Katie Robinson

Over more than five years of follow-up, thin-walled unilocular thymic cysts, as initially defined by magnetic resonance imaging (MRI), never developed irregular wall thickening, mural nodularity or septations, according to a study recently published in 

. The cysts, however, did show mural calcification, along with change in size, computed tomography (CT) attenuation and MRI signal intensity over the same period.

“Our investigation highlights the intrinsic and longitudinal characteristics of unilocular thymic cysts that contribute to their misinterpretation as thymic neoplasms and lymphadenopathy at CT,” wrote Wariya Chintanapakdee, M.D., Chulalongkorn University in Bangkok, Thailand, and colleagues.

In this retrospective study, the researchers evaluated the imaging characteristics and longitudinal change of unilocular thymic cysts on CT and MRI over time. All chest MRI studies showing thymic cysts between July 2008 and December 2019 were included. If initial CT showed a thymic lesion, the patient was referred for MRI for characterization and the baseline MRI indicated a cystic lesion. 

The researchers evaluated 244 chest MRI studies in 140 patients with 142 unique cysts and 392 CT examinations. The median follow-up duration was 2.2 years, but a subcohort of 33 patients and 34 cysts underwent CT and/or MR imaging follow-up for more than five years. 

The results in the subcohort showed that unilocular thymic cysts can arise de novo, as did with one of 34 cysts. Additionally, volume changed in 91%, T1- weighted MRI signal intensity changed in 67% and CT attenuation changed in 43%. Further, 16% of cysts had or developed mural calcifications. In 98% of thymic cysts, wall enhancement was present at MRI, with a median wall thickness of two millimeters. The results of the full study cohort of 142 unilocular thymic cysts matched those of the subcohort.

The authors explained that fluctuation in thymic cyst characteristics over time may be due to chronic recurrent intralesional hemorrhage and subsequent resorption. “The variable CT attenuation, MRI signal characteristics, cyst wall thickness, occasional findings of T1 and/or T2 hypointensity of the thymic cyst wall, wall calcification and intracyst fluid levels that we found support this hypothesis,” the authors wrote.

Limitations to the study included the absence of pathologic proof of benignity in all but the two lesions that were resected and proven to represent thymic cysts.

“Awareness of the variable imaging characteristics and behavior of thymic cysts over time should help guide clinical management. In indeterminate cases, imaging surveillance or surgical resection may be appropriate to exclude malignancy,” the authors wrote.

Articles

Awareness of the variable imaging characteristics and behavior of thymic cysts over time should help guide clinical management. 

Unilocular Thymic Cysts May Rarely Undergo Malignant Transformation 

Breast magnetic resonance imaging (MRI) that shows a lack of background parenchymal enhancement (BPE) suppression may indicate inferior response to neoadjuvant chemotherapy primarily in patients with hormone receptor (HR)-positive breast cancer. 

, Nola M. Hylton, Ph.D., of the department of radiology and biomedical imaging at the University of California San Francisco, and colleagues, noted that the findings could enable “discontinuation of ineffective treatment and initiation of a more promising alternative,” adding that “improved automated segmentation methods may enable the development of background parenchymal enhancement as a reliable biomarker of treatment response.”

This retrospective review included 3528 MRI scans from 882 women (mean age 48) who were treated for breast cancer by one of 10 drug arms between May 2010 and November 2016 in

the I-SPY 2 TRIAL. Patients underwent MRI at four points: before treatment, early treatment, inter-regimen and before surgery. BPE change from before treatment was defined as suppressed or nonsuppressed for each point. Researchers examined the relationship between BPE suppression and pathologic complete response in HR-positive and HR-negative cohorts.

In HR-positive patients, lack of BPE suppression was associated with lower pathologic complete response rate at the inter-regimen point: 11.8% compared with 28.9% for BPE suppressed (p = .02), and at the pre-surgery point: 5.3% versus 27.4%, respectively, (p = .003). BPE suppression was observed in both premenopausal women, 82% to 90% of the HR-positive cohort and 73% to 84% of the HR-negative cohort, and peri- or postmenopausal women, 62% to 73% of the HR-positive cohort and 72% to 77% of the HR-negative cohort.

, Liane E. Philpotts, M.D., a professor of radiology and biomedical imaging at Yale School of Medicine in New Haven, Connecticut, wrote that previous studies assessing the relationship between BPE and pathologic complete response rate included smaller cohorts and showed varying results. Still, she noted that while the researchers showed a significant correlation between BPE and pathologic complete response rate for HR-positive tumors, the results do not identify the cause for the BPE changes during neoadjuvant chemotherapy.

“The fact that BPE suppression was not significantly related to menopausal status suggests the effect is more than simply ovarian suppression,” Dr. Philpotts wrote. “The actual physiologic changes that alter BPE occurring with various chemotherapeutic treatments is an area that will need more investigation.”

“The data in this study contribute to the growing realization that differences among breast cancer subtypes and among individual women require an ever-increasing variety in treatment regimens,” Dr. Philpotts wrote. “Imaging has an increasingly important role in contributing to the better understanding and treatment of clinical breast cancer patient care.”

Nonsuppressed background parenchymal enhancement could indicate patients who will show inferior response and allow for a personalized redirection of treatment. 

Breast MRI Background Parenchymal Enhancement May Indicate Neoadjuvant Chemotherapy Response 

A newly developed algorithm was able to accurately classify six different types of brain tumor and discriminate pathologic from healthy magnetic resonance imaging (MRI) scans, according to a study published in 

Moreover, this was achieved from a single scan per patient without the need of any additional manual input.

In this retrospective study, Satrajit Chakrabarty, Ph.D. candidate in electrical engineering at Washington University in St Louis, and colleagues developed a three-dimensional (3D) convolutional neural network model for classifying MRI scans into a healthy class and six tumor classes: high-grade glioma, low-grade glioma, brain metastases, meningioma, pituitary adenoma and acoustic neuroma.

“This is the first study to address the most common intracranial tumor-types and directly determine the tumor class as well as detect the absence of tumor from a 3D MR volume,” the authors said. “Without the assistance of any manually segmented tumor or bounding box, the convolutional neural network model…could classify six brain tumor types and discriminate

healthy from pathological scans from a single post-contrast T1-weighted scan per patient.”

The researchers included 2105 preoperative post-contrast T1-weighted MRI scans from four publicly available datasets, and 1396 scans were used to train the model, while 361 scans were assigned to an internal test dataset and 348 scans were assigned to an external test dataset. 

On the internal test dataset, across the seven different classes, the model achieved sensitivity of 87% to 100%, a positive predictive value (PPV) of 85% to100%, area under the receiver operating characteristic curve (AUC) of 0.98 to 1.00 and precision-recall curve (AUPRC) of 0.91 to 1.00. On the external test dataset, comprising high-grade glioma and low-grade glioma, the model achieved sensitivity of 91% to 97%, a PPV of 73% to 99%, AUC of 0.97 to 0.98 and AUPRC of 0.9 to 1.0.

Of the seven classes addressed, more errors were observed for high-grade glioma, low-grade glioma and healthy. The authors explained that the reason for the misclassification between low-grade glioma and healthy classes may be due to less contrast enhancement of low-grade gliomas in post-contrast T1-weighted MRI scans due to less disruption of the blood-brain barrier.

“The model can be extended to other brain tumor types or neurologic disorders that exhibit

anomalous intensity profiles in MRI scans,” the authors said. “The network is the first step toward developing an artificial intelligence-augmented radiology workflow that can support image interpretation by providing quantitative information and statistics to the clinician to help improve diagnosis and prognosis.”

The algorithm is the first step towards developing an artificial intelligence-augmented radiology workflow that can support image interpretation to improve diagnosis and prognosis.

Algorithm Can Classify Major Intracranial Tumor Types from a Single MRI Scan 

Opportunistic computed tomography (CT) screening of osteoporosis on the thoracic and lumbar spine showed “reasonable” pooled sensitivity and specificity when using specific Hounsfield unit (HU) values as a threshold, according to a meta-analysis published recently in 

“We recommend radiologists use this simple measure to detect possible osteoporosis when reading chest or abdominal 

 and refer to orthopedic specialists for follow-up with positive findings,” said Claudia Henschke, M.D., Ph.D., of the department of radiology at Mount Sinai School of Medicine in New York, and colleagues.

While osteoporotic fractures are a major contributor to morbidity and mortality in later life, these fractures are underdiagnosed and undertreated. In this meta-analysis, the researchers aimed to assess the pooled diagnostic sensitivity and specificity of CT images for diagnosing osteoporosis in patients who meet the World Health Organization dual X-ray absorptiometry (DXA) osteoporosis criteria using specific HU values as a threshold.

For more expert insights and the latest news, subscribe to the Diagnostic Imaging e-Newsletter 

Ten eligible studies were identified from the earliest available date through July 1, 2018, in which participants underwent CT scans of the lumbar and/or thoracic spine. These scans had been carried out for different indications, with HU measurements were used to identify osteoporosis. DXA scans functioned as the reference standard.

For diagnosing osteoporosis, the mean area under the hierarchical summary receiver operating characteristic (ROC) curve was 0.84 (95% CI: 0.81, 0.87). The pooled diagnostic odds ratio suggested that patients with osteoporosis were 16.4 times more likely to have positive CT results compared with those without osteoporosis. However, the authors cited evidence of possible publication bias. The pooled diagnostic sensitivity and specificity of CT images to identify osteoporosis were 0.83 and 0.74, respectively, and 0.73 and 0.71 after adjusting for this bias.

Further, the researchers found that DXA criteria and country of the study were significant factors associated with the sensitivity of CT in detecting osteoporosis, while scanner manufacturer was associated with both sensitivity and specificity of CT.

“Radiologists should be aware of the vendors of the scanners and make sure they are using the correct CT HU threshold values in order to have a reasonable accuracy for detecting osteoporosis,” the authors said.

The authors noted that the meta-analysis was limited by the small number of studies.

Radiologists should use this simple measure to detect osteoporosis when reading chest or abdominal CT scans and refer to orthopedic specialists for follow-up with positive findings. 

Katie Robinson

Articles

Unilocular Thymic Cysts May Rarely Undergo Malignant Transformation

Breast MRI Background Parenchymal Enhancement May Indicate Neoadjuvant Chemotherapy Response

Algorithm Can Classify Major Intracranial Tumor Types from a Single MRI Scan

Threshold Values on CT Scans Help Identify Osteoporosis Opportunistically