Search

Purpose: FDG PET imaging is often recommended for the diagnosis of pulmonary nodules after indeterminate low dose CT lung cancer screening. Lowering FDG injecting is desirable for PET imaging. In this work, we aimed to investigate the performance of a deep learning framework in the automatic diagnoses of pulmonary nodules at different count levels of PET imaging., Materials and Methods: Twenty patients with 18F-FDG-avid pulmonary nodules were included and divided into independent training (60%), validation (20%), and test (20%) subsets. We trained a convolutional neural network (ResNet-50) on original DICOM images and used ImageNet pre-trained weight to fine-tune the model. Simulated low-dose PET images at the 9 count levels (20 × 10 6 , 15 × 10 6 , 10 × 10 6 , 7.5 × 10 6 , 5 × 10 6 , 2 × 10 6 , 1 × 10 6 , 0.5 × 10 6 , and 0.25 × 10 6 counts) were obtained by randomly discarding events in the PET list mode data for each subject. For the test dataset with 4 patients at the 9 count levels, 3,307 and 3,384 image patches were produced for lesion and background, respectively. The receiver-operator characteristic (ROC) curve of the proposed model under the different count levels with different lesion size groups were assessed and the areas under the ROC curve (AUC) were compared., Results: The AUC values were >0.98 for all count levels except for 0.5 and 0.25 million true counts (0.975 (CL 95%, 0.953-0.992) and 0.963 (CL 95%, 0.941-0.982), respectively). The AUC values were 0.941(CL 95%, 0.923-0.956), 0.993(CL 95%, 0.990-0.996) and 0.998(CL 95%, 0.996-0.999) for different groups of lesion size with effective diameter (R) <10 mm, 10-20 mm, and >20 mm, respectively. The count limit for achieving high AUC (≥0.96) for lesions with size R < 10 mm and R > 10 mm were 2 million (equivalent to an effective dose of 0.08 mSv) and 0.25 million true counts (equivalent to an effective dose of 0.01 mSv), respectively., Conclusion: All of the above results suggest that the proposed deep learning based method may detect small lesions <10 mm at an effective radiation dose <0.1 mSv., Advances in Knowledge: We investigated the advantages and limitations of a fully automated lung cancer detection method based on deep learning models for data with different lesion sizes and different count levels, and gave guidance for clinical application., Competing Interests: The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest., (Copyright © 2022 Guo, Wu, Xie, Tham, Zhou and Yan.)

PET imaging involves radiotracer injections, raising concerns about the risk of radiation exposure. To minimize the potential risk, one way is to reduce the injected tracer. However, this will lead to poor image quality with conventional image reconstruction and processing. In this paper, we proposed a supervised deep learning model, CycleWGANs, to boost low-dose PET image quality. Validations were performed on a low dose dataset simulated from a real dataset with biopsy-proven primary lung cancer or suspicious radiological abnormalities. Low dose PET images were reconstructed on reduced PET raw data by randomly discarding events in the PET list mode data towards the count level of 1 million. Traditional image denoising methods (Non-Local Mean (NLM) and block-matching 3D(BM3D)) and two recently-published deep learning methods (RED-CNN and 3D-cGAN) were included for comparisons. At the count level of 1 million (true counts), the proposed model can accurately estimate full-dose PET image from low-dose input image, which is superior to the other four methods in terms of the mean and maximum standardized uptake value (SUV mean and SUV max ) bias for lesions and normal tissues. The bias of SUV (SUV mean , SUV max ) for lesions and normal tissues are (-2.06±3.50%,-0.84±6.94%) and (-0.45±5.59%, N/A) in the estimated PET images, respectively. However, the RED-CNN achieved the best score in traditional metrics, such as structure similarity (SSIM), peak signal to noise ratio (PSNR) and normalized root mean square error (NRMSE). Correlation and profile analyses have successfully explained this phenomenon and further suggested that our method could effectively preserve edge and also SUV values than RED-CNN, 3D-cGAN and NLM with a slightly higher noise., Competing Interests: Declaration of Competing Interest We declare that we have no financial and personal relationships with other people or organizations that can inappropriately influence our work, there is no professional or other personal interest of any nature or kind in any product, service and/or company that could be construed as influencing the position presented in, or the review of, the manuscript entitled., (Copyright © 2020. Published by Elsevier B.V.)

Background: 18 F-FDG PET based radiomics is promising for precision oncology imaging. This work aims to explore quantitative accuracies of radiomic features (RFs) for low-dose 18 F-FDG PET imaging., Methods: Twenty lung cancer patients were prospectively enrolled and underwent 18 F-FDG PET/CT scans. Low-dose PET situations (true counts: 20×10 6 , 15×10 6 , 10×10 6 , 7.5×10 6 , 5×10 6 , 2×10 6 , 1×10 6 , 0.5×10 6 , 0.25×10 6 ) were simulated by randomly discarding counts from the acquired list-mode data. Each PET image was created using the scanner default reconstruction parameters. Each lesion volume of interest (VOI) was obtained via an adaptive contouring method with a threshold of 50% peak standardized uptake value (SUVpeak) in the PET images with full count data and VOIs were copied to the PET images at the reduced count level. Conventional SUV measures, features calculated from first-order statistics (FOS) and texture features (TFs) were calculated. Texture based RF include features calculated from gray-level co-occurrence matrix (GLCM), gray-level run length matrix (GLRLM), gray-level size zone matrix (GLSZM), neighboring gray-level dependence matrix (NGLDM) and neighbor gray-tone difference matrix (NGTDM). Bias percentage (BP) at different count levels for each RF was calculated., Results: Fifty-seven lesions with a volume greater than 1.5 cm 3 were found (mean volume, 25.7 cm 3 , volume range, 1.5-245.4 cm 3 ). In comparison with normal total counts, mean SUV (SUVmean) in the lesions, normal lungs and livers, Entropy and sum entropy from GLCM, busyness from NGTDM and run-length non-uniformity from GLRLM were the most robust features, with a BP of 5% at the count level of 1×10 6 (equivalent to an effective dose of 0.04 mSv) RF including cluster shade from GLCM, long-run low grey-level emphasis, high grey-level run emphasis and short-run low grey-level emphasis from GLRM exhibited the worst performance with 50% of bias with 20×10 6 counts (equivalent to an effective dose of 0.8 mSv)., Conclusions: In terms of the lesions included in this study, SUVmean, entropy and sum entropy from GLCM, busyness from NGTDM and run-length non-uniformity from GLRLM were the least sensitive features to lowering count., Competing Interests: Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at http://dx.doi.org/10.21037/tcr-20-1715). The authors have no conflicts of interest to declare., (2020 Translational Cancer Research. All rights reserved.)

The aim of this retrospective national cohort study is to assess the association between various radiation heart dosimetric parameters (RHDPs), acute myocardial infarct (AMI) and overall survival (OS) outcomes in non-small cell lung cancer (NSCLC) patients treated with post-operative thoracic radiotherapy (PORT) using contemporary radiation techniques.We identified patients with stage I to III NSCLC treated with PORT at the 2 national cancer institutions from 2007 to 2014. We linked their electronic medical records to the national AMI and death registries. Univariable Cox regression was performed to assess the association between various RHDPs, AMI, and OS.We included 43 eligible patients with median follow-up of 36.6 months. Median age was 64 years. Majority of the patients had pathological stage III disease (72%). Median prescription dose was 60Gy. Median mean heart dose (MHD) was 9.4Gy. There were no AMI events. The 5-year OS was 34%. Univariable Cox regression showed that age was significantly associated with OS (hazard ratio, 1.06; 95% confidence interval, 1.01 to 1.10; P = .008). Radiation heart doses, including MHD, volume of heart receiving at least 5, 25, 30, 40, 50Gy and dose to 30% of heart volume, were not significantly associated with OS.There is insufficient evidence to conclude that RHDPs are associated with OS for patients with NSCLC treated with PORT in this study. Studies with larger sample size and longer term follow-up are needed to assess AMI outcome.

Purpose: To evaluate the rate of discordance of epidermal growth factor receptor (EGFR) mutation between primary lung tumor and paired distant metastases in non-small-cell lung cancer (NSCLC)., Methods: We performed a meta-analysis of 17 studies (518 cases) assessing discordance rates of EGFR mutation in primary tumors and paired distant metastases. We performed subgroup analyses based on EGFR mutation status in primary tumor (mutant or wildtype), site of distant metastasis (bone, central nervous system (CNS) or lung/ pleural), methods of testing (direct sequencing or allele-specific testing) and timing of metastasis (synchronous or metachronous)., Results: The overall discordance rate in EGFR mutation was low at 10.36% (95% CI = 4.23% to 18.79%) and varied widely between studies (I2 = 83.18%). The EGFR discordance rate was statistically significantly higher in bone metastases (45.49%, 95% CI = 14.13 to 79.02) than CNS (17.26%, 95% CI = 7.64 to 29.74; P = 0.002) and lung/ pleural metastases (8.17%, 95% CI = 3.35 to 14.85; P < 0.001). Subgroup analyses did not demonstrate any significant effect modification on the discordance rates by the EGFR mutation status in primary lung tumor, methods of testing and timing of metastasis., Conclusion: The overall discordance rate in EGFR mutation between primary lung tumor and paired distant metastases in NSCLC is low, although higher discordance rates were observed in bone metastases compared with CNS and lung/pleural metastases. Future studies assessing the impact of EGFR mutation discordance on treatment outcomes are required., Competing Interests: The authors have declared that no competing interests exist.

To evaluate the response and quality of life of palliative gastric radiotherapy in patients with symptomatic locally advanced gastric cancer. Patients with bleeding, pain or obstruction and were treated with palliative gastric radiotherapy to a dose of 36 Gy in 12 daily fractions. The primary outcomes were symptom response rates. Secondary outcomes included overall survival, adverse events and proportion of patients with ≥10-point absolute improvement in the fatigue, nausea/vomiting and pain subscales in the EORTC Qualify of Life Questionnaire C30 (EORTC QLQ-C30) and dysphagia/pain subscales in the gastric specific module (STO22) at the end of RT and 1 month after the completion of radiotherapy. Fifty patients were accrued. Median survival duration was 85 days. 40/50 patients (80%) with bleeding, 2/2 (100%) patients with obstruction and 1/1 (100%) patient with pain responded to radiotherapy. Improvements fatigue, nausea/vomiting and pain subscales of the EORTC QLQ-C30 was seen in 50%, 28% and 44% of patients at the end of RT and in 63%, 31% and 50% of patients 1 month after RT. Improvements in dysphagia/pain subscales of the STO22 was seen in 42% and 28% of patients at then end of RT and 44% and 19% of patients 1 month after RT. Two patients (5%) had grade 3 anorexia and gastritis. Palliative gastric radiotherapy was effective, well tolerated and resulted in improvement in fatigue, dysphagia and pain at the end of radiotherapy and 1 month after the completion of radiotherapy in a significant proportion of patients., (© 2019 The Authors. Cancer Medicine published by John Wiley & Sons Ltd.)

Background: The survival benefit of PCI in ES-SCLC reported by a European randomized trial (RCT) in 2007 was not replicated by a Japanese RCT published in 2017. This study aimed to evaluate the uptake of PCI before and after publication of the European RCT and its association with survival in ES-SCLC., Methods: We identified eligible patients in the only two Singapore national cancer centres from 2003 to 2010. We linked their electronic medical records to the national death registry. We described the utilization of PCI in patients diagnosed from 2003 to 2006 (pre-adoption cohort) with patients diagnosed from 2007 to 2010 (post-adoption cohort). We performed univariable and multivariable Cox regression analysis to assess the association between PCI and survival., Results: We identified 224 patients with ES-SCLC with no brain metastases. Among the 71 patients who had at least stable disease after first line chemotherapy, there was an increase in the use of PCI from the period 2007 to 2010 compared with 2003 to 2006 (32% versus 10%, P = 0.044). PCI was associated with improved OS (hazard ratio 0.22, 95% CI 0.10 to 0.47, P < 0.001) compared to no PCI in the multivariable analysis., Conclusion: There was an increase in the adoption of PCI for ES-SCLC since 2007. PCI was associated with improved survival in patients who did not have mandatory MRI brain imaging prior to PCI and had stable disease or better after first line chemotherapy, suggesting that the results of the European RCT are reproducible in the real-world practice.

Purpose: To determine if the addition of induction chemotherapy (IC) to concurrent chemoradiation (CCRT) in locally advanced nasopharyngeal carcinoma (LA-NPC) can improve survival., Methods: We performed a meta-analysis of both randomized controlled trials (RCTs) and observational studies (OBS) to compare the effects of addition of IC to CCRT versus (vs) CCRT alone on overall survival (OS), progression free survival (PFS), distant metastasis-free survival (DMFS) and adverse events (AE) in LA-NPC. We searched MEDLINE for eligible studies comparing IC plus CCRT vs CCRT for LA-NPC from Jan 1996 to May 2017. We selected RCTs and OBS that included patients with non-metastatic, LA-NPC who received IC followed by CCRT or CCRT alone. Three reviewers independently assessed the abstracts for eligibility. We assessed the methodological quality of the included studies using the MERGE criteria. We performed the meta-analysis with random effects model. We used the GRADE approach to appraise the quality of evidence from RCTs. The primary outcome was OS; secondary outcomes included PFS, DMFS and AE., Results: We found six RCTs and five OBS including 2802 patients with low to moderate risk of bias in their methodological quality. There was high quality evidence from the RCTs that IC improved PFS (HR 0.69, 95% CI 0.57-0.84, P = 0.0003, I 2  = 0%) and OS (HR 0.77, 95% CI 0.60-0.98, P = 0.03, I 2  = 0%) significantly and was associated with more frequent AE. The estimates of IC effects from RCTs and OBS were similar (PFS HR 0.69 vs 0.71, interaction P (IP) = 0.92; OS HR 0.77 vs 0.58, IP = 0.27)., Conclusions: IC delays disease progression and improves survival significantly for LA-NPC treated with CCRT, and was associated with more toxicity. There were no divergent results between RCTs and OBS. IC followed by CCRT can be considered one of the standard treatment options for LA-NPC., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Temporal lobe necrosis as result of radiation for nasopharyngeal cancer (NPC) occurs up to 28% of NPC patients. The only effective mitigation is by strict adherence to temporal lobe dose tolerances during radiotherapy planning, which in turn hinges on accurate temporal lobe delineation. We aim to improve the accuracy and to standardize temporal lobe contouring for patients receiving head and neck radiotherapy for NPC in a tertiary teaching hospital in Singapore.The baseline data were obtained from 10 patients in the diagnostic phase and the effect of interventions were measured in 37 patients who underwent head and neck radiotherapy over a 6-month period.We conducted the project based on the Clinical Practice Improvement Program methodology. The baseline pooled mean percentage variation in temporal lobe contouring was 39.9% (0.8%-60.2%). There was a low level of temporal lobe contouring concordance and this provided the impetus for implementation of strategies to improve the accuracy and reproducibility of temporal lobe contouring. The interventions included supervision and training of radiation therapists and residents in temporal lobe contouring, and standardization of temporal lobe contouring with a protocol and contouring atlas.Thirty-seven patients were treated during the study period from June to November 2014. Following implementation of the first set of interventions, the pooled mean percentage variation in temporal lobe contouring decreased but was not sustained. The implementation of the second set of interventions resulted in a decrease from 39.9% (January to September 2014) to 17.3% (October to November 2014) where P = .004 using t test. Weekly variation was seen throughout the study period but the decrease was sustained after standardizing and providing a contouring atlas for temporal lobe contouring.Temporal lobe contouring can be standardized through effective implementation of a temporal lobe contouring protocol and atlas.

Objectives: The aim of this retrospective observational study is to assess the association between various radiation heart dosimetric parameters (RHDPs) and acute myocardial infarct (AMI) and overall survival (OS) outcomes in stage III non-small cell lung cancer (NSCLC) treated with definitive radiotherapy with or without chemotherapy., Materials and Methods: We identified eligible patients treated at two institutions from 2007 to 2014. We linked their electronic medical records to the national AMI and death registries. We performed univariable and multivariable Cox regressions analysis to assess the association between various RHDPs, AMI and OS., Results: 120 eligible patients were included with a median follow-up of 17.6 months. Median age was 65.5 years. Median prescription dose was 60 Gy. Median mean heart dose (MHD) was 12.6 Gy. Univariable analysis showed that higher MHD (hazard ratio (HR), 1.03; 95% confidence interval (CI), 1.01-1.06; P = .008) and volume of heart receiving at least 5 Gy (V5) (HR, 1.01; 95% CI, 1.00-1.03; P = .042) were associated with increased hazards for AMI. Univariable analysis showed that higher MHD, V5, V25, V30, V40, V50 and dose to 30% of heart volume were associated with increased hazards for death. Multivariable analysis showed that there was no statistically significant association between various RHDPs and OS., Conclusion: The incidence of AMI is low among stage III NSCLC treated with definitive radiotherapy with or without chemotherapy. There is insufficient evidence to conclude that RHDPs are associated with AMI or OS in our study., (Copyright © 2018 Elsevier B.V. All rights reserved.)

Background: The purpose of this clinical review was to summate the published data for the long-term outcomes of reirradiation with intensity-modulated radiotherapy (IMRT) for locally recurrent nasopharyngeal carcinoma (NPC)., Methods: We searched biomedical literature databases for eligible studies published from January 2005 to September 2016. Outcomes of interests were 5-year local failure-free survival, distant failure-free survival, overall survival (OS), and toxicities. Meta-analysis was performed using a random effects model., Results: We found 4 comparative and 8 noncomparative studies (n = 1768). Reirradiation was associated with pooled event rates of 72% (95% confidence interval [CI] 66%-78%; I 2  = 84%), 85% (95% CI 82%-88%; I 2  = 69%), and 41% (95% CI 36%-47%; I 2  = 80%) for 5-year local failure-free survival, distant failure-free survival, and OS, respectively, with significant heterogeneity among the study results. The pooled event rate for grade 5 toxicities was 33% (95% CI 30%-35%; I 2  = 0%) with minimal heterogeneity., Conclusion: Reirradiation with IMRT for locally recurrent NPC could confer long-term disease control and survival but is associated with significant mortality., (© 2017 Wiley Periodicals, Inc.)

Background: To determine if the presence of epidermal growth factor receptor (EGFR) sensitizing mutations improves tumor control and survival outcomes in patients with non-metastatic non-small cell lung cancer (NSCLC) who received definitive thoracic radiation therapy (TRT) with or without chemotherapy., Materials and Methods: We searched MEDLINE for eligible comparative studies which compared the outcomes of patients treated with definitive TRT according to EGFR mutation status. Meta-analysis was performed using random effects model. Outcomes of interest were tumor overall response rate (ORR), loco-regional (LRR), distant recurrence rates (DRR), relapse-free survival (RFS), overall survival (OS) and adverse events (AE)., Results: We found seven studies including 537 patients with stage III NSCLC. Up to 45% of patients in the studies had mutations in exon 19 and 21. Patients harbouring EGFR sensitizing mutations had a trend towards improvement in ORR (risk ratio 1.17, 95% confidence interval 0.99-1.37, P = 0.06) compared to EGFR wild type status. There were no significant differences in LRR, DRR, RFS, OS and AE outcomes between the EGFR mutant and EGFR wild type groups., Conclusions: The presence of EGFR sensitizing mutations may improve tumour response rate but not survival in patients with localized NSCLC treated with definitive thoracic radiation therapy with or without chemotherapy., Competing Interests: CONFLICTS OF INTEREST None.

Objectives: To determine the pathological response rates and toxicity and in patients with locally advanced rectal cancer treated with concurrent capecitabine and dose escalated intensity modulated radiotherapy (IMRT) Methods: Patients with stage II or III adenocarcinoma of the rectum were treated with preoperative concurrent capecitabine and IMRT. Dose of capecitabine was 825mg/m 2 , 5 days a week for 5 weeks. IMRT was used to deliver a dose of 45Gy in 25 fractions (1.8Gy per fraction daily, 5 days a week over 5 weeks) to the regional lymphatics and areas at risk of harbouring microscopic disease. A concomitant synchronous integrated boost (SIB) to the gross tumour with a margin to a total dose of 55Gy in 25 fractions was also delivered in the same period. TME surgery was performed 8 weeks after preoperative therapy. The primary endpoint is pathological complete response rate (pCR) and the secondary endpoint was downstaging rates, Sphincter preservation rates (SPR), disease free survival (DFS) at 2 years and toxicity graded using the CTCAE v3.0. Results: Twenty three patients were enrolled. Three were not evaluable; one did not complete treatment due to logistic issues and two declined surgery. The remaining 20 patients completed preoperative chemoIMRT followed by TME surgery. At a median follow-up of 38.2 months (17.5-53.2 months), 90% (18 of 20) patients were alive. The 2 year overall survival and DFS were 90% and 90% respectively. 35%(7/20) of patients had a pCR. 65% (13 of 20) patients had successful downstaging of their rectal tumours. There was no local recurrence. Sphincter preservation rate was 85%. Treatment was well tolerated with only one patient (5%) having Grade 3 radiation proctitis. Conclusions: Preoperative concurrent capecitabine and dose escalated IMRT is well tolerated and results in high rates of pCR. A randomized trial comparing this regimen with standard 3D conformal chemoradiotherapy is warranted., Competing Interests: Competing Interests: The authors have declared that no competing interest exists.