Development of programs to predict the occurrence of mucositis from digital imaging and communications in medicine data by machine learning in head and neck volumetric modulated radiotherapy.

Volumetric modulated arc therapy (VMAT) with cisplatin for head and neck cancer is often accompanied by symptoms of pharyngeal and oral mucositis. However, no standard medical program exists for the prevention and treatment of mucositis, and the mechanisms of mucositis have not yet been fully proven. Therefore, adaptive radiotherapy (ART), which is a re-planning process, is administered when severe mucositis develops during the treatment period. We extracted the treatment plans of patients who developed severe mucositis from DICOM data and used machine learning to determine its quantitative features. This study aimed to develop a machine learning program that can predict the development of mucositis requiring ART. This study included 61 patients who received concurrent chemotherapy and radiotherapy (RT). For each patient, the equivalent square field size of each segmental irradiation field used for VMAT, dose per segment (Gy), clinical target volume high, and mean dose of the oral cavity (Gy) were calculated. Furthermore, 671 five-dimensional lists were generated from the acquired data. Support vector machine (SVM) and K-nearest neighbor (KNN) were used for machine learning. For the accuracy score, the test size was varied from 10% to 90%, and the random number of data extracted in each test size was further varied from 1 to 100 to calculate a mean accuracy score. The mean accuracy scores of SVM and KNN were 0.981 ± 0.020 and 0.972 ± 0.033, respectively. The presence or absence of ART for mucositis was classified with high accuracy. The classification of the five-dimensional list was implemented with high accuracy, and a program was constructed to predict the onset of mucositis requiring ART before treatment began. This study suggests that it may support preventive measures against mucositis and the completion of RT without having to re-plan.

Development of Machine-Learning Prediction Programs for Delivering Adaptive Radiation Therapy With Tumor Geometry and Body Shape Changes in Head and Neck Volumetric Modulated Arc Therapy.

Purpose: During radiation therapy for head and neck cancer using volumetric modulated arc therapy, excessive dosing or underdosing occurs as a result of the decrease in tumor volume and changes in body weight. Adaptive radiation therapy (ART) is performed when significant changes are observed; however, the decision to implement ART depends on the oncologist's subjective judgment. The purpose of this study was to present objective indicators for ART and develop a program to predict the need for ART. Methods and Materials: The study included 47 patients in the non-ART group and 21 patients in the ART group with shape changes. Patients who received ART could not be covered with the prescribed radiation therapy dose due to shape changes. For each patient, 1112 6-dimensional lists were created, including the number of irradiations, amount of change in the clinical target volume (CTV), rate of change in CTV, mean oral cavity dose, age, and body mass index. Support vector machine and k-nearest neighbor were used for machine learning. The random number of test data to be extracted varied from 1 to 9, and a mean accuracy score was calculated. These programs could predict the need for ART if the accuracy score was high. Results: The classification accuracy of the list, including the amount of change in the CTV and rate of change in CTV up to 20 fractions, was 0.963 and 0.967 for support vector machine and k-nearest neighbor, respectively. Conclusions: This program predicted the need for ART with more than 90% accuracy based on shape changes over time in cone beam computed tomography analysis for up to 20 fractions. This may provide significant support for objective decisions to implement ART based on the amount of change over time during treatment.

Examination of conversion method of dose distribution of lung cancer IMRT using fluence reversible calculation function in O-ring type linac and C-type linac.

Generally, converting irradiation plans between C-arm linacs (C-linac) when the linac fails is possible without recalculating the dose distribution using a treatment planning system (TPS), because they have similar mechanical structure. However, the O-ring-type linac (O-linac) differs from the C-linac in forming the dose distribution. Therefore, if O-linac breaks down, it is necessary to formulate a treatment plan from scratch. In this study, we investigated a method for converting irradiation from an O-linac to a C-linac. Thirty patients with lung cancer who underwent volumetric-modulated arc therapy with an O-linac were included in this study. The O-linac dose distribution was converted into energy fluence by the function of the TPS. The alternative linac multi-leaf collimator (MLC) was then optimized to achieve energy fluence. The homogeneity index, conformity index, and planning treatment volume (D95%, D2%) of the converted plan were compared with the original plan. For organ at risk (OAR), the dose-volume histograms (DVHs) of the lung, esophagus, heart, and spinal cord were evaluated. Additionally, the shapes of the isodose curves were compared using the Dice similarity coefficient (DSC). There was no significant difference between the target and OARs (p > 0.05). The mean DSCs of 30% to 100% isodose curves of the prescribed dose and the isodose ≥ 105% and ≤ 20%were > 0.8 and < 0.8, respectively. Due to the structural differences of MLC, the dose-volume and generation positions were different in the dose range of ≥ 105% and ≤ 20%; hence, DSCs decreased. However, no statistically significant difference in the DVH was identified for either treatment plan. Based on this result, we propose a simple replanning method for performing MLC fitting after converting the dose to the energy fluence.

[A Preliminary Study of Optimal Imaging Acquisition Parameters for Fiducial Markers in Liver Stereotactic Body Radiotherapy].

In liver stereotactic body radiotherapy (SBRT) using fiducial markers, the accuracy of automatic image recognition of fiducial markers is important, and the imaging dose cannot be neglected in image-guided radiotherapy. Optimal imaging parameters of fiducial markers were investigated for automatic image recognition and imaging dose. We investigated automatic recognition with fiducial markers of different shapes and sizes. In addition, the optimum imaging conditions were examined based on the automatic recognition when the presence or absence of a filter, focal spot size, and phantom thickness were altered using the fiducial markers with a high automatic recognition. The results for different shapes and sizes of fiducial markers showed that larger markers were recognized more automatically, whereas shorter markers were recognized in the correct position. By using the filter, we were able to reduce the imaging dose by one third or one half compared to the case without the filter. The results for the focal spot size showed that using a larger size resulted in higher automatic recognition accuracy than using a smaller size. For the relationship between the automatically recognized imaging conditions and the air kerma when the phantom thickness was altered, it was necessary to keep the tube current-time product constant and increase the tube voltage in order to avoid poor recognition accuracy. The parameters we proposed are effective in shortening the treatment time and reducing the imaging dose because they allow us to acquire images with low doses and high accuracy of automatic recognition.

Comparison of a Hybrid IMRT/VMAT technique with non-coplanar VMAT and non-coplanar IMRT for unresectable olfactory neuroblastoma using the RayStation treatment planning system-EUD, NTCP and planning study.

The purpose of this study was to compare hybrid intensity-modulated radiotherapy (IMRT) and volumetric-modulated arc therapy (Hybrid IMRT/VMAT), with non-coplanar (nc) IMRT and nc-VMAT treatment plans for unresectable olfactory neuroblastoma (ONB). Hybrid IMRT/VMAT, nc-IMRT and nc-VMAT plans were optimized for 12 patients with modified Kadish C stage ONB. Dose prescription was 65 Gy in 26 fractions. Dose-volume histogram parameters, conformation number (CN), homogeneity index (HI), integral dose and monitor units (MUs) delivered per fraction were assessed. Equivalent uniform dose (EUD) and normal tissue complication probability (NTCP) based on the EUD model (NTCPLogit) and the Lyman-Kutcher-Burman model (NTCPLKB) were also evaluated. We found that the Hybrid IMRT/VMAT plan significantly improved the CN for clinical target volume (CTV) and planning treatment volume (PTV) compared with the nc-VMAT plan. In general, sparing of organs at risk (OARs) is similar with the three techniques, although the Hybrid IMRT/VMAT plan resulted in a significantly reduced Dmax to contralateral (C/L) optic nerve compared with the nc-IMRT plan. The Hybrid IMRT/VMAT plan significantly reduce EUD to the ipsilateral (I/L) and C/L optic nerve in comparison with the nc-IMRT plan and nc-VMAT plan, but the difference in NTCP between the three technique was <1%. We concluded that the Hybrid IMRT/VMAT technique can offer improvement in terms of target conformity and EUD for optic nerves, while achieving equal or better OAR sparing compared with nc-IMRT and nc-VMAT, and can be a viable radiation technique for treating unresectable ONB. However, the clinical benefit of these small differences in dosimetric data, EUD and NTCP of optic nerves may be minimal.

Dose-Volume and Radiobiological Model-Based Comparative Evaluation of the Gastrointestinal Toxicity Risk of Photon and Proton Irradiation Plans in Localized Pancreatic Cancer Without Distant Metastasis.

Background: Radiobiological model-based studies of photon-modulated radiotherapy for pancreatic cancer have reported reduced gastrointestinal (GI) toxicity, although the risk is still high. The purpose of this study was to investigate the potential of 3D-passive scattering proton beam therapy (3D-PSPBT) in limiting GI organ at risk (OAR) toxicity in localized pancreatic cancer based on dosimetric data and the normal tissue complication probability (NTCP) model. Methods: The data of 24 pancreatic cancer patients were retrospectively analyzed, and these patients were planned with intensity-modulated radiotherapy (IMRT), volume-modulated arc therapy (VMAT), and 3D-PSPBT. The tumor was targeted without elective nodal coverage. All generated plans consisted of a 50.4-GyE (Gray equivalent) dose in 28 fractions with equivalent OAR constraints, and they were normalized to cover 50% of the planning treatment volume (PTV) with 100% of the prescription dose. Physical dose distributions were evaluated. GI-OAR toxicity risk for different endpoints was estimated by using published NTCP Lyman-Kutcher-Burman (LKB) models. Analysis of variance (ANOVA) was performed to compare the dosimetric data, and ΔNTCPIMRT-PSPBT and ΔNTCPVMAT-PSPBT were also computed. Results: Similar homogeneity and conformity for the clinical target volume (CTV) and PTV were exhibited by all three planning techniques (P > 0.05). 3D-PSPBT resulted in a significant dose reduction for GI-OARs in both the low-intermediate dose range (below 30 GyE) and the highest dose region (D max and V 50 GyE) in comparison with IMRT and VMAT (P < 0.05). Based on the NTCP evaluation, the NTCP reduction for GI-OARs by 3D-PSPBT was minimal in comparison with IMRT and VMAT. Conclusion: 3D-PSPBT results in minimal NTCP reduction and has less potential to substantially reduce the toxicity risk of upper GI bleeding, ulceration, obstruction, and perforation endpoints compared to IMRT and VMAT. 3D-PSPBT may have the potential to reduce acute dose-limiting toxicity in the form of nausea, vomiting, and diarrhea by reducing the GI-OAR treated volume in the low-to-intermediate dose range. However, this result needs to be further evaluated in future clinical studies.

Radiobiological model-based approach to determine the potential of dose-escalated robust intensity-modulated proton radiotherapy in reducing gastrointestinal toxicity in the treatment of locally advanced unresectable pancreatic cancer of the head.

BACKGROUND: The purpose of this study was to determine the potential of escalated dose radiation (EDR) robust intensity-modulated proton radiotherapy (ro-IMPT) in reducing GI toxicity risk in locally advanced unresectable pancreatic cancer (LAUPC) of the head in term of normal tissue complication probability (NTCP) predictive model. METHODS: For 9 patients, intensity-modulated radiotherapy (IMRT) was compared with ro-IMPT. For all plans, the prescription dose was 59.4GyE (Gray equivalent) in 33 fractions with an equivalent organ at risk (OAR) constraints. Physical dose distribution was evaluated. GI toxicity risk for different endpoints was estimated using published NTCP Lyman Kutcher Burman (LKB) models for stomach, duodenum, small bowel, and combine stomach and duodenum (Stoduo). A Wilcoxon signed-rank test was used for dosimetry parameters and NTCP values comparison. RESULT: The dosimetric results have shown that, with similar target coverage, ro-IMPT achieves a significant dose-volume reduction in the stomach, small bowel, and stoduo in low to high dose range in comparison to IMRT. NTCP evaluation for the endpoint gastric bleeding of stomach (10.55% vs. 13.97%, P = 0.007), duodenum (1.87% vs. 5.02%, P = 0.004), and stoduo (5.67% vs. 7.81%, P = 0.008) suggest reduced toxicity by ro-IMPT compared to IMRT. ∆NTCP IMRT - ro-IMPT (using parameter from Pan et al. for gastric bleed) of ≥5 to < 10% was seen in 3 patients (33%) for stomach and 2 patients (22%) for stoduo. An overall GI toxicity relative risk (NTCPro-IMPT/NTCPIMRT) reduction was noted (0.16-0.81) for all GI-OARs except for duodenum (> 1) with endpoint grade ≥ 3 GI toxicity (using parameters from Holyoake et al.). CONCLUSION: With similar target coverage and better conformity, ro-IMPT has the potential to substantially reduce the risk of GI toxicity compared to IMRT in EDR of LAUPC of the head. This result needs to be further evaluated in future clinical studies.

Prospective evaluation of the setup errors and its impact on safety margin for cervical cancer pelvic conformal radiotherapy.

AIM: The primary objective was to assess set-up errors (SE) and secondary objective was to determine optimal safety margin (SM). BACKGROUND: To evaluate the SE and its impact on the SM utilizing electronic portal imaging (EPI) for pelvic conformal radiotherapy. MATERIAL AND METHODS: 20 cervical cancer patients were enrolled in this prospective study. Supine position with ankle and knee rest was used during CT simulation. The contouring was done using consensus guideline for intact uterus. 50 Gy in 25 fractions were delivered at the isocenter with ≥95% PTV coverage. Two orthogonal (Anterior and Lateral) digitally reconstructed radiograph (DRR) was constructed as a reference image. The pair of orthogonal [Anterior-Posterior and Right Lateral] single exposure EPIs during radiation was taken. The reference DRR and EPIs were compared for shifts, and SE was calculated in the X-axis, Y-axis, and Z-axis directions. RESULTS: 320 images (40 DRRs and 280 EPIs) were assessed. The systematic error in the Z-axis (AP EPI), X-axis (AP EPI), and Y-axis (Lat EPI) ranged from -12.0 to 11.8 mm, -10.3 to 7.5 mm, and -8.50 to 9.70 mm, while the random error ranged from 1.60 to 6.15 mm, 0.59 to 4.93 mm, and 1.02 to -4.35 mm. The SM computed were 7.07, 6.36, and 7.79 mm in the Y-axis, X-axis, and Z-axis by Van Herk's equation, and 6.0, 5.51, and 6.74 mm by Stroom's equation. CONCLUSION: The computed SE helps defining SM, and it may differ between institutions. In our study, the calculated SM was approximately 8 mm in the Z-axis, 7 mm in X and Y axis for pelvic conformal radiotherapy.

Could excision repair cross-complementing group-1 mRNA expression from peripheral blood lymphocytes predict locoregional failure with cisplatin chemoradiation for locally advanced laryngeal cancer?

AIM: The association of excision repair cross-complementing 1 mRNA (ERCC-1 mRNA) expression with the outcome has been reported with immunohistochemistry (IHC) using tumor tissue in head and neck cancer. We evaluated ERCC-1 mRNA expression by reverse transcription polymerase chain reaction (RT-PCR) from peripheral blood lymphocytes (PBLs) as bio-predictor of locoregional failure (LRF) to chemoradiation (CRT) for locally advanced laryngeal squamous cell cancer (LALSCC). METHODS: A total of 107 male patients with LALSCC were enrolled in this prospective study. ERCC-1 mRNA expression by PBLs was determined by RT-PCR. Definitive CRT was delivered with 35 mg/m2 weekly cisplatin. Response Evaluation Criteria in Solid Tumor 1.1 (RECIST 1.1) were used in evaluating treatment response. The primary objective was to assess LRF. The influence of patient characteristics, treatment response, weekly cisplatin cycles, ERCC mRNA expression was determined for LRF, progression-free survival (PFS) and overall survival (OS). RESULTS: A total of 98 patients completed definitive CRT. The median value of 2-ΔΔCT ERCC-1 mRNA expression was 3.9; based on which it was categorized as low and high. Correlation of ERCC-1 expression with treatment response was insignificant (P- .38). With a median follow-up of 33 months; 2-year LRF, PFS, and OS was 63.3%, 34.7% and 79.4%. The 2-year LRF, PFS and OS for low versus high expression were 53.1% versus 73.5% (P-value = 0.036), 44.9% versus 24.4% (P-value = 0.047) and 81.6% versus 77.2% (P-value = 0.33), respectively. In multivariate analysis, ERCC-1 expression, T-stage, N-stage and tumor subsite are predictive factors for LRF; T-stage and nodal recurrence for OS; stage and treatment response for PFS. CONCLUSION: LALSCC patient with ERCC-1 mRNA low expression was associated with lower LRF rate, and improved PFS.

Microdosimetric calculation of penumbra for biological dose in wobbled carbon-ion beams with Monte Carlo Method.

In carbon-ion radiotherapy, it is important to evaluate the biological dose because the relative biological effectiveness values vary greatly in a patient's body. The microdosimetric kinetic model (MKM) is a method of estimating the biological effect of radiation by use of microdosimetry. The lateral biological dose distributions were estimated with a modified MKM, in which we considered the overkilling effect in the high linear-energy-transfer region. In this study, we used the Monte Carlo calculation of the Geant4 code to simulate a horizontal port at the Heavy Ion Medical Accelerator in Chiba of the National Institute of Radiological Sciences. The lateral biological dose distributions calculated by Geant4 were almost flat as the lateral absorbed dose in the flattened area. However, in the penumbra region, the lateral biological dose distributions were sharper than the lateral absorbed dose distributions. Furthermore, the differences between the lateral absorbed dose and biological dose distributions were dependent on the depth for each multi-leaf collimator opening size. We expect that the lateral biological dose distribution presented here will enable high-precision calculations for a treatment-planning system.