Development and external validation of the multichannel deep learning model based on unenhanced CT for differentiating fat-poor angiomyolipoma from renal cell carcinoma: a two-center retrospective study.

PURPOSE: There are undetectable levels of fat in fat-poor angiomyolipoma. Thus, it is often misdiagnosed as renal cell carcinoma. We aimed to develop and evaluate a multichannel deep learning model for differentiating fat-poor angiomyolipoma (fp-AML) from renal cell carcinoma (RCC). METHODS: This two-center retrospective study included 320 patients from the First Affiliated Hospital of Sun Yat-Sen University (FAHSYSU) and 132 patients from the Sun Yat-Sen University Cancer Center (SYSUCC). Data from patients at FAHSYSU were divided into a development dataset (n = 267) and a hold-out dataset (n = 53). The development dataset was used to obtain the optimal combination of CT modality and input channel. The hold-out dataset and SYSUCC dataset were used for independent internal and external validation, respectively. RESULTS: In the development phase, models trained on unenhanced CT images performed significantly better than those trained on enhanced CT images based on the fivefold cross-validation. The best patient-level performance, with an average area under the receiver operating characteristic curve (AUC) of 0.951 ± 0.026 (mean ± SD), was achieved using the "unenhanced CT and 7-channel" model, which was finally selected as the optimal model. In the independent internal and external validation, AUCs of 0.966 (95% CI 0.919-1.000) and 0.898 (95% CI 0.824-0.972), respectively, were obtained using the optimal model. In addition, the performance of this model was better on large tumors (≥ 40 mm) in both internal and external validation. CONCLUSION: The promising results suggest that our multichannel deep learning classifier based on unenhanced whole-tumor CT images is a highly useful tool for differentiating fp-AML from RCC.

Multimodal recurrence scoring system for prediction of clear cell renal cell carcinoma outcome: a discovery and validation study.

BACKGROUND: Improved markers for predicting recurrence are needed to stratify patients with localised (stage I-III) renal cell carcinoma after surgery for selection of adjuvant therapy. We developed a novel assay integrating three modalities-clinical, genomic, and histopathological-to improve the predictive accuracy for localised renal cell carcinoma recurrence. METHODS: In this retrospective analysis and validation study, we developed a histopathological whole-slide image (WSI)-based score using deep learning allied to digital scanning of conventional haematoxylin and eosin-stained tumour tissue sections, to predict tumour recurrence in a development dataset of 651 patients with distinctly good or poor disease outcome. The six single nucleotide polymorphism-based score, which was detected in paraffin-embedded tumour tissue samples, and the Leibovich score, which was established using clinicopathological risk factors, were combined with the WSI-based score to construct a multimodal recurrence score in the training dataset of 1125 patients. The multimodal recurrence score was validated in 1625 patients from the independent validation dataset and 418 patients from The Cancer Genome Atlas set. The primary outcome measured was the recurrence-free interval (RFI). FINDINGS: The multimodal recurrence score had significantly higher predictive accuracy than the three single-modal scores and clinicopathological risk factors, and it precisely predicted the RFI of patients in the training and two validation datasets (areas under the curve at 5 years: 0·825-0·876 vs 0·608-0·793; p<0·05). The RFI of patients with low stage or grade is usually better than that of patients with high stage or grade; however, the RFI in the multimodal recurrence score-defined high-risk stage I and II group was shorter than in the low-risk stage III group (hazard ratio [HR] 4·57, 95% CI 2·49-8·40; p<0·0001), and the RFI of the high-risk grade 1 and 2 group was shorter than in the low-risk grade 3 and 4 group (HR 4·58, 3·19-6·59; p<0·0001). INTERPRETATION: Our multimodal recurrence score is a practical and reliable predictor that can add value to the current staging system for predicting localised renal cell carcinoma recurrence after surgery, and this combined approach more precisely informs treatment decisions about adjuvant therapy. FUNDING: National Natural Science Foundation of China, and National Key Research and Development Program of China.

Rod-like Cellulose Regenerated by Bottom-Up Assembly in Natural Rubber Latex and Its Reinforcement.

As a renewable biomass material, nano-cellulose has been investigated as a reinforcing filler in rubber composites but has seen little success because of its strong inclination towards aggregating. Here, a bottom-up self-assembly approach was proposed by regenerating cellulose crystals from a mixture of cellulose solution and natural rubber (NR) latex. Different co-coagulants of both cellulose solution and natural rubber latex were added to break the dissolution equilibrium and in-situ regenerate cellulose in the NR matrix. The SEM images showed that the sizes and morphologies of regenerated cellulose (RC) varied greatly with the addition of different co-coagulants. Only when a 5 wt% acetic acid aqueous solution was used, the RC particles showed an ideal rod-like structure with small sizes of about 100 nm in diameter and 1.0 µm in length. The tensile test showed that rod-like RC (RRC)-endowed NR vulcanizates with pronounced reinforcement had a drastic upturn in stress after stretching to 200% strain. The results of XRD and the Mullins effect showed that this drastic upturn in stress was mainly attributed to the formation of rigid RRC-RRC networks during stretching instead of the strain-induced crystallization of NR. This bottom-up approach provided a simple way to ensure the effective utilization of cellulosic materials in the rubber industry.

The size effect in adhesive contact on a gradient nanostructured coating.

The adhesive contact problem between a rigid cylindrical punch and a gradient nanostructured (GNS) coating is investigated by considering the size effect. The laminated plate model is applied to characterize the material properties of a GNS coating in plane strain couple stress elasticity. By using the Fourier integral transform and transfer matrix method, the governing integral equation(s) for the two-dimensional adhesive contact problem are obtained. Numerically calculated results are presented to analyse the effect of characteristic material length, the adhesion parameter and nonhomogeneous parameters on the mechanical response of the GNS coating for the adhesive contact problem. We explore the nanoscale contact of a GNS coating with shear modulus varying as a function of depth according to an exponential function or the power-law function. The present results provide a way to improve the contact deformation and damage to nanoelectromechanical systems by adjusting the gradient index of the GNS coating.

Predictive value of single-nucleotide polymorphism signature for recurrence in localised renal cell carcinoma: a retrospective analysis and multicentre validation study.

BACKGROUND: Identification of high-risk localised renal cell carcinoma is key for the selection of patients for adjuvant treatment who are at truly higher risk of reccurrence. We developed a classifier based on single-nucleotide polymorphisms (SNPs) to improve the predictive accuracy for renal cell carcinoma recurrence and investigated whether intratumour heterogeneity affected the precision of the classifier. METHODS: In this retrospective analysis and multicentre validation study, we used paraffin-embedded specimens from the training set of 227 patients from Sun Yat-sen University (Guangzhou, Guangdong, China) with localised clear cell renal cell carcinoma to examine 44 potential recurrence-associated SNPs, which were identified by exploratory bioinformatics analyses of a genome-wide association study from The Cancer Genome Atlas (TCGA) Kidney Renal Clear Cell Carcinoma (KIRC) dataset (n=114, 906 600 SNPs). We developed a six-SNP-based classifier by use of LASSO Cox regression, based on the association between SNP status and patients' recurrence-free survival. Intratumour heterogeneity was investigated from two other regions within the same tumours in the training set. The six-SNP-based classifier was validated in the internal testing set (n=226), the independent validation set (Chinese multicentre study; 428 patients treated between Jan 1, 2004 and Dec 31, 2012, at three hospitals in China), and TCGA set (441 retrospectively identified patients who underwent resection between 1998 and 2010 for localised clear cell renal cell carcinoma in the USA). The main outcome was recurrence-free survival; the secondary outcome was overall survival. FINDINGS: Although intratumour heterogeneity was found in 48 (23%) of 206 cases in the internal testing set with complete SNP information, the predictive accuracy of the six-SNP-based classifier was similar in the three different regions of the training set (areas under the curve [AUC] at 5 years: 0·749 [95% CI 0·660-0·826] in region 1, 0·734 [0·651-0·814] in region 2, and 0·736 [0·649-0·824] in region 3). The six-SNP-based classifier precisely predicted recurrence-free survival of patients in three validation sets (hazard ratio [HR] 5·32 [95% CI 2·81-10·07] in the internal testing set, 5·39 [3·38-8·59] in the independent validation set, and 4·62 [2·48-8·61] in the TCGA set; all p<0·0001), independently of patient age or sex and tumour stage, grade, or necrosis. The classifier and the clinicopathological risk factors (tumour stage, grade, and necrosis) were combined to construct a nomogram, which had a predictive accuracy significantly higher than that of each variable alone (AUC at 5 years 0·811 [95% CI 0·756-0·861]). INTERPRETATION: Our six-SNP-based classifier could be a practical and reliable predictor that can complement the existing staging system for prediction of localised renal cell carcinoma recurrence after surgery, which might enable physicians to make more informed treatment decisions about adjuvant therapy. Intratumour heterogeneity does not seem to hamper the accuracy of the six-SNP-based classifier as a reliable predictor of recurrence. The classifier has the potential to guide treatment decisions for patients at differing risks of recurrence. FUNDING: National Key Research and Development Program of China, National Natural Science Foundation of China, Guangdong Provincial Science and Technology Foundation of China, and Guangzhou Science and Technology Foundation of China.

Morphological Type Identification of Self-Incompatibility in Dendrobium and Its Phylogenetic Evolution Pattern.

Self-incompatibility (SI) is a type of reproductive barrier within plant species and is one of the mechanisms for the formation and maintenance of the high diversity and adaptation of angiosperm species. Approximately 40% of flowering plants are SI species, while only 10% of orchid species are self-incompatible. Intriguingly, as one of the largest genera in Orchidaceae, 72% of Dendrobium species are self-incompatible, accounting for nearly half of the reported SI species in orchids, suggesting that SI contributes to the high diversity of orchid species. However, few studies investigating SI in Dendrobium have been published. This study aimed to address the following questions: (1) How many SI phenotypes are in Dendrobium, and what are they? (2) What is their distribution pattern in the Dendrobium phylogenetic tree? We investigated the flowering time, the capsule set rate, and the pollen tube growth from the representative species of Dendrobium after artificial pollination and analysed their distribution in the Asian Dendrobium clade phylogenetic tree. The number of SI phenotypes exceeded our expectations. The SI type of Dendrobium chrysanthum was the primary type in the Dendrobium SI species. We speculate that there are many different SI determinants in Dendrobium that have evolved recently and might be specific to Dendrobium or Orchidaceae. Overall, this work provides new insights and a comprehensive understanding of Dendrobium SI.

Lack of S-RNase-Based Gametophytic Self-Incompatibility in Orchids Suggests That This System Evolved after the Monocot-Eudicot Split.

Self-incompatibility (SI) is found in approximately 40% of flowering plant species and at least 100 families. Although orchids belong to the largest angiosperm family, only 10% of orchid species present SI and have gametophytic SI (GSI). Furthermore, a majority (72%) of Dendrobium species, which constitute one of the largest Orchidaceae genera, show SI and have GSI. However, nothing is known about the molecular mechanism of GSI. The S-determinants of GSI have been well characterized at the molecular level in Solanaceae, Rosaceae, and Plantaginaceae, which use an S-ribonuclease (S-RNase)-based system. Here, we investigate the hypothesis that Orchidaceae uses a similar S-RNase to those described in Rosaceae, Solanaceae, and Plantaginaceae SI species. In this study, two SI species (Dendrobium longicornu and D. chrysanthum) were identified using fluorescence microscopy. Then, the S-RNase- and SLF-interacting SKP1-like1 (SSK1)-like genes present in their transcriptomes and the genomes of Phalaenopsis equestris, D. catenatum, Vanilla shenzhenica, and Apostasia shenzhenica were investigated. Sequence, phylogenetic, and tissue-specific expression analyses revealed that none of the genes identified was an S-determinant, suggesting that Orchidaceae might have a novel SI mechanism. The results also suggested that RNase-based GSI might have evolved after the split of monocotyledons (monocots) and dicotyledons (dicots) but before the split of Asteridae and Rosidae. This is also the first study to investigate S-RNase-based GSI in monocots. However, studies on gene identification, differential expression, and segregation analyses in controlled crosses are needed to further evaluate the genes with high expression levels in GSI tissues.

A scoring system based on artificial neural network for predicting 10-year survival in stage II A colon cancer patients after radical surgery.

Nearly 20% patients with stage II A colon cancer will develop recurrent disease post-operatively. The present study aims to develop a scoring system based on Artificial Neural Network (ANN) model for predicting 10-year survival outcome. The clinical and molecular data of 117 stage II A colon cancer patients from Sun Yat-sen University Cancer Center were used for training set and test set; poor pathological grading (score 49), reduced expression of TGFBR2 (score 33), over-expression of TGF-ß (score 45), MAPK (score 32), pin1 (score 100), ß-catenin in tumor tissue (score 50) and reduced expression of TGF-ß in normal mucosa (score 22) were selected as the prognostic risk predictors. According to the developed scoring system, the patients were divided into 3 subgroups, which were supposed with higher, moderate and lower risk levels. As a result, for the 3 subgroups, the 10-year overall survival (OS) rates were 16.7%, 62.9% and 100% (P < 0.001); and the 10-year disease free survival (DFS) rates were 16.7%, 61.8% and 98.8% (P < 0.001) respectively. It showed that this scoring system for stage II A colon cancer could help to predict long-term survival and screen out high-risk individuals for more vigorous treatment.

A CpG-methylation-based assay to predict survival in clear cell renal cell carcinoma.

Clear cell renal cell carcinomas (ccRCCs) display divergent clinical behaviours. Molecular markers might improve risk stratification of ccRCC. Here we use, based on genome-wide CpG methylation profiling, a LASSO model to develop a five-CpG-based assay for ccRCC prognosis that can be used with formalin-fixed paraffin-embedded specimens. The five-CpG-based classifier was validated in three independent sets from China, United States and the Cancer Genome Atlas data set. The classifier predicts the overall survival of ccRCC patients (hazard ratio=2.96-4.82; P=3.9 × 10(-6)-2.2 × 10(-9)), independent of standard clinical prognostic factors. The five-CpG-based classifier successfully categorizes patients into high-risk and low-risk groups, with significant differences of clinical outcome in respective clinical stages and individual 'stage, size, grade and necrosis' scores. Moreover, methylation at the five CpGs correlates with expression of five genes: PITX1, FOXE3, TWF2, EHBP1L1 and RIN1. Our five-CpG-based classifier is a practical and reliable prognostic tool for ccRCC that can add prognostic value to the staging system.