Purpose: Our purpose was to validate and compare the performance of 4 organ dose reconstruction approaches for historical radiation treatment planning based on 2-dimensional radiographs.Methods and... Show morePurpose: Our purpose was to validate and compare the performance of 4 organ dose reconstruction approaches for historical radiation treatment planning based on 2-dimensional radiographs.Methods and materials: We considered 10 patients with Wilms tumor with planning computed tomography images for whom we developed typical historic Wilms tumor radiation treatment plans, using anteroposterior and posteroanterior parallel-opposed 6 MV flank fields, normalized to 14.4 Gy. Two plans were created for each patient, with and without corner blocking. Regions of interest (lungs, heart, nipples, liver, spleen, contralateral kidney, and spinal cord) were delineated, and dose-volume metrics including organ mean and minimum dose (Dmean and Dmin) were computed as the reference baseline for comparison. Dosimetry for the 20 plans was then independently reconstructed using 4 different approaches. Three approaches involved surrogate anatomy, among which 2 used demographic-matching criteria for phantom selection/building, and 1 used machine learning. The fourth approach was also machine learning-based, but used no surrogate anatomies. Absolute differences in organ dose-volume metrics between the reconstructed and the reference values were calculated.Results: For Dmean and Dmin (average and minimum point dose) all 4 dose reconstruction approaches performed within 10% of the prescribed dose (≤1.4 Gy). The machine learning-based approaches showed a slight advantage for several of the considered regions of interest. For Dmax (maximum point dose), the absolute differences were much higher, that is, exceeding 14% (2 Gy), with the poorest agreement observed for near-beam and out-of-beam organs for all approaches.Conclusions: The studied approaches give comparable dose reconstruction results, and the choice of approach for cohort dosimetry for late effects studies should still be largely driven by the available resources (data, time, expertise, and funding). Show less
Virgolin, M.; Wang, Z.; Balgobind, B.V.; Dijk, I.W.E.M. van; Wiersma, J.; Kroon, P.S.; ... ; Alderliesten, T. 2020
To study radiotherapy-related adverse effects, detailed dose information (3D distribution) is needed for accurate dose-effect modeling. For childhood cancer survivors who underwent radiotherapy in... Show moreTo study radiotherapy-related adverse effects, detailed dose information (3D distribution) is needed for accurate dose-effect modeling. For childhood cancer survivors who underwent radiotherapy in the pre-CT era, only 2D radiographs were acquired, thus 3D dose distributions must be reconstructed from limited information. State-of-the-art methods achieve this by using 3D surrogate anatomies. These can however lack personalization and lead to coarse reconstructions. We present and validate a surrogate-free dose reconstruction method based on Machine Learning (ML). Abdominal planning CTs (n = 142) of recently-treated childhood cancer patients were gathered, their organs at risk were segmented, and 300 artificial Wilms' tumor plans were sampled automatically. Each artificial plan was automatically emulated on the 142 CTs, resulting in 42,600 3D dose distributions from which dose-volume metrics were derived. Anatomical features were extracted from digitally reconstructed radiographs simulated from the CTs to resemble historical radiographs. Further, patient and radiotherapy plan features typically available from historical treatment records were collected. An evolutionary ML algorithm was then used to link features to dose-volume metrics. Besides 5-fold cross validation, a further evaluation was done on an independent dataset of five CTs each associated with two clinical plans. Cross-validation resulted in mean absolute errors <= 0.6 Gy for organs completely inside or outside the field. For organs positioned at the edge of the field, mean absolute errors <= 1.7 Gy for D-mean, <= 2.9 Gy for D-2cc, and <= 13% for V-5 Gy and V-10 Gy, were obtained, without systematic bias. Similar results were found for the independent dataset. To conclude, we proposed a novel organ dose reconstruction method that uses ML models to predict dose-volume metric values given patient and plan features. Our approach is not only accurate, but also efficient, as the setup of a surrogate is no longer needed. Show less
Background: While four-dimensional computed tomography (4DCT) is extensively used in adults, reluctance remains to use 4DCT in children. Day-to-day (interfractional) variability and irregular... Show moreBackground: While four-dimensional computed tomography (4DCT) is extensively used in adults, reluctance remains to use 4DCT in children. Day-to-day (interfractional) variability and irregular respiration (intrafractional variability) have shown to be limiting factors of 4DCT effectiveness in adults. In order to evaluate 4DCT applicability in children, the purpose of this study is to quantify inter- and intrafractional variability of respiratory motion in children and adults. The pooled analysis enables a solid comparison to reveal if 4DCT application for planning purposes in children could be valid.Methods/materials: We retrospectively included 90 patients (45 children and 45 adults), for whom the diaphragm was visible on abdominal/thoracic free-breathing cone beam CTs (480 pediatric, 524 adult CBCTs). For each CBCT, the cranial caudal position of end-exhale and end-inhale positions of the right diaphragm dome were manually selected in the projection images. The difference in position between both phases defines the amplitude. Cycle time equaled inspiratory plus expiratory time. We analyzed the variability of the inter- and intrafractional respiratory-induced diaphragm motion.Results: Ranges of respiratory motion characteristics were large in both children and adults (amplitude: 4-17 vs 5-24 mm, cycle time 2.1-3.9 vs 2.7-6.5 s). The mean amplitude was slightly smaller in children than in adults (10.7 vs 12.3 mm; P = 0.06). Interfractional amplitude variability was statistically significantly smaller in children than in adults (1.4 vs 2.2 mm; P = 0.00). Mean cycle time was statistically significantly shorter in children (2.9 vs 3.6 s; P = 0.00). Additionally, intrafractional cycle time variability was statistically significantly smaller in children (0.5 vs 0.7 s; P = 0.00).Conclusions: Overall variability is smaller in children than in adults, indicating that respiratory motion is more regular in children than in adults. This implies that a single pretreatment 4DCT could be a good representation of daily respiratory motion in children and will be at least equally beneficial for planning purposes as it is in adults. Show less
Recurrent pediatric tumors pose a challenge since treatment options may be limited, particularly after previous irradiation. Positive results have been reported for chemotherapy and hyperthermia,... Show moreRecurrent pediatric tumors pose a challenge since treatment options may be limited, particularly after previous irradiation. Positive results have been reported for chemotherapy and hyperthermia, but the combination of re-irradiation and hyperthermia has not been investigated thus far, although it is a proven treatment strategy in adults. The theoretical feasibility of re-irradiation plus hyperthermia was investigated for infield recurrent pediatric sarcoma in the pelvic region and the extremities. A total of 46 recurrent pediatric sarcoma cases diagnosed at the Academic Medical Center (Amsterdam, The Netherlands) between 2002 and 2017 were evaluated. Patients not previously irradiated, outfield recurrences and locations other than the pelvis and extremities were excluded, ultimately yielding four eligible patients: Two with sarcomas in the pelvis and two in an extremity. Re-irradiation and hyperthermia treatment plans were simulated for 23x2 Gy treatment schedules and weekly hyperthermia. The radiosensitizing effect of hyperthermia was quantified using biological modelling with a temperature-dependent change in the parameters of the linear-quadratic model. The possible effectiveness of re-irradiation plus hyperthermia was estimated by calculating the equivalent radiotherapy dose distribution. Treatment planning revealed that tumors located in the pelvis and the extremities can be effectively heated in children. Equivalent dose distributions indicated that hyperthermic radiosensitization can be quantified as a target-selective additional D95% of typically 10 Gy, thereby delivering a possibly curative dose of 54 Gy, without substantially increasing the equivalent dose to the organs at risk. Therefore, re-irradiation plus hyperthermia is a theoretically feasible and possibly effective treatment option for recurrent pediatric sarcoma in the pelvic region and the extremities, and its clinical feasibility is worthy of evaluation. Show less
Background: In adults, a single pre-treatment four-dimensional CT (4D-CT) acquisition is often used to account for respiratory-induced target motion during radiotherapy. However, studies have... Show moreBackground: In adults, a single pre-treatment four-dimensional CT (4D-CT) acquisition is often used to account for respiratory-induced target motion during radiotherapy. However, studies have indicated that a 4D-CT is not always representative for respiratory motion. Our aim was to investigate whether respiratory-induced diaphragm motion in children on a single pre-treatment 4DCT can accurately predict respiratory-induced diaphragm motion as observed on cone beam CTs (CBCTs).Methods: Twelve patients (mean age 14.5 yrs.; range 8.6-17.9 yrs) were retrospectively included based on visibility of the diaphragm on abdominal or thoracic imaging data acquired during free breathing. A 4DCT for planning purposes and daily/weekly CBCTs (total 125; range 4-29 per patient) acquired prior to dose delivery were available. The amplitude, corresponding to the difference in position of the diaphragm in cranial-caudal direction in end-inspiration and end-expiration phases, was extracted from the 4DCT (A(4DCT)). The amplitude in CBCTs (A(CBCT)) was defined as displacement between averaged in- and expiration diaphragm positions on corresponding projection images, and the distribution of A(CBCT) was compared to A(4DCT) (one-sample t-test, significance level p<0.05).Results: Over all patients, the mean A(4DCT) was 10.4 mm and the mean A(CBCT) 11.6 mm. For 9/12 patients, A(4DCT) differed significantly (p<0.05) from A(CBCT). Differences >3 mm were found in 69/125 CBCTs (55%), with A(4DCT) mostly underestimating A(CBCT). For 7/12 patients, diaphragm positions differed significantly from the baseline position.Conclusion: Respiratory-induced diaphragm motion determined on 4DCT does not accurately predict the daily respiratory-induced diaphragm motion observed on CBCTs, as the amplitude and baseline position differed statistically significantly in the majority of patients. Regular monitoring of respiratory motion during the treatment course using CBCTs could yield a higher accuracy when a daily adaptation to the actual breathing amplitude takes place. Show less
Huijskens, S.C.; Dijk, I.W.E.M. van; Visser, J.; Balgobind, B.V.; Lindert, D. te; Rasch, C.R.N.; ... ; Bel, A. 2018
Background: Interfractional organ position variation might differ for abdominal organs and this could have consequences for defining safety margins. Therefore, the purpose of this study is to... Show moreBackground: Interfractional organ position variation might differ for abdominal organs and this could have consequences for defining safety margins. Therefore, the purpose of this study is to quantify interfractional position variations of abdominal organs in children in order to investigate possible correlations between abdominal organs and determine whether position variation is location-dependent.Methods: For 20 children (2.2-17.8 years), we retrospectively analyzed 113 CBCTs acquired during the treatment course, which were registered to the reference CT to assess interfractional position variation of the liver, spleen, kidneys, and both diaphragm domes. Organ position variation was assessed in three orthogonal directions and relative to the bony anatomy. Diaphragm dome position variation was assessed in the cranial-caudal (CC) direction only. We investigated possible correlations between position variations of the organs (Spearman's correlation test, p), and tested if organ position variations in the CC direction are related to the diaphragm dome position variations (linear regression analysis, R-2) (both tests: significance level p < 0.05). Differences of variations of systematic (Sigma) and random errors (sigma) between organs were tested (Bonferroni significance level p < 0.004).Results: In all directions, correlations between liver and spleen position variations, and between right and left kidney position variations were weak (p <= 0.43). In the CC direction, the position variations of the right and left diaphragm domes were significantly, and stronger, correlated with position variations of the liver (R-2 = 0.55) and spleen (R-2 = 0.63), respectively, compared to the right (R-2 = 0.00) and left kidney (R-2 = 0.25). Differences in Sigma and sigma between all organs were small and insignificant.Conclusions: No (strong) correlations between interfractional position variations of abdominal organs in children were observed. From present results, we concluded that diaphragm dome position variations could be more representative for superiorly located abdominal (liver, spleen) organ position variations than for inferiorly located (kidneys) organ position variations. Differences of systematic and random errors between abdominal organs were small, suggesting that for margin definitions, there was insufficient evidence of a dependence of organ position variation on anatomical location. Show less
Background and purpose: Radiotherapy involving the thoracic region is associated with cardiotoxicity in long-term childhood cancer survivors. We quantified heart volume changes during radiotherapy... Show moreBackground and purpose: Radiotherapy involving the thoracic region is associated with cardiotoxicity in long-term childhood cancer survivors. We quantified heart volume changes during radiotherapy in children (<18 years) and investigated correlations with patient and treatment related characteristics. Material and methods: Between 2010 and 2016, 34 children received radiotherapy involving the thoracic region. We delineated heart contours and measured heart volumes on 114 CBCTs. Relative volume changes were quantified with respect to the volume on the first CBCT (i.e., 100%). Cardiac radiation dose parameters expressed as 2 Gy/fraction equivalent doses were calculated from DVHs. Chemotherapy was categorized as treatment with anthracyclines, alkylating agents, vinca-alkaloids, and other. Results: The overall median heart volume reduction from the first to the last CBCT was 5.5% (interquartile range1.6-9.7%; p < 0.001). Heart volumes decreased significantly between the baseline measurement and the first week (Bonferroni's adjusted p = 0.002); volume changes were not significant during the following weeks. Univariate analysis showed a significant correlation between heart volume reduction and alkylating agents; however, no multivariate analyses could be done to further confirm this. Conclusions: We found a significant heart volume reduction in children during radiotherapy. Elucidation of underlying mechanisms, clinical relevance, and possible long-term consequences of early heart volume reduction require a prospective follow-up study. (C) 2018 Elsevier B.V. All rights reserved. Show less
Background and purpose: To analyse the variability of respiratory motion during image-guided radiotherapy in paediatric cancer patients and to investigate possible relationships thereof with... Show moreBackground and purpose: To analyse the variability of respiratory motion during image-guided radiotherapy in paediatric cancer patients and to investigate possible relationships thereof with patient-specific factors.Material and methods: Respiratory-induced diaphragm motion was retrospectively analysed on 480 cone beam CTs acquired during the treatment course of 45 children (<18 years). The cranial-caudal positions of the top of the right diaphragm in exhale and inhale phases were manually selected in the projection images. The difference in position between both phases defines the amplitude. The cycle time equalled inspiratory plus expiratory time. We analysed the variability of the intra- and interfractional respiratory motion and studied possible correlations between respiratory-induced diaphragm motion and age, height, and weight.Results: Over all patients, mean amplitude and cycle time were 10.7 mm (range 4.1-17.4 mm) and 2.9 s (range 2.1-3.9 s). Intrafractional variability was larger than interfractional variability (2.4 mm vs. 1.4 mm and 0.5 s vs. 0.4 s for amplitude and cycle time, respectively). Correlations between mean amplitude and patient-specific factors were significant but weak (p < 0.05, p <= 0.45).Conclusions: Large ranges of amplitude and cycle time and weak correlations confirm that respiratory motion is patient-specific and requires an individualized approach to account for. Since interfractional variability was small, we suggest that a pre-treatment 4DCT in children could be sufficiently predictive to quantify the respiratory motion. (C) 2017 Elsevier B.V. All rights reserved. Show less
Dijk, I.W.E.M. van; Huijskens, S.C.; Jong, R. de; Visser, J.; Fajardo, R.D.; Rasch, C.R.N.; ... ; Bel, A. 2017
Background: Pediatric safety margins are generally based on data from adult studies; however, adult-based margins might be too large for children. The aim of this study was to quantify and compare... Show moreBackground: Pediatric safety margins are generally based on data from adult studies; however, adult-based margins might be too large for children. The aim of this study was to quantify and compare interfractional organ position variation in children and adults.Material and methods: For 35 children and 35 adults treated with thoracic/abdominal irradiation, 850 (range 5-30 per patient) retrospectively collected cone beam CT images were registered to the reference CT that was used for radiation treatment planning purposes. Renal position variation was assessed in three orthogonal directions and summarized as 3D vector lengths. Diaphragmatic position variation was assessed in the cranio-caudal (CC) direction only. We calculated means and SDs to estimate group systematic (sigma) and random errors (sigma) of organ position variation. Finally, we investigated possible correlations between organ position variation and patients' height.Results: Interfractional organ position variation was different in children and adults. Median 3D right and left kidney vector lengths were significantly smaller in children than in adults (2.8, 2.9mm vs. 5.6, 5.2mm, respectively; p<.05). Generally, the pediatric sigma and sigma were significantly smaller than in adults (p<.007). Overall and within both subgroups, organ position variation and patients' height were only negligibly correlated.Conclusions: Interfractional renal and diaphragmatic position variation in children is smaller than in adults indicating that pediatric margins should be defined differently from adult margins. Underlying mechanisms and other components of geometrical uncertainties need further investigation to explain differences and to appropriately define pediatric safety margins. Show less
Huijskens, S.C.; Dijk, I.W.E.M. van; Jong, R. de; Visser, J.; Fajardo, R.D.; Ronckers, C.M.; ... ; Bel, A. 2015
Background and purpose: To quantify renal and diaphragmatic interfractional motion in order to estimate systematic and random errors, and to investigate the correlation between interfractional... Show moreBackground and purpose: To quantify renal and diaphragmatic interfractional motion in order to estimate systematic and random errors, and to investigate the correlation between interfractional motion and patient-specific factors.Material and methods: We used 527 retrospective abdominal-thoracic cone beam CT scans of 39 childhood cancer patients (<18 years) to quantify renal motion relative to bony anatomy in the left-right (LR), cranio-caudal (CC) and anterior-posterior (AP) directions, and diaphragmatic motion in the CC direction only. Interfractional motion was quantified by distributions of systematic and random errors in each direction (standard deviations Sigma and sigma, respectively). Also, correlation between organ motion and height was analyzed.Results: Inter-patient organ motion varied widely, with the largest movements in the CC direction. Values of Sigma in LR, CC, and AP directions were 1.1, 3.8, 2.1 mm for the right, and 1.3, 3.0, 1.5 mm for the left kidney, respectively. The sigma in these three directions was 1.1, 3.1, 1.7 mm for the right, and 1.2, 2.9, 2.1 mm for the left kidney, respectively. For the diaphragm we estimated Sigma = 5.2 mm and sigma = 4.0 mm. No correlations were found between organ motion and height.Conclusions: The large inter-patient organ motion variations and the lack of correlation between motion and patient-related factors, suggest that individualized margin approaches might be required. (C) 2015 Elsevier Ireland Ltd. All rights reserved. Show less
