External validation of a deep learning-based algorithm for detection of tall cells in papillary thyroid carcinoma: A multicenter study.

The tall cell subtype (TC-PTC) is an aggressive subtype of papillary thyroid carcinoma (PTC). The TC-PTC is defined as a PTC comprising at least 30% epithelial cells that are three times as tall as they are wide. In practice, this definition is difficult to adhere to, resulting in high inter-observer variability. In this multicenter study, we validated a previously trained deep learning (DL)-based algorithm for detection of tall cells on 160 externally collected hematoxylin and eosin (HE)-stained PTC whole-slide images. In a test set of 360 manual annotations of regions of interest from 18 separate tissue sections in the external dataset, the DL-based algorithm detected TCs with a sensitivity of 90.6% and a specificity of 88.5%. The DL algorithm detected non-TC areas with a sensitivity of 81.6% and a specificity of 92.9%. In the validation datasets, 20% and 30% TC thresholds correlated with a significantly shorter relapse-free survival. In conclusion, the DL algorithm detected TCs in unseen, external scanned HE tissue slides with high sensitivity and specificity without any retraining.

A deep learning-based algorithm for tall cell detection in papillary thyroid carcinoma.

INTRODUCTION: According to the World Health Organization, the tall cell variant (TCV) is an aggressive subtype of papillary thyroid carcinoma (PTC) comprising at least 30% epithelial cells two to three times as tall as they are wide. In practice, applying this definition is difficult causing substantial interobserver variability. We aimed to train a deep learning algorithm to detect and quantify the proportion of tall cells (TCs) in PTC. METHODS: We trained the deep learning algorithm using supervised learning, testing it on an independent dataset, and further validating it on an independent set of 90 PTC samples from patients treated at the Hospital District of Helsinki and Uusimaa between 2003 and 2013. We compared the algorithm-based TC percentage to the independent scoring by a human investigator and how those scorings associated with disease outcomes. Additionally, we assessed the TC score in 71 local and distant tumor relapse samples from patients with aggressive disease. RESULTS: In the test set, the deep learning algorithm detected TCs with a sensitivity of 93.7% and a specificity of 94.5%, whereas the sensitivity fell to 90.9% and specificity to 94.1% for non-TC areas. In the validation set, the deep learning algorithm TC scores correlated with a diminished relapse-free survival using cutoff points of 10% (p = 0.044), 20% (p < 0.01), and 30% (p = 0.036). The visually assessed TC score did not statistically significantly predict survival at any of the analyzed cutoff points. We observed no statistically significant difference in the TC score between primary tumors and relapse tumors determined by the deep learning algorithm or visually. CONCLUSIONS: We present a novel deep learning-based algorithm to detect tall cells, showing that a high deep learning-based TC score represents a statistically significant predictor of less favorable relapse-free survival in PTC.

Outcome and Biomarker Supervised Deep Learning for Survival Prediction in Two Multicenter Breast Cancer Series.

BACKGROUND: Prediction of clinical outcomes for individual cancer patients is an important step in the disease diagnosis and subsequently guides the treatment and patient counseling. In this work, we develop and evaluate a joint outcome and biomarker supervised (estrogen receptor expression and ERBB2 expression and gene amplification) multitask deep learning model for prediction of outcome in breast cancer patients in two nation-wide multicenter studies in Finland (the FinProg and FinHer studies). Our approach combines deep learning with expert knowledge to provide more accurate, robust, and integrated prediction of breast cancer outcomes. MATERIALS AND METHODS: Using deep learning, we trained convolutional neural networks (CNNs) with digitized tissue microarray (TMA) samples of primary hematoxylin-eosin-stained breast cancer specimens from 693 patients in the FinProg series as input and breast cancer-specific survival as the endpoint. The trained algorithms were tested on 354 TMA patient samples in the same series. An independent set of whole-slide (WS) tumor samples from 674 patients in another multicenter study (FinHer) was used to validate and verify the generalization of the outcome prediction based on CNN models by Cox survival regression and concordance index (c-index). Visual cancer tissue characterization, i.e., number of mitoses, tubules, nuclear pleomorphism, tumor-infiltrating lymphocytes, and necrosis was performed on TMA samples in the FinProg test set by a pathologist and combined with deep learning-based outcome prediction in a multitask algorithm. RESULTS: The multitask algorithm achieved a hazard ratio (HR) of 2.0 (95% confidence interval [CI] 1.30-3.00), P < 0.001, c-index of 0.59 on the 354 test set of FinProg patients, and an HR of 1.7 (95% CI 1.2-2.6), P = 0.003, c-index 0.57 on the WS tumor samples from 674 patients in the independent FinHer series. The multitask CNN remained a statistically independent predictor of survival in both test sets when adjusted for histological grade, tumor size, and axillary lymph node status in a multivariate Cox analyses. An improved accuracy (c-index 0.66) was achieved when deep learning was combined with the tissue characteristics assessed visually by a pathologist. CONCLUSIONS: A multitask deep learning algorithm supervised by both patient outcome and biomarker status learned features in basic tissue morphology predictive of survival in a nationwide, multicenter series of patients with breast cancer. The algorithms generalized to another independent multicenter patient series and whole-slide breast cancer samples and provide prognostic information complementary to that of a comprehensive series of established prognostic factors.

HLA-G expression correlates with histological grade but not with prognosis in colorectal carcinoma.

Trophoblast-specific expression of HLA-G induces immune tolerance for the developing fetus. Pathological HLA-G expression later in life might contribute to immune escape of various cancers. We studied the still controversial role of HLA-G in colorectal carcinoma (CRC) using the MEM-G/1 antibody and a tissue microarray series of CRC tumors (n = 317). HLA-G expression appeared in 20% of the tumors and showed high intratumoral heterogeneity. HLA-G positivity was associated with better differentiation (p = 0.002) and non-mucinous histology (p = 0.008). However, HLA-G expression alone showed no prognostic value: 5-years disease-specific survival among patients with HLA-G expression was 68.9% (95% CI: 62.7%-75.0%) compared to 74.8% (95% CI: 63.2%-86.3%) among those without expression. These results support a modulatory role of HLA-G in CRC.

Point-of-Care Digital Cytology With Artificial Intelligence for Cervical Cancer Screening in a Resource-Limited Setting.

Importance: Cervical cancer is highly preventable but remains a common and deadly cancer in areas without screening programs. The creation of a diagnostic system to digitize Papanicolaou test samples and analyze them using a cloud-based deep learning system (DLS) may provide needed cervical cancer screening to resource-limited areas. Objective: To determine whether artificial intelligence-supported digital microscopy diagnostics can be implemented in a resource-limited setting and used for analysis of Papanicolaou tests. Design, Setting, and Participants: In this diagnostic study, cervical smears from 740 HIV-positive women aged between 18 and 64 years were collected between September 1, 2018, and September 30, 2019. The smears were digitized with a portable slide scanner, uploaded to a cloud server using mobile networks, and used to train and validate a DLS for the detection of atypical cervical cells. This single-center study was conducted at a local health care center in rural Kenya. Exposures: Detection of squamous cell atypia in the digital samples by analysis with the DLS. Main Outcomes and Measures: The accuracy of the DLS in the detection of low- and high-grade squamous intraepithelial lesions in Papanicolaou test whole-slide images. Results: Papanicolaou test results from 740 HIV-positive women (mean [SD] age, 41.8 [10.3] years) were collected. The DLS was trained using 350 whole-slide images and validated on 361 whole-slide images (average size, 100 387 × 47 560 pixels). For detection of cervical cellular atypia, sensitivities were 95.7% (95% CI, 85.5%-99.5%) and 100% (95% CI, 82.4%-100%), and specificities were 84.7% (95% CI, 80.2%-88.5%) and 78.4% (95% CI, 73.6%-82.4%), compared with the pathologist assessment of digital and physical slides, respectively. Areas under the receiver operating characteristic curve were 0.94 and 0.96, respectively. Negative predictive values were high (99%-100%), and accuracy was high, particularly for the detection of high-grade lesions. Interrater agreement was substantial compared with the pathologist assessment of digital slides (κ = 0.72) and fair compared with the assessment of glass slides (κ = 0.36). No samples that were classified as high grade by manual sample analysis had false-negative assessments by the DLS. Conclusions and Relevance: In this study, digital microscopy with artificial intelligence was implemented at a rural clinic and used to detect atypical cervical smears with a high sensitivity compared with visual sample analysis.

Deep learning identifies morphological features in breast cancer predictive of cancer ERBB2 status and trastuzumab treatment efficacy.

The treatment of patients with ERBB2 (HER2)-positive breast cancer with anti-ERBB2 therapy is based on the detection of ERBB2 gene amplification or protein overexpression. Machine learning (ML) algorithms can predict the amplification of ERBB2 based on tumor morphological features, but it is not known whether ML-derived features can predict survival and efficacy of anti-ERBB2 treatment. In this study, we trained a deep learning model with digital images of hematoxylin-eosin (H&E)-stained formalin-fixed primary breast tumor tissue sections, weakly supervised by ERBB2 gene amplification status. The gene amplification was determined by chromogenic in situ hybridization (CISH). The training data comprised digitized tissue microarray (TMA) samples from 1,047 patients. The correlation between the deep learning-predicted ERBB2 status, which we call H&E-ERBB2 score, and distant disease-free survival (DDFS) was investigated on a fully independent test set, which included whole-slide tumor images from 712 patients with trastuzumab treatment status available. The area under the receiver operating characteristic curve (AUC) in predicting gene amplification in the test sets was 0.70 (95% CI, 0.63-0.77) on 354 TMA samples and 0.67 (95% CI, 0.62-0.71) on 712 whole-slide images. Among patients with ERBB2-positive cancer treated with trastuzumab, those with a higher than the median morphology-based H&E-ERBB2 score derived from machine learning had more favorable DDFS than those with a lower score (hazard ratio [HR] 0.37; 95% CI, 0.15-0.93; P = 0.034). A high H&E-ERBB2 score was associated with unfavorable survival in patients with ERBB2-negative cancer as determined by CISH. ERBB2-associated morphology correlated with the efficacy of adjuvant anti-ERBB2 treatment and can contribute to treatment-predictive information in breast cancer.

Antibody Supervised Training of a Deep Learning Based Algorithm for Leukocyte Segmentation in Papillary Thyroid Carcinoma.

The quantity of leukocytes in papillary thyroid carcinoma (PTC) potentially have prognostic and treatment predictive value. Here, we propose a novel method for training a convolutional neural network (CNN) algorithm for segmenting leukocytes in PTCs. Tissue samples from two retrospective PTC cohort were obtained and representative tissue slides from twelve patients were stained with hematoxylin and eosin (HE) and digitized. Then, the HE slides were destained and restained immunohistochemically (IHC) with antibodies to the pan-leukocyte anti CD45 antigen and scanned again. The two stain-pairs of all representative tissue slides were registered, and image tiles of regions of interests were exported. The image tiles were processed and the 3,3'-diaminobenzidine (DAB) stained areas representing anti CD45 expression were turned into binary masks. These binary masks were applied as annotations on the HE image tiles and used in the training of a CNN algorithm. Ten whole slide images (WSIs) were used for training using a five-fold cross-validation and the remaining two slides were used as an independent test set for the trained model. For visual evaluation, the algorithm was run on all twelve WSIs, and in total 238,144 tiles sized 500 × 500 pixels were analyzed. The trained CNN algorithm had an intersection over union of 0.82 for detection of leukocytes in the HE image tiles when comparing the prediction masks to the ground truth anti CD45 mask. We conclude that this method for generating antibody supervised annotations using the destain-restain IHC guided annotations resulted in high accuracy segmentations of leukocytes in HE tissue images.

Osteoid Metaplasia in Femoral Artery Plaques Is Associated With the Clinical Severity of Lower Extremity Artery Disease in Men.

Lamellar metaplastic bone, osteoid metaplasia (OM), is found in atherosclerotic plaques, especially in the femoral arteries. In the carotid arteries, OM has been documented to be associated with plaque stability. This study investigated the clinical impact of OM load in femoral artery plaques of patients with lower extremity artery disease (LEAD) by using a deep learning-based image analysis algorithm. Plaques from 90 patients undergoing endarterectomy of the common femoral artery were collected and analyzed. After decalcification and fixation, 4-µm-thick longitudinal sections were stained with hematoxylin and eosin, digitized, and uploaded as whole-slide images on a cloud-based platform. A deep learning-based image analysis algorithm was trained to analyze the area percentage of OM in whole-slide images. Clinical data were extracted from electronic patient records, and the association with OM was analyzed. Fifty-one (56.7%) sections had OM. Females with diabetes had a higher area percentage of OM than females without diabetes. In male patients, the area percentage of OM inversely correlated with toe pressure and was significantly associated with severe symptoms of LEAD including rest pain, ulcer, or gangrene. According to our results, OM is a typical feature of femoral artery plaques and can be quantified using a deep learning-based image analysis method. The association of OM load with clinical features of LEAD appears to differ between male and female patients, highlighting the need for a gender-specific approach in the study of the mechanisms of atherosclerotic disease. In addition, the role of plaque characteristics in the treatment of atherosclerotic lesions warrants further consideration in the future.

Deep Learning Algorithms for Corneal Amyloid Deposition Quantitation in Familial Amyloidosis.

OBJECTIVES: The aim of this study was to train and validate deep learning algorithms to quantitate relative amyloid deposition (RAD; mean amyloid deposited area per stromal area) in corneal sections from patients with familial amyloidosis, Finnish (FAF), and assess its relationship with visual acuity. METHODS: Corneal specimens were obtained from 42 patients undergoing penetrating keratoplasty, stained with Congo red, and digitally scanned. Areas of amyloid deposits and areas of stromal tissue were labeled on a pixel level for training and validation. The algorithms were used to quantify RAD in each cornea, and the association of RAD with visual acuity was assessed. RESULTS: In the validation of the amyloid area classification, sensitivity was 86%, specificity 92%, and F-score 81. For corneal stromal area classification, sensitivity was 74%, specificity 82%, and F-score 73. There was insufficient evidence to demonstrate correlation (Spearman's rank correlation, -0.264, p = 0.091) between RAD and visual acuity (logMAR). CONCLUSIONS: Deep learning algorithms can achieve a high sensitivity and specificity in pixel-level classification of amyloid and corneal stromal area. Further modeling and development of algorithms to assess earlier stages of deposition from clinical images is necessary to better assess the correlation between amyloid deposition and visual acuity. The method might be applied to corneal dystrophies as well.

Breast cancer outcome prediction with tumour tissue images and machine learning.

PURPOSE: Recent advances in machine learning have enabled better understanding of large and complex visual data. Here, we aim to investigate patient outcome prediction with a machine learning method using only an image of tumour sample as an input. METHODS: Utilising tissue microarray (TMA) samples obtained from the primary tumour of patients (N = 1299) within a nationwide breast cancer series with long-term-follow-up, we train and validate a machine learning method for patient outcome prediction. The prediction is performed by classifying samples into low or high digital risk score (DRS) groups. The outcome classifier is trained using sample images of 868 patients and evaluated and compared with human expert classification in a test set of 431 patients. RESULTS: In univariate survival analysis, the DRS classification resulted in a hazard ratio of 2.10 (95% CI 1.33-3.32, p = 0.001) for breast cancer-specific survival. The DRS classification remained as an independent predictor of breast cancer-specific survival in a multivariate Cox model with a hazard ratio of 2.04 (95% CI 1.20-3.44, p = 0.007). The accuracy (C-index) of the DRS grouping was 0.60 (95% CI 0.55-0.65), as compared to 0.58 (95% CI 0.53-0.63) for human expert predictions based on the same TMA samples. CONCLUSIONS: Our findings demonstrate the feasibility of learning prognostic signals in tumour tissue images without domain knowledge. Although further validation is needed, our study suggests that machine learning algorithms can extract prognostically relevant information from tumour histology complementing the currently used prognostic factors in breast cancer.

Detection of breast cancer lymph node metastases in frozen sections with a point-of-care low-cost microscope scanner.

BACKGROUND: Detection of lymph node metastases is essential in breast cancer diagnostics and staging, affecting treatment and prognosis. Intraoperative microscopy analysis of sentinel lymph node frozen sections is standard for detection of axillary metastases but requires access to a pathologist for sample analysis. Remote analysis of digitized samples is an alternative solution but is limited by the requirement for high-end slide scanning equipment. OBJECTIVE: To determine whether the image quality achievable with a low-cost, miniature digital microscope scanner is sufficient for detection of metastases in breast cancer lymph node frozen sections. METHODS: Lymph node frozen sections from 79 breast cancer patients were digitized using a prototype miniature microscope scanner and a high-end slide scanner. Images were independently reviewed by two pathologists and results compared between devices with conventional light microscopy analysis as ground truth. RESULTS: Detection of metastases in the images acquired with the miniature scanner yielded an overall sensitivity of 91% and specificity of 99% and showed strong agreement when compared to light microscopy (k = 0.91). Strong agreement was also observed when results were compared to results from the high-end slide scanner (k = 0.94). A majority of discrepant cases were micrometastases and sections of which no anticytokeratin staining was available. CONCLUSION: Accuracy of detection of metastatic cells in breast cancer sentinel lymph node frozen sections by visual analysis of samples digitized using low-cost, point-of-care microscopy is comparable to analysis of digital samples scanned using a high-end, whole slide scanner. This technique could potentially provide a workflow for digital diagnostics in resource-limited settings, facilitate sample analysis at the point-of-care and reduce the need for trained experts on-site during surgical procedures.

Deep learning for detecting tumour-infiltrating lymphocytes in testicular germ cell tumours.

AIMS: To evaluate if a deep learning algorithm can be trained to identify tumour-infiltrating lymphocytes (TILs) in tissue samples of testicular germ cell tumours and to assess whether the TIL counts correlate with relapse status of the patient. METHODS: TILs were manually annotated in 259 tumour regions from 28 whole-slide images (WSIs) of H&E-stained tissue samples. A deep learning algorithm was trained on half of the regions and tested on the other half. The algorithm was further applied to larger areas of tumour WSIs from 89 patients and correlated with clinicopathological data. RESULTS: A correlation coefficient of 0.89 was achieved when comparing the algorithm with the manual TIL count in the test set of images in which TILs were present (n=47). In the WSI regions from the 89 patient samples, the median TIL density was 1009/mm2. In seminomas, none of the relapsed patients belonged to the highest TIL density tertile (>2011/mm2). TIL quantifications performed visually by three pathologists on the same tumours were not significantly associated with outcome. The average interobserver agreement between the pathologists when assigning a patient into TIL tertiles was 0.32 (Kappa test) compared with 0.35 between the algorithm and the experts, respectively. A higher TIL density was associated with a lower clinical tumour stage, seminoma histology and lack of lymphovascular invasion. CONCLUSIONS: Deep learning-based image analysis can be used for detecting TILs in testicular germ cell cancer more objectively and it has potential for use as a prognostic marker for disease relapse.

Spatial aspects of oncogenic signalling determine the response to combination therapy in slice explants from Kras-driven lung tumours.

A key question in precision medicine is how functional heterogeneity in solid tumours informs therapeutic sensitivity. We demonstrate that spatial characteristics of oncogenic signalling and therapy response can be modelled in precision-cut slices from Kras-driven non-small-cell lung cancer with varying histopathologies. Unexpectedly, profiling of in situ tumours demonstrated that signalling stratifies mostly according to histopathology, showing enhanced AKT and SRC activity in adenosquamous carcinoma, and mitogen-activated protein kinase (MAPK) activity in adenocarcinoma. In addition, high intertumour and intratumour variability was detected, particularly of MAPK and mammalian target of rapamycin (mTOR) complex 1 activity. Using short-term treatment of slice explants, we showed that cytotoxic responses to combination MAPK and phosphoinositide 3-kinase-mTOR inhibition correlate with the spatially defined activities of both pathways. Thus, whereas genetic drivers determine histopathology spectra, histopathology-associated and spatially variable signalling activities determine drug sensitivity. Our study is in support of spatial aspects of signalling heterogeneity being considered in clinical diagnostic settings, particularly to guide the selection of drug combinations. © 2018 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland.

Deep learning based tissue analysis predicts outcome in colorectal cancer.

Image-based machine learning and deep learning in particular has recently shown expert-level accuracy in medical image classification. In this study, we combine convolutional and recurrent architectures to train a deep network to predict colorectal cancer outcome based on images of tumour tissue samples. The novelty of our approach is that we directly predict patient outcome, without any intermediate tissue classification. We evaluate a set of digitized haematoxylin-eosin-stained tumour tissue microarray (TMA) samples from 420 colorectal cancer patients with clinicopathological and outcome data available. The results show that deep learning-based outcome prediction with only small tissue areas as input outperforms (hazard ratio 2.3; CI 95% 1.79-3.03; AUC 0.69) visual histological assessment performed by human experts on both TMA spot (HR 1.67; CI 95% 1.28-2.19; AUC 0.58) and whole-slide level (HR 1.65; CI 95% 1.30-2.15; AUC 0.57) in the stratification into low- and high-risk patients. Our results suggest that state-of-the-art deep learning techniques can extract more prognostic information from the tissue morphology of colorectal cancer than an experienced human observer.

Antibody-supervised deep learning for quantification of tumor-infiltrating immune cells in hematoxylin and eosin stained breast cancer samples.

BACKGROUND: Immune cell infiltration in tumor is an emerging prognostic biomarker in breast cancer. The gold standard for quantification of immune cells in tissue sections is visual assessment through a microscope, which is subjective and semi-quantitative. In this study, we propose and evaluate an approach based on antibody-guided annotation and deep learning to quantify immune cell-rich areas in hematoxylin and eosin (H&E) stained samples. METHODS: Consecutive sections of formalin-fixed parafin-embedded samples obtained from the primary tumor of twenty breast cancer patients were cut and stained with H&E and the pan-leukocyte CD45 antibody. The stained slides were digitally scanned, and a training set of immune cell-rich and cell-poor tissue regions was annotated in H&E whole-slide images using the CD45-expression as a guide. In analysis, the images were divided into small homogenous regions, superpixels, from which features were extracted using a pretrained convolutional neural network (CNN) and classified with a support of vector machine. The CNN approach was compared to texture-based classification and to visual assessments performed by two pathologists. RESULTS: In a set of 123,442 labeled superpixels, the CNN approach achieved an F-score of 0.94 (range: 0.92-0.94) in discrimination of immune cell-rich and cell-poor regions, as compared to an F-score of 0.88 (range: 0.87-0.89) obtained with the texture-based classification. When compared to visual assessment of 200 images, an agreement of 90% (κ = 0.79) to quantify immune infiltration with the CNN approach was achieved while the inter-observer agreement between pathologists was 90% (κ = 0.78). CONCLUSIONS: Our findings indicate that deep learning can be applied to quantify immune cell infiltration in breast cancer samples using a basic morphology staining only. A good discrimination of immune cell-rich areas was achieved, well in concordance with both leukocyte antigen expression and pathologists' visual assessment.

Phase I study with ONCOS-102 for the treatment of solid tumors - an evaluation of clinical response and exploratory analyses of immune markers.

BACKGROUND: We conducted a phase I study with a granulocyte macrophage colony stimulating factor (GMCSF)-expressing oncolytic adenovirus, ONCOS-102, in patients with solid tumors refractory to available treatments. The objectives of the study were to determine the optimal dose for further use and to assess the safety, tolerability and adverse event (AE) profile of ONCOS-102. Further, the response rate and overall survival were evaluated as well as preliminary evidence of disease control. As an exploratory endpoint, the effect of ONCOS 102 on biological correlates was examined. METHODS: The study was conducted using a classic 3 + 3 dose escalation study design involving 12 patients. Patients were repeatedly treated intratumorally with ONCOS-102 plus daily low-dose oral cyclophosphamide (CPO). Tumor response was evaluated with diagnostic positron emission tomography (PET) and computed tomography (CT). Tumor biopsies were collected at baseline and after treatment initiation for analysis of immunological correlates. Peripheral blood mononuclear cells (PBMCs) were collected at baseline and during the study to assess antigen specificity of CD8+ T cells by interferon gamma (IFNγ) enzyme linked immunospot assay (ELISPOT). RESULTS: No dose limiting toxicity (DLT) or maximum tolerated dose (MTD) was identified for ONCOS-102. Four out of ten (40 %) evaluable patients had disease control based on PET/CT scan at 3 months and median overall survival was 9.3 months. A short-term increase in systemic pro-inflammatory cytokines and a prominent infiltration of TILs to tumors was seen post-treatment in 11 out of 12 patients. Two patients showed marked infiltration of CD8+ T cells to tumors and concomitant systemic induction of tumor-specific CD8+ T cells. Interestingly, high expression levels of genes associated with activated TH1 cells and TH1 type immune profile were observed in the post-treatment biopsies of these two patients. CONCLUSIONS: ONCOS-102 is safe and well tolerated at the tested doses. All three examined doses may be used in further development. There was evidence of antitumor immunity and signals of clinical efficacy. Importantly, treatment resulted in infiltration of CD8+ T cells to tumors and up-regulation of PD-L1, highlighting the potential of ONCOS-102 as an immunosensitizing agent for combinatory therapies with checkpoint inhibitors. TRIAL REGISTRATION: NCT01598129. Registered 19/04/2012.

Chronic Activation of Innate Immunity Correlates With Poor Prognosis in Cancer Patients Treated With Oncolytic Adenovirus.

Despite many clinical trials conducted with oncolytic viruses, the exact tumor-level mechanisms affecting therapeutic efficacy have not been established. Currently there are no biomarkers available that would predict the clinical outcome to any oncolytic virus. To assess the baseline immunological phenotype and find potential prognostic biomarkers, we monitored mRNA expression levels in 31 tumor biopsy or fluid samples from 27 patients treated with oncolytic adenovirus. Additionally, protein expression was studied from 19 biopsies using immunohistochemical staining. We found highly significant changes in several signaling pathways and genes associated with immune responses, such as B-cell receptor signaling (P < 0.001), granulocyte macrophage colony-stimulating factor (GM-CSF) signaling (P < 0.001), and leukocyte extravasation signaling (P < 0.001), in patients surviving a shorter time than their controls. In immunohistochemical analysis, markers CD4 and CD163 were significantly elevated (P = 0.020 and P = 0.016 respectively), in patients with shorter than expected survival. Interestingly, T-cell exhaustion marker TIM-3 was also found to be significantly upregulated (P = 0.006) in patients with poor prognosis. Collectively, these data suggest that activation of several functions of the innate immunity before treatment is associated with inferior survival in patients treated with oncolytic adenovirus. Conversely, lack of chronic innate inflammation at baseline may predict improved treatment outcome, as suggested by good overall prognosis.

Quantification of Estrogen Receptor-Alpha Expression in Human Breast Carcinomas With a Miniaturized, Low-Cost Digital Microscope: A Comparison with a High-End Whole Slide-Scanner.

INTRODUCTION: A significant barrier to medical diagnostics in low-resource environments is the lack of medical care and equipment. Here we present a low-cost, cloud-connected digital microscope for applications at the point-of-care. We evaluate the performance of the device in the digital assessment of estrogen receptor-alpha (ER) expression in breast cancer samples. Studies suggest computer-assisted analysis of tumor samples digitized with whole slide-scanners may be comparable to manual scoring, here we study whether similar results can be obtained with the device presented. MATERIALS AND METHODS: A total of 170 samples of human breast carcinoma, immunostained for ER expression, were digitized with a high-end slide-scanner and the point-of-care microscope. Corresponding regions from the samples were extracted, and ER status was determined visually and digitally. Samples were classified as ER negative (<1% ER positivity) or positive, and further into weakly (1-10% positivity) and strongly positive. Interobserver agreement (Cohen's kappa) was measured and correlation coefficients (Pearson's product-momentum) were calculated for comparison of the methods. RESULTS: Correlation and interobserver agreement (r = 0.98, p < 0.001, kappa = 0.84, CI95% = 0.75-0.94) were strong in the results from both devices. Concordance of the point-of-care microscope and the manual scoring was good (r = 0.94, p < 0.001, kappa = 0.71, CI95% = 0.61-0.80), and comparable to the concordance between the slide scanner and manual scoring (r = 0.93, p < 0.001, kappa = 0.69, CI95% = 0.60-0.78). Fourteen (8%) discrepant cases between manual and device-based scoring were present with the slide scanner, and 16 (9%) with the point-of-care microscope, all representing samples of low ER expression. CONCLUSIONS: Tumor ER status can be accurately quantified with a low-cost imaging device and digital image-analysis, with results comparable to conventional computer-assisted or manual scoring. This technology could potentially be expanded for other histopathological applications at the point-of-care.

Assessment of tumour viability in human lung cancer xenografts with texture-based image analysis.

AIMS: To build and evaluate an automated method for assessing tumour viability in histological tissue samples using texture features and supervised learning. METHODS: H&E-stained sections (n=56) of human non-small cell lung adenocarcinoma xenografts were digitised with a whole-slide scanner. A novel image analysis method based on local binary patterns and a support vector machine classifier was trained with a set of sample regions (n=177) extracted from the whole-slide images and tested with another set of images (n=494). The extracted regions, or single-tissue entity images, were chosen to represent as pure as possible examples of three morphological tissue entities: viable tumour tissue, non-viable tumour tissue and mouse host tissue. RESULTS: An agreement of 94.5% (area under the curve=0.995, kappa=0.90) was achieved to classify the single-tissue entity images in the test set (n=494) into the viable tumour and non-viable tumour tissue categories. The algorithm assigned 250 of the 252 non-viable and 219 of the 242 of viable sample regions to the correct categories, respectively. This corresponds to a sensitivity of 90.5% and specificity of 99.2%. CONCLUSIONS: The proposed image analysis-based tumour viability assessment resulted in a high agreement with expert annotations. By providing extraction of detailed information of the tumour microenvironment, the automated method can be used in preclinical research settings. The method could also have implications in cancer diagnostics, cancer outcome prognostics and prediction.

T-cell subsets in peripheral blood and tumors of patients treated with oncolytic adenoviruses.

The quality of the antitumor immune response is decisive when developing new immunotherapies for cancer. Oncolytic adenoviruses cause a potent immunogenic stimulus and arming them with costimulatory molecules reshapes the immune response further. We evaluated peripheral blood T-cell subsets of 50 patients with refractory solid tumors undergoing treatment with oncolytic adenovirus. These data were compared to changes in antiviral and antitumor T cells, treatment efficacy, overall survival, and T-cell subsets in pre- and post-treatment tumor biopsies. Treatment caused a significant (P < 0.0001) shift in T-cell subsets in blood, characterized by a proportional increase of CD8(+) cells, and decrease of CD4(+) cells. Concomitant treatment with cyclophosphamide and temozolomide resulted in less CD4(+) decrease (P = 0.041) than cyclophosphamide only. Interestingly, we saw a correlation between T-cell changes in peripheral blood and the tumor site. This correlation was positive for CD8(+) and inverse for CD4(+) cells. These findings give insight to the interconnections between peripheral blood and tumor-infiltrating lymphocyte (TIL) populations regarding oncolytic virotherapy. In particular, our data suggest that induction of T-cell response is not sufficient for clinical response in the context of immunosuppressive tumors, and that peripheral blood T cells have a complicated and potentially misleading relationship with TILs.