Distinct fibroblast subpopulations associated with bone, brain or intrapulmonary metastasis in advanced non-small-cell lung cancer.

BACKGROUND: Bone or brain metastases may develop in 20-40% of individuals with late-stage non-small-cell lung cancer (NSCLC), resulting in a median overall survival of only 4-6 months. However, the primary lung cancer tissue's distinctions between bone, brain and intrapulmonary metastases of NSCLC at the single-cell level have not been underexplored. METHODS: We conducted a comprehensive analysis of 14 tissue biopsy samples obtained from treatment-naïve advanced NSCLC patients with bone (n = 4), brain (n = 6) or intrapulmonary (n = 4) metastasis using single-cell sequencing originating from the lungs. Following quality control and the removal of doublets, a total of 80 084 cells were successfully captured. RESULTS: The most significant inter-group differences were observed in the fraction and function of fibroblasts. We identified three distinct cancer-associated fibroblast (CAF) subpopulations: myofibroblastic CAF (myCAF), inflammatory CAF (iCAF) and antigen-presenting CAF (apCAF). Notably, apCAF was prevalent in NSCLC with bone metastasis, while iCAF dominated in NSCLC with brain metastasis. Intercellular signalling network analysis revealed that apCAF may play a role in bone metastasis by activating signalling pathways associated with cancer stemness, such as SPP1-CD44 and SPP1-PTGER4. Conversely, iCAF was found to promote brain metastasis by activating invasion and metastasis-related molecules, such as MET hepatocyte growth factor. Furthermore, the interaction between CAFs and tumour cells influenced T-cell exhaustion and signalling pathways within the tumour microenvironment. CONCLUSIONS: This study unveils the direct interplay between tumour cells and CAFs in NSCLC with bone or brain metastasis and identifies potential therapeutic targets for inhibiting metastasis by disrupting these critical cell-cell interactions.

Attention-Gated Deep-Learning-Based Automatic Digitization of Interstitial Needles in High-Dose-Rate Brachytherapy for Cervical Cancer.

Purpose: Deep learning can be used to automatically digitize interstitial needles in high-dose-rate (HDR) brachytherapy for patients with cervical cancer. The aim of this study was to design a novel attention-gated deep-learning model, which may further improve the accuracy of and better differentiate needles. Methods and Materials: Seventeen patients with cervical cancer with 56 computed tomography-based interstitial HDR brachytherapy plans from the local hospital were retrospectively chosen with the local institutional review board's approval. Among them, 50 plans were randomly selected as the training set and the rest as the validation set. Spatial and channel attention gates (AGs) were added to 3-dimensional convolutional neural networks (CNNs) to highlight needle features and suppress irrelevant regions; this was supposed to facilitate convergence and improve accuracy of automatic needle digitization. Subsequently, the automatically digitized needles were exported to the Oncentra treatment planning system (Elekta Solutions AB, Stockholm, Sweden) for dose evaluation. The geometric and dosimetric accuracy of automatic needle digitization was compared among 3 methods: (1) clinically approved plans with manual needle digitization (ground truth); (2) the conventional deep-learning (CNN) method; and (3) the attention-added deep-learning (CNN + AG) method, in terms of the Dice similarity coefficient (DSC), tip and shaft positioning errors, dose distribution in the high-risk clinical target volume (HR-CTV), organs at risk, and so on. Results: The attention-gated CNN model was superior to CNN without AGs, with a greater DSC (approximately 94% for CNN + AG vs 89% for CNN). The needle tip and shaft errors of the CNN + AG method (1.1 mm and 1.8 mm, respectively) were also much smaller than those of the CNN method (2.0 mm and 3.3 mm, respectively). Finally, the dose difference for the HR-CTV D90 using the CNN + AG method was much more accurate than that using CNN (0.4% and 1.7%, respectively). Conclusions: The attention-added deep-learning model was successfully implemented for automatic needle digitization in HDR brachytherapy, with clinically acceptable geometric and dosimetric accuracy. Compared with conventional deep-learning neural networks, attention-gated deep learning may have superior performance and great clinical potential.

Efficacy and safety of durvalumab + chemotherapy vs. atezolizumab + chemotherapy in the treatment of small­cell lung cancer: a retrospective comparative cohort study.

Background: Durvalumab and atezolizumab have recently been approved in extensive small cell lung cancer (SCLC) with moderate median overall survival (OS) improvements. However, only limited data exist regarding the impact of immunotherapy in real-world SCLC patients. This study sought to assess the efficacy and safety of atezolizumab plus chemotherapy and durvalumab plus chemotherapy in the treatment of SCLC in a real-world setting. Methods: A retrospective cohort study of all patients treated for SCLC with chemotherapy with PD-L1 inhibitor, at 3 centers in China between February 1, 2020 and April 30, 2022. Patient characteristics, adverse-events and survival analyses were conducted. Results: A total of 143 patients were enrolled in this study, 100 were treated with durvalumab and the remainder with atezolizumab. The baseline characteristics of the two groups were fundamentally balanced before using PD-L1 inhibitors (P>0.05). The median OS (mOS) of the patients who received durvalumab or atezolizumab as the first-line treatment were 22.0 and 10.0 months, respectively (P=0.03). Survival analysis of patients with brain metastasis (BM) revealed that the median progression-free survival (mPFS) of patients without BM treated with durvalumab plus chemotherapy (5.5 months) was longer than that of those with BM (4.0 months) (P=0.03). In contrast, in the atezolizumab plus chemotherapy regimen, BM did not affect survival. In addition, the addition of radiotherapy to treatment with PD-L1 inhibitors in combination with chemotherapy has a tendency to improve long-term survival. As for safety analysis, there was no significant difference in the incidence of immune-related adverse events (IRAEs) during PD-L1 inhibitor therapy between the 2 groups (P>0.05). And during treatment with immunochemotherapy, radiotherapy was not associated with the development of IRAE (P=0.42) but increased the risk of immune-related pneumonitis (P=0.026). Conclusions: The implication of this study for clinical practice is a preference for durvalumab in first-line immunotherapy for SCLC. In addition, appropriate radiotherapy during treatment with PD-L1 inhibitors in combination with chemotherapy may prolong long-term survival, but the occurrence of immune-related pneumonitis should be vigilant. Data from this study are limited and the baseline characteristics of the two populations still need to be more finely classified.

Advantages and robustness of partial VMAT with prone position for neoadjuvant rectal cancer evaluated by CBCT-based offline adaptive radiotherapy.

BACKGROUND AND PURPOSE: This study aims to explore the advantages and robustness of the partial arc combined with prone position planning technique for radiotherapy in rectal cancer patients. Adaptive radiotherapy is recalculated and accumulated on the synthesis CT (sCT) obtained by deformable image registration between planning CT and cone beam CT (CBCT). Full and partial volume modulation arc therapy (VMAT) with the prone position on gastrointestinal and urogenital toxicity, based on the probability of normal tissue complications (NTCP) model in rectal cancer patients were evaluated. MATERIALS AND METHODS: Thirty-one patients were studied retrospectively. The contours of different structures were outlined in 155 CBCT images. First, full VMAT (F-VMAT) and partial VMAT (P-VMAT) planning techniques were designed and calculated using the same optimization constraints for each individual patient. The Acuros XB (AXB) algorithm was used in order to generate more realistic dose distributions and DVH, considering the air cavities. Second, the Velocity 4.0 software was used to fuse the planning CT and CBCT to obtain the sCT. Then, the AXB algorithm was used in the Eclipse 15.6 software to conduct re-calculation based on the sCT to obtain the corresponding dose. Furthermore, the NTCP model was used to analyze its radiobiological side effects on the bladder and the bowel bag. RESULTS: With a CTV coverage of 98%, when compared with F-VMAT, P-VMAT with the prone position technique can effectively reduce the mean dose of the bladder and the bowel bag. The NTCP model showed that the P-VMAT combined with the prone planning technique resulted in a significantly lower complication probability of the bladder (1.88 ± 2.08 vs 1.62 ± 1.41, P = 0.041) and the bowel bag (1.28 ± 1.70 vs 0.95 ± 1.52, P < 0.001) than the F-VMAT. In terms of robustness, P-VMAT was more robust than F-VMAT, considering that less dose and NTCP variation was observed in the CTV, bladder and bowel bag. CONCLUSION: This study analyzed the advantages and robustness of the P-VMAT in the prone position from three aspects, based on the sCT fused by CBCT. Whether it is in regards to dosimetry, radiobiological effects or robustness, P-VMAT in the prone position has shown comparative advantages.

X-Ray Activatable Au/Ag Nanorods for Tumor Radioimmunotherapy Sensitization and Monitoring of the Therapeutic Response Using NIR-II Photoacoustic Imaging.

Radioimmunotherapy (RIT) is an advanced physical therapy used to kill primary cancer cells and inhibit the growth of distant metastatic cancer cells. However, challenges remain because RIT generally has low efficacy and serious side effects, and its effects are difficult to monitor in vivo. This work reports that Au/Ag nanorods (NRs) enhance the effectiveness of RIT against cancer while allowing the therapeutic response to be monitored using activatable photoacoustic (PA) imaging in the second near-infrared region (NIR-II, 1000-1700 nm). The Au/Ag NRs can be etched using high-energy X-ray to release silver ions (Ag+ ), which promotes dendritic cell (DC) maturation, enhances T-cell activation and infiltration, and effectively inhibits primary and distant metastatic tumor growth. The survival time of metastatic tumor-bearing mice treated with Au/Ag NR-enhanced RIT is 39 days compared with 23 days in the PBS control group. Furthermore, the surface plasmon absorption intensity at 1040 nm increases fourfold after Ag+ are released from the Au/Ag NRs, allowing X-ray activatable NIR-II PA imaging to monitor the RIT response with a high signal-to-background ratio of 24.4. Au/Ag NR-based RIT has minimal side effects and shows great promise for precise cancer RIT.

Real-world data on efficacy and safety of osimertinib in non-small cell lung cancer patients with EGFR T790M mutation detected by first and repeat rebiopsy.

BACKGROUND: Osimertinib could effectively target epidermal growth factor receptor (EGFR) T790M resistance mutations in non-small cell lung cancer (NSCLC), indicating that rebiopsy may be particularly important. However, the clinical benefit of repeat rebiopsy in T790M-negative patients with NSCLC detected by the first rebiopsy is still unclear, and data on the efficacy and safety of osimertinib in patients with NSCLC who are T790M-positive patients on a repeat rebiopsy remain rare. METHODS: We retrospectively collected the clinical data of advanced NSCLC patients with common EGFR mutation who were treated with 1/2-generation (1/2G) EGFR-tyrosine kinase inhibitors (TKIs) in first-line therapy in our center from January 2018 to December 2020. The detection rate of T790M by first and repeat rebiopsy was recorded, and we also analyzed the efficacy and safety of osimertinib for T790M-positive patients. RESULTS: Among 190 common EGFR-mutant patients who received 1/2G EGFR-TKIs with advanced NSCLC in the first-line treatment, 141 patients developed progressive disease. In total, 110 of 141 accepted the first rebiopsy, with a T790M prevalence of 50.9% (56/110). In total, 43 T790M-positive patients who received osimertinib were included in first rebiopsy group. Of 54 T790M-negative patients detected by the first rebiopsy, 28 underwent repeated rebiopsy in subsequent clinical treatment, and 10 (35.7%) T790M-positive cases were confirmed. In total, eight T790M-positive patients treated with osimertinib were included in repeat rebiopsy group. Overall, 66 (60%) of 110 patients acquired a T790M mutation. In patients with the T790M mutation discovered by the first and repeat rebiopsy, osimertinib resulted in median progression-free survival of 7 (95% confidence interval [CI]: 5.3-8.7) and 6 (95% CI: 4.7-7.3) months, respectively (p = .656). The median overall survival since osimertinib initiation for T790M-positive patients at first rebiopsy was 20 (95% CI: 15.1-24.9) months and 19 (95% CI: 16.9-21.1) months, for those at repeated rebiopsy (p = .888). Adverse events of grade 3 or higher were similar in the two groups (25.6% vs. 12.5%, p = .616). There was no treatment-related death in the two groups. CONCLUSIONS: Repeat rebiopsy can increase the T790M mutation positivity rate. Osimertinib showed similar efficacy and safety in T790M-positive patients whether detected by the first or repeat rebiopsy.

Evaluation of Hybrid VMAT Advantages and Robustness Considering Setup Errors Using Surface Guided Dose Accumulation for Internal Lymph Mammary Nodes Irradiation of Postmastectomy Radiotherapy.

Objectives: Setup error is a key factor affecting postmastectomy radiotherapy (PMRT) and irradiation of the internal mammary lymph nodes is the most investigated aspect for PMRT patients. In this study, we evaluated the robustness, radiobiological, and dosimetric benefits of the hybrid volumetric modulated arc therapy (H-VMAT) planning technique based on the setup error in dose accumulation using a surface-guided system for radiation therapy. Methods: We retrospectively selected 32 patients treated by a radiation oncologist and evaluated the clinical target volume (CTV), including internal lymph node irradiation (IMNIs), and considered the planning target volume (PTV) margin to be 5 mm. Three different planning techniques were evaluated: tangential-VMAT (T-VMAT), intensity-modulated radiation therapy (IMRT), and H-VMAT. The interfraction and intrafraction setup errors were analyzed in each field and the accumulated dose was evaluated as the patients underwent daily surface-guided monitoring. These parameters were included while evaluating CTV coverage, the dose required for the left anterior descending artery (LAD) and the left ventricle (LV), the normal tissue complication probability (NTCP) for the heart and lungs, and the second cancer complication probability (SCCP) for contralateral breast (CB). Results: When the setup error was accounted for dose accumulation, T-VMAT (95.51%) and H-VMAT (95.48%) had a higher CTV coverage than IMRT (91.25%). In the NTCP for the heart, H-VMAT (0.04%) was higher than T-VMAT (0.01%) and lower than IMRT (0.2%). However, the SCCP (1.05%) of CB using H-VMAT was lower than that using T-VMAT (2%) as well as delivery efficiency. And T-VMAT (3.72) and IMRT (10.5).had higher plan complexity than H-VMAT (3.71). Conclusions: In this study, based on the dose accumulation of setup error for patients with left-sided PMRT with IMNI, we found that the H-VMAT technique was superior for achieving an optimum balance between target coverage, OAR dose, complication probability, plan robustness, and complexity.

Advances of Patient-Derived Organoids in Personalized Radiotherapy.

Patient-derived organoids (PDO), based on the advanced three-dimensional (3D) culture technology, can provide more relevant physiological and pathological cancer models, which is especially beneficial for developing and optimizing cancer therapeutic strategies. Radiotherapy (RT) is a cornerstone of curative and palliative cancer treatment, which can be performed alone or integrated with surgery, chemotherapy, immunotherapy, or targeted therapy in clinical care. Among all cancer therapies, RT has great local control, safety and effectiveness, and is also cost-effective per life-year gained for patients. It has been reported that combing RT with chemotherapy or immunotherapy or radiosensitizer drugs may enhance treatment efficacy at faster rates and lower cost. However, very few FDA-approved combinations of RT with drugs or radiosensitizers exist due to the lack of accurate and relevant preclinical models. Meanwhile, radiation dose escalation may increase treatment efficacy and induce more toxicity of normal tissue as well, which has been studied by conducting various clinical trials, very expensive and time-consuming, often burdensome on patients and sometimes with controversial results. The surged PDO technology may help with the preclinical test of RT combination and radiation dose escalation to promote precision radiation oncology, where PDO can recapitulate individual patient' tumor heterogeneity, retain characteristics of the original tumor, and predict treatment response. This review aims to introduce recent advances in the PDO technology and personalized radiotherapy, highlight the strengths and weaknesses of the PDO cancer models, and finally examine the existing RT-related PDO trials or applications to harness personalized and precision radiotherapy.

Artificial intelligence-based bone-enhanced magnetic resonance image-a computed tomography/magnetic resonance image composite image modality in nasopharyngeal carcinoma radiotherapy.

BACKGROUND: In the radiotherapy of nasopharyngeal carcinoma (NPC), magnetic resonance imaging (MRI) is widely used to delineate tumor area more accurately. While MRI offers the higher soft tissue contrast, patient positioning and couch correction based on bony image fusion of computed tomography (CT) is also necessary. There is thus an urgent need to obtain a high image contrast between bone and soft tissue to facilitate target delineation and patient positioning for NPC radiotherapy. In this paper, our aim is to develop a novel image conversion between the CT and MRI modalities to obtain clear bone and soft tissue images simultaneously, here called bone-enhanced MRI (BeMRI). METHODS: Thirty-five patients were retrospectively selected for this study. All patients underwent clinical CT simulation and 1.5T MRI within the same week in Shenzhen Second People's Hospital. To synthesize BeMRI, two deep learning networks, U-Net and CycleGAN, were constructed to transform MRI to synthetic CT (sCT) images. Each network used 28 patients' images as the training set, while the remaining 7 patients were used as the test set (~1/5 of all datasets). The bone structure from the sCT was then extracted by the threshold-based method and embedded in the corresponding part of the MRI image to generate the BeMRI image. To evaluate the performance of these networks, the following metrics were applied: mean absolute error (MAE), structural similarity index (SSIM), and peak signal-to-noise ratio (PSNR). RESULTS: In our experiments, both deep learning models achieved good performance and were able to effectively extract bone structure from MRI. Specifically, the supervised U-Net model achieved the best results with the lowest overall average MAE of 125.55 (P<0.05) and produced the highest SSIM of 0.89 and PSNR of 23.84. These results indicate that BeMRI can display bone structure in higher contrast than conventional MRI. CONCLUSIONS: A new image modality BeMRI, which is a composite image of CT and MRI, was proposed. With high image contrast of both bone structure and soft tissues, BeMRI will facilitate tumor localization and patient positioning and eliminate the need to frequently check between separate MRI and CT images during NPC radiotherapy.

Metabolizable Near-Infrared-II Nanoprobes for Dynamic Imaging of Deep-Seated Tumor-Associated Macrophages in Pancreatic Cancer.

Tumor-associated macrophages (TAMs) play a crucial part in cancer evolution. Dynamic imaging of TAMs is of great significance for treatment outcome evaluation and precision tumor therapy. Currently, most fluorescence nanoprobes tend to accumulate in the liver and are difficult to metabolize, which leads to strong background signals and inadequate imaging quality of TAMs nearby the liver such as pancreatic cancer. Herein, we aim to develop metabolizable dextran-indocyanine green (DN-ICG) nanoprobes in the second near-infrared window (NIR-II, 1 000-1 700 nm) for dynamic imaging of TAMs in pancreatic cancer. Compared to free ICG, the NIR-II fluorescence intensity of DN-ICG nanoprobes increased by 279% with significantly improved stability. We demonstrated that DN-ICG nanoprobes could specifically target TAMs through the interaction of dextran with specific ICAM-3-grabbing nonintegrin related 1 (SIGN-R1), which were highly expressed in TAMs. Subsequently, DN-ICG nanoprobes gradually metabolized in the liver yet remained in pancreatic tumor stroma in mouse models, achieving a high signal-to-background ratio (SBR = 7) in deep tissue (∼0.5 cm) NIR-II imaging of TAMs. Moreover, DN-ICG nanoprobes could detect dynamic changes of TAMs induced by low-dose radiotherapy and zoledronic acid. Therefore, the highly biocompatible and biodegradable DN-ICG nanoprobes harbor great potential for precision therapy in pancreatic cancer.

DVH Prediction for VMAT in NPC with GRU-RNN: An Improved Method by Considering Biological Effects.

PURPOSE: A recurrent neural network (RNN) and its variants such as gated recurrent unit-based RNN (GRU-RNN) were found to be very suitable for dose-volume histogram (DVH) prediction in our previously published work. Using the dosimetric information generated by nonmodulated beams of different orientations, the GRU-RNN model was capable of accurate DVH prediction for nasopharyngeal carcinoma (NPC) treatment planning. On the basis of our previous work, we proposed an improved approach and aimed to further improve the DVH prediction accuracy as well as study the feasibility of applying the proposed method to relatively small-size patient data. METHODS: Eighty NPC volumetric modulated arc therapy (VMAT) plans with local IRB's approval in recent two years were retrospectively and randomly selected in this study. All these original plans were created using the Eclipse treatment planning system (V13.5, Varian Medical Systems, USA) with ≥95% of PGTVnx receiving the prescribed doses of 70 Gy, ≥95% of PGTVnd receiving 66 Gy, and ≥95% of PTV receiving 60 Gy. Among them, fifty plans were used to train the DVH prediction model, and the remaining were used for testing. On the basis of our previously published work, we simplified the 3-layer GRU-RNN model to a single-layer model and further trained every organ at risk (OAR) separately with an OAR-specific equivalent uniform dose- (EUD-) based loss function. RESULTS: The results of linear least squares regression obtained by the new proposed method showed the excellent agreements between the predictions and the original plans with the correlation coefficient r = 0.976 and 0.968 for EUD results and maximum dose results, respectively, and the coefficient r of our previously published method was 0.957 and 0.946, respectively. The Wilcoxon signed-rank test results between the proposed and the previous work showed that the proposed method could significantly improve the EUD prediction accuracy for the brainstem, spinal cord, and temporal lobes with a p value < 0.01. CONCLUSIONS: The accuracy of DVH prediction achieved in different OARs showed the great improvements compared to the previous works, and more importantly, the effectiveness and robustness showed by the simplified GRU-RNN trained from relatively small-size DVH samples, fully demonstrated the feasibility of applying the proposed method to small-size patient data. Excellent agreements in both EUD results and maximum dose results between the predictions and original plans indicated the application prospect in a physically and biologically related (or a mixture of both) model for treatment planning.

Dose calculation in proton therapy using a discovery cross-domain generative adversarial network (DiscoGAN).

PURPOSE: Accurate dose calculation is a critical step in proton therapy. A novel machine learning-based approach was proposed to achieve comparable accuracy to that of Monte Carlo simulation while reducing the computational time. METHODS: Computed tomography-based patient phantoms were used and three treatment sites were selected (thorax, head, and abdomen), comprising different beam pathways and beam energies. The training data were generated using Monte Carlo simulations. A discovery cross-domain generative adversarial network (DiscoGAN) was developed to perform the mapping between two domains: stopping power and dose, with HU values from CT images incorporated as auxiliary features. The accuracy of dose calculation was quantitatively evaluated in terms of mean relative error (MRE) and mean absolute error (MAE). The relationship between the DiscoGAN performance and other factors such as absolute dose, beam energy and location within the beam cross-section (center and off-center lines) was examined. RESULTS: The DiscoGAN model is found to be effective in dose calculation. For the abdominal case, the MRE is found to 1.47% (mean), 3.30% (maximum) and 0.67% (minimum). For the thoracic case, the MRE is found to ~2.43% (mean), 4.80% (maximum) and 0.71% (minimum). For the head case, the MRE is found to ~2.83% (mean), 4.84% (maximum) and 1.01% (minimum). Comparable accuracy is found in the independent validation dataset (different CT images), achieving a mean MRE of ~1.65% (thorax), 4.02% (head) and 1.64% (abdomen). For the energy span between 80 and 130 MeV, no strong dependency of accuracy on beam energy is found. The results imply that no systematic deviation, either over-dose or under-dose, occurs between the predicted dose and raw dose. CONCLUSION: The DiscoGAN framework demonstrates great potential as a tool for dose calculation in proton therapy, achieving comparable accuracy yet being more efficient relative to Monte Carlo simulation. Its comparison with the pencil beam algorithm (PBA) will be the next step of our research. If successful, our proposed approach is expected to find its use in more advanced applications such as inverse planning and adaptive proton therapy.

Proton FLASH: passive scattering or pencil beam scanning?

This study focused on a direct comparison of dose delivery efficiency between two proton FLASH delivery modes: passive scattering and pencil beam scanning (PBS). Monte-Carlo simulation of the beamline was performed using the Geant4 package. Two proton energies (63 and 230 MeV) were selected, targeting for shallow and deep-seated tumors, respectively. Two irradiation field sizes were selected: 13 × 13 mm2 and 50 × 50 mm2. For each delivery mode, two cases were investigated: shoot-through and Bragg peak, yielding a total of 4 delivery scenarios. For the passive scattering mode, the impact on dose rate by multiple components along the beamline were investigated, including ridge-filter, scatterer, range shifter and collimator. A quantitative comparison among four scenarios was made in terms of field size, dose, dose rate and treatment plan quality (dose volume histogram). For the 230 MeV case, the dose rate (for 1 nA current) is 0.05 Gy s-1 (passive with Bragg peak, field size: 50 × 50 mm2) and 2.6 Gy s-1 (PBS with shoot-through). Dose rate comparison is made between passive scattering and PBS as the delivery changes from spot-layer to shoot-through. In conclusion, the study successfully established a benchmark reference for dose rate performance for different scenarios, taking into account components along the beamline, field size and beam current. The results allow us to predict and compare the required beam current to yield a dose rate sufficiently high, above the threshold of the FLASH effect.

Mitigation of the Interplay Effects of Combining 4D Robust With Layer Repainting Techniques in Proton-Based SBRT for Patients With Early-Stage Non-small Cell Lung Cancer.

OBJECTIVE: The objective of this study was to evaluate the interplay effects in proton-based stereotactic body radiotherapy (SBRT) using 4D robust optimization combined with iso-energy layer repainting techniques for non-small cell lung cancer (NSCLC). MATERIALS AND METHODS: Twelve patients with early-stage NSCLC who underwent 4DCT were retrospectively selected. A robust CTV-based 4D plan was generated for each based on commercial Treatment planning system (TPS), considering patient setup errors, range uncertainties, and organ motion. The 4D static dose (4DSD) and 4D dynamic dose (4DDD) were calculated using a hybrid deformable algorithm and simulated proton delivery system. An index Δ I M R ( % ) was developed to quantitatively evaluate the interplay effects. The interplay effects of the 4D robust plan and multiple iso-energy layers (3, 4, 5, 6, and 7) of the robust repainting 4D plan were calculated based on Δ I M R ( % ) to select the optimal times for layer repainting. RESULTS: Due to the interplay effects, the mean target values D2 and D5 increased to 1.28 and 1.01%, and the target values D98 and D95 decreased to 2.01 and 1.77%, respectively, for the 4D Robust SBRT plan. After multiple iso-energy repainting times, the interplay effects of the target values D98 and D95 tended to migrate, from 2.01 to 0.92% in target value D98 and from 1.77 to 0.89% in target value D95, respectively. Moreover, a positive linear correlation was observed between the optimal interplay effect mitigation and target range of motion. CONCLUSION: In proton-based 4D Robust SBRT, the interplay effects degraded the target dose distribution but were mitigated using iso-energy layer repainting techniques. However, there was no significant correlation between the number of repainting layers and improvements in the dose distributions. The optimal layer repainting times based on the interplay effect index were ascertained according to the patient characteristics.

Feasibility study of patient-specific dose verification in proton therapy utilizing positron emission tomography (PET) and generative adversarial network (GAN).

PURPOSE: Online dose verification based on proton-induced positron emitters is a promising strategy for quality assurance in proton therapy. Because of the nonlinear correlation between dose and the activity distributions, a machine learning-based approach was developed to establish their relationship. METHODS: Simulations were carried out using a pencil beam scanning system and a computed tomography (CT) image-based phantom. A DiscoGAN model was developed to perform dose verification for both central and off-center lines. Besides the activity as input, HU information from CT images and stopping power (SP) prior were incorporated as auxiliary features for the model. The performance was quantitatively studied in terms of mean absolute error (MAE) and mean relative error (MRE), under different signal-to-noise ratios (SNRs). In addition to a dataset comprising monoenergetic beams, two additional datasets were generated to evaluate the model's generalization capability: five reconstructed PET images based on an in-beam PET system and a dataset comprising spread-out Bragg peaks (SOBPs). RESULTS: The feasibility of dose verification was successfully demonstrated for all three datasets. For the monoenergetic case (i.e., raw activity of positron emitters), the MRE is found to be <1% for the central lines and 5% for the off-center lines, respectively. The range uncertainty is found to be less than 1 mm. The prediction based on five PET images, which take into account the detection of 511-keV photons and image reconstruction, yields slightly inferior performance. For the SOBP case, the MRE of the center lines is found to be <3% and the range uncertainty is <1 mm. The inclusion of anatomical information (HU and SP) improves both accuracy and generalization of the DiscoGAN model. CONCLUSION: The combination of proton-induced positron emitters, in-beam PET, and machine learning may become a useful tool allowing for patient-specific online dose verification in proton therapy.

Feasibility study of range verification based on proton-induced acoustic signals and recurrent neural network.

Range verification in proton therapy is a critical quality assurance task. We studied the feasibility of online range verification based on proton-induced acoustic signals, using a bidirectional long-short-term-memory recurrent neural network and various signal processing techniques. Dose distribution of 1D pencil proton beams inside a CT image-based phantom was analytically calculated. The propagation of acoustic signal inside the phantom was modeled using the k-Wave toolbox. For signal processing, five methods were investigated: down-sampling (DS), DS + HT (Hilbert transform), Wavelet decomposition (Wavedec db1, db4 and db20). The performances were quantitatively evaluated in terms of mean absolute error, mean relative error (MRE) and the Bragg peak localization error ([Formula: see text]). In addition, the study analyzed the impact of noise levels, the number of sensors, as well as the location of sensors. For the noiseless case (32 sensors), the Wavedec db1 method demonstrates the best performance: [Formula: see text] is less than one pixel and the dose accuracy over the region adjacent to the Bragg peak (MRE50) is ∼3.04%. With the presence of noise, the Wavedec db1 method demonstrates the best noise immunity, achieving [Formula: see text] less than 1 mm and an MRE50 of ∼12%. The proposed machine learning framework may become a useful tool allowing for online range verification in proton therapy.

Robust Optimization of SBRT Planning for Patients With Early Stage Non-Small Cell Lung Cancer.

PURPOSE: Setup uncertainty is a known challenge for stereotactic body radiotherapy planning. Using the internal target volume-based robust optimization was proposed as a more accurate way than the conventional planning target volume-based optimization when considering the robustness criteria. In this study, we aim to investigate the feasibility of internal target volume-based robust optimization in stereotactic body radiotherapy planning using 4-dimensional computed tomography and develop a novel dose-volume histogram band width metric to quantitatively evaluate robustness. METHOD AND MATERIALS: A total of 50 patients with early stage non-small cell lung cancer, who underwent stereotactic body radiotherapy, were retrospectively selected. Each of the 50 patients had 2 stereotactic body radiotherapy plans: one with the conventional planning target volume-based optimization and the other with patient-specific robustly optimized internal target volume and with a uniform 5 mm setup error. These were compared with the planning target volume-based optimization method based on both plan quality and robustness. The quality was evaluated using dosimetric parameters and radiobiology parameters, such as high-dose spillage (V90%RX, conformity index), intermediate-dose spillage (dose falloff products), low-dose spillage (normal tissue: V50%RX), and lung tissue complication probability. The robustness was evaluated under a uniform 3 to 5 mm setup errors with a novel proposed metric: dose-volume histogram band width. RESULTS: When compared with planning target volume-based optimization plans, the internal target volume-based robust optimization plans have better conformity of internal target volume coverage (conformity index: 1.17 vs 1.27, P < .001), intermediate-dose spillage (dose falloff product: 129 vs 167, P < .001), low-dose spillage in normal tissue (V50%RX: 0.8% vs 1.5%, P < .05), and lower risk of radiation pneumonitis (lung tissue complication probability: 4.2% vs 5.5%, P < .001). For the robustness, dose-volume histogram band width analysis shows that the average values in internal target volume, D95%, D98%, and D99%, of internal target volume-based robust optimization are smaller than that of planning target volume-based optimization (unit cGy) under 3-, 4-, and 5-mm setup uncertainties (3-mm setup uncertainty: 42 vs 73 cGy; 4-mm setup uncertainty: 88 vs 176 cGy; 5-mm setup uncertainty: 229 vs 490 cGy), which might indicate that internal target volume-based robust optimization harbored a greater robustness regardless of the setup errors. CONCLUSIONS: Internal target volume-based robust optimization may have clinical potential in offering better plan quality in both target and organs at risk and lower risk of radiation pneumonitis. In addition, the proposed internal target volume-based robust optimization may demonstrate robustness regardless of different setup uncertainties in the stereotactic body radiotherapy planning. REGISTRATION: Retrospective study with local ethics committee approval.

Synthetic CT Generation Based on T2 Weighted MRI of Nasopharyngeal Carcinoma (NPC) Using a Deep Convolutional Neural Network (DCNN).

Purpose: There is an emerging interest of applying magnetic resonance imaging (MRI) to radiotherapy (RT) due to its superior soft tissue contrast for accurate target delineation as well as functional information for evaluating treatment response. MRI-based RT planning has great potential to enable dose escalation to tumors while reducing toxicities to surrounding normal tissues in RT treatments of nasopharyngeal carcinoma (NPC). Our study aims to generate synthetic CT from T2-weighted MRI using a deep learning algorithm. Methods: Thirty-three NPC patients were retrospectively selected for this study with local IRB's approval. All patients underwent clinical CT simulation and 1.5T MRI within the same week in our hospital. Prior to CT/MRI image registration, we had to normalize two different modalities to a similar intensity scale using the histogram matching method. Then CT and T2 weighted MRI were rigidly and deformably registered using intensity-based registration toolbox elastix (version 4.9). A U-net deep learning algorithm with 23 convolutional layers was developed to generate synthetic CT (sCT) using 23 NPC patients' images as the training set. The rest 10 NPC patients were used as the test set (~1/3 of all datasets). Mean absolute error (MAE) and mean error (ME) were calculated to evaluate HU differences between true CT and sCT in bone, soft tissue and overall region. Results: The proposed U-net algorithm was able to create sCT based on T2-weighted MRI in NPC patients, which took 7 s per patient on average. Compared to true CT, MAE of sCT in all tested patients was 97 ± 13 Hounsfield Unit (HU) in soft tissue, 131 ± 24 HU in overall region, and 357 ± 44 HU in bone, respectively. ME was -48 ± 10 HU in soft tissue, -6 ± 13 HU in overall region, and 247 ± 44 HU in bone, respectively. The majority soft tissue and bone region was reconstructed accurately except the interface between soft tissue and bone and some delicate structures in nasal cavity, where the inaccuracy was induced by imperfect deformable registration. One patient example was shown with almost no difference in dose distribution using true CT vs. sCT in the PTV regions in the sinus area with fine bone structures. Conclusion: Our study indicates that it is feasible to generate high quality sCT images based on T2-weighted MRI using the deep learning algorithm in patients with nasopharyngeal carcinoma, which may have great clinical potential for MRI-only treatment planning in the future.

The value of the computer-aided diagnosis system for thyroid lesions based on computed tomography images.

BACKGROUND: Thyroid nodules are commonly found at palpation amounting to 4-7% of the asymptomatic population and 50% of the cases are found at autopsy. Only a small proportion of thyroid nodules are malignant. The major challenge is the differential diagnosis of benign or malignant thyroid nodules, so we aim to develop the computer-assisted diagnostic method based on computed tomography (CT) images for thyroid lesions. METHODS: In this study, we retrospectively collected 52 benign and 46 malignant thyroid nodules from 90 patients in CT examinations, together with the pathologist findings and radiology diagnosis. The first-order statistic and gray-level co-occurrence matrix features were extracted from thyroid computed tomography images. These texture features were used to assess the malignancy risk of the thyroid nodules. Several classification algorithms, including support vector machine, linear discriminant analysis, random forest, and bootstrap aggregating, were applied in the prediction. Leave-one-out cross-validation was used to evaluate the performance of thyroid cancer recognition. RESULTS: In thyroid cancer identification based on a computed tomography image, we found the system using 17 texture features and support vector machine performed well. The accuracy, area under the receiver operating characteristic curve, sensitivity, specificity, positive predictive value, and negative predictive value, were 0.8673, 0.9105, 0.9130, 0.8269, 0.8235 and 0.9146, respectively. CONCLUSIONS: The proposed computer-aided diagnosis system provides a good assessment of the malignancy-risk of the thyroid nodules, which may help radiologists to improve the accuracy and efficiency of thyroid diagnosis.

Simulation of dosimetry impact of 4DCT uncertainty in 4D dose calculation for lung SBRT.

BACKGROUND: Due to the heterogeneity of patient's individual respiratory motion pattern in lung stereotactic body radiotherapy (SBRT), treatment planning dose assessment using a traditional four-dimensional computed tomography (4DCT_traditional) images based on a uniform breathing curve may not represent the true treatment dose delivered to the patient. The purpose of this study was to evaluate the accumulated dose discrepancy between based on the 4DCT_traditional and true 4DCT (4DCT_true) that incorporated with the patient's real entire breathing motion. The study also explored a novel 4D robust planning strategy to compensate for such heterogeneity respiratory motion uncertainties. METHODS: Simulated and measured patient specific breathing curves were used to generate 4D targets motion CT images. Volumetric-modulated arc therapy (VMAT) was planned using two arcs. Accumulated dose was obtained by recalculating the plan dose on each individual phase image and then deformed the dose from each phase image to the reference image. The "4 D dose" (D4D) and "true dose" (Dtrue) were the accumulated dose based on the 4DCT_traditional and 4DCT_true respectively. The average worse case dose discrepancy ([Formula: see text]) between D4D and Dtrue in all treatment fraction was calculated to evaluate dosimetric /planning parameters and correlate them with the heterogeneity of respiratory-induced motion patterns. A novel 4D robust optimization strategy for VMAT (4D Ro-VMAT) based on the probability density function(pdf) of breathing curve was proposed to improve the target coverage in the presence of heterogeneity respiratory motion. The data were assessed with a paired t-tests. RESULTS: With increasing breathing amplitude from 5 to 20 mm, target [Formula: see text], [Formula: see text] increased from 1.59,1.39 to 10.15%,8.66% respectively. When the standard deviation of breathing amplitude increased from 15 to 35% of the mean amplitude, [Formula: see text], [Formula: see text] increased from 4.06,3.48 to 10.25%,6.63% respectively. The 4D Ro-VMAT plan significantly improve the target dose compared to VMAT plan. CONCLUSION: When the breathing curve amplitude is more than 10 mm and standard deviation of amplitude is higher than 25% of mean amplitude, special care is needed to choose an appropriated dose accumulation approach to evaluate lung SBRT plan target coverage robustness. The proposed 4D Ro_VMAT strategy based on the pdf of patient specific breathing curve could effectively compensate such uncertainties.