Optimized b-value selection for the discrimination of prostate cancer grades, including the cribriform pattern, using diffusion weighted imaging

Quantitative Histomorphometric Features of Prostate Cancer Predict Patients Who Biochemically Recur Following Prostatectomy

Prostate cancer is the most commonly diagnosed cancer in men, accounting for 27% of the new male cancer diagnoses in 2022. If organ-confined, removal of the prostate through radical prostatectomy is considered curative; however, distant metastases may occur, resulting in a poor patient prognosis. This study sought to determine whether quantitative pathomic features of prostate cancer differ in patients who biochemically experience biological recurrence after surgery. Whole-mount prostate histology from 78 patients was analyzed for this study. In total, 614 slides were hematoxylin and eosin stained and digitized to produce whole slide images (WSI). Regions of differing Gleason patterns were digitally annotated by a genitourinary fellowship-trained pathologist, and high-resolution tiles were extracted from each annotated region of interest for further analysis. Individual glands within the prostate were identified using automated image processing algorithms, and histomorphometric features were calculated on a per-tile basis and across WSI and averaged by patients. Tiles were organized into cancer and benign tissues. Logistic regression models were fit to assess the predictive value of the calculated pathomic features across tile groups and WSI; additionally, models using clinical information were used for comparisons. Logistic regression classified each pathomic feature model at accuracies >80% with areas under the curve of 0.82, 0.76, 0.75, and 0.72 for all tiles, cancer only, noncancer only, and across WSI. This was comparable with standard clinical information, Gleason Grade Groups, and CAPRA score, which achieved similar accuracies but areas under the curve of 0.80, 0.77, and 0.70, respectively. This study demonstrates that the use of quantitative pathomic features calculated from digital histology of prostate cancer may provide clinicians with additional information beyond the traditional qualitative pathologist assessment. Further research is warranted to determine possible inclusion in treatment guidance.

T2-Weighted MRI Radiomic Features Predict Prostate Cancer Presence and Eventual Biochemical Recurrence

Prostate cancer (PCa) is the most diagnosed non-cutaneous cancer in men. Despite therapies such as radical prostatectomy, which is considered curative, distant metastases may form, resulting in biochemical recurrence (BCR). This study used radiomic features calculated from multi-parametric magnetic resonance imaging (MP-MRI) to evaluate their ability to predict BCR and PCa presence. Data from a total of 279 patients, of which 46 experienced BCR, undergoing MP-MRI prior to surgery were assessed for this study. After surgery, the prostate was sectioned using patient-specific 3D-printed slicing jigs modeled using the T2-weighted imaging (T2WI). Sectioned tissue was stained, digitized, and annotated by a GU-fellowship trained pathologist for cancer presence. Digitized slides and annotations were co-registered to the T2WI and radiomic features were calculated across the whole prostate and cancerous lesions. A tree regression model was fitted to assess the ability of radiomic features to predict BCR, and a tree classification model was fitted with the same radiomic features to classify regions of cancer. We found that 10 radiomic features predicted eventual BCR with an AUC of 0.97 and classified cancer at an accuracy of 89.9%. This study showcases the application of a radiomic feature-based tool to screen for the presence of prostate cancer and assess patient prognosis, as determined by biochemical recurrence.

Lesion-specific 3D-printed moulds for image-guided tissue multi-sampling of ovarian tumours: A prospective pilot study

Background

High-Grade Serous Ovarian Carcinoma (HGSOC) is the most prevalent and lethal subtype of ovarian cancer, but has a paucity of clinically-actionable biomarkers due to high degrees of multi-level heterogeneity. Radiogenomics markers have the potential to improve prediction of patient outcome and treatment response, but require accurate multimodal spatial registration between radiological imaging and histopathological tissue samples. Previously published co-registration work has not taken into account the anatomical, biological and clinical diversity of ovarian tumours.

Methods

In this work, we developed a research pathway and an automated computational pipeline to produce lesion-specific three-dimensional (3D) printed moulds based on preoperative cross-sectional CT or MRI of pelvic lesions. Moulds were designed to allow tumour slicing in the anatomical axial plane to facilitate detailed spatial correlation of imaging and tissue-derived data. Code and design adaptations were made following each pilot case through an iterative refinement process.

Results

Five patients with confirmed or suspected HGSOC who underwent debulking surgery between April and December 2021 were included in this prospective study. Tumour moulds were designed and 3D-printed for seven pelvic lesions, covering a range of tumour volumes (7 to 133 cm3) and compositions (cystic and solid proportions). The pilot cases informed innovations to improve specimen and subsequent slice orientation, through the use of 3D-printed tumour replicas and incorporation of a slice orientation slit in the mould design, respectively. The overall research pathway was compatible with implementation within the clinically determined timeframe and treatment pathway for each case, involving multidisciplinary clinical professionals from Radiology, Surgery, Oncology and Histopathology Departments.

Conclusions

We developed and refined a computational pipeline that can model lesion-specific 3D-printed moulds from preoperative imaging for a variety of pelvic tumours. This framework can be used to guide comprehensive multi-sampling of tumour resection specimens.

Comparison of a machine and deep learning model for automated tumor annotation on digitized whole slide prostate cancer histology

One in eight men will be affected by prostate cancer (PCa) in their lives. While the current clinical standard prognostic marker for PCa is the Gleason score, it is subject to inter-reviewer variability. This study compares two machine learning methods for discriminating between cancerous regions on digitized histology from 47 PCa patients. Whole-slide images were annotated by a GU fellowship-trained pathologist for each Gleason pattern. High-resolution tiles were extracted from annotated and unlabeled tissue. Patients were separated into a training set of 31 patients (Cohort A, n = 9345 tiles) and a testing cohort of 16 patients (Cohort B, n = 4375 tiles). Tiles from Cohort A were used to train a ResNet model, and glands from these tiles were segmented to calculate pathomic features to train a bagged ensemble model to discriminate tumors as (1) cancer and noncancer, (2) high- and low-grade cancer from noncancer, and (3) all Gleason patterns. The outputs of these models were compared to ground-truth pathologist annotations. The ensemble and ResNet models had overall accuracies of 89% and 88%, respectively, at predicting cancer from noncancer. The ResNet model was additionally able to differentiate Gleason patterns on data from Cohort B while the ensemble model was not. Our results suggest that quantitative pathomic features calculated from PCa histology can distinguish regions of cancer; however, texture features captured by deep learning frameworks better differentiate unique Gleason patterns.

Homologous point transformer for multi-modality prostate image registration

Registration is the process of transforming images so they are aligned in the same coordinate space. In the medical field, image registration is often used to align multi-modal or multi-parametric images of the same organ. A uniquely challenging subset of medical image registration is cross-modality registration-the task of aligning images captured with different scanning methodologies. In this study, we present a transformer-based deep learning pipeline for performing cross-modality, radiology-pathology image registration for human prostate samples. While existing solutions for multi-modality prostate image registration focus on the prediction of transform parameters, our pipeline predicts a set of homologous points on the two image modalities. The homologous point registration pipeline achieves better average control point deviation than the current state-of-the-art automatic registration pipeline. It reaches this accuracy without requiring masked MR images which may enable this approach to achieve similar results in other organ systems and for partial tissue samples.

Bridging the gap between prostate radiology and pathology through machine learning

Background

Prostate cancer remains the second deadliest cancer for American men despite clinical advancements. Currently, magnetic resonance imaging (MRI) is considered the most sensitive non‐invasive imaging modality that enables visualization, detection, and localization of prostate cancer, and is increasingly used to guide targeted biopsies for prostate cancer diagnosis. However, its utility remains limited due to high rates of false positives and false negatives as well as low inter‐reader agreements.

Purpose

Machine learning methods to detect and localize cancer on prostate MRI can help standardize radiologist interpretations. However, existing machine learning methods vary not only in model architecture, but also in the ground truth labeling strategies used for model training. We compare different labeling strategies and the effects they have on the performance of different machine learning models for prostate cancer detection on MRI.

Methods

Four different deep learning models (SPCNet, U‐Net, branched U‐Net, and DeepLabv3+) were trained to detect prostate cancer on MRI using 75 patients with radical prostatectomy, and evaluated using 40 patients with radical prostatectomy and 275 patients with targeted biopsy. Each deep learning model was trained with four different label types: pathology‐confirmed radiologist labels, pathologist labels on whole‐mount histopathology images, and lesion‐level and pixel‐level digital pathologist labels (previously validated deep learning algorithm on histopathology images to predict pixel‐level Gleason patterns) on whole‐mount histopathology images. The pathologist and digital pathologist labels (collectively referred to as pathology labels) were mapped onto pre‐operative MRI using an automated MRI‐histopathology registration platform.

Results

Radiologist labels missed cancers (ROC‐AUC: 0.75‐0.84), had lower lesion volumes (~68% of pathology lesions), and lower Dice overlaps (0.24‐0.28) when compared with pathology labels. Consequently, machine learning models trained with radiologist labels also showed inferior performance compared to models trained with pathology labels. Digital pathologist labels showed high concordance with pathologist labels of cancer (lesion ROC‐AUC: 0.97‐1, lesion Dice: 0.75‐0.93). Machine learning models trained with digital pathologist labels had the highest lesion detection rates in the radical prostatectomy cohort (aggressive lesion ROC‐AUC: 0.91‐0.94), and had generalizable and comparable performance to pathologist label‐trained‐models in the targeted biopsy cohort (aggressive lesion ROC‐AUC: 0.87‐0.88), irrespective of the deep learning architecture. Moreover, machine learning models trained with pixel‐level digital pathologist labels were able to selectively identify aggressive and indolent cancer components in mixed lesions on MRI, which is not possible with any human‐annotated label type.

Conclusions

Machine learning models for prostate MRI interpretation that are trained with digital pathologist labels showed higher or comparable performance with pathologist label‐trained models in both radical prostatectomy and targeted biopsy cohort. Digital pathologist labels can reduce challenges associated with human annotations, including labor, time, inter‐ and intra‐reader variability, and can help bridge the gap between prostate radiology and pathology by enabling the training of reliable machine learning models to detect and localize prostate cancer on MRI.

Tumor Connectomics: Mapping the Intra-Tumoral Complex Interaction Network Using Machine Learning

The high-level relationships that form complex networks within tumors and between surrounding tissue is challenging and not fully understood. To better understand these tumoral networks, we developed a tumor connectomics framework (TCF) based on graph theory with machine learning to model the complex interactions within and around the tumor microenvironment that are detectable on imaging. The TCF characterization model was tested with independent datasets of breast, brain, and prostate lesions with corresponding validation datasets in breast and brain cancer. The TCF network connections were modeled using graph metrics of centrality, average path length (APL), and clustering from multiparametric MRI with IsoSVM. The Matthews Correlation Coefficient (MCC), Area Under the Curve-ROC, and Precision-Recall (AUC-ROC and AUC-PR) were used for statistical analysis. The TCF classified the breast and brain tumor cohorts with an IsoSVM AUC-PR and MCC of 0.86, 0.63 and 0.85, 0.65, respectively. The TCF benign breast lesions had a significantly higher clustering coefficient and degree centrality than malignant TCFs. Grade 2 brain tumors demonstrated higher connectivity compared to Grade 4 tumors with increased degree centrality and clustering coefficients. Gleason 7 prostate lesions had increased betweenness centrality and APL compared to Gleason 6 lesions with AUC-PR and MCC ranging from 0.90 to 0.99 and 0.73 to 0.87, respectively. These TCF findings were similar in the validation breast and brain datasets. In conclusion, we present a new method for tumor characterization and visualization that results in a better understanding of the global and regional connections within the lesion and surrounding tissue.

Diffusion Restriction Comparison between Gleason 4 Fused Glands and Cribriform Glands within Patient Using Whole-Mount Prostate Pathology as Ground Truth

The presence and extent of cribriform patterned Gleason 4 (G4) glands are associated with poor prognosis following radical prostatectomy. This study used whole-mount prostate histology and multiparametric magnetic resonance imaging (MP-MRI) to evaluate diffusion differences in G4 gland morphology. Fourty-eight patients underwent MP-MRI prior to prostatectomy, of whom 22 patients had regions of both G4 cribriform glands and G4 fused glands (G4CG and G4FG, respectively). After surgery, the prostate was sliced using custom, patient-specific 3D-printed slicing jigs modeled according to the T2-weighted MR image, processed, and embedded in paraffin. Whole-mount hematoxylin and eosin-stained slides were annotated by our urologic pathologist and digitally contoured to differentiate the lumen, epithelium, and stroma. Digitized slides were co-registered to the T2-weighted MRI scan. Linear mixed models were fitted to the MP-MRI data to consider the different hierarchical structures at the patient and slide level. We found that Gleason 4 cribriform glands were more diffusion-restricted than fused glands.

Selective identification and localization of indolent and aggressive prostate cancers via CorrSigNIA: an MRI-pathology correlation and deep learning framework

Automated methods for detecting prostate cancer and distinguishing indolent from aggressive disease on Magnetic Resonance Imaging (MRI) could assist in early diagnosis and treatment planning. Existing automated methods of prostate cancer detection mostly rely on ground truth labels with limited accuracy, ignore disease pathology characteristics observed on resected tissue, and cannot selectively identify aggressive (Gleason Pattern≥4) and indolent (Gleason Pattern=3) cancers when they co-exist in mixed lesions. In this paper, we present a radiology-pathology fusion approach, CorrSigNIA, for the selective identification and localization of indolent and aggressive prostate cancer on MRI. CorrSigNIA uses registered MRI and whole-mount histopathology images from radical prostatectomy patients to derive accurate ground truth labels and learn correlated features between radiology and pathology images. These correlated features are then used in a convolutional neural network architecture to detect and localize normal tissue, indolent cancer, and aggressive cancer on prostate MRI. CorrSigNIA was trained and validated on a dataset of 98 men, including 74 men that underwent radical prostatectomy and 24 men with normal prostate MRI. CorrSigNIA was tested on three independent test sets including 55 men that underwent radical prostatectomy, 275 men that underwent targeted biopsies, and 15 men with normal prostate MRI. CorrSigNIA achieved an accuracy of 80% in distinguishing between men with and without cancer, a lesion-level ROC-AUC of 0.81±0.31 in detecting cancers in both radical prostatectomy and biopsy cohort patients, and lesion-levels ROC-AUCs of 0.82±0.31 and 0.86±0.26 in detecting clinically significant cancers in radical prostatectomy and biopsy cohort patients respectively. CorrSigNIA consistently outperformed other methods across different evaluation metrics and cohorts. In clinical settings, CorrSigNIA may be used in prostate cancer detection as well as in selective identification of indolent and aggressive components of prostate cancer, thereby improving prostate cancer care by helping guide targeted biopsies, reducing unnecessary biopsies, and selecting and planning treatment.

Using IVIM Parameters to Differentiate Prostate Cancer and Contralateral Normal Tissue through Fusion of MRI Images with Whole-Mount Pathology Specimen Images by Control Point Registration Method

The intravoxel incoherent motion (IVIM) model may enhance the clinical value of multiparametric magnetic resonance imaging (mpMRI) in the detection of prostate cancer (PCa). However, while past IVIM modeling studies have shown promise, they have also reported inconsistent results and limitations, underscoring the need to further enhance the accuracy of IVIM modeling for PCa detection. Therefore, this study utilized the control point registration toolbox function in MATLAB to fuse T2-weighted imaging (T2WI) and diffusion-weighted imaging (DWI) MRI images with whole-mount pathology specimen images in order to eliminate potential bias in IVIM calculations. Sixteen PCa patients underwent prostate MRI scans before undergoing radical prostatectomies. The image fusion method was then applied in calculating the patients’ IVIM parameters. Furthermore, MRI scans were also performed on 22 healthy young volunteers in order to evaluate the changes in IVIM parameters with aging. Among the full study cohort, the f parameter was significantly increased with age, while the D* parameter was significantly decreased. Among the PCa patients, the D and ADC parameters could differentiate PCa tissue from contralateral normal tissue, while the f and D* parameters could not. The presented image fusion method also provided improved precision when comparing regions of interest side by side. However, further studies with more standardized methods are needed to further clarify the benefits of the presented approach and the different IVIM parameters in PCa characterization.

Feasibility of diffusion weighting with a local inside-out nonlinear gradient coil for prostate MRI

PURPOSE

Prostate cancer remains the 2nd leading cancer killer of men, yet it is also a disease with a high rate of overtreatment. Diffusion weighted imaging (DWI) has shown promise as a reliable, grade-sensitive imaging method, but it is limited by low image quality. Currently, DWI quality image is directly related to low gradient amplitudes, since weak gradients must be compensated with long echo times.

METHODS

We propose a new type of MRI accessory, an "inside-out" and nonlinear gradient, whose sole purpose is to deliver diffusion encoding to a region of interest. Performance was simulated in OPERA and the resulting fields were used to simulate DWI with two compartment and kurtosis models. Experiments with a nonlinear head gradient prove the accuracy of DWI and ADC maps diffusion encoded with nonlinear gradients.

RESULTS

Simulations validated thermal and mechanical safety while showing a 5 to 10-fold increase in gradient strength over prostate. With these strengths, lesion CNR in ADC maps approximately doubled for a range of anatomical positions. Proof-of-principle experiments show that spatially varying b-values can be corrected for accurate DWI and ADC.

CONCLUSIONS

Dedicated nonlinear diffusion encoding hardware could improve prostate DWI.

Automated detection of aggressive and indolent prostate cancer on magnetic resonance imaging

PURPOSE

While multi-parametric magnetic resonance imaging (MRI) shows great promise in assisting with prostate cancer diagnosis and localization, subtle differences in appearance between cancer and normal tissue lead to many false positive and false negative interpretations by radiologists. We sought to automatically detect aggressive cancer (Gleason pattern ≥ 4) and indolent cancer (Gleason pattern 3) on a per-pixel basis on MRI to facilitate the targeting of aggressive cancer during biopsy.

METHODS

We created the Stanford Prostate Cancer Network (SPCNet), a convolutional neural network model, trained to distinguish between aggressive cancer, indolent cancer, and normal tissue on MRI. Ground truth cancer labels were obtained by registering MRI with whole-mount digital histopathology images from patients who underwent radical prostatectomy. Before registration, these histopathology images were automatically annotated to show Gleason patterns on a per-pixel basis. The model was trained on data from 78 patients who underwent radical prostatectomy and 24 patients without prostate cancer. The model was evaluated on a pixel and lesion level in 322 patients, including six patients with normal MRI and no cancer, 23 patients who underwent radical prostatectomy, and 293 patients who underwent biopsy. Moreover, we assessed the ability of our model to detect clinically significant cancer (lesions with an aggressive component) and compared it to the performance of radiologists.

RESULTS

Our model detected clinically significant lesions with an area under the receiver operator characteristics curve of 0.75 for radical prostatectomy patients and 0.80 for biopsy patients. Moreover, the model detected up to 18% of lesions missed by radiologists, and overall had a sensitivity and specificity that approached that of radiologists in detecting clinically significant cancer.

CONCLUSIONS

Our SPCNet model accurately detected aggressive prostate cancer. Its performance approached that of radiologists, and it helped identify lesions otherwise missed by radiologists. Our model has the potential to assist physicians in specifically targeting the aggressive component of prostate cancers during biopsy or focal treatment.

Ex Vivo MR Histology and Cytometric Feature Mapping Connect Three-dimensional in Vivo MR Images to Two-dimensional Histopathologic Images of Murine Sarcomas

Purpose To establish a platform for quantitative tissue-based interpretation of cytoarchitecture features from tumor MRI measurements. Materials and Methods In a pilot preclinical study, multicontrast in vivo MRI of murine soft-tissue sarcomas in 10 mice, followed by ex vivo MRI of fixed tissues (termed MR histology), was performed. Paraffin-embedded limb cross-sections were stained with hematoxylin-eosin, digitized, and registered with MRI. Registration was assessed by using binarized tumor maps and Dice similarity coefficients (DSCs). Quantitative cytometric feature maps from histologic slides were derived by using nuclear segmentation and compared with registered MRI, including apparent diffusion coefficients and transverse relaxation times as affected by magnetic field heterogeneity (T2* maps). Cytometric features were compared with each MR image individually by using simple linear regression analysis to identify the features of interest, and the goodness of fit was assessed on the basis of R² values. Results Registration of MR images to histopathologic slide images resulted in mean DSCs of 0.912 for ex vivo MR histology and 0.881 for in vivo MRI. Triplicate repeats showed high registration repeatability (mean DSC, >0.9). Whole-slide nuclear segmentations were automated to detect nuclei on histopathologic slides (DSC = 0.8), and feature maps were generated for correlative analysis with MR images. Notable trends were observed between cell density and in vivo apparent diffusion coefficients (best line fit: R² = 0.96, P < .001). Multiple cytoarchitectural features exhibited linear relationships with in vivo T2* maps, including nuclear circularity (best line fit: R² = 0.99, P < .001) and variance in nuclear circularity (best line fit: R² = 0.98, P < .001). Conclusion An infrastructure for registering and quantitatively comparing in vivo tumor MRI with traditional histologic analysis was successfully implemented in a preclinical pilot study of soft-tissue sarcomas. Keywords: MRI, Pathology, Animal Studies, Tissue Characterization Supplemental material is available for this article. © RSNA, 2021.

Effect of Echo Times on Prostate Cancer Detection on T2-Weighted Images

PURPOSE

To compare the effect of different echo times (TE) on the detection of prostate cancer (PCa) on T2-weighted MR images.

MATERIALS AND METHODS

This study recruited patients (n = 38) with histologically confirmed PCa who underwent preoperative 3T MRI. Three radiologists independently marked region on interests (ROIs) on suspected PCa lesions on T2-weighted images at different TEs: 90, 150, and 180 ms obtained with Turbo Spin Echo imaging protocol with multiple echoes. The ROIs were assigned a value 1-5 indicating the reviewer's confidence in accurately detecting PCa. These ROIs were compared to histologically confirmed PCa (n = 95) on whole mount prostatectomy sections to calculate sensitivity, positive predictive value (PPV), and confidence score.

RESULTS

Two radiologists (R1, R2) showed significantly increased sensitivity for PCa detection at 180 ms TE compared to 90 ms (R1: 43.2, 50.5, 50.5%, R2: 45.3, 44.2, 53.7% at TE of 90, 150, 180 ms, respectively) (p = 0.048, 0.033 for R1 and R2). Sensitivity was similar for radiologist 3 (45.3%-46.3%) at different TE values (p = 0.953). No significant difference in the PPV (R1: 64.1%-70.6%, R2: 46.7%-56.0%, R3: 70.5%-81.5%) and the confidence score assigned (R1: 4.6-4.8, R2: 4.6-4.8 R3: 4.3-4.4) was found for either of the radiologists.

CONCLUSION

Our results suggest improved detection of PCa with similar PPV and confidence scores when higher TE values are utilized for T2-weighted image acquisition.

Accurate segmentation of prostate cancer histomorphometric features using a weakly supervised convolutional neural network

Purpose: Prostate cancer primarily arises from the glandular epithelium. Histomophometric techniques have been used to assess the glandular epithelium in automated detection and classification pipelines; however, they are often rigid in their implementation, and their performance suffers on large datasets where variation in staining, imaging, and preparation is difficult to control. The purpose of this study is to quantify performance of a pixelwise segmentation algorithm that was trained using different combinations of weak and strong stroma, epithelium, and lumen labels in a prostate histology dataset.

Approach: We have combined weakly labeled datasets generated using simple morphometric techniques and high-quality labeled datasets from human observers in prostate biopsy cores to train a convolutional neural network for use in whole mount prostate labeling pipelines. With trained networks, we characterize pixelwise segmentation of stromal, epithelium, and lumen (SEL) regions on both biopsy core and whole-mount H&E-stained tissue.

Results: We provide evidence that by simply training a deep learning algorithm on weakly labeled data generated from rigid morphometric methods, we can improve the robustness of classification over the morphometric methods used to train the classifier.

Conclusions: We show that not only does our approach of combining weak and strong labels for training the CNN improve qualitative SEL labeling within tissue but also the deep learning generated labels are superior for cancer classification in a higher-order algorithm over the morphometrically derived labels it was trained on.

Radio-pathomic mapping model generated using annotations from five pathologists reliably distinguishes high-grade prostate cancer

Purpose: Our study predictively maps epithelium density in magnetic resonance imaging (MRI) space while varying the ground truth labels provided by five pathologists to quantify the downstream effects of interobserver variability. Approach: Clinical imaging and postsurgical tissue from 48 recruited prospective patients were used in our study. Tissue was sliced to match the MRI orientation and whole-mount slides were stained and digitized. Data from 28 patients ( n = 33 slides) were sent to five pathologists to be annotated. Slides from the remaining 20 patients ( n = 123 slides) were annotated by one of the five pathologists. Interpathologist variability was measured using Krippendorff's alpha. Pathologist-specific radiopathomic mapping models were trained using a partial least-squares regression using MRI values to predict epithelium density, a known marker for disease severity. An analysis of variance characterized intermodel means difference in epithelium density. A consensus model was created and evaluated using a receiver operator characteristic classifying high grade versus low grade and benign, and was statistically compared to apparent diffusion coefficient (ADC). Results: Interobserver variability ranged from low to acceptable agreement (0.31 to 0.69). There was a statistically significant difference in mean predicted epithelium density values ( p < 0.001 ) between the five models. The consensus model outperformed ADC (areas under the curve = 0.80 and 0.71, respectively, p < 0.05 ). Conclusion: We demonstrate that radiopathomic maps of epithelium density are sensitive to the pathologist annotating the dataset; however, it is unclear if these differences are clinically significant. The consensus model produced the best maps, matched the performance of the best individual model, and outperformed ADC.

Registration of presurgical MRI and histopathology images from radical prostatectomy via RAPSODI

PURPOSE

Magnetic resonance imaging (MRI) has great potential to improve prostate cancer diagnosis; however, subtle differences between cancer and confounding conditions render prostate MRI interpretation challenging. The tissue collected from patients who undergo radical prostatectomy provides a unique opportunity to correlate histopathology images of the prostate with preoperative MRI to accurately map the extent of cancer from histopathology images onto MRI. We seek to develop an open-source, easy-to-use platform to align presurgical MRI and histopathology images of resected prostates in patients who underwent radical prostatectomy to create accurate cancer labels on MRI.

METHODS

Here, we introduce RAdiology Pathology Spatial Open-Source multi-Dimensional Integration (RAPSODI), the first open-source framework for the registration of radiology and pathology images. RAPSODI relies on three steps. First, it creates a three-dimensional (3D) reconstruction of the histopathology specimen as a digital representation of the tissue before gross sectioning. Second, RAPSODI registers corresponding histopathology and MRI slices. Third, the optimized transforms are applied to the cancer regions outlined on the histopathology images to project those labels onto the preoperative MRI.

RESULTS

We tested RAPSODI in a phantom study where we simulated various conditions, for example, tissue shrinkage during fixation. Our experiments showed that RAPSODI can reliably correct multiple artifacts. We also evaluated RAPSODI in 157 patients from three institutions that underwent radical prostatectomy and have very different pathology processing and scanning. RAPSODI was evaluated in 907 corresponding histpathology-MRI slices and achieved a Dice coefficient of 0.97 ± 0.01 for the prostate, a Hausdorff distance of 1.99 ± 0.70 mm for the prostate boundary, a urethra deviation of 3.09 ± 1.45 mm, and a landmark deviation of 2.80 ± 0.59 mm between registered histopathology images and MRI.

CONCLUSION

Our robust framework successfully mapped the extent of cancer from histopathology slices onto MRI providing labels from training machine learning methods to detect cancer on MRI.

Integrative radiogenomics for virtual biopsy and treatment monitoring in ovarian cancer

BACKGROUND

Ovarian cancer survival rates have not changed in the last 20 years. The majority of cases are High-grade serous ovarian carcinomas (HGSOCs), which are typically diagnosed at an advanced stage with multiple metastatic lesions. Taking biopsies of all sites of disease is infeasible, which challenges the implementation of stratification tools based on molecular profiling.

MAIN BODY

In this review, we describe how these challenges might be overcome by integrating quantitative features extracted from medical imaging with the analysis of paired genomic profiles, a combined approach called radiogenomics, to generate virtual biopsies. Radiomic studies have been used to model different imaging phenotypes, and some radiomic signatures have been associated with paired molecular profiles to monitor spatiotemporal changes in the heterogeneity of tumours. We describe different strategies to integrate radiogenomic information in a global and local manner, the latter by targeted sampling of tumour habitats, defined as regions with distinct radiomic phenotypes.

CONCLUSION

Linking radiomics and biological correlates in a targeted manner could potentially improve the clinical management of ovarian cancer. Radiogenomic signatures could be used to monitor tumours during the course of therapy, offering additional information for clinical decision making. In summary, radiogenomics may pave the way to virtual biopsies and treatment monitoring tools for integrative tumour analysis.

Gleason Probability Maps: A Radiomics Tool for Mapping Prostate Cancer Likelihood in MRI Space

Prostate cancer is the most common noncutaneous cancer in men in the United States. The current paradigm for screening and diagnosis is imperfect, with relatively low specificity, high cost, and high morbidity. This study aims to generate new image contrasts by learning a distribution of unique image signatures associated with prostate cancer. In total, 48 patients were prospectively recruited for this institutional review board–approved study. Patients underwent multiparametric magnetic resonance imaging 2 weeks before surgery. Postsurgical tissues were annotated by a pathologist and aligned to the in vivo imaging. Radiomic profiles were generated by linearly combining 4 image contrasts (T2, apparent diffusion coefficient [ADC] 0-1000, ADC 50-2000, and dynamic contrast-enhanced) segmented using global thresholds. The distribution of radiomic profiles in high-grade cancer, low-grade cancer, and normal tissues was recorded, and the generated probability values were applied to a naive test set. The resulting Gleason probability maps were stable regardless of training cohort, functioned independent of prostate zone, and outperformed conventional clinical imaging (area under the curve [AUC] = 0.79). Extensive overlap was seen in the most common image signatures associated with high- and low-grade cancer, indicating that low- and high-grade tumors present similarly on conventional imaging.

Spatial density and diversity of architectural histology in prostate cancer: influence on diffusion weighted magnetic resonance imaging

Background

To assess the influence of specific histopathologic patterns on MRI diffusion characteristics by performing rigorous whole-mount/imaging registration and correlating histologic architectures observed in prostate cancer with diffusion characteristics in prostate MRIs.

Methods

Fifty-two whole-mount pathology blocks from 15 patients who underwent multiparametric MRI (mpMRI) at a single institution prior to radical prostatectomy were retrospectively analyzed. Regions containing individual morphologic patterns (N=21 patterns, including variations of cribriforming, expansile sheets, single cells, patterns of early intraluminal complexity, and mucin rupture patterns) were digitally annotated by an expert genitourinary pathologist. Distinct tumor foci on each slide were also assigned a Gleason grade and scored as having any high-risk histologic pattern. Digital sections were aligned to MRI using a patient-specific mold and registered using local mean weighted piecewise transformation based on anatomic control points. Density and presence of morphological patterns was correlated to apparent diffusion coefficient (ADC) signal intensity using mixed effects model accounting for nested intra-foci, intra-patient correlation. Influence of intra-tumoral heterogeneity was assessed by affinity propagation clustering (APC) of morphology features and correlated to foci- and cluster-level ADC metrics.

Results

One hundred eleven distinct tumor foci were evaluated. Beta diversity, reflecting average morphology representation across inter- and intra-foci areas, demonstrated higher intra-tumor diversity within high-risk foci (P<0.05). ADC signal demonstrated an inverse correlation with foci-level Gleason grade (P>0.05), which was strengthened in cluster-level analysis for intra-foci regions containing high-risk morphologies (P=0.017). In voxel-based analysis, dense regions demonstrate lower ADC, but the presence and density for each morphology influenced ADC independently (ANOVA P<0.001).

Conclusions

Architectural features influence ADC characteristics of MRI, with more complex tumors having lower ADC values regulated by presence and density of specific morphologies.

Biologic Significance of Magnetic Resonance Imaging Invisibility in Localized Prostate Cancer

PURPOSE

Multiparametric magnetic resonance imaging (mpMRI) is used widely for prostate cancer (PCa) evaluation. Approximately 35% of aggressive tumors, however, are not visible on mpMRI. We sought to identify the molecular alterations associated with mpMRI-invisible tumors and determine whether mpMRI visibility is associated with PCa prognosis.

METHODS

Discovery and validation cohorts included patients who underwent mpMRI before radical prostatectomy and were found to harbor both mpMRI-visible (Prostate Imaging and Reporting Data System 3 to 5) and -invisible (Prostate Imaging and Reporting Data System 1 or 2) foci on surgical pathology. Next-generation sequencing was performed to determine differential gene expression between mpMRI-visible and -invisible foci. A genetic signature for tumor mpMRI visibility was derived in the discovery cohort and assessed in an independent validation cohort. Its association with long-term oncologic outcomes was evaluated in a separate testing cohort.

RESULTS

The discovery cohort included 10 patients with 26 distinct PCa foci on surgical pathology, of which 12 (46%) were visible and 14 (54%) were invisible on preoperative mpMRI. Next-generation sequencing detected prioritized genetic mutations in 14 (54%) tumor foci (n = 8 mpMRI visible, n = 6 mpMRI invisible). A nine-gene signature (composed largely of cell organization/structure genes) associated with mpMRI visibility was derived (area under the curve = 0.89), and the signature predicted MRI visibility with 75% sensitivity and 100% specificity (area under the curve = 0.88) in the validation cohort. In the testing cohort (n = 375, median follow-up 8 years) there was no significant difference in biochemical recurrence, distant metastasis, or cancer-specific mortality in patients with predicted mpMRI-visible versus -invisible tumors (all P > .05).

CONCLUSION

Compared with mpMRI-invisible disease, mpMRI-visible tumors are associated with underexpression of cellular organization genes. mpMRI visibility does not seem to be predictive of long-term cancer outcomes, highlighting the need for biopsy strategies that detect mpMRI-invisible tumors.

Predictive role of PI-RADSv2 and ADC parameters in differentiating Gleason pattern 3 + 4 and 4 + 3 prostate cancer

PURPOSE

To compare the predictive roles of qualitative (PI-RADSv2) and quantitative assessment (ADC metrics), in differentiating Gleason pattern (GP) 3 + 4 from the more aggressive GP 4 + 3 prostate cancer (PCa) using radical prostatectomy (RP) specimen as the reference standard.

METHODS

We retrospectively identified treatment-naïve peripheral (PZ) and transitional zone (TZ) Gleason Score 7 PCa patients who underwent multiparametric 3T prostate MRI (DWI with b value of 0,1400 and where unavailable, 0,500) and subsequent RP from 2011 to 2015. For each lesion identified on MRI, a PI-RADSv2 score was assigned by a radiologist blinded to pathology data. A PI-RADSv2 score ≤ 3 was defined as "low risk," a PI-RADSv2 score ≥ 4 as "high risk" for clinically significant PCa. Mean tumor ADC (ADCT), ADC of adjacent normal tissue (ADCN), and ADCratio (ADCT/ADCN) were calculated. Stepwise regression analysis using tumor location, ADCT and ADCratio, b value, low vs. high PI-RADSv2 score was performed to differentiate GP 3 + 4 from 4 + 3.

RESULTS

119 out of 645 cases initially identified met eligibility requirements. 76 lesions were GP 3 + 4, 43 were 4 + 3. ADCratio was significantly different between the two GP groups (p = 0.001). PI-RADSv2 score ("low" vs. "high") was not significantly different between the two GP groups (p = 0.17). Regression analysis selected ADCT (p = 0.03) and ADCratio (p = 0.0007) as best predictors to differentiate GP 4 + 3 from 3 + 4. Estimated sensitivity, specificity, and accuracy of the predictive model in differentiating GP 4 + 3 from 3 + 4 were 37, 82, and 66%, respectively.

CONCLUSIONS

ADC metrics could differentiate GP 3 + 4 from 4 + 3 PCa with high specificity and moderate accuracy while PI-RADSv2, did not differentiate between these patterns.

Radio-pathomic Maps of Epithelium and Lumen Density Predict the Location of High-Grade Prostate Cancer

PURPOSE

This study aims to combine multiparametric magnetic resonance imaging (MRI) and digitized pathology with machine learning to generate predictive maps of histologic features for prostate cancer localization.

METHODS AND MATERIALS

Thirty-nine patients underwent MRI prior to prostatectomy. After surgery, tissue was sliced according to MRI orientation using patient-specific 3-dimensionally printed slicing jigs. Whole-mount sections were annotated by our pathologist and digitally contoured to differentiate the lumen and epithelium. Slides were co-registered to the T2-weighted MRI scan. A learning curve was generated to determine the number of patients required for a stable machine-learning model. Patients were randomly stratified into 2 training sets and 1 test set. Two partial least-squares regression models were trained, each capable of predicting lumen and epithelium density. Predicted density values were calculated for each patient in the test dataset, mapped into the MRI space, and compared between regions confirmed as high-grade prostate cancer.

RESULTS

The learning-curve analysis showed that a stable fit was achieved with data from 10 patients. Maps indicated that regions of increased epithelium and decreased lumen density, generated from each independent model, corresponded with pathologist-annotated regions of high-grade cancer.

CONCLUSIONS

We present a radio-pathomic approach to mapping prostate cancer. We find that the maps are useful for highlighting high-grade tumors. This technique may be relevant for dose-painting strategies in prostate radiation therapy.