Machine learning-based prediction of future breast cancer using algorithmically measured background parenchymal enhancement on high-risk screening MRI.

BACKGROUND: Preliminary work has demonstrated that background parenchymal enhancement (BPE) assessed by radiologists is predictive of future breast cancer in women undergoing high-risk screening MRI. Algorithmically assessed measures of BPE offer a more precise and reproducible means of measuring BPE than human readers and thus might improve the predictive performance of future cancer development. PURPOSE: To determine if algorithmically extracted imaging features of BPE on screening breast MRI in high-risk women are associated with subsequent development of cancer. STUDY TYPE: Case-control study. POPULATION: In all, 133 women at high risk for developing breast cancer; 46 of these patients developed breast cancer subsequently over a follow-up period of 2 years. FIELD STRENGTH/SEQUENCE: 5 T or 3.0 T T1 -weighted precontrast fat-saturated and nonfat-saturated sequences and postcontrast nonfat-saturated sequences. ASSESSMENT: Automatic features of BPE were extracted with a computer algorithm. Subjective BPE scores from five breast radiologists (blinded to clinical outcomes) were also available. STATISTICAL TESTS: Leave-one-out crossvalidation for a multivariate logistic regression model developed using the automatic features and receiver operating characteristic (ROC) analysis were performed to calculate the area under the curve (AUC). Comparison of automatic features and subjective features was performed using a generalized regression model and the P-value was obtained. Odds ratios for automatic and subjective features were compared. RESULTS: The multivariate model discriminated patients who developed cancer from the patients who did not, with an AUC of 0.70 (95% confidence interval: 0.60-0.79, P < 0.001). The imaging features remained independently predictive of subsequent development of cancer (P < 0.003) when compared with the subjective BPE assessment of the readers. DATA CONCLUSION: Automatically extracted BPE measurements may potentially be used to further stratify risk in patients undergoing high-risk screening MRI. LEVEL OF EVIDENCE: 3 Technical Efficacy: Stage 5 J. Magn. Reson. Imaging 2019;50:456-464.

A machine learning approach to radiogenomics of breast cancer: a study of 922 subjects and 529 DCE-MRI features.

BACKGROUND: Recent studies showed preliminary data on associations of MRI-based imaging phenotypes of breast tumours with breast cancer molecular, genomic, and related characteristics. In this study, we present a comprehensive analysis of this relationship. METHODS: We analysed a set of 922 patients with invasive breast cancer and pre-operative MRI. The MRIs were analysed by a computer algorithm to extract 529 features of the tumour and the surrounding tissue. Machine-learning-based models based on the imaging features were trained using a portion of the data (461 patients) to predict the following molecular, genomic, and proliferation characteristics: tumour surrogate molecular subtype, oestrogen receptor, progesterone receptor and human epidermal growth factor status, as well as a tumour proliferation marker (Ki-67). Trained models were evaluated on the set of the remaining 461 patients. RESULTS: Multivariate models were predictive of Luminal A subtype with AUC = 0.697 (95% CI: 0.647-0.746, p < .0001), triple negative breast cancer with AUC = 0.654 (95% CI: 0.589-0.727, p < .0001), ER status with AUC = 0.649 (95% CI: 0.591-0.705, p < .001), and PR status with AUC = 0.622 (95% CI: 0.569-0.674, p < .0001). Associations between individual features and subtypes we also found. CONCLUSIONS: There is a moderate association between tumour molecular biomarkers and algorithmically assessed imaging features.

The nuclear envelope protein, LAP1B, is a novel protein phosphatase 1 substrate.

Protein phosphatase 1 (PP1) binding proteins are quintessential regulators, determining substrate specificity and defining subcellular localization and activity of the latter. Here, we describe a novel PP1 binding protein, the nuclear membrane protein lamina associated polypeptide 1B (LAP1B), which interacts with the DYT1 dystonia protein torsinA. The PP1 binding domain in LAP1B was here identified as the REVRF motif at amino acids 55-59. The LAP1B:PP1 complex can be immunoprecipitated from cells in culture and rat cortex and the complex was further validated by yeast co-transformations and blot overlay assays. PP1, which is enriched in the nucleus, binds to the N-terminal nuclear domain of LAP1B, as shown by immunocolocalization and domain specific binding studies. PP1 dephosphorylates LAP1B, confirming the physiological relevance of this interaction. These findings place PP1 at a key position to participate in the pathogenesis of DYT1 dystonia and related nuclear envelope-based diseases.

Using the BI-RADS lexicon in a restrictive form of double reading as a strategy for minimizing screening mammography recall rates.

OBJECTIVE: The purpose of this article is to determine the potential reduction in screening recall rates by strictly following standardized BI-RADS lexicon for lesions seen on screening mammography. MATERIALS AND METHODS: Of 3084 consecutive mammograms performed at our screening facilities, 345 women with 437 lesions were recalled for additional imaging and constituted our study population. Three radiologists retrospectively classified lesions using the standard BI-RADS lexicon and assigned each to one of four groups: group A, the finding met criteria for recall by the BI-RADS lexicon; group B, the finding did not meet strict BI-RADS criteria for recall but was sufficiently indeterminate to warrant recall by the majority of the study panel; group C, the finding was classifiable by the BI-RADS lexicon but was not recalled because it was benign or stable; and group D, the questioned finding was not considered an abnormality by our study panel. Recall rates and the cancer detection rate were determined. The adjusted recall rate was calculated for lesions considered appropriate for recall (group A), and the reduction in the recall rate was determined. RESULTS: Nineteen malignancies were detected in our recalled population, for a cancer detection rate of 0.65%. All 19 malignancies were lesions considered appropriate for recall (group A). If only group A lesions had been recalled, the recall rate would have decreased from 11.4% to 6.2%, representing a 46% reduction in recalls without affecting the cancer detection rate. CONCLUSION: Using the BI-RADS lexicon as a decision-making aid may help adjust thresholds for recalling indeterminate or suspicious lesions and reduce recall rates from screening mammography.

Sputum-based molecular biomarkers for the early detection of lung cancer: limitations and promise.

Lung cancer is the leading cause of cancer deaths, with an overall survival of 15% at five years. Biomarkers that can sensitively and specifically detect lung cancer at early stage are crucial for improving this poor survival rate. Sputum has been the target for the discovery of non-invasive biomarkers for lung cancer because it contains airway epithelial cells, and molecular alterations identified in sputum are most likely to reflect tumor-associated changes or field cancerization caused by smoking in the lung. Sputum-based molecular biomarkers include morphology, allelic imbalance, promoter hypermethylation, gene mutations and, recently, differential miRNA expression. To improve the sensitivity and reproducibility of sputum-based biomarkers, we recommend standardization of processing protocols, bronchial epithelial cell enrichment, and identification of field cancerization biomarkers.

A molecular mechanism underlying the neural-specific defect in torsinA mutant mice.

A striking but poorly understood feature of many diseases is the unique involvement of neural tissue. One example is the CNS-specific disorder DYT1 dystonia, caused by a 3-bp deletion ("DeltaE") in the widely expressed gene TOR1A. Disease mutant knockin mice (Tor1a(DeltaE/DeltaE)) exhibit disrupted nuclear membranes selectively in neurons, mimicking the tissue specificity of the human disease and providing a model system in which to dissect the mechanisms underlying neural selectivity. Our in vivo studies demonstrate that lamina-associated polypeptide 1 (LAP1) and torsinB function with torsinA to maintain normal nuclear membrane morphology. Moreover, we show that nonneuronal cells express dramatically higher levels of torsinB and that RNAi-mediated depletion of torsinB (but not other torsin family members) causes nuclear membrane abnormalities in Tor1a(DeltaE/DeltaE) nonneuronal cells. The Tor1a(DeltaE/DeltaE) neural selective phenotype therefore arises because high levels of torsinB protect nonneuronal cells from the consequences of torsinA dysfunction, demonstrating how tissue specificity may result from differential susceptibility of cell types to insults that disrupt ubiquitous biological pathways.

Primary dystonia: molecules and mechanisms.

Primary dystonia is characterized by abnormal, involuntary twisting and turning movements that reflect impaired motor system function. The dystonic brain seems normal, in that it contains no overt lesions or evidence of neurodegeneration, but functional brain imaging has uncovered abnormalities involving the cortex, striatum and cerebellum, and diffusion tensor imaging suggests the presence of microstructural defects in white matter tracts of the cerebellothalamocortical circuit. Clinical electrophysiological studies show that the dystonic CNS exhibits aberrant plasticity--perhaps related to deficient inhibitory neurotransmission--in a range of brain structures, as well as the spinal cord. Dystonia is, therefore, best conceptualized as a motor circuit disorder, rather than an abnormality of a particular brain structure. None of the aforementioned abnormalities can be strictly causal, as they are not limited to regions of the CNS subserving clinically affected body parts, and are found in seemingly healthy patients with dystonia-related mutations. The study of dystonia-related genes will, hopefully, help researchers to unravel the chain of events from molecular to cellular to system abnormalities. DYT1 mutations, for example, cause abnormalities within the endoplasmic reticulum-nuclear envelope endomembrane system. Other dystonia-related gene products traffic through the endoplasmic reticulum, suggesting a potential cell biological theme underlying primary dystonia.

CTNNB1 mutations and overexpression of Wnt/beta-catenin target genes in WT1-mutant Wilms' tumors.

Gain-of-function mutations in exon 3 of beta-catenin (CTNNB1) are specific for Wilms' tumors that have lost WT1, but 50% of WT1-mutant cases lack such "hot spot" mutations. To ask whether stabilization of beta-catenin might be essential after WT1 loss, and to identify downstream target genes, we compared expression profiles in WT1-mutant versus WT1 wild-type Wilms' tumors. Supervised and nonsupervised hierarchical clustering of the expression data separated these two classes of Wilms' tumor. The WT1-mutant tumors overexpressed genes encoding myogenic and other transcription factors (MOX2, LBX1, SIM2), signaling molecules (TGFB2, FST, BMP2A), extracellular Wnt inhibitors (WIF1, SFRP4), and known beta-catenin/TCF targets (FST, CSPG2, CMYC). Beta-Catenin/TCF target genes were overexpressed in the WT1-mutant tumors even in the absence of CTNNB1 exon 3 mutations, and complete sequencing revealed gain-of-function mutations elsewhere in the CTNNB1 gene in some of these tumors, increasing the overall mutation frequency to 75%. Lastly, we identified and validated a novel direct beta-catenin target gene, GAD1, among the WT1-mutant signature genes. These data highlight two molecular classes of Wilms' tumor, and indicate strong selection for stabilization of beta-catenin in the WT1-mutant class. Beta-Catenin stabilization can initiate tumorigenesis in other systems, and this mechanism is likely critical in tumor formation after loss of WT1.

Fortuitous detection of papillary carcinoma of the thyroid with F-18 FDG positron emission tomography in a patient with non-Hodgkin lymphoma.

Use of F-18 fluorodeoxyglucose positron emission tomography (F-18 FDG PET) in patients with known thyroid cancer is well documented, but the role of this imaging modality in the initial workup of a thyroid nodule has not been defined. The incidental finding of a hypermetabolic focus in the thyroid on F-18 FDG PET in patients with a variety of primary malignancies is reported. Based on the authors' literature search, this is the first documented case of a thyroid cancer resulting from papillary carcinoma detected in a patient with a history of non-Hodgkin lymphoma.