Evaluating the Diagnostic Performance of Large Language Models on Complex Multimodal Medical Cases.

Large language models showed interpretative reasoning in solving diagnostically challenging medical cases.

Construction of a radiogenomic association map of pancreatic ductal adenocarcinoma.

BACKGROUND: Pancreatic adenocarcinoma (PDAC) persists as a malignancy with high morbidity and mortality that can benefit from new means to characterize and detect these tumors, such as radiogenomics. In order to address this gap in the literature, constructed a transcriptomic-CT radiogenomic (RG) map for PDAC. METHODS: In this Institutional Review Board approved study, a cohort of subjects (n = 50) with gene expression profile data paired with histopathologically confirmed resectable or borderline resectable PDAC were identified. Studies with pre-operative contrast-enhanced CT images were independently assessed for a set of 88 predefined imaging features. Microarray gene expression profiling was then carried out on the histopathologically confirmed pancreatic adenocarcinomas and gene networks were constructed using Weighted Gene Correlation Network Analysis (WCGNA) (n = 37). Data were analyzed with bioinformatics analyses, multivariate regression-based methods, and Kaplan-Meier survival analyses. RESULTS: Survival analyses identified multiple features of interest that were significantly associated with overall survival, including Tumor Height (P = 0.014), Tumor Contour (P = 0.033), Tumor-stroma Interface (P = 0.014), and the Tumor Enhancement Ratio (P = 0.047). Gene networks for these imaging features were then constructed using WCGNA and further annotated according to the Gene Ontology (GO) annotation framework for a biologically coherent interpretation of the imaging trait-associated gene networks, ultimately resulting in a PDAC RG CT-transcriptome map composed of 3 stage-independent imaging traits enriched in metabolic processes, telomerase activity, and podosome assembly (P < 0.05). CONCLUSIONS: A CT-transcriptomic RG map for PDAC composed of semantic and quantitative traits with associated biology processes predictive of overall survival, was constructed, that serves as a reference for further mechanistic studies for non-invasive phenotyping of pancreatic tumors.

Multi-center validation of an artificial intelligence system for detection of COVID-19 on chest radiographs in symptomatic patients.

OBJECTIVES: While chest radiograph (CXR) is the first-line imaging investigation in patients with respiratory symptoms, differentiating COVID-19 from other respiratory infections on CXR remains challenging. We developed and validated an AI system for COVID-19 detection on presenting CXR. METHODS: A deep learning model (RadGenX), trained on 168,850 CXRs, was validated on a large international test set of presenting CXRs of symptomatic patients from 9 study sites (US, Italy, and Hong Kong SAR) and 2 public datasets from the US and Europe. Performance was measured by area under the receiver operator characteristic curve (AUC). Bootstrapped simulations were performed to assess performance across a range of potential COVID-19 disease prevalence values (3.33 to 33.3%). Comparison against international radiologists was performed on an independent test set of 852 cases. RESULTS: RadGenX achieved an AUC of 0.89 on 4-fold cross-validation and an AUC of 0.79 (95%CI 0.78-0.80) on an independent test cohort of 5,894 patients. Delong's test showed statistical differences in model performance across patients from different regions (p < 0.01), disease severity (p < 0.001), gender (p < 0.001), and age (p = 0.03). Prevalence simulations showed the negative predictive value increases from 86.1% at 33.3% prevalence, to greater than 98.5% at any prevalence below 4.5%. Compared with radiologists, McNemar's test showed the model has higher sensitivity (p < 0.001) but lower specificity (p < 0.001). CONCLUSION: An AI model that predicts COVID-19 infection on CXR in symptomatic patients was validated on a large international cohort providing valuable context on testing and performance expectations for AI systems that perform COVID-19 prediction on CXR. KEY POINTS: • An AI model developed using CXRs to detect COVID-19 was validated in a large multi-center cohort of 5,894 patients from 9 prospectively recruited sites and 2 public datasets. • Differences in AI model performance were seen across region, disease severity, gender, and age. • Prevalence simulations on the international test set demonstrate the model's NPV is greater than 98.5% at any prevalence below 4.5%.

Tissue clearing techniques for three-dimensional optical imaging of intact human prostate and correlations with multi-parametric MRI.

BACKGROUND: Tissue clearing technologies have enabled remarkable advancements for in situ characterization of tissues and exploration of the three-dimensional (3D) relationships between cells, however, these studies have predominantly been performed in non-human tissues and correlative assessment with clinical imaging has yet to be explored. We sought to evaluate the feasibility of tissue clearing technologies for 3D imaging of intact human prostate and the mapping of structurally and molecularly preserved pathology data with multi-parametric volumetric MR imaging (mpMRI). METHODS: Whole-mount prostates were processed with either hydrogel-based CLARITY or solvent-based iDISCO. The samples were stained with a nuclear dye or fluorescently labeled with antibodies against androgen receptor, alpha-methylacyl coenzyme-A racemase, or p63, and then imaged with 3D confocal microscopy. The apparent diffusion coefficient and Ktrans maps were computed from preoperative mpMRI. RESULTS: Quantitative analysis of cleared normal and tumor prostate tissue volumes displayed differences in 3D tissue architecture, marker-specific cell staining, and cell densities that were significantly correlated with mpMRI measurements in this initial, pilot cohort. CONCLUSIONS: 3D imaging of human prostate volumes following tissue clearing is a feasible technique for quantitative radiology-pathology correlation analysis with mpMRI and provides an opportunity to explore functional relationships between cellular structures and cross-sectional clinical imaging.

Frequency and Distribution of Chest Radiographic Findings in Patients Positive for COVID-19.

Background Current coronavirus disease 2019 (COVID-19) radiologic literature is dominated by CT, and a detailed description of chest radiography appearances in relation to the disease time course is lacking. Purpose To describe the time course and severity of findings of COVID-19 at chest radiography and correlate these with real-time reverse transcription polymerase chain reaction (RT-PCR) testing for severe acute respiratory syndrome coronavirus 2, or SARS-CoV-2, nucleic acid. Materials and Methods This is a retrospective study of patients with COVID-19 confirmed by using RT-PCR and chest radiographic examinations who were admitted across four hospitals and evaluated between January and March 2020. Baseline and serial chest radiographs (n = 255) were reviewed with RT-PCR. Correlation with concurrent CT examinations (n = 28) was performed when available. Two radiologists scored each chest radiograph in consensus for consolidation, ground-glass opacity, location, and pleural fluid. A severity index was determined for each lung. The lung scores were summed to produce the final severity score. Results The study was composed of 64 patients (26 men; mean age, 56 years ± 19 [standard deviation]). Of these, 58 patients had initial positive findings with RT-PCR (91%; 95% confidence interval: 81%, 96%), 44 patients had abnormal findings at baseline chest radiography (69%; 95% confidence interval: 56%, 80%), and 38 patients had initial positive findings with RT-PCR testing and abnormal findings at baseline chest radiography (59%; 95% confidence interval: 46%, 71%). Six patients (9%) showed abnormalities at chest radiography before eventually testing positive for COVID-19 with RT-PCR. Sensitivity of initial RT-PCR (91%; 95% confidence interval: 83%, 97%) was higher than that of baseline chest radiography (69%; 95% confidence interval: 56%, 80%) (P = .009). Radiographic recovery (mean, 6 days ± 5) and virologic recovery (mean, 8 days ± 6) were not significantly different (P = .33). Consolidation was the most common finding (30 of 64; 47%) followed by ground-glass opacities (21 of 64; 33%). Abnormalities at chest radiography had a peripheral distribution (26 of 64; 41%) and lower zone distribution (32 of 64; 50%) with bilateral involvement (32 of 64; 50%). Pleural effusion was uncommon (two of 64; 3%). The severity of findings at chest radiography peaked at 10-12 days from the date of symptom onset. Conclusion Findings at chest radiography in patients with coronavirus disease 2019 frequently showed bilateral lower zone consolidation, which peaked at 10-12 days from symptom onset. © RSNA, 2020.

Receptor-based Surrogate Subtypes and Discrepancies with Breast Cancer Intrinsic Subtypes: Implications for Image Biomarker Development.

Purpose To determine the concordance and accuracy of imaging surrogates of immunohistochemical (IHC) markers and the molecular classification of breast cancer. Materials and Methods A total of 3050 patients from 17 public breast cancer data sets containing IHC marker receptor status (estrogen receptor/progesterone receptor/human epidermal growth factor receptor 2 [HER2]) and their molecular classification (basal-like, HER2-enriched, luminal A or B) were analyzed. Diagnostic accuracy and concordance as measured with the κ statistic were calculated between the IHC and molecular classifications. Simulations were performed to assess the relationship between accuracy of imaging-based IHC markers to predict molecular classification. A simulation was performed to examine effects of misclassification of molecular type on patient survival. Results Accuracies of intrinsic subtypes based on IHC subtype were 71.7% (luminal A), 53.7% (luminal B), 64.8% (HER2-enriched), and 81.7% (basal-like). The κ agreement was fair (κ = 0.36) for luminal A and HER2-enriched subtypes, good (κ = 0.65) for the basal-like subtype, and poor (κ = 0.09) for the luminal B subtypes. Introduction of image misclassification by simulation lowered image-true subtype accuracies and κ values. Simulation analysis showed that misclassification caused survival differences between luminal A and basal-like subtypes to decrease. Conclusion There is poor concordance between triple-receptor status and intrinsic molecular subtype in breast cancer, arguing against their use in the design of prognostic genomic-based image biomarkers. © RSNA, 2018 Online supplemental material is available for this article.

Multiregional Radiogenomic Assessment of Prostate Microenvironments with Multiparametric MR Imaging and DNA Whole-Exome Sequencing of Prostate Glands with Adenocarcinoma.

Purpose To assess the underlying genomic variation of prostate gland microenvironments of patients with prostate adenocarcinoma in the context of colocalized multiparametric magnetic resonance (MR) imaging and histopathologic assessment of normal and abnormal regions by using whole-exome sequencing. Materials and Methods Six patients with prostate adenocarcinoma who underwent robotic prostatectomy with whole-mount preservation of the prostate were identified, which enabled spatial mapping between preoperative multiparametric MR imaging and the gland. Four regions of interest were identified within each gland, including regions found to be normal and abnormal via histopathologic analysis. Whole-exome DNA sequencing (>50 times coverage) was performed on each of these spatially targeted regions. Radiogenomic analysis of imaging and mutation data were performed with hierarchical clustering, phylogenetic analysis, and principal component analysis. Results Radiogenomic multiparametric MR imaging and whole-exome spatial characterization in six patients with prostate adenocarcinoma (three patients, Gleason score of 3 + 4; and three patients, Gleason score of 4 + 5) was performed across 23 spatially distinct regions. Hierarchical clustering separated histopathologic analysis-proven high-grade lesions from the normal regions, and this reflected concordance between multiparametric MR imaging and resultant histopathologic analysis in all patients. Seventy-seven mutations involving 29 cancer-associated genes across the 23 spatially distinct prostate samples were identified. There was no significant difference in mutation load in cancer-associated genes between regions that were proven to be normal via histopathologic analysis (34 mutations per sample ± 19), mildly suspicious via multiparametric MR imaging (37 mutations per sample ± 21), intermediately suspicious via multiparametric MR imaging (31 mutations per sample ± 15), and high-grade cancer (33 mutations per sample ± 18) (P = .30). Principal component analysis resolved samples from different patients and further classified samples (regardless of histopathologic status) from prostate glands with Gleason score 3 + 4 versus 4 + 5 samples. Conclusion Multiregion spatial multiparametric MR imaging and whole-exome radiogenomic analysis of prostate glands with adenocarcinoma shows a continuum of mutations across regions that were found via histologic analysis to be high grade and normal. © RSNA, 2017 Online supplemental material is available for this article.

Genomic Adequacy from Solid Tumor Core Needle Biopsies of ex Vivo Tissue and in Vivo Lung Masses: Prospective Study.

Purpose To identify the variables and factors that affect the quantity and quality of nucleic acid yields from imaging-guided core needle biopsy. Materials and Methods This study was approved by the institutional review board and compliant with HIPAA. The authors prospectively obtained 232 biopsy specimens from 74 patients (177 ex vivo biopsy samples from surgically resected masses were obtained from 49 patients and 55 in vivo lung biopsy samples from computed tomographic [CT]-guided lung biopsies were obtained from 25 patients) and quantitatively measured DNA and RNA yields with respect to needle gauge, number of needle passes, and percentage of the needle core. RNA quality was also assessed. Significance of correlations among variables was assessed with analysis of variance followed by linear regression. Conditional probabilities were calculated for projected sample yields. Results The total nucleic acid yield increased with an increase in the number of needle passes or a decrease in needle gauge (two-way analysis of variance, P < .0001 for both). However, contrary to calculated differences in volume yields, the effect of needle gauge was markedly greater than the number of passes. For example, the use of an 18-gauge versus a 20-gauge biopsy needle resulted in a 4.8-5.7 times greater yield, whereas a double versus a single pass resulted in a 2.4-2.8 times greater yield for 18- versus 20-gauge needles, respectively. Ninety-eight of 184 samples (53%) had an RNA integrity number of at least 7 (out of a possible score of 10). Conclusion With regard to optimizing nucleic acid yields in CT-guided lung core needle biopsies used for genomic analysis, there should be a preference for using lower gauge needles over higher gauge needles with more passes. ©RSNA, 2016 Online supplemental material is available for this article. An earlier incorrect version of this article appeared online. This article was corrected on October 21, 2016.

Radiogenomic Analysis Demonstrates Associations between (18)F-Fluoro-2-Deoxyglucose PET, Prognosis, and Epithelial-Mesenchymal Transition in Non-Small Cell Lung Cancer.

Purpose To investigate whether non-small cell lung cancer (NSCLC) tumors that express high normalized maximum standardized uptake value (SUVmax) are associated with a more epithelial-mesenchymal transition (EMT)-like phenotype. Materials and Methods In this institutional review board-approved study, a public NSCLC data set that contained fluorine 18 ((18)F) fluoro-2-deoxyglucose positron emission tomography (PET) and messenger RNA expression profile data (n = 26) was obtained, and patients were categorized on the basis of measured normalized SUVmax values. Significance analysis of microarrays was then used to create a radiogenomic signature. The prognostic ability of this signature was assessed in a second independent data set that consisted of clinical and messenger RNA expression data (n = 166). Signature concordance with EMT was evaluated by means of validation in a publicly available cell line data set. Finally, by establishing an in vitro EMT lung cancer cell line model, an attempt was made to substantiate the radiogenomic signature with quantitative polymerase chain reaction, and functional assays were performed, including Western blot, cell migration, glucose transporter, and hexokinase assays (paired t test), as well as pharmacologic assays against chemotherapeutic agents (half-maximal effective concentration). Results Differential expression analysis yielded a 14-gene radiogenomic signature (P < .05, false discovery rate [FDR] < 0.20), which was confirmed to have differences in disease-specific survival (log-rank test, P = .01). This signature also significantly overlapped with published EMT cell line gene expression data (P < .05, FDR < 0.20). Finally, an EMT cell line model was established, and cells that had undergone EMT differentially expressed this signature and had significantly different EMT protein expression (P < .05, FDR < 0.20), cell migration, glucose uptake, and hexokinase activity (paired t test, P < .05). Cells that had undergone EMT also had enhanced chemotherapeutic resistance, with a higher half-maximal effective concentration than that of cells that had not undergone EMT (P < .05). Conclusion Integrative radiogenomic analysis demonstrates an association between increased normalized (18)F fluoro-2-deoxyglucose PET SUVmax, outcome, and EMT in NSCLC. (©) RSNA, 2016 Online supplemental material is available for this article.

A computed tomography radiogenomic biomarker predicts microvascular invasion and clinical outcomes in hepatocellular carcinoma.

UNLABELLED: Microvascular invasion (MVI) in hepatocellular carcinoma (HCC) is an independent predictor of poor outcomes subsequent to surgical resection or liver transplantation (LT); however, MVI currently cannot be adequately determined preoperatively. Radiogenomic venous invasion (RVI) is a contrast-enhanced computed tomography (CECT) biomarker of MVI derived from a 91-gene HCC "venous invasion" gene expression signature. Preoperative CECTs of 157 HCC patients who underwent surgical resection (N = 72) or LT (N = 85) between 2000 and 2009 at three institutions were evaluated for the presence or absence of RVI. RVI was assessed for its ability to predict MVI and outcomes. Interobserver agreement for scoring RVI was substantial among five radiologists (κ = 0.705; P < 0.001). The diagnostic accuracy, sensitivity, and specificity of RVI in predicting MVI was 89%, 76%, and 94%, respectively. Positive RVI score was associated with lower overall survival (OS) than negative RVI score in the overall cohort (P < 0.001; 48 vs. >147 months), American Joint Committee on Cancer tumor-node-metastasis stage II (P < 0.001; 34 vs. >147 months), and in LT patients within Milan criteria (P < 0.001; 69 vs. >147 months). Positive RVI score also portended lower recurrence-free survival at 3 years versus negative RVI score (P = 0.001; 27% vs. 62%). CONCLUSION: RVI is a noninvasive radiogenomic biomarker that accurately predicts histological MVI in HCC surgical candidates. Its presence on preoperative CECT is associated with early disease recurrence and poor OS and may be useful for identifying patients less likely to derive a durable benefit from surgical treatment.

The radiogenomic risk score stratifies outcomes in a renal cell cancer phase 2 clinical trial.

OBJECTIVES: To characterize a radiogenomic risk score (RRS), a previously defined biomarker, and to evaluate its potential for stratifying radiological progression-free survival (rPFS) in patients with metastatic renal cell carcinoma (mRCC) undergoing pre-surgical treatment with bevacizumab. METHODOLOGY: In this IRB-approved study, prospective imaging analysis of the RRS was performed on phase II clinical trial data of mRCC patients (n = 41) evaluating whether patient stratification according to the RRS resulted in groups more or less likely to have a rPFS to pre-surgical bevacizumab prior to cytoreductive nephrectomy. Survival times of RRS subgroups were analyzed using Kaplan-Meier survival analysis. RESULTS: The RRS is enriched in diverse molecular processes including drug response, stress response, protein kinase regulation, and signal transduction pathways (P < 0.05). The RRS successfully stratified rPFS to bevacizumab based on pre-treatment computed tomography imaging with a median progression-free survival of 6 versus >25 months (P = 0.005) and overall survival of 25 versus >37 months in the high and low RRS groups (P = 0.03), respectively. Conventional prognostic predictors including the Motzer and Heng criteria were not predictive in this cohort (P > 0.05). CONCLUSIONS: The RRS stratifies rPFS to bevacizumab in patients from a phase II clinical trial with mRCC undergoing cytoreductive nephrectomy and pre-surgical bevacizumab. KEY POINTS: • The RRS SOMA stratifies patient outcomes in a phase II clinical trial. • RRS stratifies subjects into prognostic groups in a discrete or continuous fashion. • RRS is biologically enriched in diverse processes including drug response programs.

Breast Cancer: Radiogenomic Biomarker Reveals Associations among Dynamic Contrast-enhanced MR Imaging, Long Noncoding RNA, and Metastasis.

PURPOSE: To perform a radiogenomic analysis of women with breast cancer to study the multiscale relationships among quantitative computer vision-extracted dynamic contrast material-enhanced (DCE) magnetic resonance (MR) imaging phenotypes, early metastasis, and long noncoding RNA (lncRNA) expression determined by means of high-resolution next-generation RNA sequencing. MATERIALS AND METHODS: In this institutional review board-approved study, an automated image analysis platform extracted 47 computational quantitative features from DCE MR imaging data in a training set (n = 19) to screen for MR imaging biomarkers indicative of poor metastasis-free survival (MFS). The lncRNA molecular landscape of the candidate feature was defined by using an RNA sequencing-specific negative binomial distribution differential expression analysis. Then, this radiogenomic biomarker was applied prospectively to a validation set (n = 42) to allow prediction of MFS and lncRNA expression by using quantitative polymerase chain reaction analysis. RESULTS: The quantitative MR imaging feature, enhancing rim fraction score, was predictive of MFS in the training set (P = .007). RNA sequencing analysis yielded an average of 55.7 × 10(6) reads per sample and identified 14 880 lncRNAs from a background of 189 883 transcripts per sample. Radiogenomic analysis allowed identification of three previously uncharacterized and five named lncRNAs significantly associated with high enhancing rim fraction, including Homeobox transcript antisense intergenic RNA (HOTAIR) (P < .05), a known predictor of poor MFS in patients with breast cancer. Independent validation confirmed the association of the enhancing rim fraction phenotype with both MFS (P = .002) and expression of four of the top five differentially expressed lncRNAs (P < .05), including HOTAIR. CONCLUSION: The enhancing rim fraction score, a quantitative DCE MR imaging lncRNA radiogenomic biomarker, is associated with early metastasis and expression of the known predictor of metastatic progression, HOTAIR.

The Radiogenomic Risk Score: Construction of a Prognostic Quantitative, Noninvasive Image-based Molecular Assay for Renal Cell Carcinoma.

PURPOSE: To evaluate the feasibility of constructing radiogenomic-based surrogates of molecular assays (SOMAs) in patients with clear-cell renal cell carcinoma (CCRCC) by using data extracted from a single computed tomographic (CT) image. MATERIALS AND METHODS: In this institutional review board approved study, gene expression profile data and contrast material-enhanced CT images from 70 patients with CCRCC in a training set were independently assessed by two radiologists for a set of predefined imaging features. A SOMA for a previously validated CCRCC-specific supervised principal component (SPC) risk score prognostic gene signature was constructed and termed the radiogenomic risk score (RRS). It uses the microarray data and a 28-trait image array to evaluate each CT image with multiple regression of gene expression analysis. The predictive power of the RRS SOMA was then prospectively validated in an independent dataset to confirm its relationship to the SPC gene signature (n = 70) and determination of patient outcome (n = 77). Data were analyzed by using multivariate linear regression-based methods and Cox regression modeling, and significance was assessed with receiver operator characteristic curves and Kaplan-Meier survival analysis. RESULTS: Our SOMA faithfully represents the tissue-based molecular assay it models. The RRS scaled with the SPC gene signature (R = 0.57, P < .001, classification accuracy 70.1%, P < .001) and predicted disease-specific survival (log rank P < .001). Independent validation confirmed the relationship between the RRS and the SPC gene signature (R = 0.45, P < .001, classification accuracy 68.6%, P < .001) and disease-specific survival (log-rank P < .001) and that it was independent of stage, grade, and performance status (multivariate Cox model P < .05, log-rank P < .001). CONCLUSION: A SOMA for the CCRCC-specific SPC prognostic gene signature that is predictive of disease-specific survival and independent of stage was constructed and validated, confirming that SOMA construction is feasible.

Illuminating radiogenomic characteristics of glioblastoma multiforme through integration of MR imaging, messenger RNA expression, and DNA copy number variation.

PURPOSE: To perform a multilevel radiogenomics study to elucidate the glioblastoma multiforme (GBM) magnetic resonance (MR) imaging radiogenomic signatures resulting from changes in messenger RNA (mRNA) expression and DNA copy number variation (CNV). MATERIALS AND METHODS: Radiogenomic analysis was performed at MR imaging in 23 patients with GBM in this retrospective institutional review board-approved HIPAA-compliant study. Six MR imaging features-contrast enhancement, necrosis, contrast-to-necrosis ratio, infiltrative versus edematous T2 abnormality, mass effect, and subventricular zone (SVZ) involvement-were independently evaluated and correlated with matched genomic profiles (global mRNA expression and DNA copy number profiles) in a significant manner that also accounted for multiple hypothesis testing by using gene set enrichment analysis (GSEA), resampling statistics, and analysis of variance to gain further insight into the radiogenomic signatures in patients with GBM. RESULTS: GSEA was used to identify various oncogenic pathways with MR imaging features. Correlations between 34 gene loci were identified that showed concordant variations in gene dose and mRNA expression, resulting in an MR imaging, mRNA, and CNV radiogenomic association map for GBM. A few of the identified gene-to-trait associations include association of the contrast-to-necrosis ratio with KLK3 and RUNX3; association of SVZ involvement with Ras oncogene family members, such as RAP2A, and the metabolic enzyme TYMS; and association of vasogenic edema with the oncogene FOXP1 and PIK3IP1, which is a member of the PI3K signaling network. CONCLUSION: Construction of an MR imaging, mRNA, and CNV radiogenomic association map has led to identification of MR traits that are associated with some known high-grade glioma biomarkers and association with genomic biomarkers that have been identified for other malignancies but not GBM. Thus, the traits and genes identified on this map highlight new candidate radiogenomic biomarkers for further evaluation in future studies.

ALK molecular phenotype in non-small cell lung cancer: CT radiogenomic characterization.

PURPOSE: To present a radiogenomic computed tomographic (CT) characterization of anaplastic lymphoma kinase (ALK)-rearranged non-small cell lung cancer (NSCLC) (ALK+). MATERIALS AND METHODS: In this HIPAA-compliant institutional review board-approved retrospective study, CT studies, ALK status, and clinical-pathologic data in 172 patients with NSCLC from three institutions were analyzed. A screen of 24 CT image traits was performed in a training set of 59 patients, followed by random forest variable selection incorporating 24 CT traits plus six clinical-pathologic covariates to identify a radiogenomic predictor of ALK+ status. This predictor was then validated in an independent cohort (n = 113). Test-for-accuracy and subset analyses were performed. A similar analysis was performed to identify a biomarker associated with shorter progression-free survival (PFS) after therapy with the ALK inhibitor crizotinib. RESULTS: ALK+ status was associated with central tumor location, absence of pleural tail, and large pleural effusion. An ALK+ radiogenomic CT status biomarker consisting of these three imaging traits with patient age of younger than 60 years showed strong discriminatory power for ALK+ status, with a sensitivity of 83.3% (15 of 18), a specificity of 77.9% (74 of 95), and an accuracy of 78.8% (89 of 113) in independent testing. The discriminatory power was particularly strong in patients with operable disease (stage IIIA or lower), with a sensitivity of 100.0% (five of five), a specificity of 88.1% (37 of 42), and an accuracy of 89.4% (42 of 47). Tumors with a disorganized vessel pattern had a shorter PFS with crizotinib therapy than tumors without this trait (11.4 vs 20.2 months, P = .041). CONCLUSION: ALK+ NSCLC has distinct characteristics at CT imaging that, when combined with clinical covariates, discriminate ALK+ from non-ALK tumors and can potentially identify patients with a shorter durable response to crizotinib.

Evaluating patency rates of an ultralow-porosity expanded polytetrafluoroethylene covered stent in the treatment of venous stenosis in arteriovenous dialysis circuits.

PURPOSE: To evaluate the efficacy of an ultralow-porosity expanded polytetrafluoroethylene (ePTFE) covered stent in the treatment of autogenous arteriovenous fistula (AVF) and prosthetic arteriovenous graft (AVG) venous outflow stenoses. MATERIALS AND METHODS: Clinical and angiographic outcomes of 20 consecutive patients with arteriovenous dialysis circuits treated with the endoprosthesis were reviewed following institutional review board approval. Patients were followed routinely at 2 months and 6 months after stent placement, or earlier if clinically warranted. The primary endpoint was 2- and 6-month primary treatment area patency. Secondary endpoints included primary circuit patency, primary assisted patency, and secondary patency. RESULTS: Eleven patients with AVFs and nine patients with AVGs were treated successfully with the covered stent. Primary treatment area patency rates were 85% ± 16 at both 2 months and 6 months. Primary circuit patency rates were was 65% ± 21 and 45% ± 22, respectively; primary assisted patency rates were 90% ± 13 and 85% ± 16, respectively; and secondary patency rates were 100% and 90% ± 13, respectively. Of the three cases of lost primary treatment area patency, two developed thrombosis and one developed recurrent stenosis. No significant differences were found between patients with AVFs and AVGs. CONCLUSIONS: Data from this preliminary study suggests that the ultralow-porosity ePTFE covered stent may be a clinically viable option for treatment of venous outflow stenoses in arteriovenous vascular access circuits.

Multi-Modality, Multi-Dimensional Characterization of Pediatric Non-Alcoholic Fatty Liver Disease.

Non-alcoholic fatty liver disease is a multifaceted disease that progresses through multiple phases; it involves metabolic as well as structural changes. These alterations can be measured directly or indirectly through blood, non-invasive imaging, and/or tissue analyses. While some studies have evaluated the correlations between two sets of measurements (e.g., histopathology with cross-sectional imaging or blood biomarkers), the interrelationships, if any, among histopathology, clinical blood profiles, cross-sectional imaging, and metabolomics in a pediatric cohort remain unknown. We created a multiparametric clinical MRI-histopathologic NMR network map of pediatric NAFLD through multimodal correlation networks, in order to gain insight into how these different sets of measurements are related. We found that leptin and other blood markers were correlated with many other measurements; however, upon filtering out the blood biomarkers, the network was decomposed into three independent hubs centered around histopathological features, each with associated MRI and plasma metabolites. These multi-modality maps could serve as a framework for characterizing disease status and progression and could potentially guide medical interventions.

Radiogenomic analysis of breast cancer using MRI: a preliminary study to define the landscape.

OBJECTIVE: Molecular profiling studies have defined the increasing importance of gene expression phenotyping in breast cancer. However, the relationship between global transcriptomic profiles and the information provided by breast MRI remains to be examined. In this pilot study, our aim was to provide a preliminary radiogenomic association map linking MR image phenotypes to underlying global gene expression patterns in breast cancer. MATERIALS AND METHODS: From a multiinstitutional study, a total of 353 patients with a diagnosis of breast cancer were examined for gene expression analysis. Radiogenomic analysis was then performed on a subset of these patients (n = 10) who also underwent breast MRI. Two radiologists evaluated each MRI study across 26 predefined imaging phenotypes. Analyses were performed to correlate the expression and imaging data and to define associations between specific MR image phenotypes and gene sets of interest. RESULTS: High-level analysis revealed 21 imaging traits that were globally correlated (p < 0.05), with 71% of the total genes measured in patients with breast cancer. A significant correlation was identified between heterogeneous enhancement patterns and a previously described interferon breast cancer subtype (p < 0.01). We also identified 12 imaging traits that significantly correlated (false discovery rate < 0.25) with gene sets related to breast cancer and 11 traits correlated (false discovery rate < 0.25) to prognostic gene sets (van't Veer, wound response, and hypoxia metagene signatures) using gene enrichment analysis. CONCLUSION: Radiogenomic analysis of breast cancer with MRI is a novel approach to understanding the underlying molecular biology of breast cancers.