Tissue biomarker scoring by pathologists is central to defining the appropriate therapy for patients with cancer. Yet, inter-pathologist variability in the interpretation of ambiguous cases can ...affect diagnostic accuracy. Modern artificial intelligence methods such as deep learning have the potential to supplement pathologist expertise to ensure constant diagnostic accuracy. We developed a computational approach based on deep learning that automatically scores HER2, a biomarker that defines patient eligibility for anti-HER2 targeted therapies in breast cancer. In a cohort of 71 breast tumour resection samples, automated scoring showed a concordance of 83% with a pathologist. The twelve discordant cases were then independently reviewed, leading to a modification of diagnosis from initial pathologist assessment for eight cases. Diagnostic discordance was found to be largely caused by perceptual differences in assessing HER2 expression due to high HER2 staining heterogeneity. This study provides evidence that deep learning aided diagnosis can facilitate clinical decision making in breast cancer by identifying cases at high risk of misdiagnosis.
Histology is the gold standard to unveil microscopic brain structures and pathological alterations in humans and animal models of disease. However, due to tedious manual interventions, quantification ...of histopathological markers is classically performed on a few tissue sections, thus restricting measurements to limited portions of the brain. Recently developed 3D microscopic imaging techniques have allowed in-depth study of neuroanatomy. However, quantitative methods are still lacking for whole-brain analysis of cellular and pathological markers. Here, we propose a ready-to-use, automated, and scalable method to thoroughly quantify histopathological markers in 3D in rodent whole brains. It relies on block-face photography, serial histology and 3D-HAPi (Three Dimensional Histology Analysis Pipeline), an open source image analysis software. We illustrate our method in studies involving mouse models of Alzheimer's disease and show that it can be broadly applied to characterize animal models of brain diseases, to evaluate therapeutic interventions, to anatomically correlate cellular and pathological markers throughout the entire brain and to validate in vivo imaging techniques.
Programmed cell death ligand-1 (PD-L1) expression levels in patients’ tumors have demonstrated clinical utility across many cancer types and are used to determine treatment eligibility. Several ...independently developed PD-L1 immunohistochemical (IHC) predictive assays are commercially available and have demonstrated different levels of staining between assays, generating interest in understanding the similarities and differences between assays. Previously, we identified epitopes in the internal and external domains of PD-L1, bound by antibodies in routine clinical use (SP263, SP142, 22C3, and 28-8). Variance in performance of assays utilizing these antibodies, observed following exposure to preanalytical factors such as decalcification, cold ischemia, and duration of fixation, encouraged additional investigation of antibody-binding sites, to understand whether binding site structures/conformations contribute to differential PD-L1 IHC assay staining. We proceeded to further investigate the epitopes on PD-L1 bound by these antibodies, alongside the major clones utilized in laboratory-developed tests (E1L3N, QR1, and 73-10). Characterization of QR1 and 73-10 clones demonstrated that both bind the PD-L1 C-terminal internal domain, similar to SP263/SP142. Our results also demonstrate that under suboptimal decalcification or fixation conditions, the performance of internal domain antibodies is less detrimentally affected than that of external domain antibodies 22C3/28-8. Furthermore, we show that the binding sites of external domain antibodies are susceptible to deglycosylation and conformational structural changes, which directly result in IHC staining reduction or loss. The binding sites of internal domain antibodies were unaffected by deglycosylation or conformational structural change. This study demonstrates that the location and conformation of binding sites, recognized by antibodies employed in PD-L1 diagnostic assays, differ significantly and exhibit differing degrees of robustness. These findings should reinforce the need for vigilance when performing clinical testing with different PD-L1 IHC assays, particularly in the control of cold ischemia and the selection of fixation and decalcification conditions.
Recently developed techniques to visualize immunostained tissues in 3D and in large samples have expanded the scope of microscopic investigations at the level of the whole brain. Here, we propose to ...adapt voxel-based statistical analysis to 3D high-resolution images of the immunostained rodent brain. The proposed approach was first validated with a simulation dataset with known cluster locations. Then, it was applied to characterize the effect of ADAM30, a gene involved in the metabolism of the amyloid precursor protein, in a mouse model of Alzheimer's disease. This work introduces voxel-based analysis of 3D immunostained microscopic brain images and, therefore, opens the door to localized whole-brain exploratory investigation of pathological markers and cellular alterations.
Because they bridge the genetic gap between rodents and humans, non-human primates (NHPs) play a major role in therapy development and evaluation for neurological disorders. However, translational ...research success from NHPs to patients requires an accurate phenotyping of the models. In patients, magnetic resonance imaging (MRI) combined with automated segmentation methods has offered the unique opportunity to assess in vivo brain morphological changes. Meanwhile, specific challenges caused by brain size and high field contrasts make existing algorithms hard to use routinely in NHPs. To tackle this issue, we propose a complete pipeline, Primatologist, for multi-region segmentation. Tissue segmentation is based on a modular statistical model that includes random field regularization, bias correction and denoising and is optimized by expectation-maximization. To deal with the broad variety of structures with different relaxing times at 7 T, images are segmented into 17 anatomical classes, including subcortical regions. Pre-processing steps insure a good initialization of the parameters and thus the robustness of the pipeline. It is validated on 10 T2-weighted MRIs of healthy macaque brains. Classification scores are compared with those of a non-linear atlas registration, and the impact of each module on classification scores is thoroughly evaluated.
•A segmentation pipeline is proposed to the non-human primate neuroimaging community.•It allows automatic segmentation of Macaque brain MRIs into 17 anatomical classes.•It relies on a generative model of intensity and a 3D digital atlas.•We showed that Primatologist performs better than a conventional atlas registration.
Extracellular deposition of β amyloid plaques is an early event associated to Alzheimer's disease. Here, we have used in vivo gadolinium-stained high resolution (29(∗)29(∗)117 μm(3)) magnetic ...resonance imaging (MRI) to follow-up in a longitudinal way individual amyloid plaques in APP/PS1 mice and evaluate the efficacy of a new immunotherapy (SAR255952) directed against protofibrillar and fibrillary forms of Aβ. APP/PS1 mice were treated for 5 months between the age of 3.5 and 8.5 months. SAR255952 reduced amyloid load in 8.5-months-old animals, but not in 5.5-months animals compared to mice treated with a control antibody (DM4). Histological evaluation confirmed the reduction of amyloid load and revealed a lower density of amyloid plaques in 8.5-months SAR255952-treated animals. The longitudinal follow-up of individual amyloid plaques by MRI revealed that plaques that were visible at 5.5 months were still visible at 8.5 months in both SAR255952 and DM4-treated mice. This suggests that the amyloid load reduction induced by SAR255952 is related to a slowing down in the formation of new plaques rather than to the clearance of already formed plaques.
The c-Met/hepatocyte growth factor receptor pathway is frequently dysregulated in multiple cancer types, including non-small cell lung cancer (NSCLC). MET amplification has been shown to develop as a ...resistance mechanism to treatment in NSCLC. The identification of increased MET copy number within tumour cells is increasingly important to stratify those tumours and patients which are susceptible to treatment targetting MET kinase inhibition. Fluorescence in situ hybridisation (FISH) has been successfully employed to identify patients with abnormal MET gene copy number with numerous probes available for use. Here we report a FISH protocol that reduces probe hybridisation time in NSCLC tissue to 1 hour and compare the results with other protocols. MET gene copy number was determined in 20 NSCLC cases using 3 FISH probes: 1. Kreatech FISH, MET (7q31) SE 7 ready to use probes, hybridised using an overnight protocol; 2. Dako MET IQFISH probe with CEP7 ready to use probe, hybridised for 2 hours; 3. Kreatech MET (7q31) SE 7 XL FISH probe, prepared in SwiftFISH buffer and hybridised for 1 hour. The MET gene copy number and MET: centromere 7 gene ratio were determined for each tissue and cases categorised as having MET high or MET low status. All three FISH probes were shown to demonstrate good agreement with each other. Overall percentage agreement between probes was ≥90%. Intraclass correlation showed good agreement (ICC ≥0.80) between all three assays for MET gene copy number and MET: centromere 7 gene ratio. These FISH protocols provide evidence that rapid laboratory developed FISH assays with short turnaround time perform consistently with standard protocols, potentially enabling faster treatment decisions.
Validation data for segmentation algorithms dedicated to preclinical images is fiercely lacking, especially when compared to the large number of databases of Human brain images and segmentations ...available to the academic community. Not only is such data essential for validating methods, it is also needed for objectively comparing concurrent algorithms and detect promising paths, as segmentation challenges have shown for clinical images.
The dataset we present here is a first step in this direction. It comprises 10 T2-weighted MRIs of healthy adult macaque brains, acquired on a 7T magnet, along with corresponding manual segmentations into 17 brain anatomic labelled regions spread over 5 hierarchical levels based on a previously published macaque atlas (Calabrese et al., 2015) 1.
By giving access to this unique dataset, we hope to provide a reference needed by the non-human primate imaging community. This dataset was used in an article presenting a new primate brain morphology analysis pipeline, Primatologist (Balbastre et al., 2017) 2. Data is available through a NITRC repository (https://www.nitrc.org/projects/mircen_macset).
In biomedical research, cell counting is important to assess physiological and pathophysiological information. However, the automated analysis of microscopic images of tissues remains extremely ...challenging. We propose an automated processing protocol for proper segmentation of individual cells in microscopic images. A Gaussian filter is applied to improve signal to noise ratio (SNR) then an original minmax method is proposed to produce an image in which information describing both cell centers (minima) and boundaries are enhanced. Finally, a contour-based model initialized from minima in the min-max cartography is carried out to achieve cell individualization. This method is evaluated on a NeuN-stained macaque brain section in sub-regions presenting various levels of fraction of neuron surface occupation. Comparison with several methods of reference demonstrates that the performances of our method are superior. A first application to the segmentation of neurons in the hippocampus illustrates the ability of our approach to deal with massive and complex data.