Automation of pattern recognition analysis of dynamic contrast-enhanced MRI data to characterize intratumoral vascular heterogeneity

Distinguishing metastatic triple-negative breast cancer from nonmetastatic breast cancer using second harmonic generation imaging and resonance Raman spectroscopy

Triple-negative breast cancer (TNBC) is an aggressive subset of breast cancer that is more common in African-American and Hispanic women. Early detection followed by intensive treatment is critical to improving poor survival rates. The current standard to diagnose TNBC from histopathology of biopsy samples is invasive and time-consuming. Imaging methods such as mammography and magnetic resonance (MR) imaging, while covering the entire breast, lack the spatial resolution and specificity to capture the molecular features that identify TNBC. Two nonlinear optical modalities of second harmonic generation (SHG) imaging of collagen, and resonance Raman spectroscopy (RRS) potentially offer novel rapid, label-free detection of molecular and morphological features that characterize cancerous breast tissue at subcellular resolution. In this study, we first applied MR methods to measure the whole-tumor characteristics of metastatic TNBC (4T1) and nonmetastatic estrogen receptor positive breast cancer (67NR) models, including tumor lactate concentration and vascularity. Subsequently, we employed for the first time in vivo SHG imaging of collagen and ex vivo RRS of biomolecules to detect different microenvironmental features of these two tumor models. We achieved high sensitivity and accuracy for discrimination between these two cancer types by quantitative morphometric analysis and nonnegative matrix factorization along with support vector machine. Our study proposes a new method to combine SHG and RRS together as a promising novel photonic and optical method for early detection of TNBC.

Pattern recognition analysis of dynamic susceptibility contrast (DSC)‐MRI curves automatically segments tissue areas with intact blood–brain barrier in a rat stroke model: A feasibility and comparison study

Background

The manual segmentation of intact blood–brain barrier (BBB) regions in the stroke brain is cumbersome, due to the coexistence of infarction, large blood vessels, ventricles, and intact BBB regions, specifically in areas with weak signal enhancement following contrast agent injection.

Hypothesis

That from dynamic susceptibility contrast (DSC)‐MRI alone, without user intervention, regions of weak BBB damage can be segmented based on the leakage‐related parameter K 2 and the extent of intact BBB regions, needed to estimate K 2 values, determined.

Study Type

Feasibility.

Animal Model

Ten female Sprague–Dawley rats (SD, 200–250g) underwent 1‐hour middle carotid artery occlusion (MCAO) and 1‐day reperfusion. Two SD rats underwent 1‐hour MCAO with 3‐day and 5‐day reperfusion.

Field Strength/Sequence

7T; ADC and T1 maps using diffusion‐weighted echo planar imaging (EPI) and relaxation enhancement (RARE) with variable repetition time (TR), respectively. dynamic contrast‐enhanced (DCE)‐MRI using FLASH. DSC‐MRI using gradient‐echo EPI.

Assessment

Constrained nonnegative matrix factorization (cNMF) was applied to the dynamic ‐curves of DSC‐MRI (<4 min) in a BBB‐disrupted rat model. Areas of voxels with intact BBB, classified by automated cNMF analyses, were then used in estimating K 1 and K 2 values, and compared with corresponding values from manually‐derived areas.

Statistical Tests

Mean ± standard deviation of ΔT1‐differences between ischemic and healthy areas were displayed with unpaired Student's t‐tests. Scatterplots were displayed with slopes and intercepts and Pearson's r values were evaluated between K 2 maps obtained with automatic (cNMF)‐ and manually‐derived regions of interest (ROIs) of the intact BBB region.

Results

Mildly BBB‐damaged areas (indistinguishable from DCE‐MRI (10 min) parameters) were automatically segmented. Areas of voxels with intact BBB, classified by automated cNMF, matched closely the corresponding, manually‐derived areas when respective areas were used in estimating K 2 maps (Pearson's r = 0.97, 12 slices).

Data Conclusion

Automatic segmentation of short DSC‐MRI data alone successfully identified areas with intact and compromised BBB in the stroke brain and compared favorably with manual segmentation.

Level of Evidence: 3

Technical Efficacy: Stage 1

J. Magn. Reson. Imaging 2020;51:1369–1381.

Quantifying Reoxygenation in Pancreatic Cancer During Stereotactic Body Radiotherapy

Hypoxia, the state of low oxygenation that often arises in solid tumours due to their high metabolism and irregular vasculature, is a major contributor to the resistance of tumours to radiation therapy (RT) and other treatments. Conventional RT extends treatment over several weeks or more, and nominally allows time for oxygen levels to increase ("reoxygenation") as cancer cells are killed by RT, mitigating the impact of hypoxia. Recent advances in RT have led to an increase in the use stereotactic body radiotherapy (SBRT), which delivers high doses in five or fewer fractions. For cancers such as pancreatic adenocarcinoma for which hypoxia varies significantly between patients, SBRT might not be optimal, depending on the extent to which reoxygenation occurs during its short duration. We used fluoro-5-deoxy-α-D-arabinofuranosyl)-2-nitroimidazole positron-emission tomography (FAZA-PET) imaging to quantify hypoxia before and after 5-fraction SBRT delivered to patient-derived pancreatic cancer xenografts orthotopically implanted in mice. An imaging technique using only the pre-treatment FAZA-PET scan and repeat dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) scans throughout treatment was able to predict the change in hypoxia. Our results support the further testing of this technique for imaging of reoxygenation in the clinic.

Cortical bone vessel identification and quantification on contrast-enhanced MR images

Background

Cortical bone porosity is a major determinant of bone strength. Despite the biomechanical importance of cortical bone porosity, the biological drivers of cortical porosity are unknown. The content of cortical pore space can indicate pore expansion mechanisms; both of the primary components of pore space, vessels and adipocytes, have been implicated in pore expansion. Dynamic contrast-enhanced MRI (DCE-MRI) is widely used in vessel detection in cardiovascular studies, but has not been applied to visualize vessels within cortical bone. In this study, we have developed a multimodal DCE-MRI and high resolution peripheral QCT (HR-pQCT) acquisition and image processing pipeline to detect vessel-filled cortical bone pores.

Methods

For this in vivo human study, 19 volunteers (10 males and 9 females; mean age =63±5) were recruited. Both distal and ultra-distal regions of the non-dominant tibia were imaged by HR-pQCT (82 µm nominal resolution) for bone structure segmentation and by 3T DCE-MRI (Gadavist; 9 min scan time; temporal resolution =30 sec; voxel size 230×230×500 µm³) for vessel visualization. The DCE-MRI was registered to the HR-pQCT volume and the voxels within the MRI cortical bone region were extracted. Features of the DCE data were calculated and voxels were categorized by a 2-stage hierarchical kmeans clustering algorithm to determine which voxels represent vessels. Vessel volume fraction (volume ratio of vessels to cortical bone), vessel density (average vessel count per cortical bone volume), and average vessel volume (mean volume of vessels) were calculated to quantify the status of vessel-filled pores in cortical bone. To examine spatial resolution and perform validation, a virtual phantom with 5 channel sizes and an applied pseudo enhancement curve was processed through the proposed image processing pipeline. Overlap volume ratio and Dice coefficient was calculated to measure the similarity between the detected vessel map and ground truth.

Results

In the human study, mean vessel volume fraction was 2.2%±1.0%, mean vessel density was 0.68±0.27 vessel/mm³, and mean average vessel volume was 0.032±0.012 mm³/vessel. Signal intensity for detected vessel voxels increased during the scan, while signal for non-vessel voxels within pores did not enhance. In the validation phantom, channels with diameter 250 µm or greater were detected successfully, with volume ratio equal to 1 and Dice coefficient above 0.6. Both statistics decreased dramatically for channel sizes less than 250 µm.

Conclusions

We have a developed a multi-modal image acquisition and processing pipeline that successfully detects vessels within cortical bone pores. The performance of this technique degrades for vessel diameters below the in-plane spatial resolution of the DCE-MRI acquisition. This approach can be applied to investigate the biological systems associated with cortical pore expansion.

Multiparametric MRI and Coregistered Histology Identify Tumor Habitats in Breast Cancer Mouse Models

It is well-recognized that solid tumors are genomically, anatomically, and physiologically heterogeneous. In general, more heterogeneous tumors have poorer outcomes, likely due to the increased probability of harboring therapy-resistant cells and regions. It is hypothesized that the genomic and physiologic heterogeneity are related, because physiologically distinct regions will exert variable selection pressures leading to the outgrowth of clones with variable genomic/proteomic profiles. To investigate this, methods must be in place to interrogate and define, at the microscopic scale, the cytotypes that exist within physiologically distinct subregions ("habitats") that are present at mesoscopic scales. MRI provides a noninvasive approach to interrogate physiologically distinct local environments, due to the biophysical principles that govern MRI signal generation. Here, we interrogate different physiologic parameters, such as perfusion, cell density, and edema, using multiparametric MRI (mpMRI). Signals from six different acquisition schema were combined voxel-by-voxel into four clusters identified using a Gaussian mixture model. These were compared with histologic and IHC characterizations of sections that were coregistered using MRI-guided 3D printed tumor molds. Specifically, we identified a specific set of MRI parameters to classify viable-normoxic, viable-hypoxic, nonviable-hypoxic, and nonviable-normoxic tissue types within orthotopic 4T1 and MDA-MB-231 breast tumors. This is the first coregistered study to show that mpMRI can be used to define physiologically distinct tumor habitats within breast tumor models. SIGNIFICANCE: This study demonstrates that noninvasive imaging metrics can be used to distinguish subregions within heterogeneous tumors with histopathologic correlation.

Quantitative imaging of cancer in the postgenomic era: Radio(geno)mics, deep learning, and habitats

Although cancer often is referred to as "a disease of the genes," it is indisputable that the (epi)genetic properties of individual cancer cells are highly variable, even within the same tumor. Hence, preexisting resistant clones will emerge and proliferate after therapeutic selection that targets sensitive clones. Herein, the authors propose that quantitative image analytics, known as "radiomics," can be used to quantify and characterize this heterogeneity. Virtually every patient with cancer is imaged radiologically. Radiomics is predicated on the beliefs that these images reflect underlying pathophysiologies, and that they can be converted into mineable data for improved diagnosis, prognosis, prediction, and therapy monitoring. In the last decade, the radiomics of cancer has grown from a few laboratories to a worldwide enterprise. During this growth, radiomics has established a convention, wherein a large set of annotated image features (1-2000 features) are extracted from segmented regions of interest and used to build classifier models to separate individual patients into their appropriate class (eg, indolent vs aggressive disease). An extension of this conventional radiomics is the application of "deep learning," wherein convolutional neural networks can be used to detect the most informative regions and features without human intervention. A further extension of radiomics involves automatically segmenting informative subregions ("habitats") within tumors, which can be linked to underlying tumor pathophysiology. The goal of the radiomics enterprise is to provide informed decision support for the practice of precision oncology.