Image and statistical analysis of melanocytic histology

Visual Intratumor Heterogeneity and Breast Tumor Progression

High intratumoral heterogeneity is thought to be a poor prognostic indicator. However, the source of heterogeneity may also be important, as genomic heterogeneity is not always reflected in histologic or 'visual' heterogeneity. We aimed to develop a predictor of histologic heterogeneity and evaluate its association with outcomes and molecular heterogeneity. We used VGG16 to train an image classifier to identify unique, patient-specific visual features in 1655 breast tumors (5907 core images) from the Carolina Breast Cancer Study (CBCS). Extracted features for images, as well as the epithelial and stromal image components, were hierarchically clustered, and visual heterogeneity was defined as a greater distance between images from the same patient. We assessed the association between visual heterogeneity, clinical features, and DNA-based molecular heterogeneity using generalized linear models, and we used Cox models to estimate the association between visual heterogeneity and tumor recurrence. Basal-like and ER-negative tumors were more likely to have low visual heterogeneity, as were the tumors from younger and Black women. Less heterogeneous tumors had a higher risk of recurrence (hazard ratio = 1.62, 95% confidence interval = 1.22-2.16), and were more likely to come from patients whose tumors were comprised of only one subclone or had a TP53 mutation. Associations were similar regardless of whether the image was based on stroma, epithelium, or both. Histologic heterogeneity adds complementary information to commonly used molecular indicators, with low heterogeneity predicting worse outcomes. Future work integrating multiple sources of heterogeneity may provide a more comprehensive understanding of tumor progression.

Evaluation of medical decision support systems (DDX generators) using real medical cases of varying complexity and origin

Background

Medical decision support systems (CDSSs) are increasingly used in medicine, but their utility in daily medical practice is difficult to evaluate. One variant of CDSS is a generator of differential diagnoses (DDx generator). We performed a feasibility study on three different, publicly available data sets of medical cases in order to identify the frequency in which two different DDx generators provide helpful information (either by providing a list of differential diagnosis or recognizing the expert diagnosis if available) for a given case report.

Methods

Used data sets were n = 105 cases from a web-based forum of telemedicine with real life cases from Afghanistan (Afghan data set; AD), n = 124 cases discussed in a web-based medical forum (Coliquio data set; CD). Both websites are restricted for medical professionals only. The third data set consisted 50 special case reports published in the New England Journal of Medicine (NEJM). After keyword extraction, data were entered into two different DDx generators (IsabelHealth (IH), Memem7 (M7)) to examine differences in target diagnosis recognition and physician-rated usefulness between DDx generators.

Results

Both DDx generators detected the target diagnosis equally successfully (all cases: M7, 83/170 (49%); IH 90/170 (53%), NEJM: M7, 28/50 (56%); IH, 34/50 (68%); differences n.s.). Differences occurred in AD, where detection of an expert diagnosis was less successful with IH than with M7 (29.7% vs. 54.1%, p = 0.003). In contrast, in CD IH performed significantly better than M7 (73.9% vs. 32.6%, p = 0.021). Congruent identification of target diagnosis occurred in only 46/170 (27.1%) of cases. However, a qualitative analysis of the DDx results revealed useful complements from using the two systems in parallel.

Conclusion

Both DDx systems IsabelHealth and Memem7 provided substantial help in finding a helpful list of differential diagnoses or identifying the target diagnosis either in standard cases or complicated and rare cases. Our pilot study highlights the need for different levels of complexity and types of real-world medical test cases, as there are significant differences between DDx generators away from traditional case reports. Combining different results from DDx generators seems to be a possible approach for future review and use of the systems.

Nuclear morphometric analysis in tissue as an objective tool with potential use to improve melanoma staging

Alterations in nuclear size and shape are commonly observed in cancers, and its objective evaluation may provide valuable clinical information about the outcome of the disease. Here, we applied the nuclear morphometric analysis in tissues in hematoxylin and eosin-digitized slides of nevi and melanoma, to objectively contribute to the prognostic evaluation of these tumors. To this, we analyzed the nuclear morphometry of 34 melanomas classified according to the TNM stage. Eight cases of melanocytic nevi were used as non-neoplastic tissues to set the non-neoplastic parameters of nuclear morphology. Our samples were set as G1 (control, nevi), G2 (T1T2N0M0), G3 (T3T4N0M0), G4 (T1T2N1M1), and G5 (T3T4N1M1). Image-Pro Plus 6.0 software was used to acquire measurements related to nuclear size (variable: Area) and shape (variables: Aspect, AreaBox, Roundness, and RadiusRatio, which were used to generate the Nuclear Irregularity Index). From these primary variables, a set of secondary variables were generated. All the seven primary and secondary variables related to the nuclear area were different among groups (Pillai's trace P<0.001), whereas Nuclear Irregularity Index, which is the variable related to nuclear shape, did not differ among groups. The secondary variable 'Average Area of Large Nuclei' was able to differ all pairwise comparisons, including thin nonmetastatic from thin metastatic tumors. In conclusion, the objective quantification of nuclear area in hematoxylin and eosin slides may provide objective information about the risk stratification of these tumors and has the potential to be used as an additional method in clinical decision making.

Image analysis with deep learning to predict breast cancer grade, ER status, histologic subtype, and intrinsic subtype

RNA-based, multi-gene molecular assays are available and widely used for patients with ER-positive/HER2-negative breast cancers. However, RNA-based genomic tests can be costly and are not available in many countries. Methods for inferring molecular subtype from histologic images may identify patients most likely to benefit from further genomic testing. To identify patients who could benefit from molecular testing based on H&E stained histologic images, we developed an image analysis approach using deep learning. A training set of 571 breast tumors was used to create image-based classifiers for tumor grade, ER status, PAM50 intrinsic subtype, histologic subtype, and risk of recurrence score (ROR-PT). The resulting classifiers were applied to an independent test set (n = 288), and accuracy, sensitivity, and specificity of each was assessed on the test set. Histologic image analysis with deep learning distinguished low-intermediate vs. high tumor grade (82% accuracy), ER status (84% accuracy), Basal-like vs. non-Basal-like (77% accuracy), Ductal vs. Lobular (94% accuracy), and high vs. low-medium ROR-PT score (75% accuracy). Sampling considerations in the training set minimized bias in the test set. Incorrect classification of ER status was significantly more common for Luminal B tumors. These data provide proof of principle that molecular marker status, including a critical clinical biomarker (i.e., ER status), can be predicted with accuracy >75% based on H&E features. Image-based methods could be promising for identifying patients with a greater need for further genomic testing, or in place of classically scored variables typically accomplished using human-based scoring.

Use of computer-assisted image analysis for semi-quantitative histology of the hepatopancreas in whiteleg shrimp Penaeus vannamei (Boone)

Despite the increasing use of novel molecular techniques in pathology, histology remains the standard method for monitoring tissue alterations and for assessing pathology. Histopathological evaluation is generally laborious and subjective with risk of discrepancies in semi-quantitative scoring between pathologists. In contrast, computer-assisted image analysis (CAIA) is potentially faster, more objective and thus suitable for routine screening. Limited research has been carried out on CAIA in crustacean histopathology, and the methods described were not fully automated. Therefore, the aim of this study was to develop CAIA in whiteleg shrimp (Penaeus vannamei) for the study of the hepatopancreas. Paraffin sections were immunohistochemically stained with monoclonal antibodies WSH8 against haemocytes and counterstained with Mayer's haematoxylin for detection of haemocytes and B-cell vacuoles, and modified toluidine blue protocol was used for detection of F-cells; frozen sections were stained with Oil Red O for detection of lipid droplets within R-cells. Visiopharm^® software was used to develop and validated protocols for the quantification of morphological parameters (areas of haemocyte infiltration, F-cells, B-cell vacuoles, lipid droplets and their ratios to total tissue area and total lumen area). These protocols enable the future use of CAIA for determination of the nutritional and pathological condition of this organ.

Automatic measurement of melanoma depth of invasion in skin histopathological images

Measurement of melanoma depth of invasion (DoI) in skin tissues is of great significance in grading the severity of skin disease and planning patient's treatment. However, accurate and automatic measurement of melanocytic tumor depth is a challenging problem mainly due to the difficulty of skin granular identification and melanoma detection. In this paper, we propose a technique for measuring melanoma DoI in microscopic images digitized from MART1 (i.e., meleanoma-associated antigen recognized by T cells) stained skin histopathological sections. The technique consists of four modules. First, skin melanoma areas are detected by combining color features with the Mahalanobis distance measure. Next, skin epidermis is segmented by a multi-thresholding method. The skin granular layer is then identified based on Bayesian classification of segmented skin epidermis pixels. Finally, the melanoma DoI is computed using a multi-resolution approach with Hausdorff distance measurement. Experimental results show that the proposed technique provides a superior performance in measuring the melanoma DoI than two closely related techniques.

New Insights into Mechanisms and Functions of Nuclear Size Regulation

Nuclear size is generally maintained within a defined range in a given cell type. Changes in cell size that occur during cell growth, development, and differentiation are accompanied by dynamic nuclear size adjustments in order to establish appropriate nuclear-to-cytoplasmic volume relationships. It has long been recognized that aberrations in nuclear size are associated with certain disease states, most notably cancer. Nuclear size and morphology must impact nuclear and cellular functions. Understanding these functional implications requires an understanding of the mechanisms that control nuclear size. In this review, we first provide a general overview of the diverse cellular structures and activities that contribute to nuclear size control, including structural components of the nucleus, effects of DNA amount and chromatin compaction, signaling, and transport pathways that impinge on the nucleus, extranuclear structures, and cell cycle state. We then detail some of the key mechanistic findings about nuclear size regulation that have been gleaned from a variety of model organisms. Lastly, we review studies that have implicated nuclear size in the regulation of cell and nuclear function and speculate on the potential functional significance of nuclear size in chromatin organization, gene expression, nuclear mechanics, and disease. With many fundamental cell biological questions remaining to be answered, the field of nuclear size regulation is still wide open.

Appearance normalization of histology slides

This paper presents a method for automatic color and intensity normalization of digitized histology slides stained with two different agents. In comparison to previous approaches, prior information on the stain vectors is used in the plane estimation process, resulting in improved stability of the estimates. Due to the prevalence of hematoxylin and eosin staining for histology slides, the proposed method has significant practical utility. In particular, it can be used as a first step to standardize appearance across slides and is effective at countering effects due to differing stain amounts and protocols and counteracting slide fading. The approach is validated against non-prior plane-fitting using synthetic experiments and 13 real datasets. Results of application of the method to adjustment of faded slides are given, and the effectiveness of the method in aiding statistical classification is shown.

Loss of 5-hydroxymethylcytosine correlates with increasing morphologic dysplasia in melanocytic tumors

DNA methylation is the most well-studied epigenetic modification in cancer biology. 5-hydroxymethylcytosine is an epigenetic mark that can be converted from 5-methylcytosine by the ten-eleven translocation gene family. We recently reported the loss of 5-hydroxymethylcytosine in melanoma compared with benign nevi and suggested that loss of this epigenetic marker is correlated with tumor virulence based on its association with a worse prognosis. In this study, we further characterize the immunoreactivity patterns of 5-hydroxymethylcytosine in the full spectrum of melanocytic lesions to further validate the potential practical application of this epigenetic marker. One hundred and seventy-five cases were evaluated: 18 benign nevi, 20 dysplastic nevi (10 low-grade and 10 high-grade lesions), 10 atypical Spitz nevi, 20 borderline tumors, 5 melanomas arising within nevi, and 102 primary melanomas. Progressive loss of 5-hydroxymethylcytosine from benign dermal nevi to high-grade dysplastic nevi to borderline melanocytic neoplasms to melanoma was observed. In addition, an analysis of the relationship of nuclear diameter with 5-hydroxymethylcytosine staining intensity within lesional cells revealed a significant correlation between larger nuclear diameter and decreased levels of 5-hydroxymethylcytosine. Furthermore, borderline lesions uniquely exhibited a diverse spectrum of staining of each individual case. This study further substantiates the association of 5-hydroxymethylcytosine loss with dysplastic cytomorphologic features and tumor progression and supports the classification of borderline lesions as a biologically distinct category of melanocytic lesions.

Sizing and shaping the nucleus: mechanisms and significance

The size and shape of the nucleus are tightly regulated, indicating the physiological significance of proper nuclear morphology, yet the mechanisms and functions of nuclear size and shape regulation remain poorly understood. Correlations between altered nuclear morphology and certain disease states have long been observed, most notably many cancers are diagnosed and staged based on graded increases in nuclear size. Here we review recent studies investigating the mechanisms regulating nuclear size and shape, how mitotic events influence nuclear morphology, and the role of nuclear size and shape in subnuclear chromatin organization and cancer progression.