Early relapse is an adverse prognostic factor for survival outcomes in patients with oral cavity squamous cell carcinoma: results from a nationwide registry study.

BACKGROUND: The prognostic significance of the relapse interval in patients with resected oral cavity squamous cell carcinoma (OCSCC) is a matter of ongoing debate. In this large-scale, registry-based, nationwide study, we examined whether the time interval between surgery and the first disease relapse may affect survival outcomes in Taiwanese patients with OCSCC. METHODS: Data made available by the Taiwan Health Promotion Administration as of 2004 were obtained. The study cohort consisted of patients who were included in the registry between 2011 and 2017. Disease staging was performed according to the American Joint Committee on Cancer (AJCC) Staging Manual, Eight Edition. We retrospectively reviewed the clinical records of 13,789 patients with OCSCC who received surgical treatment. A total of 2327 (16.9%) patients experienced a first disease relapse. The optimal cutoff value for the relapse interval was 330 days when both 5-year disease-specific survival (DSS) and overall survival (OS) (≤ 330/>330 days, n = 1630/697) were taken into account. In addition, we undertook a propensity score (PS)-matched analysis of patients (n = 654 each) with early (≤ 330 days) versus late (> 330 days) relapse. RESULTS: The median follow-up time in the entire study cohort was 702 days (433 and 2001 days in the early and late relapse groups, respectively). Compared with patients who experienced late relapse, those with early relapse showed a higher prevalence of the following adverse prognostic factors: pT4, pN3, pStage IV, poor differentiation, depth of invasion ≥ 10 mm, and extra-nodal extension. Multivariable analysis revealed that early relapse was an independent adverse prognostic factor for both 5-year DSS and OS (average hazard ratios [AHRs]: 3.24 and 3.91, respectively). In the PS-matched cohort, patients who experienced early relapse showed less favorable 5-year DSS: 58% versus 30%, p < 0.0001 (AHR: 3.10 [2.69 - 3.57]) and OS: 49% versus 22%, p < 0.0001 (AHR: 3.32 [2.89 - 3.81]). CONCLUSION: After adjustment for potential confounders and PS matching, early relapse was an adverse prognostic factor for survival outcomes in patients with OCSCC. Our findings may have significant implications for risk stratification.

Using T1 mapping indices to evaluate muscle function and predict conservative treatment outcomes in diabetic patients with peripheral arterial disease.

OBJECTIVES: To investigate interstitial muscle fibrosis via T1 mapping indices and its relationships with muscle function and conservative treatment outcomes. METHODS: A total of 49 DM patients with PAD were prospectively recruited from 2016 to 2017. All PAD patients underwent pre-treatment MRI with conservative treatment via a rehabilitation program and antiplatelet therapy. The need to require percutaneous transluminal angioplasty intervention was recorded as intolerance to conservative treatment outcomes. We quantified calf interstitial muscle fibrosis using T1 mapping indices (native T1, post-contrast T1, and the extracellular volume fraction [ECV]). Muscle function was evaluated using a 6-min walking test (6MWT) and a 3-min stepping test (3MST). PAD patients were divided into two groups according to their tolerance or intolerance of the conservative treatment. Pearson's correlation, reproducibility, and multivariable Cox hazard analyses were performed with p < 0.05 indicating statistical significance. RESULTS: Among the T1 mapping indices in the posterior compartment of the calf in PAD patients, the native T1 value was significantly correlated with 6MWT (r = -0.422, p = 0.010) and 3MST (r = -0.427, p = 0.009). All T1 mapping indices showed excellent intra-observer and inter-observer correlations. ECV was an independent predictor of conservative treatment intolerance (average ECV, hazard ratio: 1.045, 95% confidence interval: 1.011-1.079, p = 0.009). CONCLUSIONS: T1 mapping measurements are reproducible with excellent intra-observer and inter-observer correlations. T1 mapping indices may be predictive of treatment and functional outcomes and carry promise in patient evaluation. TRIAL REGISTRATION: Clinical Trials Identifier: NCT02850432 . KEY POINTS: • T1 mapping measurements of the calf muscles are reproducible with excellent intra-observer and inter-observer correlations (0.98 and 0.95 for anterior and posterior compartment muscle extracellular volume matrix [ECV] measurements, respectively). • ECV is shown to independently predict conservative treatment intolerance. • T1 mapping indices may be predictive of treatment and functional outcomes and carry promise in patient evaluation.

Fully automated segmentation and radiomics feature extraction of hypopharyngeal cancer on MRI using deep learning.

OBJECTIVES: To use convolutional neural network for fully automated segmentation and radiomics features extraction of hypopharyngeal cancer (HPC) tumor in MRI. METHODS: MR images were collected from 222 HPC patients, among them 178 patients were used for training, and another 44 patients were recruited for testing. U-Net and DeepLab V3 + architectures were used for training the models. The model performance was evaluated using the dice similarity coefficient (DSC), Jaccard index, and average surface distance. The reliability of radiomics parameters of the tumor extracted by the models was assessed using intraclass correlation coefficient (ICC). RESULTS: The predicted tumor volumes by DeepLab V3 + model and U-Net model were highly correlated with those delineated manually (p < 0.001). The DSC of DeepLab V3 + model was significantly higher than that of U-Net model (0.77 vs 0.75, p < 0.05), particularly in those small tumor volumes of < 10 cm3 (0.74 vs 0.70, p < 0.001). For radiomics extraction of the first-order features, both models exhibited high agreement (ICC: 0.71-0.91) with manual delineation. The radiomics extracted by DeepLab V3 + model had significantly higher ICCs than those extracted by U-Net model for 7 of 19 first-order features and for 8 of 17 shape-based features (p < 0.05). CONCLUSION: Both DeepLab V3 + and U-Net models produced reasonable results in automated segmentation and radiomic features extraction of HPC on MR images, whereas DeepLab V3 + had a better performance than U-Net. CLINICAL RELEVANCE STATEMENT: The deep learning model, DeepLab V3 + , exhibited promising performance in automated tumor segmentation and radiomics extraction for hypopharyngeal cancer on MRI. This approach holds great potential for enhancing the radiotherapy workflow and facilitating prediction of treatment outcomes. KEY POINTS: • DeepLab V3 + and U-Net models produced reasonable results in automated segmentation and radiomic features extraction of HPC on MR images. • DeepLab V3 + model was more accurate than U-Net in automated segmentation, especially on small tumors. • DeepLab V3 + exhibited higher agreement for about half of the first-order and shape-based radiomics features than U-Net.

Clinical Outcomes of Taiwanese Patients with Resected Oral Cavity Squamous Cell Carcinoma Who Underwent Reconstruction with Free Versus Local Flaps.

BACKGROUND: We sought to compare the clinical outcomes of Taiwanese patients with resected oral cavity squamous cell carcinoma (OCSCC) who underwent reconstruction with free versus local flaps. METHODS: From 2011 to 2017, we examined 8646 patients with first primary OCSCC who received surgery either with or without adjuvant therapy. Of these patients, 7297 and 1349 received free and local flap reconstruction, respectively. Two propensity score-matched groups of patients who underwent free versus local flap (n = 1268 each) reconstructions were examined. Margin status was not included as a propensity score-matched variable. RESULTS: Compared with local flaps, patients who received free flaps had a higher prevalence of the following variables: male sex, age < 65 years, pT3-4, pN1-3, p-Stage III-IV, depth ≥ 10 mm, margin > 4 mm, extranodal extension (ENE), and adjuvant therapy (all p < 0.0001). Multivariable analysis identified the reconstruction method (local vs. free flaps, only overall survival [OS]), age ≥ 65 years, pT3-4, pN1-3, p-Stage III-IV, depth ≥ 10 mm (only OS), margins ≤ 4 mm, and ENE as independent adverse prognosticators for disease-specific survival (DSS) and OS. The results of propensity score-matched analyses revealed that, compared with free flaps, patients who underwent local flap reconstruction showed less favorable 5-year DSS (hazard ratio [HR] 1.26, 82%/77%; p = 0.0100) and OS (HR 1.21, 73%/68%; p = 0.0079). CONCLUSIONS: After adjusting for covariates using multivariate models, and also by propensity score modeling, OCSCC patients who underwent free flap reconstruction showed a higher frequency of clear margins and a significant survival advantage compared with those who received local flaps.

Clinical impact of a deep learning system for automated detection of missed pulmonary nodules on routine body computed tomography including the chest region.

OBJECTIVES: To evaluate the clinical impact of a deep learning system (DLS) for automated detection of pulmonary nodules on computed tomography (CT) images as a second reader. METHODS: This single-centre retrospective study screened 21,150 consecutive body CT studies from September 2018 to February 2019. Pulmonary nodules detected by the DLS on axial CT images but not mentioned in initial radiology reports were flagged. Flagged images were scored by four board-certificated radiologists each with at least 5 years of experience. Nodules with scores of 2 (understandable miss) or 3 (should not be missed) were then categorised as unlikely to be clinically significant (2a or 3a) or likely to be clinically significant (2b or 3b) according to the 2017 Fleischner guidelines for pulmonary nodules. The miss rate was defined as the total number of studies receiving scores of 2 or 3 divided by total screened studies. RESULTS: Among 172 nodules flagged by the DLS, 60 (35%) missed nodules were confirmed by the radiologists. The nodules were further categorised as 2a, 2b, 3a, and 3b in 24, 14, 10, and 12 studies, respectively, with an overall positive predictive value of 35%. Missed pulmonary nodules were identified in 0.3% of all CT images, and one-third of these lesions were considered clinically significant. CONCLUSIONS: Use of DLS-assisted automated detection as a second reader can identify missed pulmonary nodules, some of which may be clinically significant. CLINICAL RELEVANCE/APPLICATION: Use of DLS to help radiologists detect pulmonary lesions may improve patient care. KEY POINTS: • DLS-assisted automated detection as a second reader is feasible in a large consecutive cohort. • Performance of combined radiologists and DLS was better than DLS or radiologists alone. • Pulmonary nodules were missed more frequently in abdomino-pelvis CT than the thoracic CT.

Prospective comparison of early interim 18F-FDG-PET with 18F-FLT-PET for predicting treatment response and survival in metastatic breast cancer.

BACKGROUND: To compare the value of interim 18F-FLT-PET and 18F-FDG-PET for predicting treatment outcomes in patients with metastatic breast cancer after salvage therapy. METHODS: Patients with metastatic breast cancer received PET/CT using 18F-FLT and 18F-FDG at baseline, after the 1st and 2nd cycle of systemic chemotherapy. The clinical response was classified according to Response Evaluation Criteria in Solid Tumors 1.1 based on contrast-enhanced CT after 3 months of systemic chemotherapy. The metabolic response on PET was assessed according to European Organization for Research and Treatment of Cancer criteria or PET Response Criteria in Solid Tumors (PERCIST) and was correlated to the clinical response, overall survival (OS), and progression-free survival (PFS). RESULTS: Twenty-five patients entered final analysis. On 18F-FDG-PET, clinical responders after 2 chemotherapy cycles (post-2c) had a significantly greater reduction of maximal standardized uptake value (SUV) and the peak SUV corrected for lean body mass (SULpeak) of the tumor than non-responders (P = 0.030 and 0.003). Metabolic response determined by PERCIST on post-2c 18F-FDG-PET showed a high area under the receiver operating characteristics curve of 0.801 in predicting clinical response (P = 0.011). Patients who were metabolic responders by PERCIST on post-2c 18F-FDG-PET had a significantly longer PFS (53.8% vs. 16.7%, P = 0.014) and OS (100% vs. 47.6%, P = 0.046) than non-responders. Survival differences between responders and non-responders in the interim 18F-FLT-PET were not significant. CONCLUSIONS: 18F-FLT-PET failed to show an advantage over 18F-FDG-PET in predicting the treatment response and survival in patients with metastatic breast cancer. Assessment of treatment outcome by interim 18F-FDG-PET may aid treatment. TRIAL REGISTRATION: The study was retrospectively registered on 02/06/2020 on Clinicaltrials.gov (identifier NCT04411966 ).

Comparison of 18F-FDG PET/MRI, MRI, and 18F-FDG PET/CT for the detection of synchronous cancers and distant metastases in patients with oropharyngeal and hypopharyngeal squamous cell carcinoma.

PURPOSE: In this prospective study, we sought to compare the clinical utility of fluorodeoxyglucose PET/MRI, MRI, and PET/CT in the detection of synchronous cancers and distant metastases in patients with oropharyngeal and hypopharyngeal squamous cell carcinoma (OHSCC). METHODS: We examined 198 consecutive patients with biopsy-proven OHSCC who agreed to receive chemoradiation. All patients underwent pretreatment PET/MRI and PET/CT on the same day. Patients were followed-up for a minimum of 12 months or until death. The McNemar's test and receiver-operating characteristic (ROC) curves were used to compare sensitivity/specificity and the diagnostic capabilities of PET/MRI, MRI, and PET/CT, respectively. RESULTS: We identified 55 patients (27.7%) who had synchronous cancers and/or distant metastases (number of involved sites: 83). The results of site-based analysis revealed that the sensitivity of PET/MRI was 15.7% higher than that of MRI (73.5% versus 57.8%, p < 0.001) and 3.6% higher compared with PET/CT (73.5% versus 69.9%, p = 0.083), whereas the sensitivity of PET/CT was 12.1% higher than that of MRI (69.9% versus 57.8%, p = 0.012). On a patient-basis, ROC curve analysis demonstrated that PET/MRI yielded a greater area under curve than MRI (0.930 versus 0.905, p = 0.023). There were no significant differences in terms of diagnostic capability between MRI and PET/CT (0.905 versus 0.917, p = 0.469) and between PET/MRI and PET/CT (0.930 versus 0.917, p = 0.062). CONCLUSIONS: In our cohort, PET/MRI showed a significantly higher diagnostic capability than MRI and no significant difference compared with PET/CT for the detection of synchronous cancers or distant metastases in patients with OHSCC.

A combined analysis of maximum standardized uptake value on FDG-PET, genetic markers, and clinicopathological risk factors in the prognostic stratification of patients with resected oral cavity squamous cell carcinoma.

OBJECTIVE: Clinical outcomes of patients with resected oral cavity squamous cell carcinoma (OCSCC) chiefly depend on the presence of specific clinicopathological risk factors (RFs). Here, we performed a combined analysis of FDG-PET, genetic markers, and clinicopathological RFs in an effort to improve prognostic stratification. METHODS: We retrospectively reviewed the clinical records of 2036 consecutive patients with first primary OCSCC who underwent surgery between 1996 and 2016. Of them, 345 underwent ultra-deep targeted sequencing (UDTS, between 1996 and 2011) and 168 whole exome sequencing (WES, between 2007 and 2016). Preoperative FDG-PET imaging was performed in 1135 patients from 2001 to 2016. Complete data on FDG-PET, genetic markers, and clinicopathological RFs were available for 327 patients. RESULTS: Using log-ranked tests based on 5-year disease-free survival (DFS), the optimal cutoff points for maximum standardized uptake values (SUV-max) of the primary tumor and neck metastatic nodes were 22.8 and 9.7, respectively. The 5-year DFS rates were as follows: SUVtumor-max ≥ 22.8 or SUVnodal-max ≥ 9.7 (n = 77) versus SUVtumor-max < 22.8 and SUVnodal-max < 9.7 (n = 250), 32%/62%, P < 0.001; positive UDTS or WES gene panel (n = 64) versus negative (n = 263), 25%/62%, P < 0.001; pN3b (n = 165) versus pN1-2 (n = 162), 42%/68%, P < 0.001. On multivariate analyses, SUVtumor-max ≥ 22.8 or SUVnodal-max ≥ 9.7, a positive UDTS/WES gene panel, and pN3b disease were identified as independent prognosticators for 5-year outcomes. Based on these variables, we devised a scoring system that identified four distinct prognostic groups. The 5-year rates for patients with a score from 0 to 3 were as follows: loco-regional control, 80%/67%/47%/24% (P < 0.001); distant metastases, 13%/23%/55%/92% (P < 0.001); DFS, 74%/58%/28%/7% (P < 0.001); and disease-specific survival, 80%/64%/35%/7% (P < 0.001) respectively. CONCLUSIONS: The combined assessment of tumor and nodal SUV-max, genetic markers, and pathological node status may refine the prognostic stratification of OCSCC patients.

Peripheral arterial disease: the role of extracellular volume measurements in lower limb muscles with MRI.

OBJECTIVES: Peripheral arterial disease (PAD) is characterised by arterial occlusion and fibrosis in the lower extremities. Extracellular volume matrix fraction (ECV) is a biomarker of skeletal muscle fibrosis, but has not been applied to the lower extremities with PAD. This study investigated the clinical feasibility of using ECV for calf muscle fibrosis quantification by comparing normal controls (NC) and PAD patients. METHODS: From October 2016 to December 2017, we recruited patients with PAD, and patients with head and neck cancer receiving fibular flap as NC group. All participants underwent magnetic resonance imaging (MRI) to determine the ECV of the calves and the differences between the NC and PAD groups. ECV was calculated from T1 values at steady-state equilibrium, defined as the point in time after contrast agent injection when the variance of T1 relaxation time in blood and muscle becomes less than 5%. RESULTS: A total of 46 patients (18 in the NC group and 28 in the PAD group) were recruited. Steady-state equilibrium was reached at 11-12 min after contrast agent injection. The NC group had significantly lower mean ECV than the PAD group (12.71% vs. 31.92%, respectively, p < 0.001). In the PAD group, the mean ECV was slightly lower in patients with collateral vessels than in those without (26.58% vs. 34.88%, respectively, p = 0.047). CONCLUSION: Evaluation of skeletal fibrosis in PAD using ECV is feasible. ECV can help identify PAD patients with collateral vessel formation and lay the foundation for future research in PAD management. KEY POINTS: • Steady-state equilibrium for ECV measurement of the lower limbs can be reached at around 11-12 min. • Quantification of lower limb muscle fibrosis by measuring ECV is clinically feasible and can be used to differentiate between patients with PAD and histologically proven normal controls. • ECV can differentiate PAD patients with or without visible collateral vessels, further expanding its role in identifying the presence of collateral supply in clinical decision-making.

Deep learning for fully automated tumor segmentation and extraction of magnetic resonance radiomics features in cervical cancer.

OBJECTIVE: To develop and evaluate the performance of U-Net for fully automated localization and segmentation of cervical tumors in magnetic resonance (MR) images and the robustness of extracting apparent diffusion coefficient (ADC) radiomics features. METHODS: This retrospective study involved analysis of MR images from 169 patients with cervical cancer stage IB-IVA captured; among them, diffusion-weighted (DW) images from 144 patients were used for training, and another 25 patients were recruited for testing. A U-Net convolutional network was developed to perform automated tumor segmentation. The manually delineated tumor region was used as the ground truth for comparison. Segmentation performance was assessed for various combinations of input sources for training. ADC radiomics were extracted and assessed using Pearson correlation. The reproducibility of the training was also assessed. RESULTS: Combining b0, b1000, and ADC images as a triple-channel input exhibited the highest learning efficacy in the training phase and had the highest accuracy in the testing dataset, with a dice coefficient of 0.82, sensitivity 0.89, and a positive predicted value 0.92. The first-order ADC radiomics parameters were significantly correlated between the manually contoured and fully automated segmentation methods (p < 0.05). Reproducibility between the first and second training iterations was high for the first-order radiomics parameters (intraclass correlation coefficient = 0.70-0.99). CONCLUSION: U-Net-based deep learning can perform accurate localization and segmentation of cervical cancer in DW MR images. First-order radiomics features extracted from whole tumor volume demonstrate the potential robustness for longitudinal monitoring of tumor responses in broad clinical settings. U-Net-based deep learning can perform accurate localization and segmentation of cervical cancer in DW MR images. KEY POINTS: • U-Net-based deep learning can perform accurate fully automated localization and segmentation of cervical cancer in diffusion-weighted MR images. • Combining b0, b1000, and apparent diffusion coefficient (ADC) images exhibited the highest accuracy in fully automated localization. • First-order radiomics feature extraction from whole tumor volume was robust and could thus potentially be used for longitudinal monitoring of treatment responses.

Classifying Neck Lymph Nodes of Head and Neck Squamous Cell Carcinoma in MRI Images with Radiomic Features.

The aim of this study is to develop a computer-aided diagnosis method to help classify medical images of neck lymph nodes in head and neck cancer patients. According to the current practice guidelines, the classification of lymph node status is critical for patient stratification before treatment. Take extra-nodal extension (ENE) of metastatic neck lymph nodes, the status of ENE has been considered a single factor affecting the decision of whether systemic treatment with toxicity should be given to patients with otherwise non-advanced cancer status. Medical imaging prior to surgery serves as tools for clinical staging and determining the extent of neck lymph node dissection during the tumor resection surgery. The information contained in these images may also help determine the status of ENE and thus stratify patients for more precise treatment. In the current practice, there has been not a reliable computer-aided tool for this task. In this study, we used open-source software to investigate radiomic features that help distinguish malignant from benign and ENE from non-ENE lymph nodes. We have identified 89 features that can differentiate malignant from benign and 4 features that can differentiate ENE from non-ENE lymph nodes. Furthermore, we fed the significant features to a multilayer perceptron neural network to predict malignancy and ENE of lymph nodes and achieved 84% and 77% of accuracy in each task, respectively.

Tumor Depth of Invasion (Tumor > 4 cm/Depth > 10 mm and Depth > 20 mm) and Through Cortex/Skin Invasion are Both Valid Criteria for Classifying Tumors as pT4a in AJCC 2018 Oral Cavity Cancer Staging System.

BACKGROUND: According to the AJCC third to seventh edition staging manuals (1988-2010), the presence of through cortex and/or skin invasion in oral cavity squamous cell carcinoma (OCSCC) identifies T4a tumors. The AJCC eighth edition (2018) introduced a depth of invasion (DOI) > 20 mm as a criterion for pT4a. Subsequently, a revision maintained that tumors > 4 cm with a DOI > 10 mm should be classified as pT4a. We sought to analyze the prognostic impact of the three distinct criteria identifying pT4a disease. METHODS: We examined 667 consecutive patients with pT3-4 buccal/gum/hard palate/retromolar SCC who underwent surgery between 1996 and 2016. pT1/pT2 (n = 108/359) disease were included for comparison purposes. RESULTS: The 5-year outcomes of patients with pT1/pT2/without (n = 406)/with tumor > 4 cm/DOI > 10 mm (n = 261), pT1/pT2/DOI ≤ 20 mm (n = 510)/> 20 mm (n = 157), and pT1/pT2/without (n = 305)/with through cortex/skin invasion (n = 362) were as follows: disease-specific survival (DSS), 98%/89%/79%/65%, p < 0.001, 98%/89%/78%/59%, p < 0.001, and 98%/89%79%/69%, p < 0.001; overall survival (OS), 90%/79%/63%/51%, p < 0.001, 90%/79%/63%/42%, p < 0.001, and 90%/79%/65%/52%, p < 0.001. In pT3-4 disease, a tumor > 4 cm/DOI > 10 mm was an independent adverse prognosticator for 5-year DSS rate, DOI > 20 mm was an independent adverse prognosticator for 5-year DSS and OS rates, whereas through cortex/skin invasion independently predicted 5-year OS rates. CONCLUSIONS: All of the three criteria (tumor > 4 cm/DOI > 10 mm, DOI > 20 mm, and through cortex/skin invasion) identify high-risk patients, which should be reflected in further revisions of pT4a classification in OCSCC.

Value of early evaluation of treatment response using 18F-FDG PET/CT parameters and the Epstein-Barr virus DNA load for prediction of outcome in patients with primary nasopharyngeal carcinoma.

PURPOSE: To determine the value of early evaluation of response to concurrent chemoradiotherapy (CCRT) using 18F-FDG PET-derived parameters and the Epstein-Barr virus (EBV) DNA titre in outcome prediction in patients with primary nasopharyngeal carcinoma (NPC). METHODS: Sixty patients with primary NPC were prospectively enrolled. All patients underwent 18F-FDG PET/CT before and during CCRT. The plasma EBV DNA titre was measured along with the PET/CT-derived parameters. Changes in EBV DNA titre and PET/CT-derived parameters during CCRT were analysed in relation to response to treatment, recurrence-free survival (RFS) and overall survival (OS). RESULTS: A total lesion glycolysis (TLG) reduction ratio of ≤0.6 and a detectable EBV DNA titre during CCRT were predictors of an unfavourable response to treatment, RFS and OS. In multivariate analysis, a TLG reduction ratio of ≤0.6 predicted incomplete remission (p = 0.002) and decreased RFS (p = 0.003). The proportion of patients with a TLG reduction ratio of >0.6 who achieved a complete response was more than twice that of patients with a TLG reduction ratio of ≤0.6. A detectable EBV DNA titre, a TLG reduction ratio of ≤0.6 and older age were independently associated with a poorer OS (p = 0.037, 0.009 and 0.016, respectively). A scoring system was developed based on these independent predictors of OS. Patients with a score of 1 and 2/3 had poorer survival outcomes than those with a score of 0 (hazard ratio 4.756, p = 0.074, and hazard ratio 18.973, p = 0.001, respectively). This scoring system appeared to be superior to the traditional TNM staging system (p < 0.001 versus p = 0.175). CONCLUSION: Early evaluation of response to CCRT using 18F-FDG PET-derived parameters and the EBV DNA titre can predict outcome in patients with primary NPC. A combination of interim PET parameters and the EBV DNA titre enables better stratification of patients into subgroups with different survival rates.

Effect of Chemotherapy on Dual-Energy X-ray Absorptiometry (DXA) Body Composition Precision Error in Head and Neck Cancer Patients.

BACKGROUND: Precision error in dual-energy X-ray absorptiometry (DXA) is defined as difference in results due to instrumental and technical factors given no biologic change. The aim of this study is to compare precision error in DXA body composition scans in head and neck cancer patients before and 2 months after chemotherapy. METHODOLOGY: A total of 34 male head and neck cancer patients with normal body mass index (BMI) were prospectively enrolled and all patients received 2 consecutive DXA scans both before and after 2 months of chemotherapy for a total of 4 scans. The precision error of 3 DXA body composition values (lean mass, fat mass, and bone mineral content) was calculated for total body and 5 body regions (arms, legs, trunk, android, and gynoid). Precision errors before and after treatment were compared using generalized estimating equation model. RESULTS: There was no significant change in precision error for the DXA total body composition values following chemotherapy; lean mass (0.33%-0.40%, p = 0.179), total fat mass (1.39%-1.70%, p = 0.259) and total bone mineral content (0.42%-0.56%, p = 0.243). However, there were significant changes in regional precision error; trunk lean mass (1.19%-1.77%, p = 0.014) and android fat mass (2.17%-3.72%, p = 0.046). CONCLUSIONS: For head and neck cancer patients, precision error of DXA total body composition values did not change significantly following chemotherapy; however, there were significant changes in fat mass in the android and lean mass in the trunk. Caution should be exercised when interpreting longitudinal DXA body composition data in those body parts.

Clinical utility of simultaneous whole-body 18F-FDG PET/MRI as a single-step imaging modality in the staging of primary nasopharyngeal carcinoma.

PURPOSE: Both head and neck magnetic resonance imaging (MRI) and 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography (PET)/computed tomography (CT) play a crucial role in the staging of primary nasopharyngeal carcinoma (NPC). In this study, we sought to prospectively investigate the clinical utility of simultaneous whole-body 18F-FDG PET/MRI for primary staging of NPC patients. METHODS: We examined 113 patients with histologically confirmed NPC who underwent pretreatment, simultaneous whole-body PET/MRI and PET/CT for primary tumor staging. The images obtained with the different imaging modalities were interpreted independently and compared with each other. RESULTS: PET/MRI increased the accuracy of head and neck MRI for assessment of primary tumor extent in four patients via addition of FDG uptake information to increase the conspicuity of morphologically subtle lesions. PET/MR images were more discernible than PET/CT images for mapping tumor extension, especially intracranial invasion. Regarding the N staging assessment, the sensitivity of PET/MRI (99.5%) was higher than that of head and neck MRI (94.2%) and PET/CT (90.9%). PET/MRI was particularly useful for distinguishing retropharyngeal nodal metastasis from adjacent nasopharyngeal tumors. For distant metastasis evaluation, PET/MRI exhibited a similar sensitivity (90% vs. 86.7% vs. 83.3%), but higher positive predictive value (93.1% vs. 78.8% vs. 83.3%) than whole-body MRI and PET/CT, respectively. CONCLUSIONS: For tumor staging of NPC, simultaneous whole-body PET/MRI was more accurate than head and neck MRI and PET/CT, and may serve as a single-step staging modality.

Clinical Outcomes in pT4 Tongue Carcinoma are Worse than in pT3 Disease: How Extrinsic Muscle Invasion Should be Considered?

BACKGROUND: The identification of extrinsic tongue muscle invasion in oral cavity cancer remains challenging. Notably, the most recent American Joint Committee on Cancer (AJCC 2017, 8th edition) staging manual indicates that extrinsic muscle invasion does not lead to the diagnosis of a T4 tumor. Because this approach carries the risk of tumor downstaging, we compared the clinical outcomes of patients with oral tongue squamous cell carcinoma (SCC) staged as pT3 vs. pT4 according to the AJCC 2010, 7th edition criteria. METHODS: We retrospectively examined the records of consecutive patients with pT3 (n = 135) and pT4 (n = 68) tongue SCC who underwent radical surgery. Of the 68 pT4 tongue SCC, 63 (93%) had extrinsic muscle involvement alone. The 5-year locoregional control (LRC), distant metastasis (DM), and disease-free survival (DFS) rates served as outcome measures. RESULTS: Compared with pT3 tongue SCC, pT4 patients presented significantly more frequently with pN2 disease, extranodal extension, poor tumor differentiation, tumor depth >15 and >20 mm, margin status ≤4 mm, perineural invasion, vascular invasion, and were more frequently treated with surgery plus concurrent chemoradiotherapy. Less favorable 5-year outcomes were observed in patients with pT4 than pT3 tumors (LRC 50 vs. 75%, p < 0.001; DM 27 vs. 14%, p = 0.013; DFS 43 vs. 69%, respectively, p < 0.001). We identified pT4 disease (vs. pT3) as an independent adverse prognostic factor for LRC and DFS. CONCLUSIONS: We suggest classifying patients with tongue SCC and extrinsic muscle invasion as having pT4 disease.

Clinical Outcomes of Taiwanese Patients with cT4 Oral Cavity Squamous Cell Carcinoma: Toward the Identification of the Optimal Initial Treatment Approach for cT4b Patients.

BACKGROUND: The National Comprehensive Cancer Network guidelines recommend that patients with oral cavity squamous cell carcinoma (OSCC) and cT4b disease should be either included in clinical trials or treated with a nonsurgical approach. However, surgery may be feasible in selected patients with adequate safety margins. Using the nationwide Taiwanese Cancer Registry Database, we examined the prognosis of cT4b OSCC patients in relation to their treatment approach. METHODS: Of the 18,910 patients with previously untreated first primary OSCC identified between 2004 and 2010, 492 (2.6 %) had cT4b tumors. Of them, 327 (66 %) received initial treatment with surgery, whereas 165 (34 %) were initially treated with a nonsurgical approach. Of the latter group, 78 patients subsequently underwent surgery. A 5-year disease-specific survival (DSS) ≥45 % was considered as a favorable outcome. RESULTS: Better 5-year DSS and overall survival (OS) rates were observed in cT4b patients initially treated with surgery (vs. nonsurgery; DSS, 51 vs. 38 %; OS, 43 vs. 27 %, respectively, p < 0.001). Of the participants initially treated with surgery, patients with cN0-2 disease had better 5-year survival rates (DSS: cN0, 59 %; cN1, 53 %; cN2, 46 %; OS: cN0, 49 %; cN1, 50 %; cN2, 37 %) than those with cN3 disease (DSS: 0 %; OS: 0 %). Among cT4b patients who initially received a nonsurgical treatment, subjects who subsequently underwent surgery showed better outcomes. CONCLUSIONS: Primary surgery is performed in approximately two-thirds of cT4b OSCC patients, with cN0-2 cases showing a good prognosis. Patients who initially received a nonsurgical approach can subsequently be treated with surgery and achieve favorable outcomes.

Diffusion radiomics analysis of intratumoral heterogeneity in a murine prostate cancer model following radiotherapy: Pixelwise correlation with histology.

PURPOSE: To investigate the biological meaning of apparent diffusion coefficient (ADC) values in tumors following radiotherapy. MATERIALS AND METHODS: Five mice bearing TRAMP-C1 tumor were half-irradiated with a dose of 15 Gy. Diffusion-weighted images, using multiple b-values from 0 to 3000 s/mm2 , were acquired at 7T on day 6. ADC values calculated by a two-point estimate and monoexponential fitting of signal decay were compared between the irradiated and nonirradiated regions of the tumor. Pixelwise ADC maps were correlated with histological metrics including nuclear counts, nuclear sizes, nuclear spaces, cytoplasmic spaces, and extracellular spaces. RESULTS: As compared with the nonirradiated region, the irradiated region exhibited significant increases in ADC, extracellular space, and nuclear size, and a significant decrease in nuclear counts (P < 0.001 for all). Optimal ADC to differentiate the irradiated from nonirradiated regions was achieved at a b-value of 800 s/mm2 by the two-point method and monoexponential curve fitting. ADC positively correlated with extracellular spaces (r = 0.74) and nuclear sizes (r = 0.72), and negatively correlated with nuclear counts (r = -0.82, P < 0.001 for all). CONCLUSION: As a radiomic biomarker, ADC maps correlating with histological metrics pixelwise could be a means of evaluating tumor heterogeneity and responses to radiotherapy. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 2 J. MAGN. RESON. IMAGING 2017;46:483-489.

1 H MR spectroscopy in cervical carcinoma using external phase array body coil at 3.0 Tesla: Prediction of poor prognostic human papillomavirus genotypes.

PURPOSE: To assess the clinical value of proton (1 H) MR spectroscopy in cervical carcinomas, in the prediction of poor prognostic human papillomavirus (HPV) genotypes as well as persistent disease following concurrent chemoradiotherapy (CCRT). MATERIALS AND METHODS: 1 H MR spectroscopy using external phase array coil was performed in 52 consecutive cervical cancer patients at 3 Tesla (T). Poor prognostic HPV genotypes (alpha-7 species or absence of HPV infection) and persistent cervical carcinoma after CCRT were recorded. Statistical significance was calculated with the Mann-Whitney two-sided nonparametric test and areas under the receiver operating characteristics curve (AUC) analysis. RESULTS: A 4.3-fold (P = 0.032) increased level of methyl resonance at 0.9 ppm was found in the poor prognostic HPV genotypes, mainly attributed to the presence of HPV18, with a sensitivity of 75%, a specificity of 81%, and an AUC of 0.76. Poor prognostic HPV genotypes were more frequently observed in patients with adeno-/adenosquamous carcinoma (Chi-square, P < 0.0001). In prediction of the four patients with persistent disease after CCRT, elevated methyl resonance demonstrated a sensitivity of 100%, a specificity of 74%, and an AUC of 0.82. CONCLUSION: 1 H MR spectroscopy at 3T can be used to depict the elevated lipid resonance levels in cervical carcinomas, as well as help to predict the poor prognostic HPV genotypes and persistent disease following CCRT. Further large studies with longer follow up times are warranted to validate our initial findings. LEVEL OF EVIDENCE: 1 J. Magn. Reson. Imaging 2017;45:899-907.

Endometrial cancer with cervical stromal invasion: diagnostic accuracy of diffusion-weighted and dynamic contrast enhanced MR imaging at 3T.

OBJECTIVES: To compare the diagnostic accuracy of diffusion-weighted (DW) and dynamic contrast-enhanced (DCE) magnetic resonance (MR) imaging for detecting cervical stromal invasion in endometrial cancer. METHODS: Eighty-three consecutive women with endometrial cancer underwent preoperative evaluation in a 3-T unit, including T2-weighted, DW (b = 0 and 1000 s/mm2), and DCE MR imaging. Two radiologists independently assessed presence of cervical stromal invasion, with histopathological reference as gold standard. RESULTS: For assessing cervical stromal invasion, the diagnostic accuracy, sensitivity, and specificity, respectively for Reader 1/Reader 2, were as follows: DW MR imaging- 95.2 %/91.6 %, 91.7 %/100 %, and 95.8 %/90.1 %; DCE MR imaging- 91.6 %/88 %, 58.3 %/50 %, and 97.2 %/94.4 %. The diagnostic performance of DW MR imaging (Reader 1: areas under the receiver operating characteristic curve (AUC) = 0.98; Reader 2: AUC = 0.97) was significantly higher than that of DCE MR imaging (p = 0.009 for Reader 2) or T2-weighted MR imaging (Reader 1: p = 0.006; Reader 2: p = 0.013). Patients with cervical stromal invasion showed a significantly greater canal width (p < 0.0001) and myometrial invasion extent (p = 0.006). CONCLUSIONS: DW MR imaging has superior diagnostic performance compared with DCE MR imaging in the detection of cervical stromal invasion. KEY POINTS: • DWI demonstrates a higher accuracy than DCE in detecting cervical stromal invasion. • Tumour ADC values are similar between patients without or with cervical invasion. • Canal widening causes false-negativity on DCE and T2W but not on DWI.