Improving Microstructural Estimation in Time-Dependent Diffusion MRI With a Bayesian Method.

BACKGROUND: Accurately fitting diffusion-time-dependent diffusion MRI (td-dMRI) models poses challenges due to complex and nonlinear formulas, signal noise, and limited clinical data acquisition. PURPOSE: Introduce a Bayesian methodology to refine microstructural fitting within the IMPULSED (Imaging Microstructural Parameters Using Limited Spectrally Edited Diffusion) model and optimize the prior distribution within the Bayesian framework. STUDY TYPE: Retrospective. POPULATION: Involving 69 pediatric patients (median age 6 years, interquartile range [IQR] 3-9 years, 61% male) with 41 low-grade and 28 high-grade gliomas, of which 76.8% were identified within the brainstem or cerebellum. FIELD STRENGTH/SEQUENCE: 3 T, oscillating gradient spin-echo (OGSE) and pulsed gradient spin-echo (PGSE). ASSESSMENT: The Bayesian method's performance in fitting cell diameter ( d $$ d $$ ), intracellular volume fraction ( f in $$ {f}_{in} $$ ), and extracellular diffusion coefficient ( D ex $$ {D}_{ex} $$ ) was compared against the NLLS method, considering simulated and experimental data. The tumor region-of-interest (ROI) were manually delineated on the b0 images. The diagnostic performance in distinguishing high- and low-grade gliomas was assessed, and fitting accuracy was validated against H&E-stained pathology. STATISTICAL TESTS: T-test, receiver operating curve (ROC), area under the curve (AUC) and DeLong's test were conducted. Significance considered at P < 0.05. RESULTS: Bayesian methodology manifested increased accuracy with robust estimates in simulation (RMSE decreased by 29.6%, 40.9%, 13.6%, and STD decreased by 29.2%, 43.5%, and 24.0%, respectively for d $$ d $$ , f in $$ {f}_{in} $$ , and D ex $$ {D}_{ex} $$ compared to NLLS), indicating fewer outliers and reduced error. Diagnostic performance for tumor grade was similar in both methods, however, Bayesian method generated smoother microstructural maps (outliers ratio decreased by 45.3% ± 19.4%) and a marginal enhancement in correlation with H&E staining result (r = 0.721 for f in $$ {f}_{in} $$ compared to r = 0.698 using NLLS, P = 0.5764). DATA CONCLUSION: The proposed Bayesian method substantially enhances the accuracy and robustness of IMPULSED model estimation, suggesting its potential clinical utility in characterizing cellular microstructure. EVIDENCE LEVEL: 3 TECHNICAL EFFICACY: Stage 1.

Histological and molecular classifications of pediatric glioma with time-dependent diffusion MRI-based microstructural mapping.

BACKGROUND: Gliomas are the most common type of central nervous system tumors in children, and the combination of histological and molecular classification is essential for prognosis and treatment. Here, we proposed a newly developed microstructural mapping technique based on diffusion-time-dependent diffusion MRI td-dMRI theory to quantify tumor cell properties and tested these microstructural markers in identifying histological grade and molecular alteration of H3K27. METHODS: This prospective study included 69 pediatric glioma patients aged 6.14 ±â€…3.25 years old, who underwent td-dMRI with pulsed and oscillating gradient diffusion sequences on a 3T scanner. dMRI data acquired at varying tds were fitted into a 2-compartment microstructural model to obtain intracellular fraction (fin), cell diameter, cellularity, etc. Apparent diffusivity coefficient (ADC) and T1 and T2 relaxation times were also obtained. H&E stained histology was used to validate the estimated microstructural properties. RESULTS: For histological classification of low- and high-grade pediatric gliomas, the cellularity index achieved the highest area under the receiver-operating-curve (AUC) of 0.911 among all markers, while ADC, T1, and T2 showed AUCs of 0.906, 0.885, and 0.886. For molecular classification of H3K27-altered glioma in 39 midline glioma patients, cell diameter showed the highest discriminant power with an AUC of 0.918, and the combination of cell diameter and extracellular diffusivity further improved AUC to 0.929. The td-dMRI estimated fin correlated well with the histological ground truth with r = 0.7. CONCLUSIONS: The td-dMRI-based microstructural properties outperformed routine MRI measurements in diagnosing pediatric gliomas, and the different microstructural features showed complementary strength in histological and molecular classifications.

A Randomized Controlled Trial Comparing Two Different Schedules for Cisplatin Treatment in Patients with Locoregionally Advanced Nasopharyngeal Cancer.

PURPOSE: Previous studies suggest that a cumulative cisplatin dose of 200 mg/m2 might be adequate in the intensity-modulated radiation therapy (IMRT) era for locoregionally advanced nasopharyngeal carcinoma (LANPC). However, two cycles of once-every-3-weeks cisplatin at 100 mg/m2 has never been prospectively compared with standard once-a-week cisplatin regimen. PATIENTS AND METHODS: This trial was conducted at three hospitals from 2011 to 2016. Patients who met the eligibility criteria were recruited (ChiCTR-TRC-12001979) and randomly assigned (1:1) via a computer-generated sequence to receive once-every-3-weeks cisplatin at 100 mg/m2 for two cycles or once-a-week cisplatin at 40 mg/m2 for six cycles concurrently with IMRT. Primary endpoint was failure-free survival and between-group absolute difference of 10% as the noninferiority margin. RESULTS: A total of 510 patients were enrolled. Median follow-up time was 58.3 months with 85.4% of 3-year failure-free survival in the once-every-3-weeks group and 85.6% in the once-a-week group. An absolute difference of -0.2% (95% confidence interval, -6.3 to 5.9; P noninferiority = 0.0016). Acute toxicities of grade 3 or higher occurred in 55.8% in the once-every-3-weeks group and 66.3% in the once-a-week group (P = 0.015). The most common acute toxicities were hematologic abnormalities, including leukopenia (16% vs. 27%; P = 0.0022) and thrombocytopenia (1% vs. 5%; P = 0.015). The late grade 3-4 auditory loss rate was significantly lower in the once-every-3-weeks group than the once-a-week group (6% vs. 13%; P = 0.0039). CONCLUSIONS: Once-every-3-weeks cisplatin as concurrent chemoradiotherapy is noninferior to once-a-week cisplatin in the treatment efficacy in the LANPC. Although both regimens are well tolerated, severe acute toxicities and late-onset auditory loss are higher in the once-a-week group.

An interpretable machine learning prognostic system for locoregionally advanced nasopharyngeal carcinoma based on tumor burden features.

OBJECTIVES: We aimed to build a survival system by combining a highly-accurate machine learning (ML) model with explainable artificial intelligence (AI) techniques to predict distant metastasis in locoregionally advanced nasopharyngeal carcinoma (NPC) patients using magnetic resonance imaging (MRI)-based tumor burden features. MATERIALS AND METHODS: 1643 patients from three hospitals were enrolled according to set criteria. We employed ML to develop a survival model based on tumor burden signatures and all clinical factors. Shapley Additive exPlanations (SHAP) was utilized to explain prediction results and interpret the complex non-linear relationship among features and distant metastasis. We also constructed other models based on routinely used cancer stages, Epstein-Barr virus (EBV) DNA, or other clinical features for comparison. Concordance index (C-index), receiver operating curve (ROC) analysis and decision curve analysis (DCA) were executed to assess the effectiveness of the models. RESULTS: Our proposed system consistently demonstrated promising performance across independent cohorts. The concordance indexes were 0.773, 0.766 and 0.760 in the training, internal validation and external validation sets. SHAP provided personalized protective and risk factors for each NPC patient and uncovered some novel non-linear relationships between features and distant metastasis. Furthermore, high-risk patients who received induction chemotherapy (ICT) and concurrent chemoradiotherapy (CCRT) had better 5-year distant metastasis-free survival (DMFS) than those who only received CCRT, whereas ICT + CCRT and CCRT had similar DMFS in low-risk patients. CONCLUSIONS: The interpretable machine learning system demonstrated superior performance in predicting metastasis in locoregionally advanced NPC. High-risk patients might benefit from ICT.

Induction chemotherapy with lobaplatin and fluorouracil versus cisplatin and fluorouracil followed by chemoradiotherapy in patients with stage III-IVB nasopharyngeal carcinoma: an open-label, non-inferiority, randomised, controlled, phase 3 trial.

BACKGROUND: Cisplatin-based induction chemotherapy plus concurrent chemoradiotherapy in the treatment of patients with locoregionally advanced nasopharyngeal carcinoma has been recommended in the National Comprehensive Cancer Network Guidelines. However, cisplatin is associated with poor patient compliance and has notable side-effects. Lobaplatin, a third-generation platinum drug, has shown promising antitumour activity against several malignancies with less toxicity. In this study, we aimed to evaluate the efficacy of lobaplatin-based induction chemotherapy plus concurrent chemoradiotherapy over a cisplatin-based regimen in patients with locoregional, advanced nasopharyngeal carcinoma. METHODS: In this open-label, non-inferiority, randomised, controlled, phase 3 trial done at five hospitals in China, patients aged 18-60 years with previously untreated, non-keratinising stage III-IVB nasopharyngeal carcinoma; Karnofsky performance-status score of at least 70; and adequate haematological, renal, and hepatic function were randomly assigned (1:1) to receive intravenously either lobaplatin-based (lobaplatin 30 mg/m2 on days 1 and 22, and fluorouracil 800 mg/m2 on days 1-5 and 22-26 for two cycles) or cisplatin-based (cisplatin 100 mg/m2 on days 1 and 22, and fluorouracil 800 mg/m2 on days 1-5 and 22-26 for two cycles) induction chemotherapy, followed by concurrent lobaplatin-based (two cycles of intravenous lobaplatin 30 mg/m2 every 3 weeks plus intensity-modulated radiotherapy) or cisplatin-based (two cycles of intravenous cisplatin 100 mg/m2 every 3 weeks plus intensity-modulated radiotherapy) chemoradiotherapy. Total radiation doses of 68-70 Gy (for the sum of the volumes of the primary tumour and enlarged retropharyngeal nodes), 62-68 Gy (for the volume of clinically involved gross cervical lymph nodes), 60 Gy (for the high-risk target volume), and 54 Gy (for the low-risk target volume), were administered in 30-32 fractions, 5 days per week. Randomisation was done centrally at the clinical trial centre of Sun Yat-sen University Cancer Centre by means of computer-generated random number allocation with a block design (block size of four) stratified according to disease stage and treatment centre. Treatment assignment was known to both clinicians and patients. The primary endpoint was 5-year progression-free survival, analysed in both the intention-to-treat and per-protocol populations. If the upper limit of the 95% CI for the difference in 5-year progression-free survival between the lobaplatin-based and cisplatin-based groups did not exceed 10%, non-inferiority was met. Adverse events were analysed in all patients who received at least one cycle of induction chemotherapy. This trial is registered with the Chinese Clinical Trial Registry, ChiCTR-TRC-13003285 and is closed. FINDINGS: From June 7, 2013, to June 16, 2015, 515 patients were assessed for eligibility and 502 patients were enrolled: 252 were randomly assigned to the lobaplatin-based group and 250 to the cisplatin-based group. After a median follow-up of 75·3 months (IQR 69·9-81·1) in the intention-to-treat population, 5-year progression-free survival was 75·0% (95% CI 69·7-80·3) in the lobaplatin-based group and 75·5% (70·0 to 81·0) in the cisplatin-based group (hazard ratio [HR] 0·98, 95% CI 0·69-1·39; log-rank p=0·92), with a difference of 0·5% (95% CI -7·1 to 8·1; pnon-inferiority=0·0070). In the per-protocol population, the 5-year progression-free survival was 74·8% (95% CI 69·3 to 80·3) in the lobaplatin-based group and 76·4% (70·9 to 81·9) in the cisplatin-based group (HR 1·04, 95% CI 0·73 to 1·49; log-rank p=0·83), with a difference of 1·6% (-6·1 to 9·3; pnon-inferiority=0·016). 63 (25%) of 252 patients in the lobaplatin-based group and 63 (25%) of 250 patients in the cisplatin-based group had a progression-free survival event in the intention-to-treat population; 62 (25%) of 246 patients in the lobaplatin-based group and 58 (25%) of 237 patients in the cisplatin-based group had a progression-free survival event in the per-protocol population. The most common grade 3-4 adverse events were mucositis (102 [41%] of 252 in the lobaplatin-based group vs 99 [40%] of 249 in the cisplatin-based group), leucopenia (39 [16%] vs 56 [23%]), and neutropenia (25 [10%] vs 59 [24%]). No treatment-related deaths were reported. INTERPRETATION: Lobaplatin-based induction chemotherapy plus concurrent chemoradiotherapy resulted in non-inferior survival and fewer toxic effects than cisplatin-based therapy. The results of our trial indicate that lobaplatin-based induction chemotherapy plus concurrent chemoradiotherapy might be a promising alternative regimen to cisplatin-based treatment in patients with locoregional, advanced nasopharyngeal carcinoma. FUNDING: National Science and Technology Pillar Program, International Cooperation Project of Science and Technology Program of Guangdong Province, Planned Science and Technology Project of Guangdong Province, and Cultivation Foundation for the Junior Teachers at Sun Yat-sen University. TRANSLATION: For the Chinese translation of the abstract see Supplementary Materials section.

Development of a Prognostic Model to Identify the Suitable Definitive Radiation Therapy Candidates in de Novo Metastatic Nasopharyngeal Carcinoma: A Real-World Study.

PURPOSE: We aimed to develop an accurate prognostic model to identify suitable candidates for definitive radiation therapy (DRT) in addition to palliative chemotherapy (PCT) among patients with de novo metastatic nasopharyngeal carcinoma (mNPC). METHODS AND MATERIALS: Patients with de novo mNPC who received first-line PCT with or without DRT were included. Overall survival for patients who received PCT alone versus PCT plus DRT was estimated using inverse probability of treatment weighting-adjusted survival analyses. We developed and validated a prognostic model to predict survival and stratify risks in de novo mNPC. A model-based trees approach was applied to estimate stratified treatment effects using prognostic scores obtained from the prognostic model and to identify suitable DRT candidates. Dominance analysis was used to determine the relative importance of each predictor of receiving DRT. RESULTS: A total of 460 patients were enrolled; 244 received PCT plus DRT and 216 received PCT alone. The 6-month conditional landmark, inverse probability of treatment weighting-adjusted Cox regression analysis showed that PCT plus DRT was associated with a significant survival benefit (hazard ratio: 0.516; 95% confidence interval, 0.403-0.660; P < .001). A prognostic model based on 5 independent prognostic factors, including serum lactate dehydrogenase, number of metastatic sites, presence of liver metastasis, posttreatment Epstein-Barr virus DNA level, and response of metastases to chemotherapy was developed and subsequently validated. Prognostic scores obtained from the prognostic model were used for risk stratification and efficacy estimation. High-risk patients identified using the proposed model would not benefit from additional DRT, whereas low-risk patients experienced significant survival benefits. Socioeconomic factors, including insurance status and education level, played an important role in receipt of DRT. CONCLUSIONS: Additional DRT after PCT was associated with increased overall survival in patients with de novo mNPC, especially low-risk patients identified with a newly developed prognostic model.

A Prognostic Predictive System Based on Deep Learning for Locoregionally Advanced Nasopharyngeal Carcinoma.

BACKGROUND: Images from magnetic resonance imaging (MRI) are crucial unstructured data for prognostic evaluation in nasopharyngeal carcinoma (NPC). We developed and validated a prognostic system based on the MRI features and clinical data of locoregionally advanced NPC (LA-NPC) patients to distinguish low-risk patients with LA-NPC for whom concurrent chemoradiotherapy (CCRT) is sufficient. METHODS: This multicenter, retrospective study included 3444 patients with LA-NPC from January 1, 2010, to January 31, 2017. A 3-dimensional convolutional neural network was used to learn the image features from pretreatment MRI images. An eXtreme Gradient Boosting model was trained with the MRI features and clinical data to assign an overall score to each patient. Comprehensive evaluations were implemented to assess the performance of the predictive system. We applied the overall score to distinguish high-risk patients from low-risk patients. The clinical benefit of induction chemotherapy (IC) was analyzed in each risk group by survival curves. RESULTS: We constructed a prognostic system displaying a concordance index of 0.776 (95% confidence interval [CI] = 0.746 to 0.806) for the internal validation cohort and 0.757 (95% CI = 0.695 to 0.819), 0.719 (95% CI = 0.650 to 0.789), and 0.746 (95% CI = 0.699 to 0.793) for the 3 external validation cohorts, which presented a statistically significant improvement compared with the conventional TNM staging system. In the high-risk group, patients who received induction chemotherapy plus CCRT had better outcomes than patients who received CCRT alone, whereas there was no statistically significant difference in the low-risk group. CONCLUSIONS: The proposed framework can capture more complex and heterogeneous information to predict the prognosis of patients with LA-NPC and potentially contribute to clinical decision making.

The diagnostic performance of quantitative mapping in breast cancer patients: a preliminary study using synthetic MRI.

BACKGROUND: Previous studies have indicated that quantitative MRI (qMR) is beneficial for diagnosis of breast cancer. As a novel qMR technology, synthetic MRI (syMRI) may be advantageous by offering simultaneous generation of T1 and T2 mapping in one scan within a few minutes and without concern to the deposition of the gadolinium contrast agent in cell nucleus. In this study, the potential of quantitative mapping derived from Synthetic MRI (SyMRI) to diagnose breast cancer was investigated. METHODS: From April 2018 to May 2019, a total of 87 patients with suspicious breast lesions underwent both conventional and SyMRI before treatment. The quantitative metrics derived from SyMRI, including T1 and T2 values, were measured in breast lesions. The diagnostic performance of SyMRI was evaluated with unpaired Student's t-tests, receiver operating characteristic curve analysis and multivariate logistic regression analysis. The AUCs of quantitative values were compared using Delong test. RESULTS: Among 77 patients who met the inclusion criteria, 48 were diagnosed with histopathological confirmed breast cancers, and the rest had benign lesions. The breast cancers showed significantly higher T1 (1611.61 ± 215.88 ms) values and lower T2 (80.93 ± 7.51 ms) values than benign lesions. The area under the ROC curve (AUC) values were 0.931 (95% CI: 0.874-0.989) and 0.883 (95% CI: 0.810-0.956) for T1 and T2 maps, respectively, in diagnostic discrimination between breast cancers and benign lesions. A slightly increased AUC of 0.978 (95% CI: 0.915-0.993) was achieved by combining those two relaxation-based quantitative metrics. CONCLUSION: In conclusion, our preliminary study showed that the quantitative T1 and T2 values obtained by SyMRI could distinguish effectively between benign and malignant breast lesions, and T1 relaxation time showed the highest diagnostic efficiency. Furthermore, combining the two quantitative relaxation metrics further improved their diagnostic performance.

Prognostic and Treatment Guiding Significance of MRI-Based Tumor Burden Features and Nodal Necrosis in Nasopharyngeal Carcinoma.

We aimed to develop a nomogram integrating MRI-based tumor burden features (MTBF), nodal necrosis, and some clinical factors to forecast the distant metastasis-free survival (DMFS) of patients suffering from non-metastatic nasopharyngeal carcinoma (NPC). A total of 1640 patients treated at Sun Yat-sen University Cancer Center (Guangzhou, China) from 2011 to 2016 were enrolled, among which 1148 and 492 patients were randomized to a training cohort and an internal validation cohort, respectively. Additionally, 200 and 257 patients were enrolled in the Foshan and Dongguan validation cohorts, respectively, which served as independent external validation cohorts. The MTBF were developed from the stepwise regression of six multidimensional tumor burden variables, based on which we developed a nomogram also integrating nodal necrosis and clinical features. This model divided the patients into high- and low-risk groups by an optimal cutoff. Compared with those of patients in the low-risk group, the DMFS [hazard ratio (HR): 4.76, 95% confidence interval (CI): 3.39-6.69; p < 0.0001], and progression-free survival (PFS; HR: 4.11, 95% CI: 3.13-5.39; p < 0.0001) of patients in the high-risk group were relatively poor. Furthermore, in the training cohort, the 3-year DMFS of high-risk patients who received induction chemotherapy (ICT) combined with concurrent chemoradiotherapy (CCRT) was better than that of those who were treated with CCRT alone (p = 0.0340), whereas low-risk patients who received ICT + CCRT had a similar DMFS to those who only received CCRT. The outcomes we obtained were all verified in the three validation cohorts. The survival model can be used as a reliable prognostic tool for NPC patients and is helpful to determine patients who will benefit from ICT.

Deep learning for risk prediction in patients with nasopharyngeal carcinoma using multi-parametric MRIs.

BACKGROUND: Magnetic resonance images (MRI) is the main diagnostic tool for risk stratification and treatment decision in nasopharyngeal carcinoma (NPC). However, the holistic feature information of multi-parametric MRIs has not been fully exploited by clinicians to accurately evaluate patients. OBJECTIVE: To help clinicians fully utilize the missed information to regroup patients, we built an end-to-end deep learning model to extract feature information from multi-parametric MRIs for predicting and stratifying the risk scores of NPC patients. METHODS: In this paper, we proposed an end-to-end multi-modality deep survival network (MDSN) to precisely predict the risk of disease progression of NPC patients. Extending from 3D dense net, this proposed MDSN extracted deep representation from multi-parametric MRIs (T1w, T2w, and T1c). Moreover, deep features and clinical stages were integrated through MDSN to more accurately predict the overall risk score (ORS) of individual NPC patient. RESULT: A total of 1,417 individuals treated between January 2012 and December 2014 were included for training and validating the end-to-end MDSN. Results were then tested in a retrospective cohort of 429 patients included in the same institution. The C-index of the proposed method with or without clinical stages was 0.672 and 0.651 on the test set, respectively, which was higher than the that of the stage grouping (0.610). CONCLUSIONS: The C-index of the model which integrated clinical stages with deep features is 0.062 higher than that of stage grouping alone (0.672 vs 0.610). We conclude that features extracted from multi-parametric MRIs based on MDSN can well assist the clinical stages in regrouping patients.

Prognostic value and the potential role of treatment options for cervical lymph node necrosis in nasopharyngeal carcinoma.

OBJECTIVE: There were few studies focused on the cervical lymph necrosis (CNN) of nasopharyngeal carcinoma (NPC) patients to develop a nomogram and guide the treatment decision at the era of intensity modulated radiation therapy (IMRT). MATERIAL AND METHODS: The prognostic accuracy of CNN in the training cohort (n = 1940) was validated in Guangzhou internal validation cohort (n = 832) and two external validation cohorts (Dongguan, n = 232; Foshan, n = 134). RESULTS: The primary end point was progression-free survival (PFS), calculated using the Kaplan-Meier method. After a median 60.0 months' follow-up, patients with CNN in the training cohort had worse 5-year PFS (70.8% vs. 89.1%, P < 0.001) than patients without CNN, which was validated in the validation cohorts. The nomogram based on CNN predicted an individual PFS risk (training: C-index 0.733; Guangzhou validation: C-index 0.736; Foshan: C-index 0.722; Dongguan: C-index 0.756). Stage N2 patients in the CNN group and stage IV patients no matter the status of CNN, PFS was better with induction chemotherapy (ICT) and CCRT than CCRT (P < 0.05). CONCLUSION: Taken together, CNN reliably predicts survival risk in NPC patients. N2 patients in the CNN group and stage IV patients may receive survival benefit from ICT.

Development of a self-constrained 3D DenseNet model in automatic detection and segmentation of nasopharyngeal carcinoma using magnetic resonance images.

OBJECTIVES: We aimed to develop a dual-task model to detect and segment nasopharyngeal carcinoma (NPC) automatically in magnetic resource images (MRI) based on deep learning method, since the differential diagnosis of NPC and atypical benign hyperplasia was difficult and the radiotherapy target contouring of NPC was labor-intensive. MATERIALS AND METHODS: A self-constrained 3D DenseNet (SC-DenseNet) architecture was improved using separated training and validation sets. A total of 4100 individuals were finally enrolled and split into the training, validation and test sets at a proximate ratio of 8:1:1 using simple randomization. The diagnostic metrics of the established model against experienced radiologists was compared in the test set. The dice similarity coefficient (DSC) of manual and model-defined tumor region was used to evaluate the efficacy of segmentation. RESULTS: Totally, 3142 nasopharyngeal carcinoma (NPC) and 958 benign hyperplasia were included. The SC-DenseNet model showed encouraging performance in detecting NPC, attained a higher overall accuracy, sensitivity and specificity than those of the experienced radiologists (97.77% vs 95.87%, 99.68% vs 99.24% and 91.67% vs 85.21%, respectively). Moreover, the model also exhibited promising performance in automatic segmentation of tumor region in NPC, with an average DSC at 0.77 ± 0.07 in the test set. CONCLUSIONS: The SC-DenseNet model showed competence in automatic detection and segmentation of NPC in MRI, indicating the promising application value as an assistant tool in clinical practice, especially in screening project.

MRI-detected residual retropharyngeal lymph node after intensity-modulated radiotherapy in nasopharyngeal carcinoma: Prognostic value and a nomogram for the pretherapy prediction of it.

BACKGROUND AND PURPOSE: To evaluate the prognostic value of MRI-detected residual retropharyngeal lymph node (RRLN) at three months after intensity-modulated radiotherapy (IMRT) in patients with nasopharyngeal carcinoma (NPC) and second, to establish a nomogram for the pretherapy prediction of RRLN. MATERIALS AND METHODS: We included 1103 patients with NPC from two hospitals (Sun Yat-Sen University Cancer Center [SYSUCC, n = 901] and Dongguan People's Hospital [DGPH, n = 202]). We evaluated the prognostic value of RRLN using Cox regression model in SYSUCC cohort. We developed a nomogram for the pretherapy prediction of RRLN using logistic regression model in SYSUCC training cohort (n = 645). We assessed the performance of this nomogram in an internal validation cohort (SYSUCC validation cohort, n = 256) and an external independent cohort (DGPH validation cohort, n = 202). RESULTS: RRLN was an independent prognostic factor for OS (HR 2.08, 95% CI 1.32-3.29), DFS (HR 2.45, 95% CI 1.75-3.42), DMFS (HR 3.31, 95% CI 2.15-5.09), and LRRFS (HR 3.04, 95% CI 1.70-5.42). We developed a nomogram based on baseline Epstein-Barr virus DNA level and three RLN status-related features (including minimum axial diameter, extracapsular nodal spread, and laterality) that predicted an individual's risk of RRLN. Our nomogram showed good discrimination in the training cohort (C-index = 0.763). The favorable performance of this nomogram was confirmed in the internal and external validation cohorts. CONCLUSION: MRI-detected RRLN at three months after IMRT was an unfavorable prognostic factor for patients with NPC. We developed and validated an easy-to-use nomogram for the pretherapy prediction of RRLN.

Prognostic value of early radiological response to first-line platinum-containing chemotherapy in patients with metastatic nasopharyngeal carcinoma.

BACKGROUND: To explore the prognostic value of early radiological response (ERR) to first-line platinum-containing chemotherapy in patients with metastatic nasopharyngeal carcinoma (mNPC), as well as its correlation with the best radiological response (BRR). PATIENTS AND METHODS: A total of 756 mNPC patients with measurable lesions who received first-line platinum-containing chemotherapy were enrolled in this study. ERR was defined as complete or partial response after 6 weeks of chemotherapy according to the Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. We performed survival analyses according to the radiological response after repeated chemotherapy. Log-rank test and Cox regression were used to analyze the survival data. RESULTS: About 470 patients achieved ERR and 78 patients achieved subsequent response (objective response after repeated chemotherapy). ERR patients had better OS (P < .001, median OS: 34.3 vs 22.2 months) and PFS (P < .001, median PFS: 10.2 vs 7.4 months) than non-ERR ones. ERR (OS: HR = 0.591, 95% CI, 0.495-0.705, P < .001, PFS: HR = 0.586, 95% CI, 0.500-0.686, P < .001) was independently prolonged survival compared with non-ERR ones. Besides, ERR was significantly correlated with the BRR (Kappa: 0.73; Pearson: 0.74, P < .001), and had significantly longer OS and PFS than patients with subsequent response, respectively. CONCLUSION: ERR is an independent prognostic factor in determining survival in mNPC patients received first-line platinum-containing chemotherapy, which may be a more sensitive predictor to assess overall efficacy of systemic treatment than BRR in mNPC. Prospective validation studies are needed.

Deep learning pathological microscopic features in endemic nasopharyngeal cancer: Prognostic value and protentional role for individual induction chemotherapy.

BACKGROUND: To explore the prognostic value and the role for treatment decision of pathological microscopic features in patients with nasopharyngeal carcinoma (NPC) using the method of deep learning. METHODS: The pathological microscopic features were extracted using the software QuPath (version 0.1.3. Queen's University) in the training cohort (Guangzhou training cohort, n = 843). We used the neural network DeepSurv to analyze the pathological microscopic features (DSPMF) and then classified patients into high-risk and low-risk groups through the time-dependent receiver operating characteristic (ROC). The prognosis accuracy of the pathological feature was validated in a validation cohort (n = 212). The primary endpoint was progression-free survival (PFS). RESULTS: We found 429 pathological microscopic features in the H&E image. Patients with high-risk scores in the training cohort had shorter 5-year PFS (HR 10.03, 6.06-16.61; P < .0001). The DSPMF (C-index: 0.723) had the higher C-index than the EBV DNA (C-index: 0.612) copies and the N stage (C-index: 0.593). Furthermore, induction chemotherapy (ICT) plus concomitant chemoradiotherapy (CCRT) had better 5-year PFS to those received CCRT (P < .0001) in the high-risk group. CONCLUSION: The DSPMF is a reliable prognostic tool for survival risk in patients with NPC and might be able to guide the treatment decision.

A deep survival analysis method based on ranking.

Survival analyses of populations and the establishment of prognoses for individual patients are important activities in the practice of medicine. Standard survival models, such as the Cox proportional hazards model, require extensive feature engineering or prior knowledge to model at an individual level. Some survival analysis models can avoid these problems by using machine learning extended the CPH model, and higher performance has been reported. In this paper, we propose an innovative loss function that is defined as the sum of an extended mean squared error loss and a pairwise ranking loss based on ranking information on survival data. We apply this loss function to optimize a deep feed-forward neural network (RankDeepSurv), which can be used to model survival data. We demonstrate that the performance of our model, RankDeepSurv, is superior to that of other state-of-the-art survival models based on an analysis of 4 public medical clinical datasets. When modelling the prognosis of nasopharyngeal carcinoma (NPC), RankDeepSurv achieved better prognostic accuracy than the CPH established by clinical experts. The difference between high and low risk groups in the RankDeepSurv model is greater than the difference in the CPH. The results show that our method has considerable potential to model survival data in medical settings.

Development and validation of an endoscopic images-based deep learning model for detection with nasopharyngeal malignancies.

BACKGROUND: Due to the occult anatomic location of the nasopharynx and frequent presence of adenoid hyperplasia, the positive rate for malignancy identification during biopsy is low, thus leading to delayed or missed diagnosis for nasopharyngeal malignancies upon initial attempt. Here, we aimed to develop an artificial intelligence tool to detect nasopharyngeal malignancies under endoscopic examination based on deep learning. METHODS: An endoscopic images-based nasopharyngeal malignancy detection model (eNPM-DM) consisting of a fully convolutional network based on the inception architecture was developed and fine-tuned using separate training and validation sets for both classification and segmentation. Briefly, a total of 28,966 qualified images were collected. Among these images, 27,536 biopsy-proven images from 7951 individuals obtained from January 1st, 2008, to December 31st, 2016, were split into the training, validation and test sets at a ratio of 7:1:2 using simple randomization. Additionally, 1430 images obtained from January 1st, 2017, to March 31st, 2017, were used as a prospective test set to compare the performance of the established model against oncologist evaluation. The dice similarity coefficient (DSC) was used to evaluate the efficiency of eNPM-DM in automatic segmentation of malignant area from the background of nasopharyngeal endoscopic images, by comparing automatic segmentation with manual segmentation performed by the experts. RESULTS: All images were histopathologically confirmed, and included 5713 (19.7%) normal control, 19,107 (66.0%) nasopharyngeal carcinoma (NPC), 335 (1.2%) NPC and 3811 (13.2%) benign diseases. The eNPM-DM attained an overall accuracy of 88.7% (95% confidence interval (CI) 87.8%-89.5%) in detecting malignancies in the test set. In the prospective comparison phase, eNPM-DM outperformed the experts: the overall accuracy was 88.0% (95% CI 86.1%-89.6%) vs. 80.5% (95% CI 77.0%-84.0%). The eNPM-DM required less time (40 s vs. 110.0 ± 5.8 min) and exhibited encouraging performance in automatic segmentation of nasopharyngeal malignant area from the background, with an average DSC of 0.78 ± 0.24 and 0.75 ± 0.26 in the test and prospective test sets, respectively. CONCLUSIONS: The eNPM-DM outperformed oncologist evaluation in diagnostic classification of nasopharyngeal mass into benign versus malignant, and realized automatic segmentation of malignant area from the background of nasopharyngeal endoscopic images.

The Clinical Effect of Metformin on the Survival of Lung Cancer Patients with Diabetes: A Comprehensive Systematic Review and Meta-analysis of Retrospective Studies.

Preclinical investigations have revealed an anti-cancer effect of metformin. Several studies of metformin treatment have demonstrated the improved clinical outcomes of lung cancer patients with diabetes; however, the results have been inconsistent among studies. Our systematic review and meta-analysis aimed to summarize the up-to-date effects of metformin on diabetic lung cancer patients. A systematic search was performed for studies published. Then, these studies were evaluated for inclusion, and relevant data was extracted. The summary risk estimates for the associations of metformin treatment with overall survival (OS) and progression-free survival (PFS) were analyzed using random/fixed-effects models. Analyses stratified by histological type were also conducted. Based on the 10 studies included in our analysis, metformin treatment was found to significantly improve survival, corresponding to reductions of 23% and 47% in OS [hazard ratio (HR)=0.77, 95% confidence interval (95%CI)=0.66-0.9, p=0.001] and PFS (HR=0.53, 95%CI=0.41-0.68, p<0.001), respectively. In addition, significant improvements in the OS for non-small cell lung cancer (NSCLC) (HR=0.77, 95%CI=0.71-0.84, p=0.002) and small cell lung cancer (SCLC) (HR=0.52, 95%CI=0.29-0.91, p=0.022) were observed in association with metformin treatment in analysis stratified by histological type. This stratified analysis also revealed a significant improvement in PFS for both NSCLC (HR=0.53, 95%CI=0.39-0.71, p<0.001) and SCLC (HR=0.54, 95%CI=0.34-0.84, p=0.007). We found that metformin treatment significantly improved the OS and PFS of diabetic lung cancer patients, and our findings suggest that metformin might be an effective treatment option for diabetic patients with lung cancer.

The Effect of Metformin on Mortality Among Diabetic Cancer Patients: A Systematic Review and Meta-analysis.

BACKGROUND: Most data suggest that cancer patients with diabetes have worse outcomes, which may be reversed with metformin. Metformin might modulate the clinical outcomes of diabetic cancer patients. We performed a systematic review and meta-analysis based on published studies over the past five years to summarize the effects of metformin on diabetic cancer patients. METHODS: We systematically searched for studies that were published over the past five years. Then, we evaluated these studies for inclusion and extracted the relevant data. The summary risk estimates for the association between metformin treatment and all-cause mortality (ACM) and cancer-specific mortality (CSM) were analyzed using random or fixed-effects models. Stratified analyses by cancer site and country were also conducted. RESULTS: Based on the 42 studies included in our analysis (37 015 diabetic cancer patients), we found a significant benefit associated with metformin treatment on survival corresponding to 27% and 26% reductions in ACM (hazard ratio [HR] = 0.73, 95% confidence interval [CI] = 0.68 to 0.79, P < .001) and CSM (HR = 0.74, 95% CI = 0.64 to 0.86, P < .001), respectively. The ACM rates for colorectal cancer, endometrial cancer, breast cancer, prostate cancer, and ovarian cancer showed significant benefits associated with metformin treatment in our stratified analyses by cancer site. Stratified analyses by cancer site also showed a significant reduction in CSM for breast cancer. This association between metformin treatment and reduced CSM for diabetic breast cancer patients was also observed in our country subgroup analyses. CONCLUSIONS: We found an association between metformin exposure and reduced ACM and CSM in diabetic patients with cancer. Our findings suggest that metformin treatment could be an effective treatment option for diabetic cancer patients.