Prediction of patient-specific quality assurance for volumetric modulated arc therapy using radiomics-based machine learning with dose distribution.

PURPOSE: We sought to develop machine learning models to predict the results of patient-specific quality assurance (QA) for volumetric modulated arc therapy (VMAT), which were represented by several dose-evaluation metrics-including the gamma passing rates (GPRs)-and criteria based on the radiomic features of 3D dose distribution in a phantom. METHODS: A total of 4,250 radiomic features of 3D dose distribution in a cylindrical dummy phantom for 140 arcs from 106 clinical VMAT plans were extracted. We obtained the following dose-evaluation metrics: GPRs with global and local normalization, the dose difference (DD) in 1% and 2% passing rates (DD1% and DD2%) for 10% and 50% dose threshold, and the distance-to-agreement in 1-mm and 2-mm passing rates (DTA1 mm and DTA2 mm) for 0.5%/mm and 1.0%.mm dose gradient threshold determined by measurement using a diode array in patient-specific QA. The machine learning regression models for predicting the values of the dose-evaluation metrics using the radiomic features were developed based on the elastic net (EN) and extra trees (ET) models. The feature selection and tuning of hyperparameters were performed with nested cross-validation in which four-fold cross-validation is used within the inner loop, and the performance of each model was evaluated in terms of the root mean square error (RMSE), the mean absolute error (MAE), and Spearman's rank correlation coefficient. RESULTS: The RMSE and MAE for the developed machine learning models ranged from <1% to nearly <10% depending on the dose-evaluation metric, the criteria, and dose and dose gradient thresholds used for both machine learning models. It was advantageous to focus on high dose region for predicating global GPR, DDs, and DTAs. For certain metrics and criteria, it was possible to create models applicable for patients' heterogeneity by training only with dose distributions in phantom. CONCLUSIONS: The developed machine learning models showed high performance for predicting dose-evaluation metrics especially for high dose region depending on the metric and criteria. Our results demonstrate that the radiomic features of dose distribution can be considered good indicators of the plan complexity and useful in predicting measured dose evaluation metrics.

Deep learning-based detection and classification of multi-leaf collimator modeling errors in volumetric modulated radiation therapy.

PURPOSE: The purpose of this study was to create and evaluate deep learning-based models to detect and classify errors of multi-leaf collimator (MLC) modeling parameters in volumetric modulated radiation therapy (VMAT), namely the transmission factor (TF) and the dosimetric leaf gap (DLG). METHODS: A total of 33 clinical VMAT plans for prostate and head-and-neck cancer were used, assuming a cylindrical and homogeneous phantom, and error plans were created by altering the original value of the TF and the DLG by ± 10, 20, and 30% in the treatment planning system (TPS). The Gaussian filters of σ = 0.5 $\sigma = 0.5$ and 1.0 were applied to the planar dose maps of the error-free plan to mimic the measurement dose map, and thus dose difference maps between the error-free and error plans were obtained. We evaluated 3 deep learning-based models, created to perform the following detections/classifications: (1) error-free versus TF error, (2) error-free versus DLG error, and (3) TF versus DLG error. Models to classify the sign of the errors were also created and evaluated. A gamma analysis was performed for comparison. RESULTS: The detection and classification of TF and DLG error were feasible for σ = 0.5 $\sigma = 0.5$ ; however, a considerable reduction of accuracy was observed for σ = 1.0 $\sigma = 1.0$ depending on the magnitude of error and treatment site. The sign of errors was detectable by the specifically trained models for σ = 0.5 $\sigma = 0.5$ and 1.0. The gamma analysis could not detect errors. CONCLUSIONS: We demonstrated that the deep learning-based models could feasibly detect and classify TF and DLG errors in VMAT dose distributions, depending on the magnitude of the error, treatment site, and the degree of mimicked measurement doses.

Radiobiological evaluation considering setup error on single-isocenter irradiation in stereotactic radiosurgery.

PURPOSE: We calculated the dosimetric indices and estimated the tumor control probability (TCP) considering six degree-of-freedom (6DoF) patient setup errors in stereotactic radiosurgery (SRS) using a single-isocenter technique. METHODS: We used simulated spherical gross tumor volumes (GTVs) with diameters of 1.0 cm (GTV 1), 2.0 cm (GTV 2), and 3.0 cm (GTV 3), and the distance (d) between the target center and isocenter was set to 0, 5, and 10 cm. We created the dose distribution by convolving the blur component to uniform dose distribution. The prescription dose was 20 Gy and the dose distribution was adjusted so that D95 (%) of each GTV was covered by 100% of the prescribed dose. The GTV was simultaneously rotated within 0°-1.0° (δR) around the x-, y-, and z-axes and then translated within 0-1.0 mm (δT) in the x-, y-, and z-axis directions. D95, conformity index (CI), and conformation number (CN) were evaluated by varying the distance from the isocenter. The TCP was estimated by translating the calculated dose distribution into a biological response. In addition, we derived the x-y-z coordinates with the smallest TCP reduction rate that minimize the sum of squares of the residuals as the optimal isocenter coordinates using the relationship between 6DoF setup error, distance from isocenter, and GTV size. RESULTS: D95, CI, and CN were decreased with increasing isocenter distance, decreasing GTV size, and increasing setup error. TCP of GTVs without 6DoF setup error was estimated to be 77.0%. TCP were 25.8% (GTV 1), 35.0% (GTV 2), and 53.0% (GTV 3) with (d, δT, δR) = (10 cm, 1.0 mm, 1.0°). The TCP was 52.3% (GTV 1), 54.9% (GTV 2), and 66.1% (GTV 3) with (d, δT, δR) = (10 cm, 1.0 mm, 1.0°) at the optimal isocenter position. CONCLUSION: The TCP in SRS for multiple brain metastases with a single-isocenter technique may decrease with increasing isocenter distance and decreasing GTV size when the 6DoF setup errors are exceeded (1.0 mm, 1.0°). Additionally, it might be possible to better maintain TCP for GTVs with 6DoF setup errors by using the optimal isocenter position.

The neurocognitive function change criteria after whole-brain radiation therapy for brain metastasis, in reference to health-related quality of life changes: a prospective observation study.

BACKGROUND: We sought to construct the optimal neurocognitive function (NCF) change criteria sensitive to health-related quality of life (HR-QOL) in patients who have undergone whole-brain radiation therapy (WBRT) for brain metastasis. METHODS: We categorized the patients by the changes of NCF into groups of improvement versus deterioration if at least one domain showed changes that exceeded the cut-off while other domains remained stable. The remaining patients were categorized as stable, and the patients who showed both significant improvement and deterioration were categorized as 'both.' We examined the clinical meaning of NCF changes using the cut-off values 1.0, 1.5, and 2.0 SD based on the percentage of patients whose HR-QOL changes were ≥ 10 points. RESULTS: Baseline, 4-month and 8-month data were available in 78, 41 (compliance; 85%), and 29 (81%) patients, respectively. At 4 months, improvement/stable/deterioration/both was seen in 15%/12%/41%/32% of the patients when 1.0 SD was used; 19%/22%/37%/22% with 1.5 SD, and 17%/37%/37%/9% with 2.0 SD. The HR-QOL scores on the QLQ-C30 functional scale were significantly worse in the deterioration group versus the others with 1.0 SD (p = 0.013) and 1.5 SD (p = 0.015). With 1.5 SD, the HR-QOL scores on the QLQ-BN20 was significantly better in the improvement group versus the others (p = 0.033). However, when 'both' was included in 'improvement' or 'deterioration,' no significant difference in HR-QOL was detected. CONCLUSIONS: The NCF cut-off of 1.5 SD and the exclusion of 'both' patients from the 'deterioration' and 'improvement' groups best reflects HR-QOL changes.

High-dose-rate brachytherapy and hypofractionated external beam radiotherapy combined with long-term androgen deprivation therapy for very high-risk prostate cancer.

OBJECTIVE: To estimate the outcomes of high-dose-rate brachytherapy combined with hypofractionated external beam radiotherapy in prostate cancer patients classified as very high risk by the National Comprehensive Cancer Network. METHODS: Between June 2009 and September 2015, 66 patients meeting the criteria for very high-risk disease received high-dose-rate brachytherapy (2 fractions of 9 Gy) as a boost of external beam radiotherapy (13 fractions of 3 Gy). Androgen deprivation therapy was administered for approximately 3 years. Biochemical failure was assessed using the Phoenix definition. RESULTS: The median follow-up period was 53 months from the completion of radiotherapy. The 5-year biochemical failure-free, distant metastasis-free, prostate cancer-specific and overall survival rates were 88.7, 89.2, 98.5 and 97.0%, respectively. The independent contribution of each component of the very high-risk criteria was assessed in multivariable models. Primary Gleason pattern 5 was associated with increased risks of biochemical failure (P = 0.017) and distant metastasis (P = 0.049), whereas clinical stage ≥T3b or >4 biopsy cores with Gleason score 8-10 had no significant impact on the two outcomes. Grade 3 genitourinary toxicities were observed in two (3.0%) patients, whereas no grade ≥3 gastrointestinal toxicities occurred. CONCLUSIONS: The present study shows that this multimodal approach provides potentially excellent cancer control and acceptable associated morbidity for very high-risk disease. Patients with primary Gleason pattern 5 are at a higher risk of poor outcomes, indicating the need for more aggressive approaches in these cases.

Radiobiological effects of the interruption time with Monte Carlo Simulation on multiple fields in photon beams.

PURPOSE: The interruption time is the irradiation interruption that occurs at sites and operations such as the gantry, collimator, couch rotation, and patient setup within the field in radiotherapy. However, the radiobiological effect of prolonging the treatment time by the interruption time for tumor cells is little evaluated. We investigated the effect of the interruption time on the radiobiological effectiveness with photon beams based on a modified microdosimetric kinetic (mMK) model. METHODS: The dose-mean lineal energy yD (keV/µm) of 6-MV photon beams was calculated by the particle and heavy ion transport system (PHITS). We set the absorbed dose to 2 or 8 Gy, and the interruption time (τ) was set to 1, 3, 5, 10, 30, and 60 min. The biological parameters such as α0, ß0, and DNA repair constant rate (a + c) values were acquired from a human non-small-cell lung cancer cell line (NCI-H460) for the mMK model. We used two-field and four-field irradiation with a constant dose rate (3 Gy/min); the photon beams were paused for interruption time τ. We calculated the relative biological effectiveness (RBE) to evaluate the interruption time's effect compared with no interrupted as a reference. RESULTS: The yD of 6-MV photon beams was 2.32 (keV/µm), and there was little effect by changing the water depth (standard deviation was 0.01). The RBE with four-field irradiation for 8 Gy was decreased to 0.997, 0.975, 0.900, and 0.836 τ = 1, 10, 30, 60 min, respectively. In addition, the RBE was affected by the repair constant rate (a + c) value, the greater the decrease in RBE with the longer the interruption time when the (a + c) value was large. CONCLUSION: The ~10-min interruption of 6-MV photon beams did not significantly impact the radiobiological effectiveness, since the RBE decrease was <3%. Nevertheless, the RBE's effect on tumor cells was decreased about 30% by increasing the 60 min interruption time at 8 Gy with four-field irradiation. It is thus necessary to make the interruption time as short as possible.

Effect of setup error in the single-isocenter technique on stereotactic radiosurgery for multiple brain metastases.

In conventional stereotactic radiosurgery (SRS), treatment of multiple brain metastases using multiple isocenters is time-consuming resulting in long dose delivery times for patients. A single-isocenter technique has been developed which enables the simultaneous irradiation of multiple targets at one isocenter. This technique requires accurate positioning of the patient to ensure optimal dose coverage. We evaluated the effect of six degrees of freedom (6DoF) setup errors in patient setups on SRS dose distributions for multiple brain metastases using a single-isocenter technique. We used simulated spherical gross tumor volumes (GTVs) with diameters ranging from 1.0 to 3.0 cm. The distance from the isocenter to the target's center was varied from 0 to 15 cm. We created dose distributions so that each target was entirely covered by 100% of the prescribed dose. The target's position vectors were rotated from 0°-2.0° and translated from 0-1.0 mm with respect to the three axes in space. The reduction in dose coverage for the targets for each setup error was calculated and compared with zero setup error. The calculated margins for the GTV necessary to satisfy the tolerance values for loss of GTV coverage of 3% to 10% were defined as coverage-based margins. In addition, the maximum isocenter to target distance for different 6DoF setup errors was calculated to satisfy the tolerance values. The dose coverage reduction and coverage-based margins increased as the target diameter decreased, and the distance and 6DoF setup error increased. An increase in setup error when a single-isocenter technique is used may increase the risk of missing the tumor; this risk increases with increasing distance from the isocenter and decreasing tumor size.

The impact of the three degrees-of-freedom fiducial marker-based setup compared to soft tissue-based setup in hypofractionated intensity-modulated radiotherapy for prostate cancer.

PURPOSE: We evaluated the setup accuracy of a three-degree-of-freedom fiducial marker (3DOF-FM)-based setup compared to a soft tissue (ST)-based setup in hypofractionated intensity-modulated radiotherapy (IMRT) for prostate cancer. MATERIALS AND METHODS: We analyzed the setup accuracy for 17 consecutive prostate cancer patients with three implanted FMs who underwent hypofractionated IMRT. The 3DOF-ST-based setup using cone-beam computed tomography (CT) was performed after a six DOF-bony structure (BS)-based setup using an ExacTrac x-ray system. The 3DOF-FM-based matching using the ExacTrac x-ray system was done during the BS- and ST-based setups. We determined the mean absolute differences and the correlation between the FM- and ST-based translational shifts relative to the BS-based setup position. The rotational mean shifts detected by the ExacTrac x-ray system were also evaluated. RESULTS: The mean differences in the anterior-posterior (AP), superior-inferior (SI), and left-right (LR) dimensions were 0.69, 0.0, and 0.30 mm, respectively. The Pearson correlation coefficients for both shifts were 0.92 for AP, 0.91 for SI, and 0.68 for LR. The percentages of shift agreements within 2 mm were 85% for AP, 93% for SI, and 99% for LR. The absolute values of rotational shifts were 0.1° for AP, 0.3°, and 1.2° for LR. CONCLUSIONS: The setup accuracy of the 3DOF-FM-based setup has the potential to be interchangeable with a ST-based setup. Our data are likely to be useful in clinical practice along with the popularization of the hypofractionated IMRT in prostate cancer.

Superiority of a soft tissue-based setup using cone-beam computed tomography over a bony structure-based setup in intensity-modulated radiotherapy for prostate cancer.

The purpose of this study was to test the superiority of a soft tissue-based setup using cone-beam computed tomography (CBCT) to a bony structure-based setup using the ExacTrac system in intensity-modulated radiotherapy (IMRT) for prostate cancer. We studied 20 patients with localized prostate cancer who received IMRT between November 2010 and February 2012. After the initial setup, the pelvic bony structure-based setup and ExacTrac system were applied. After that, CBCT and a soft tissue-based setup were used. A shift in the isocenter between the ExacTrac-based and CBCT-based setup was recorded in the anterior-posterior (AP), superior-inferior (SI), and left-right (LR) axes. The shift was considered an interfractional prostate shift. Post-treatment CBCT was also taken once a week to measure the intrafractional prostate shift, based on the coordinates of the isocenter between pre- and post-treatment CBCT. The planning target volume (PTV) margins were determined using van Herk's method. We measured the elapsed time required for soft tissue matching and the entire treatment time using CBCT. The means ± standard deviation (SD) of the inter- and intrafractional shifts were 0.9 ± 2.8 mm and -0.3 ± 1.4 mm in the AP, 0.9 ± 2.2 mm and -0.1 ± 1.2 mm in the SI, and 0.1 ± 0.7 mm and -0.1 ± 0.7 mm in the LR directions. The PTV margins in the cases of bony structure-based and soft tissue-based setups were 7.3 mm and 2.7 mm in the AP, 5.8 mm and 2.3 mm in the SI, and 1.9 mm and 1.2 mm in the LR directions. Even though the median elapsed time using CBCT was expanded in 5.9 min, the PTV margins were significantly reduced. We found the calculated PTV margins in the soft tissue-based setup using CBCT were small, and this arrangement was superior to the bony structure-based setup in prostate IMRT.

Assessing tumor volumetric reduction with consideration for setup errors based on mathematical tumor model and microdosimetric kinetic model in single-isocenter VMAT for brain metastases.

The volumetric reduction rate (VRR) was evaluated with consideration for six degrees-of-freedom (6DoF) patient setup errors based on a mathematical tumor model in single-isocenter volumetric modulated arc therapy (SI-VMAT) for brain metastases. Simulated gross tumor volumes (GTV) of 1.0 cm and dose distribution were created (27 Gy/3 fractions). The distance between the GTV center and isocenter (d) was set at 0-10 cm. The GTV was translated within 0-1.0 mm (Trans) and rotated within 0-1.0° (Rot) in the three axis directions using affine transformation. The tumor growth volume was calculated using a multicomponent mathematical model (MCTM), and lethal effects of irradiation and repair from damage during irradiation were calculated by a microdosimetric kinetic model (MKM) for non-small cell lung cancer (NSCLC) A549 and NCI-H460 (H460) cells. The VRRs were calculated 5 days after the end of irradiation using the physical dose to the GTV for varying d and 6DoF setup errors. The tolerance value of VRR, the GTV volume reduction rate, was set at 5%, based on the pre-irradiation GTV volume. With the exception of the only one A549 condition where (Trans, Rot) = (1.0 mm, 1.0°) was repeated for 3 fractions, all conditions met all the tolerance VRR values for A549 and H460 cells with varying d from 0 to 10 cm. Evaluation based on the mathematical tumor model suggested that if the 6DoF setup errors at each irradiation could be kept within 1.0 mm and 1.0°, there would be little effect on tumor volume regardless of the distance from the isocenter in SI-VMAT.

Mathematical model combined with microdosimetric kinetic model for tumor volume calculation in stereotactic body radiation therapy.

We proposed a new mathematical model that combines an ordinary differential equation (ODE) and microdosimetric kinetic model (MKM) to predict the tumor-cell lethal effect of Stereotactic body radiation therapy (SBRT) applied to non-small cell lung cancer (NSCLC). The tumor growth volume was calculated by the ODE in the multi-component mathematical model (MCM) for the cell lines NSCLC A549 and NCI-H460 (H460). The prescription doses 48 Gy/4 fr and 54 Gy/3 fr were used in the SBRT, and the effect of the SBRT on tumor cells was evaluated by the MKM. We also evaluated the effects of (1) linear quadratic model (LQM) and the MKM, (2) varying the ratio of active and quiescent tumors for the total tumor volume, and (3) the length of the dose-delivery time per fractionated dose (tinter) on the initial tumor volume. We used the ratio of the tumor volume at 1 day after the end of irradiation to the tumor volume before irradiation to define the radiation effectiveness value (REV). The combination of MKM and MCM significantly reduced REV at 48 Gy/4 fr compared to the combination of LQM and MCM. The ratio of active tumors and the prolonging of tinter affected the decrease in the REV for A549 and H460 cells. We evaluated the tumor volume considering a large fractionated dose and the dose-delivery time by combining the MKM with a mathematical model of tumor growth using an ODE in lung SBRT for NSCLC A549 and H460 cells.

The Relationship between the Contouring Time of the Metal Artifacts Area and Metal Artifacts in Head and Neck Radiotherapy.

(1) Background: The impacts of metal artifacts (MAs) on the contouring workload for head and neck radiotherapy have not yet been clarified. Therefore, this study evaluated the relationship between the contouring time of the MAs area and MAs on head and neck radiotherapy treatment planning. (2) Methods: We used treatment planning computed tomography (CT) images for head and neck radiotherapy. MAs were classified into three severities by the percentage of CT images containing MAs: mild (<25%), moderate (25−75%), and severe (>75%). We randomly selected nine patients to evaluate the relationship between MAs and the contouring time of the MAs area. (3) Results: The contouring time of MAs showed moderate positive correlations with the MAs volume and the number of CT images containing MAs. Interobserver reliability of the extracted MAs volume and contouring time were excellent and poor, respectively. (4) Conclusions: Our study suggests that the contouring time of MAs areas is related to individual commitment rather than clinical experience. Therefore, the development of software combining metal artifact reduction methods with automatic contouring methods is necessary to reducing interobserver variability and contouring workload.

Multicomponent mathematical model for tumor volume calculation with setup error using single-isocenter stereotactic radiotherapy for multiple brain metastases.

We evaluated the tumor residual volumes considering six degrees-of-freedom (6DoF) patient setup errors in stereotactic radiotherapy (SRT) with multicomponent mathematical model using single-isocenter irradiation for brain metastases. Simulated spherical gross tumor volumes (GTVs) with 1.0 (GTV 1), 2.0 (GTV 2), and 3.0 (GTV 3)-cm diameters were used. The distance between the GTV center and isocenter (d) was set at 0-10 cm. The GTV was simultaneously translated within 0-1.0 mm (T) and rotated within 0°-1.0° (R) in the three axis directions using affine transformation. We optimized the tumor growth model parameters using measurements of non-small cell lung cancer cell lines' (A549 and NCI-H460) growth. We calculated the GTV residual volume at the irradiation's end using the physical dose to the GTV when the GTV size, d, and 6DoF setup error varied. The d-values that satisfy tolerance values (10%, 35%, and 50%) of the GTV residual volume rate based on the pre-irradiation GTV volume were determined. The larger the tolerance value set for both cell lines, the longer the distance to satisfy the tolerance value. In GTV residual volume evaluations based on the multicomponent mathematical model on SRT with single-isocenter irradiation, the smaller the GTV size and the larger the distance and 6DoF setup error, the shorter the distance that satisfies the tolerance value might need to be.

Verification of Qfix Encompass™ couch modeling using the Acuros XB algorithm and HypeArc™ using a high-spatial-resolution two-dimensional diode array.

We modeled the Qfix Encompass™ immobilization system and further verified the calculated dose distribution of the AcurosXB (AXB) dose calculation algorithm using SRS MapCHECKⓇ (SRSMC) in the HyperArc™ (HA) clinical plan. An Encompass system with a StereoPHAN™ QA phantom was scanned by SOMATOM go.Sim and imported to an Eclipse™ treatment planning system to create a treatment plan for Encompass modeling. The Encompass modeling was performed in the StereoPHAN with a pinpoint ion chamber for 6 MV and 6 MV flattening filter free (6 MV FFF), and 2 × 2 cm2, 4 × 4 cm2, and 6 × 6 cm2 irradiation field sizes. The dose calculation algorithm used was AXB ver. 15.5 with a 1.0 mm calculation grid size. The Hounsfield unit (HU) values of the Encompass modeling were set to 400, -100, -200, and -300 for Encompass, and -400, -600, -700, and -800 for the Encompass base. We evaluated the dose distribution after Encompass modeling by SRSMC using gamma analysis in 12 patients. We adopted HU values of -200 for Encompass, -800 for Encompass base for 6 MV, and -200 for Encompass and -700 for Encompass. Base for 6 MV FFF was adopted as the HU values for the Encompass modeling based on the measurement results. The proposed Encompass modeling resulted in a mean pass rate evaluation >98% for both 6 MV and 6 MV FFF when the 1%/1 mm criterion was used, demonstrating that the proposed HU value can be adopted to calculate more accurate dose distributions.

The impact of target positioning error and tumor size on radiobiological parameters in robotic stereotactic radiosurgery for metastatic brain tumors.

This study aimed to evaluate the effect of target positioning error (TPE) on radiobiological parameters, such as tumor control probability (TCP) and normal tissue complication probability (NTCP), in stereotactic radiosurgery (SRS) for metastatic brain tumors of different sizes using CyberKnife. The reference SRS plans were created using the circular cone of the CyberKnife for each spherical gross tumor volume (GTV) with diameters (φ) of 5, 7.5, 10, 15, and 20 mm, contoured on computed tomography images of the head phantom. Subsequently, plans involving TPE were created by shifting the beam center by 0.1-2.0 mm in three dimensions relative to the reference plans using the same beam arrangements. Conformity index (CI), generalized equivalent uniform dose (gEUD)-based TCP, and NTCP of estimated brain necrosis were evaluated for each plan. When the gEUD parameter "a" was set to - 10, the CI and TCP for the reference plan at the φ5-mm GTV were 0.90 and 80.8%, respectively. The corresponding values for plans involving TPE of 0.5-mm, 1.0-mm, and 2.0-mm were 0.62 and 77.4%, 0.40 and 62.9%, and 0.12 and 7.2%, respectively. In contrast, the NTCP for all GTVs were the same. The TCP for the plans involving a TPE of 2-mm was 7.2% and 68.8% at the φ5-mm and φ20-mm GTV, respectively. The TPEs corresponding to a TCP reduction rate of 3% at the φ5-mm and φ20-mm GTV were 0.41 and 0.99 mm, respectively. TPE had a significant effect on TCP in SRS for metastatic brain tumors using CyberKnife, particularly for small GTVs.

Radiobiological evaluation considering the treatment time with stereotactic radiosurgery for brain metastases.

Objective: We evaluated the radiobiological effect of the irradiation time with the interruption time of stereotactic radiosurgery (SRS) using CyberKnife® (CK) systemfor brain metastases. Methods: We used the DICOM data and irradiation log file of the 10 patients with brain metastases from non-small-cell lung cancer (NSCLC) who underwent brain SRS. We defined the treatment time as the sum of the dose-delivery time and the interruption time during irradiations, and we used a microdosimetric kinetic model (MKM) to evaluate the radiobiological effects of the treatment time. The biological parameters, i.e. α0, ß0, and the DNA repair constant rate (a + c), were acquired from NCI-H460 cell for the MKM. We calculated the radiobiological dose for the gross tumor volume (GTVbio) to evaluate the treatment time's effect compared with no treatment time as a reference. The D95 (%) and the Radiation Therapy Oncology Group conformity index (RCI) and Paddick conformity index (PCI) were calculated as dosimetric indices. We used several DNA repair constant rates (a + c) (0.46, 1.0, and 2.0) to assess the radiobiological effect by varying the DNA repair date (a + c) values. Results: The mean values of D95 (%), RCI, and PCI for GTVbio were 98.8%, 0.90, and 0.80, respectively, and decreased with increasing treatment time. The mean values of D95 (%), RCI, and PCI of GTVbio at 2.0 (a+c) value were 94.9%, 0.71, and 0.49, respectively. Conclusion: The radiobiological effect of the treatment time on tumors was accurately evaluated with brain SRS using CK. Advances in knowledge: There has been no published investigation of the radiobiological impact of the longer treatment time with multiple interruptions of SRS using a CK on the target dose distribution in a comparison with the use of a linac. Radiobiological dose assessment that takes into account treatment time in the physical dose in this study may allow more accurate dose assessment in SRS for metastatic brain tumors using CK.

Calculated relative biological effectiveness (RBE) for initial DNA double-strand breaks (DSB) from flattening filter and flattening filter-free 6 MV X-ray fields.

OBJECTIVES: We evaluated the radiobiological effectiveness based on the yields of DNA double-strand breaks (DSBs) of field induction with flattening filter (FF) and FF-free (FFF) photon beams. METHODS: We used the particle and heavy ion transport system (PHITS) and a water equivalent phantom (30 × 30 × 30 cm3) to calculate the physical qualities of the dose-mean lineal energy (yD) with 6 MV FF and FFF. The relative biological effectiveness based on the yields of DNA-DSBs (RBEDSB) was calculated for standard radiation such as 220 kVp X-rays by using the estimating yields of SSBs and DSBs. The measurement points used to calculate the in-field yD and RBEDSB were located at a depth of 3, 5, and 10 cm in the water equivalent phantom on the central axis. Measurement points at 6, 8, and 10 cm in the lateral direction of each of the three depths from the central axis were set to calculate the out-of-field yD and RBEDSB. RESULTS: The RBEDSB of FFF in-field was 1.7% higher than FF at each measurement depth. The RBEDSB of FFF out-of-field was 1.9 to 6.4% higher than FF at each depth measurement point. As the distance to out-of-field increased, the RBEDSB of FFF rose higher than those of FF. FFF has a larger RBEDSB than FF based on the yields of DNA-DSBs as the distance to out-of-field increased. CONCLUSIONS: The out-of-field radiobiological effect of FFF could thus be greater than that of FF since the spreading of the radiation dose out-of-field with FFF could be a concern compared to the FF. ADVANCES IN KNOWLEDGE: The RBEDSB of FFF of out-of-field might be larger than FF.

Maximum distance in single-isocenter technique of stereotactic radiosurgery with rotational error using margin-based analysis.

Through geometrical simulation, we evaluated the effect of rotational error in patient setup on geometrical coverage and calculated the maximum distance between the isocenter and target, where the clinical PTV margin secures geometrical coverage with a single-isocenter technique. We used simulated spherical GTVs with diameters of 1.0 (GTV 1), 1.5 (GTV 2), 2.0 (GTV 3), and 3.0 cm (GTV 4). The location of the target center was set such that the distance between the target and isocenter ranged from 0 to 15 cm. We created geometrical coverage vectors so that each target was entirely covered by 100% of the prescribed dose. The vectors of the target positions were simultaneously rotated within a range of 0°-2.0° around the x-, y-, and z-axes. For each rotational error, the reduction in geometrical coverage of the targets was calculated and compared with that obtained for a rotational error of 0°. The tolerance value of the geometrical coverage reduction was defined as 5% of the GTV. The maximum distance that satisfied the 5% tolerance value for different values of rotational error at a clinical PTV margin of 0.1 cm was calculated. When the rotational errors were 0.5° for a 0.1 cm PTV margin, the maximum distances were as follows: GTV 1: 7.6 cm; GTV 2: 10.9 cm; GTV 3: 14.3 cm; and GTV 4: 21.4 cm. It might be advisable to exclude targets that are > 7.6 cm away from the isocenter with a single-isocenter technique to satisfy the tolerance value for all GTVs.

Detecting MLC modeling errors using radiomics-based machine learning in patient-specific QA with an EPID for intensity-modulated radiation therapy.

PURPOSE: We sought to develop machine learning models to detect multileaf collimator (MLC) modeling errors with the use of radiomic features of fluence maps measured in patient-specific quality assurance (QA) for intensity-modulated radiation therapy (IMRT) with an electric portal imaging device (EPID). METHODS: Fluence maps measured with EPID for 38 beams from 19 clinical IMRT plans were assessed. Plans with various degrees of error in MLC modeling parameters [i.e., MLC transmission factor (TF) and dosimetric leaf gap (DLG)] and plans with an MLC positional error for comparison were created. For a total of 152 error plans for each type of error, we calculated fluence difference maps for each beam by subtracting the calculated maps from the measured maps. A total of 837 radiomic features were extracted from each fluence difference map, and we determined the number of features used for the training dataset in the machine learning models by using random forest regression. Machine learning models using the five typical algorithms [decision tree, k-nearest neighbor (kNN), support vector machine (SVM), logistic regression, and random forest] for binary classification between the error-free plan and the plan with the corresponding error for each type of error were developed. We used part of the total dataset to perform fourfold cross-validation to tune the models, and we used the remaining test dataset to evaluate the performance of the developed models. A gamma analysis was also performed between the measured and calculated fluence maps with the criteria of 3%/2 and 2%/2 mm for all of the types of error. RESULTS: The radiomic features and its optimal number were similar for the models for the TF and the DLG error detection, which was different from the MLC positional error. The highest sensitivity was obtained as 0.913 for the TF error with SVM and logistic regression, 0.978 for the DLG error with kNN and SVM, and 1.000 for the MLC positional error with kNN, SVM, and random forest. The highest specificity was obtained as 1.000 for the TF error with a decision tree, SVM, and logistic regression, 1.000 for the DLG error with a decision tree, logistic regression, and random forest, and 0.909 for the MLC positional error with a decision tree and logistic regression. The gamma analysis showed the poorest performance in which sensitivities were 0.737 for the TF error and the DLG error and 0.882 for the MLC positional error for 3%/2 mm. The addition of another type of error to fluence maps significantly reduced the sensitivity for the TF and the DLG error, whereas no effect was observed for the MLC positional error detection. CONCLUSIONS: Compared to the conventional gamma analysis, the radiomics-based machine learning models showed higher sensitivity and specificity in detecting a single type of the MLC modeling error and the MLC positional error. Although the developed models need further improvement for detecting multiple types of error, radiomics-based IMRT QA was shown to be a promising approach for detecting the MLC modeling error.

A comparative study of high-dose-rate brachytherapy boost combined with external beam radiation therapy versus external beam radiation therapy alone for high-risk prostate cancer.

We aimed to compare the outcomes of high-dose-rate brachytherapy (HDR-BT) boost and external beam radiation therapy (EBRT) alone for high-risk prostate cancer. This was a single-center, retrospective and observational study. Consecutive patients who underwent initial radical treatment by HDR-BT boost or EBRT alone from June 2009 to May 2016 at the Niigata University Medical and Dental Hospital, Japan were included. A total of 96 patients underwent HDR-BT boost, and 61 underwent EBRT alone. The prescription dose of HDR-BT boost was set to 18 Gy twice a day with EBRT 39 Gy/13 fractions. The dose for EBRT alone was mostly 70 Gy/28 fractions. The high-risk group received >6 months of prior androgen deprivation therapy. Overall survival, biochemical-free survival, local control and distant metastasis-free survival rates at 5 years were analyzed. The incidence of urological and gastrointestinal late adverse events of Grade 2 and above was also summarized. In the National Comprehensive Cancer Network (NCCN) high-risk calssification, HDR-BT boost had a significantly higher biochemical-free survival rate at 5 years (98.9% versus 90.7%, P = 0.04). Urethral strictures were more common in the HDR-BT boost group. We will continuously observe the progress of the study patients and determine the longer term results.