Diagnostic Reproducibility: What Happens When the Same Pathologist Interprets the Same Breast Biopsy Specimen at Two Points in Time?

Dynamic whole-body [18F]FES PET/CT increases lesion visibility in patients with metastatic breast cancer

BACKGROUND

Correct classification of estrogen receptor (ER) status is essential for prognosis and treatment planning in patients with breast cancer (BC). Therefore, it is recommended to sample tumor tissue from an accessible metastasis. However, ER expression can show intra- and intertumoral heterogeneity. 16α-[18F]fluoroestradiol ([18F]FES) Positron Emission Tomography/Computed Tomography (PET/CT) allows noninvasive whole-body (WB) identification of ER distribution and is usually performed as a single static image 60 min after radiotracer injection. Using dynamic whole-body (D-WB) PET imaging, we examine [18F]FES kinetics and explore whether Patlak parametric images ( K i ) are quantitative and improve lesion visibility.

RESULTS

This prospective study included eight patients with metastatic ER-positive BC scanned using a D-WB PET acquisition protocol. The kinetics of [18F]FES were best characterized by the irreversible two-tissue compartment model in tumor lesions and in the majority of organ tissues. K i values from Patlak parametric images correlated with K i values from the full kinetic analysis, r2 = 0.77, and with the semiquantitative mean standardized uptake value (SUVmean), r2 = 0.91. Furthermore, parametric K i images had the highest target-to-background ratio (TBR) in 162/164 metastatic lesions and the highest contrast-to-noise ratio (CNR) in 99/164 lesions compared to conventional SUV images. TBR was 2.45 (95% confidence interval (CI): 2.25-2.68) and CNR 1.17 (95% CI: 1.08-1.26) times higher in K i images compared to SUV images. These quantitative differences were seen as reduced background activity in the K i images.

CONCLUSION

[18F]FES uptake is best described by an irreversible two-tissue compartment model. D-WB [18F]FES PET/CT scans can be used for direct reconstruction of parametric K i images, with superior lesion visibility and K i values comparable to K i values found from full kinetic analyses. This may aid correct ER classification and treatment decisions. Trial registration ClinicalTrials.gov: NCT04150731, https://clinicaltrials.gov/study/NCT04150731.

Photon Absorption Remote Sensing Imaging of Breast Needle Core Biopsies Is Diagnostically Equivalent to Gold Standard H&E Histologic Assessment

Photon absorption remote sensing (PARS) is a new laser-based microscope technique that permits cellular-level resolution of unstained fresh, frozen, and fixed tissues. Our objective was to determine whether PARS could provide an image quality sufficient for the diagnostic assessment of breast cancer needle core biopsies (NCB). We PARS imaged and virtually H&E stained seven independent unstained formalin-fixed paraffin-embedded breast NCB sections. These identical tissue sections were subsequently stained with standard H&E and digitally scanned. Both the 40× PARS and H&E whole-slide images were assessed by seven breast cancer pathologists, masked to the origin of the images. A concordance analysis was performed to quantify the diagnostic performances of standard H&E and PARS virtual H&E. The PARS images were deemed to be of diagnostic quality, and pathologists were unable to distinguish the image origin, above that expected by chance. The diagnostic concordance on cancer vs. benign was high between PARS and conventional H&E (98% agreement) and there was complete agreement for within-PARS images. Similarly, agreement was substantial (kappa > 0.6) for specific cancer subtypes. PARS virtual H&E inter-rater reliability was broadly consistent with the published literature on diagnostic performance of conventional histology NCBs across all tested histologic features. PARS was able to image unstained tissues slides that were diagnostically equivalent to conventional H&E. Due to its ability to non-destructively image fixed and fresh tissues, and the suitability of the PARS output for artificial intelligence assistance in diagnosis, this technology has the potential to improve the speed and accuracy of breast cancer diagnosis.

Should low-risk DCIS lose the cancer label? An evidence review

BACKGROUND

Population mammographic screening for breast cancer has led to large increases in the diagnosis and treatment of ductal carcinoma in situ (DCIS). Active surveillance has been proposed as a management strategy for low-risk DCIS to mitigate against potential overdiagnosis and overtreatment. However, clinicians and patients remain reluctant to choose active surveillance, even within a trial setting. Re-calibration of the diagnostic threshold for low-risk DCIS and/or use of a label that does not include the word 'cancer' might encourage the uptake of active surveillance and other conservative treatment options. We aimed to identify and collate relevant epidemiological evidence to inform further discussion on these ideas.

METHODS

We searched PubMed and EMBASE databases for low-risk DCIS studies in four categories: (1) natural history; (2) subclinical cancer found at autopsy; (3) diagnostic reproducibility (two or more pathologist interpretations at a single time point); and (4) diagnostic drift (two or more pathologist interpretations at different time points). Where we identified a pre-existing systematic review, the search was restricted to studies published after the inclusion period of the review. Two authors screened records, extracted data, and performed risk of bias assessment. We undertook a narrative synthesis of the included evidence within each category.

RESULTS

Natural History (n = 11): one systematic review and nine primary studies were included, but only five provided evidence on the prognosis of women with low-risk DCIS. These studies reported that women with low-risk DCIS had comparable outcomes whether or not they had surgery. The risk of invasive breast cancer in patients with low-risk DCIS ranged from 6.5% (7.5 years) to 10.8% (10 years). The risk of dying from breast cancer in patients with low-risk DCIS ranged from 1.2 to 2.2% (10 years). Subclinical cancer at autopsy (n = 1): one systematic review of 13 studies estimated the mean prevalence of subclinical in situ breast cancer to be 8.9%. Diagnostic reproducibility (n = 13): two systematic reviews and 11 primary studies found at most moderate agreement in differentiating low-grade DCIS from other diagnoses. Diagnostic drift: no studies found.

CONCLUSION

Epidemiological evidence supports consideration of relabelling and/or recalibrating diagnostic thresholds for low-risk DCIS. Such diagnostic changes would need agreement on the definition of low-risk DCIS and improved diagnostic reproducibility.

A Pragmatic Machine Learning Approach to Quantify Tumor-Infiltrating Lymphocytes in Whole Slide Images

Simple Summary

Tumor tissues sampled from patients contain prognostic and predictive information beyond what is currently being used in clinical practice. Large-scale digitization enables new ways of exploiting this information. The most promising analysis pipelines include deep learning/artificial intelligence (AI). However, to ensure success, AI often requires a time-consuming curation of data. In our approach, we repurposed AI pipelines and training data for cell segmentation and classification to identify tissue-infiltrating lymphocytes (TILs) in lung cancer tissue. We showed that our approach is able to identify TILs and provide prognostic information in an unseen dataset from lung cancer patients. Our methods can be adapted in myriad ways and may help pave the way for the large-scale deployment of digital pathology.

Abstract

Increased levels of tumor-infiltrating lymphocytes (TILs) indicate favorable outcomes in many types of cancer. The manual quantification of immune cells is inaccurate and time-consuming for pathologists. Our aim is to leverage a computational solution to automatically quantify TILs in standard diagnostic hematoxylin and eosin-stained sections (H&E slides) from lung cancer patients. Our approach is to transfer an open-source machine learning method for the segmentation and classification of nuclei in H&E slides trained on public data to TIL quantification without manual labeling of the data. Our results show that the resulting TIL quantification correlates to the patient prognosis and compares favorably to the current state-of-the-art method for immune cell detection in non-small cell lung cancer (current standard CD8 cells in DAB-stained TMAs HR 0.34, 95% CI 0.17–0.68 vs. TILs in HE WSIs: HoVer-Net PanNuke Aug Model HR 0.30, 95% CI 0.15–0.60 and HoVer-Net MoNuSAC Aug model HR 0.27, 95% CI 0.14–0.53). Our approach bridges the gap between machine learning research, translational clinical research and clinical implementation. However, further validation is warranted before implementation in a clinical setting.

Code-Free Development and Deployment of Deep Segmentation Models for Digital Pathology

Application of deep learning on histopathological whole slide images (WSIs) holds promise of improving diagnostic efficiency and reproducibility but is largely dependent on the ability to write computer code or purchase commercial solutions. We present a code-free pipeline utilizing free-to-use, open-source software (QuPath, DeepMIB, and FastPathology) for creating and deploying deep learning-based segmentation models for computational pathology. We demonstrate the pipeline on a use case of separating epithelium from stroma in colonic mucosa. A dataset of 251 annotated WSIs, comprising 140 hematoxylin-eosin (HE)-stained and 111 CD3 immunostained colon biopsy WSIs, were developed through active learning using the pipeline. On a hold-out test set of 36 HE and 21 CD3-stained WSIs a mean intersection over union score of 95.5 and 95.3% was achieved on epithelium segmentation. We demonstrate pathologist-level segmentation accuracy and clinical acceptable runtime performance and show that pathologists without programming experience can create near state-of-the-art segmentation solutions for histopathological WSIs using only free-to-use software. The study further demonstrates the strength of open-source solutions in its ability to create generalizable, open pipelines, of which trained models and predictions can seamlessly be exported in open formats and thereby used in external solutions. All scripts, trained models, a video tutorial, and the full dataset of 251 WSIs with ~31 k epithelium annotations are made openly available at https://github.com/andreped/NoCodeSeg to accelerate research in the field.

Preneoplastic Low-Risk Mammary Ductal Lesions (Atypical Ductal Hyperplasia and Ductal Carcinoma In Situ Spectrum): Current Status and Future Directions

Intraepithelial mammary ductal neoplasia is a spectrum of disease that varies from atypical ductal hyperplasia (ADH), low-grade (LG), intermediate-grade (IG), to high-grade (HG) ductal carcinoma in situ (DCIS). While ADH has the lowest prognostic significance, HG-DCIS carries the highest risk. Due to widely used screening mammography, the number of intraepithelial mammary ductal neoplastic lesions has increased. The consequence of this practice is the increase in the number of patients who are overdiagnosed and, therefore, overtreated. The active surveillance (AS) trials are initiated to separate lesions that require active treatment from those that can be safely monitored and only be treated when they develop a change in the clinical/radiologic characteristics. At the same time, the natural history of these lesions can be evaluated. This review aims to evaluate ADH/DCIS as a spectrum of intraductal neoplastic disease (risk and histomorphology); examine the controversies of distinguishing ADH vs. DCIS and the grading of DCIS; review the upgrading for both ADH and DCIS with emphasis on the variation of methods of detection and the definitions of upgrading; and evaluate the impact of all these variables on the AS trials.

Nuclear morphology in breast lesions: refining its assessment to improve diagnostic concordance

BACKGROUND

Although evaluation of nuclear morphology plays a crucial role in the diagnosis and categorisation of breast lesions, the criteria used to assess nuclear atypia rely on the subjective evaluation of several features that may result in inter- and intra-observer variation. This study aims to refine the definitions of cytonuclear features in various breast lesions.

METHODS

ImageJ was used to assess the nuclear morphological features including nuclear diameter, axis length, perimeter, area, circularity, and roundness in 160 breast lesions comprising ductal carcinoma in situ (DCIS), invasive breast carcinoma of no special type (IBC-NST), tubular carcinoma, usual ductal hyperplasia (UDH), columnar cell change (CCC) and flat epithelial atypia (FEA). Reference cells included normal epithelial cells, red blood cells (RBCs) and lymphocytes.

RESULTS

Reference cells showed size differences not only between normal epithelial cells and RBCs but also between RBCs in varied-sized blood vessels. Nottingham grade nuclear pleomorphism scores 1 and 3 cut-offs in IBC, compared to normal epithelial cells, were <1.2x and >1.4x that of mean maximum Feret's diameter and <1.6x and >2.4x that of mean nuclear area, respectively. Nuclear morphometrics were significantly different in low-grade IBC-NST vs. tubular carcinoma, low-grade DCIS vs. UDH, and in CCC vs. FEA. No differences in the nuclear features between grade matched DCIS and IBC were identified.

CONCLUSION

This study provides a guide for the assessment of nuclear atypia in breast lesions, refines the comparison with reference cells and highlights the potential diagnostic value of image analysis tools in the era of digital pathology.

On the stability of log-rank test under labeling errors

Motivation

Log-rank test is a widely used test that serves to assess the statistical significance of observed differences in survival, when comparing two or more groups. The log-rank test is based on several assumptions that support the validity of the calculations. It is naturally assumed, implicitly, that no errors occur in the labeling of the samples. That is, the mapping between samples and groups is perfectly correct. In this work, we investigate how test results may be affected when considering some errors in the original labeling.

Results

We introduce and define the uncertainty that arises from labeling errors in log-rank test. In order to deal with this uncertainty, we develop a novel algorithm for efficiently calculating a stability interval around the original log-rank P-value and prove its correctness. We demonstrate our algorithm on several datasets.

Availability and implementation

We provide a Python implementation, called LoRSI, for calculating the stability interval using our algorithm https://github.com/YakhiniGroup/LoRSI.

Supplementary information

Supplementary data are available at Bioinformatics online.

Interobserver Variability in Ductal Carcinoma In Situ of the Breast

OBJECTIVES

Since most patients with ductal carcinoma in situ (DCIS) of the breast are treated upon diagnosis, evidence on its natural progression to invasive carcinoma is limited. It is estimated that around half of the screen-detected DCIS lesions would have remained indolent if they had never been detected. Many patients with DCIS are therefore probably overtreated. Four ongoing randomized noninferiority trials explore active surveillance as a treatment option. Eligibility for these trials is mainly based on histopathologic features. Hence, the call for reproducible histopathologic assessment has never sounded louder.

METHODS

Here, the available classification systems for DCIS are discussed in depth.

RESULTS

This comprehensive review illustrates that histopathologic evaluation of DCIS is characterized by significant interobserver variability. Future digitalization of pathology, combined with development of deep learning algorithms or so-called artificial intelligence, may be an innovative solution to tackle this problem. However, implementation of digital pathology is not within reach for each laboratory worldwide. An alternative classification system could reduce the disagreement among histopathologists who use "conventional" light microscopy: the introduction of dichotomous histopathologic assessment is likely to increase interobserver concordance.

CONCLUSIONS

Reproducible histopathologic assessment is a prerequisite for robust risk stratification and adequate clinical decision-making. Two-tier histopathologic assessment might enhance the quality of care.

Integrating Artificial and Human Intelligence: A Partnership for Responsible Innovation in Biomedical Engineering and Medicine

Historically, the term "artificial intelligence" dates to 1956 when it was first used in a conference at Dartmouth College in the US. Since then, the development of artificial intelligence has in part been shaped by the field of neuroscience. By understanding the human brain, scientists have attempted to build new intelligent machines capable of performing complex tasks akin to humans. Indeed, future research into artificial intelligence will continue to benefit from the study of the human brain. While the development of artificial intelligence algorithms has been fast paced, the actual use of most artificial intelligence (AI) algorithms in biomedical engineering and clinical practice is still markedly below its conceivably broader potentials. This is partly because for any algorithm to be incorporated into existing workflows it has to stand the test of scientific validation, clinical and personal utility, application context, and is equitable as well. In this context, there is much to be gained by combining AI and human intelligence (HI). Harnessing Big Data, computing power and storage capacities, and addressing societal issues emergent from algorithm applications, demand deploying HI in tandem with AI. Very few countries, even economically developed states, lack adequate and critical governance frames to best understand and steer the AI innovation trajectories in health care. Drug discovery and translational pharmaceutical research stand to gain from AI technology provided they are also informed by HI. In this expert review, we analyze the ways in which AI applications are likely to traverse the continuum of life from birth to death, and encompassing not only humans but also all animal, plant, and other living organisms that are increasingly touched by AI. Examples of AI applications include digital health, diagnosis of diseases in newborns, remote monitoring of health by smart devices, real-time Big Data analytics for prompt diagnosis of heart attacks, and facial analysis software with consequences on civil liberties. While we underscore the need for integration of AI and HI, we note that AI technology does not have to replace medical specialists or scientists and rather, is in need of such expert HI. Altogether, AI and HI offer synergy for responsible innovation and veritable prospects for improving health care from prevention to diagnosis to therapeutics while unintended consequences of automation emergent from AI and algorithms should be borne in mind on scientific cultures, work force, and society at large.

Method for coregistration of optical measurements of breast tissue with histopathology: the importance of accounting for tissue deformations

For the validation of optical diagnostic technologies, experimental results need to be benchmarked against the gold standard. Currently, the gold standard for tissue characterization is assessment of hematoxylin and eosin (H&E)-stained sections by a pathologist. When processing tissue into H&E sections, the shape of the tissue deforms with respect to the initial shape when it was optically measured. We demonstrate the importance of accounting for these tissue deformations when correlating optical measurement with routinely acquired histopathology. We propose a method to register the tissue in the H&E sections to the optical measurements, which corrects for these tissue deformations. We compare the registered H&E sections to H&E sections that were registered with an algorithm that does not account for tissue deformations by evaluating both the shape and the composition of the tissue and using microcomputer tomography data as an independent measure. The proposed method, which did account for tissue deformations, was more accurate than the method that did not account for tissue deformations. These results emphasize the need for a registration method that accounts for tissue deformations, such as the method presented in this study, which can aid in validating optical techniques for clinical use.

A Roadmap for Foundational Research on Artificial Intelligence in Medical Imaging: From the 2018 NIH/RSNA/ACR/The Academy Workshop

Imaging research laboratories are rapidly creating machine learning systems that achieve expert human performance using open-source methods and tools. These artificial intelligence systems are being developed to improve medical image reconstruction, noise reduction, quality assurance, triage, segmentation, computer-aided detection, computer-aided classification, and radiogenomics. In August 2018, a meeting was held in Bethesda, Maryland, at the National Institutes of Health to discuss the current state of the art and knowledge gaps and to develop a roadmap for future research initiatives. Key research priorities include: 1, new image reconstruction methods that efficiently produce images suitable for human interpretation from source data; 2, automated image labeling and annotation methods, including information extraction from the imaging report, electronic phenotyping, and prospective structured image reporting; 3, new machine learning methods for clinical imaging data, such as tailored, pretrained model architectures, and federated machine learning methods; 4, machine learning methods that can explain the advice they provide to human users (so-called explainable artificial intelligence); and 5, validated methods for image de-identification and data sharing to facilitate wide availability of clinical imaging data sets. This research roadmap is intended to identify and prioritize these needs for academic research laboratories, funding agencies, professional societies, and industry.

Leveraging the variable natural history of ductal carcinoma in situ (DCIS) to select optimal therapy

PURPOSE

Ductal carcinoma in situ (DCIS) is a non-obligate precursor to invasive ductal carcinoma. The authors sought to discuss the evidence suggesting that not all DCIS will progress to invasive disease if left untreated.

RESULTS

Four lines of evidence align to suggest that not all of this in-situ disease progresses to invasive cancer: its prevalence on screening mammography, studies of missed diagnoses, incidental findings in autopsy specimens, and large retrospective reviews of those treated with excision alone.

CONCLUSION

A clearer understanding of the variable history of DCIS coupled with advances in genomic profiling of the disease holds the promise of reducing widespread over-treatment of this non-invasive cancer. Additionally, identification of higher risk of recurrence subsets may select patients for whom more aggressive treatment may be appropriate.

Ductal Carcinoma in Situ: Quantitative Preoperative Breast MR Imaging Features Associated with Recurrence after Treatment

Purpose To investigate whether specific imaging features on breast magnetic resonance (MR) images are associated with ductal carcinoma in situ (DCIS) recurrence risk after definitive treatment. Materials and Methods Patients with DCIS who underwent preoperative dynamic contrast material-enhanced (DCE) MR imaging between 2004 and 2014 with ipsilateral recurrence more than 6 months after definitive surgical treatment were retrospectively identified. For each patient, a control subject with DCIS that did not recur was identified and matched on the basis of clinical, histopathologic, and treatment features known to affect recurrence risk. On DCE MR images, lesion characteristics (longest diameter, functional tumor volume [FTV], peak percentage enhancement [PE], peak signal enhancement ratio [SER], and washout fraction) and normal tissue features (background parenchymal enhancement [BPE] volume, mean BPE) were quantitatively measured. MR imaging features were compared between patients and control subjects by using the Wilcoxon signed-rank test, with adjustment for multiple comparisons. Results Of 415 subjects with DCIS who underwent preoperative MR imaging, 14 experienced recurrence and 11 had an identifiable matching control subject (final cohort, 11 patients and 11 control subjects). Median time to recurrence was 14 months, and median follow-up for control subjects was 102 months. When compared with matched control subjects, patients with DCIS recurrence exhibited significantly greater FTV (median, 9.3 cm³ vs 1.3 cm³, P = .01), lesion peak SER (median, 1.7 vs 1.2; P = .03), and mean BPE (median, 58.3% vs 41.1%; P = .02). Conclusion Quantitative lesion and normal breast tissue characteristics at preoperative MR imaging in women with newly diagnosed DCIS show promise for association with breast cancer recurrence after treatment. ^© RSNA, 2017.