as recommended by ICRU-91 and also converted to BED10Gy. RESULTS: The intra-institutional variance in D98%, Dmean and D2% of the PTV and GTV/ITV is substantially smaller than the inter-institutional spread, indicating well protocollised planning procedures are followed. The median values per centre ranged from 56.1 Gy to 73.1 Gy (D2%), 50.4 Gy to 63.3 Gy (Dmean) and 40.5 Gy to 53.6 Gy (D98%) for the PTV and from 57.1 Gy to 73.6 Gy (D2%), 53.7 Gy to 68.7 Gy (Dmean) and 48.5 Gy to 62.3 Gy (D98%) for the GTV/ITV. Comparing the variance in PTV D98% with the variance in GTV Dmean per centre, using an F-test, shows that four centres have a larger variance in GTV Dmean, while one centre has a larger variance in PTV D98% (p values <0.01). This shows some centres focus on achieving a constant PTV coverage while others aim at a constant GTV coverage. CONCLUSION: More detailed recommendations for dose planning and reporting of lung SBRT in line with ICRU-91 were formulated, including a minimum PTV D98% of 100 Gy BED10Gy and minimum GTV/ITV mean dose of 150 Gy BED10Gy and a D2% in the range of 60-70 Gy.

Validation of a robust strategy for proton spot scanning for oesophageal cancer in the presence of anatomical changes.

SFUD strategies with one or two posterior proton beams and three target coverage strategies are compared with IMRT and tested for robustness towards anatomical changes by recalculation on surveillance CTs during treatment. We find posterior beam SFUD combining PTV coverage with robust optimization increases robustness towards anatomical changes compared to IMRT.

Difference in target definition using three different methods to include respiratory motion in radiotherapy of lung cancer.

INTRODUCTION: Minimizing the planning target volume (PTV) while ensuring sufficient target coverage during the entire respiratory cycle is essential for free-breathing radiotherapy of lung cancer. Different methods are used to incorporate the respiratory motion into the PTV. MATERIAL AND METHODS: Fifteen patients were analyzed. Respiration can be included in the target delineation process creating a respiratory GTV, denoted iGTV. Alternatively, the respiratory amplitude (A) can be measured based on the 4D-CT and A can be incorporated in the margin expansion. The GTV expanded by A yielded GTV + resp, which was compared to iGTV in terms of overlap. Three methods for PTV generation were compared. PTVdel (delineated iGTV expanded to CTV plus PTV margin), PTVσ (GTV expanded to CTV and A was included as a random uncertainty in the CTV to PTV margin) and PTV∑ (GTV expanded to CTV, succeeded by CTV linear expansion by A to CTV + resp, which was finally expanded to PTV∑). RESULTS: Deformation of tumor and lymph nodes during respiration resulted in volume changes between the respiratory phases. The overlap between iGTV and GTV + resp showed that on average 7% of iGTV was outside the GTV + resp implying that GTV + resp did not capture the tumor during the full deformable respiration cycle. A comparison of the PTV volumes showed that PTVσ was smallest and PTVΣ largest for all patients. PTVσ was in mean 14% (31 cm3) smaller than PTVdel, while PTVdel was 7% (20 cm3) smaller than PTVΣ. CONCLUSIONS: PTVσ yields the smallest volumes but does not ensure coverage of tumor during the full respiratory motion due to tumor deformation. Incorporating the respiratory motion in the delineation (PTVdel) takes into account the entire respiratory cycle including deformation, but at the cost, however, of larger treatment volumes. PTVΣ should not be used, since it incorporates the disadvantages of both PTVdel and PTVσ.

Dosimetric evaluation of anatomical changes during treatment to identify criteria for adaptive radiotherapy in oesophageal cancer patients.

BACKGROUND: Some oesophageal cancer patients undergoing chemotherapy and concomitant radiotherapy (chemoRT) show large interfractional anatomical changes during treatment. These changes may modify the dose delivered to the target and organs at risk (OARs). The aim of the presenwt study was to investigate the dosimetric consequences of anatomical changes during treatment to obtain criteria for an adaptive RT decision support system. MATERIAL AND METHODS: Twenty-nine patients were treated with chemoRT for oesophageal and gastro-oesophageal junction cancer and set up according to daily cone beam computed tomography (CBCTs) scans. All patients had an additional replanning CT scan at median fraction number 10 (9-14), which was deformably registered to the original planning CT. Gross tumour volumes (GTVs), clinical target volumes (CTVs) and OARs were transferred to the additional CT and corrected by an exwperienced physician. Treatment plans were recalculated and dose to targets and OARs was evaluated. Treatment was adapted if the volume receiving 95% of the prescribed dose (V95%) coverage of CTV decreased > 1% or planning target volume (PTV) decreased by > 3%. RESULTS: In total, nine adaptive events were observed: All nine were triggered by PTV V95% decrease > 3% [median 11% (5-41%)] and six of these were additionally triggered by CTV V95% decrease > 1% [median 5% (2-35%)]. The largest discrepancies were caused by interfractional baseline or amplitude shifts in diaphragm position (n = 5). Mediastinal (n = 6), oesophageal (n = 6) and bowel filling changes (n = 2) caused the remainder of the changes. For patients with dosimetric changes exceeding the adaptation limits, the discrepancies were confirmed by inspecting the daily CBCTs. In 31% of all patients, heart V30Gy increased more than 2% (maximum 5%). Only minor changes in lung dose or liver dose were seen. CONCLUSION: Target coverage throughout the course of chemoRT treatment is compromised in some patients due to interfractional anatomical changes. Dose to the heart may increase as well.