Search

We report the ability of two deep learning-based decision systems to stratify non-small cell lung cancer (NSCLC) patients treated with checkpoint inhibitor therapy into two distinct survival groups. Both systems analyze functional and morphological properties of epithelial regions in digital histopathology whole slide images stained with the SP263 PD-L1 antibody. The first system learns to replicate the pathologist assessment of the Tumor Cell (TC) score with a cut-point for positivity at 25% for patient stratification. The second system is free from assumptions related to TC scoring and directly learns patient stratification from the overall survival time and event information. Both systems are built on a novel unpaired domain adaptation deep learning solution for epithelial region segmentation. This approach significantly reduces the need for large pixel-precise manually annotated datasets while superseding serial sectioning or re-staining of slides to obtain ground truth by cytokeratin staining. The capacity of the first system to replicate the TC scoring by pathologists is evaluated on 703 unseen cases, with an addition of 97 cases from an independent cohort. Our results show Lin's concordance values of 0.93 and 0.96 against pathologist scoring, respectively. The ability of the first and second system to stratify anti-PD-L1 treated patients is evaluated on 151 clinical samples. Both systems show similar stratification powers (first system: HR = 0.539, p = 0.004 and second system: HR = 0.525, p = 0.003) compared to TC scoring by pathologists (HR = 0.574, p = 0.01).

Folate receptor alpha (FRα) is a cell surface GPI-anchored protein overexpressed in several solid tumors with highest prevalence in ovarian cancer and lung adenocarcinoma but restricted expression in normal tissues. An antibody drug conjugate (ADC) with a microtubule inhibitor (MTI) payload recently received accelerated approval from the FDA for FRα-expressing platinum-resistant ovarian cancer. We describe for the first time the preclinical activity of AZD5335, an FRα-targeting antibody conjugated to AZ’s proprietary topoisomerase 1 inhibitor (TOP1i) payload, AZ14170132, with a homogeneous drug-to-antibody ratio of 8 (DAR8) and potential benefits vs an MTI-based ADC. AZD5335’s primary mechanism of action is to deliver TOP1i payload into FRα-expressing cancer cells, leading to DNA damage and apoptotic cell death. The TOP1i payload mediates bystander killing, which is important for targeting tumors with less than uniformly positive expression. Here, we report that a single dose of AZD5335 at 2.5 mg/kg was sufficient to provide a robust and durable anti-tumor response in FRα-expressing ovarian cancer cell line xenografts (CDX) with a tumor growth inhibition (TGI) of 75%-94% and median best tumor volume reduction >30% in 14/17 (82%) ovarian cancer patient-derived xenografts (PDX) evaluated. FRα-expression levels (by IHC and deep-learning based image analysis) correlated with efficacy in the tested PDX models, and we observed that AZD5335 was also active in models with low levels of target expression (75% of cells with FRα staining of 2+), expected to be representative of patients who would be ineligible for treatment with the MTI-ADC. Furthermore, AZD5335 demonstrated superior activity vs an FRα-MTI benchmark ADC with respect to anti-tumor activity and duration of response in two PDX models with low-to-medium FRα expression at equal or higher drug doses (e.g., in OV0857-CIS: 96% TGI vs 24% TGI at 5 mg/kg and 95% TGI vs 2% TGI at 2.5 mg/kg of a single IV dose AZD5335 and FRα-MTI, respectively). These data indicate that AZD5335 is a promising therapeutic candidate for the treatment of ovarian cancers across the spectrum of FRα-expression. Citation Format: Marco Gymnopoulos, Tima Thomas, Diana Gasper, Judith Anderton, Ravinder Tammali, Ed Rosfjord, Nick Durham, Chris Ward, Claire Myers, Jixin Wang, Wenyan Zhong, Simon Christ, Lina Meinecke, Katharina Nekolla, Laura Sebastian Monasor, Roger Dodd, Neki Patel, Mark Albertella, Jorge Zeron-Medina, Paula Fraenkel, Puja Sapra. First disclosure of AZD5335, a TOP1i-ADC targeting low and high FRα-expressing ovarian cancer with superior preclinical activity vs FRα-MTI ADC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 2 (Clinical Trials and Late-Breaking Research); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(8_Suppl):Abstract nr LB025.

In the context of precision medicine with immunotherapies there is an increasing need for companion diagnostic tests to identify potential therapy responders and avoid treatment coming along with severe adverse events for non-responders. Here, we present a retrospective case study to discover image-based signatures for developing a potential companion diagnostic test for ipilimumab (IPI) in malignant melanoma. Signature discovery is based on digital pathology and fully automatic quantitative image analysis using virtual multiplexing as well as machine learning and deep learning on whole-slide images. We systematically correlated the patient outcome data with potentially relevant local image features using a Tissue Phenomics approach with a sound cross validation procedure for reliable performance evaluation. Besides uni-variate models we also studied combinations of signatures in several multi-variate models. The most robust and best performing model was a decision tree model based on relative densities of CD8+ tumor infiltrating lymphocytes in the intra-tumoral infiltration region. Our results are well in agreement with observations described in previously published studies regarding the predictive value of the immune contexture, and thus, provide predictive potential for future development of a companion diagnostic test.

Our aim is to complement observer-dependent approaches of immune cell evaluation in microscopy images with reproducible measures for spatial composition of lymphocytic infiltrates. Analyzing such patterns of inflammation is becoming increasingly important for therapeutic decisions, for example in transplantation medicine or cancer immunology. We developed a graph-based assessment of lymphocyte clustering in full whole slide images. Based on cell coordinates detected in the full image, a Delaunay triangulation and distance criteria are used to build neighborhood graphs. The composition of nodes and edges are used for classification, e.g. using a support vector machine. We describe the variability of these infiltrates on CD3/CD20 duplex staining in renal biopsies of long-term functioning allografts, in breast cancer cases, and in lung tissue of cystic fibrosis patients. The assessment includes automated cell detection, identification of regions of interest, and classification of lymphocytic clusters according to their degree of organization. We propose a neighborhood feature which considers the occurrence of edges with a certain type in the graph to distinguish between phenotypically different immune infiltrates. Our work addresses a medical need and provides a scalable framework that can be easily adjusted to the requirements of different research questions., Author summary In this study, we developed a workflow to detect and classify immune infiltrates in giga-pixel microscopy images. It allowed us to measure the degree of organization in lymphocyte clusters with graph-based features and finally to distinguish between tertiary lymphoid organs and other infiltrates. As a clinically relevant use case, we applied it to three different types of tissues and diseases. Our method addresses the need for observer-independent, large-scale evaluation of immune cell patterns and is a prerequisite to capture spatial composition of immune cells that can be used to parameterize mathematical models in systems immunology.

Background Introduction of checkpoint inhibitors (anti PD-1/PD-L1) have resulted in improved survival for bladder cancer patients, however, only a subset will benefit. The utility of PD-L1 expression as a prognostic or predictive biomarker is limited, motivating the search for more robust biomarkers. Here, we examined the prognostic value of densities of PD-L1, CD3, CD8 or CD68 positive cells and studied the reproducibility of the findings across two patient cohorts in a multi-assay and multicentre setting. Methods MIBC specimens (T stage 2/3) from two patient cohorts collected at Melbourne, Australia (n = 39) and University of St Andrews, UK (n = 63) were studied. Two consecutive slides were stained with a brightfield immunohistochemistry or an immunofluorescence assay in the first and second cohorts, respectively. The densities of positive cells within the tumour core were determined using assay-specific image analysis algorithms. Within each cohort, the prognostic value of each cell density was assessed using univariate and multivariate Cox regression including age, T stage, N stage and adjuvant chemotherapy as covariates. Results The Melbourne cohort is slightly older, with a poorer prognosis and a higher proportion of N2/N3 disease compared to the St Andrews cohort. Univariate and multivariate analyses identified the density of CD8 positive cells as an important prognostic factor across both cohorts (p Table: 159P . Cohort 1: Melbourne Cohort 2: St Andrews Median age 71.64 66 Median survival time (months) 23.7 - Overall 32.57 - Disease Related T stage T2 16 (41%) 26 (41%) T3 23 (59%) 37 (59%) N stage N1 3 (8%) 52 (88%) N2 8 (20%) 7 (11%) N3 28 (72%) 0 (0%) Univariate Cox regression CD8 HR 0.93 (0.88 0.98), p = 0.007 ** HR 0.90 (0.83 0.97), p = 0.005 ** PD-L1 HR 0.97 (0.94 1.00), p = 0.065 . HR 0.86 (0.75 0.98), p = 0.023 * CD3 HR 0.92 (0.87 0.97), p = 0.003 ** HR 0.92 (0.87 0.98), p = 0.010 * CD68 HR 0.93 (0.88 0.99), p = 0.025 * HR 0.97 (0.93 1.00), p = 0.088 . Multivariate Cox regression CD8 HR 0.92 (0.88 0.97), p = 0.003 ** HR 0.92 (0.86 0.99), p = 0.029 * PD-L1 HR 0.96 (0.92 1.00), p = 0.053 . HR 0.84 (0.74 0.97), p = 0.015 * CD3 HR 0.90 (0.85 0.95), p = 0.0004** HR 0.94 (0.88 1.00), p = 0.060 . CD68 HR 0.92 (0.86 0.99), p = 0.023 * HR 0.98 (0.94 1.02), p = 0.294 Conclusions Our multicentre and multi-assay study suggests that the density of CD8 positive cells in the tumour core is a robust prognostic factor in MIBC. Further investigation of PD-L1 and CD3 cell density is warranted. Legal entity responsible for the study Definiens AG; The Walter and Eliza Hall Institute of Medical Research; University of St Andrews. Funding Definiens AG. Disclosure K. Nekolla: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. N. Brieu: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. C.G. Gavriel: Research grant / Funding (self), Definiens is a full subsidiary of AstraZeneca: Definiens AG. M. Widmaier: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. A. Budco: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. D. Medrikova: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. I. Kanchev: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. M. Testori: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. J. Chan: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. D.J. Harrison: Advisory / Consultancy, Member: Scientific Advisory Board: Definiens AG; Full / Part-time employment: NuCana plc; Leadership role, Director: Industrial Centre for AI Research in Digital Diagnostics. M. Baehner: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG. P.D. Caie: Research grant / Funding (institution): Definiens AG. B. Tran: Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Research grant / Funding (self), Research grant / Funding (institution), Travel / Accommodation / Expenses: Amgen; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Research grant / Funding (self), Travel / Accommodation / Expenses: Astellas; Honoraria (self), Advisory / Consultancy, Research grant / Funding (self), Travel / Accommodation / Expenses: Bayer; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Research grant / Funding (self): BMS; Honoraria (self), Advisory / Consultancy, Speaker Bureau / Expert testimony, Research grant / Funding (self), Research grant / Funding (institution), Non-remunerated activity/ies: Janssen-Cilag; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): MSD; Honoraria (self), Advisory / Consultancy, Research grant / Funding (institution): Novartis; Honoraria (self), Advisory / Consultancy, Travel / Accommodation / Expenses: Sanofi; Honoraria (self), Advisory / Consultancy: Tolmar; Honoraria (self), Advisory / Consultancy, Research grant / Funding (self): Ipsen; Research grant / Funding (self), Research grant / Funding (institution): AstraZeneca; Research grant / Funding (self): Pfizer; Research grant / Funding (self), Research grant / Funding (institution): Servier; Research grant / Funding (institution): Aslan; Research grant / Funding (institution): GSK; Research grant / Funding (institution): Taiho; Research grant / Funding (institution): Akeso; Research grant / Funding (institution): MedImmune; Research grant / Funding (institution): Aptevo; Leadership role, Chair - GU Tumour Group: Cancer Trials Australia; Leadership role, Chair - Germ Cell Tumour Subcommittee, Member - Scientific Advisory Committee: ANZUP; Leadership role, Member - Scientific Advisory Committee: Biogrid; Leadership role, Member - Scientific Advisory Committee: Parkville Cancer Clinical Trials Unit; Leadership role, Co-Chair - Molecular Tumour Board: Victorian Comprehensive Cancer Centre. G. Schmidt: Full / Part-time employment, Definiens is a full subsidiary of AstraZeneca: Definiens AG; Shareholder / Stockholder / Stock options: AstraZeneca; Licensing / Royalties, Royalities "Tissue Phenomics" ISBN 9789814774888: Pan Stanford Publishing. All other authors have declared no conflicts of interest.

DNA-based nanoconstructs possess great potential for biomedical applications. However, the in vivo behavior of such constructs at the microscopic tissue/cell level as well as their inflammatory potential is largely unknown. Unmethylated CpG sequences of DNA are recognized by Toll-like receptor 9 (TLR9), and thus initiate an innate immune response. In this study, we investigated the use of DNA-based nanotubes as carrier systems for CpG delivery and their effect on immune cells in vivo and in real time. DNA nanotubes were microinjected into skeletal muscle of anesthetized mice. Using in vivo microscopy, we observed that the DNA tubes were internalized within minutes by tissue-resident macrophages and localized in their endosomes. Only microinjection of CpG-decorated DNA nanotubes but not of plain DNA nanotubes or CpG oligonucleotides induced a significant recruitment of leukocytes into the muscle tissue as well as activation of the NF-ĸB pathway in surrounding cells. These results suggest that DNA nanotubes are promising delivery vehicles to target tissue macrophages, whereupon the immunogenic potential depends on the decoration with CpG oligonucleotides.

The biopolymer DNA allows to create nanoscale, biocompatible structures, which can be designed in a target-specific and stimuli-responsive manner. DNA carrier systems with these characteristics hold a great potential for nanomedical applications, such as for the treatment of inflammatory diseases. Here we used a DNA-based drug carrier system for the pH-dependent delivery of the glucocorticoid dexamethasone into macrophages, a cell type with a key role in the regulation of inflammation. Dexamethasone (Dex) nanotubes were internalized within minutes by MH-S macrophages in vitro and by tissue resident macrophages in the mouse cremaster muscle in vivo and localized in their endosomes. Treatment with Dex nanotubes in vitro significantly reduced the LPS-induced TNF secretion by macrophages, as compared to equivalent amounts of free dexamethasone without affecting cell viability. Microinjection of Dex nanotubes into postischemic muscle tissue of anesthetized mice resulted in a marked reduction of ischemia-reperfusion-elicited leukocyte transmigration and diminished vascular expression of the endothelial adhesion molecules VCAM-1 and ICAM-1. Taken together, our results demonstrate that DNA nanotubes can be used as a platform for the targeted delivery of glucocorticoids and could thus foster the development of nanomedical therapeutics with reduced off-target effects.

Under steady-state conditions, aged neutrophils are removed from the circulation in bone marrow, liver, and spleen, thereby maintaining myeloid cell homeostasis. The fate of these aged immune cells under inflammatory conditions, however, remains largely obscure. Here, we demonstrate that in the acute inflammatory response during endotoxemia, aged neutrophils cease returning to the bone marrow and instead rapidly migrate to the site of inflammation. Having arrived in inflamed tissue, aged neutrophils were found to exhibit a higher phagocytic activity as compared with the subsequently recruited nonaged neutrophils. This distinct behavior of aged neutrophils under inflammatory conditions is dependent on specific age-related changes in their molecular repertoire that enable these "experienced" immune cells to instantly translate inflammatory signals into immune responses. In particular, aged neutrophils engage Toll-like receptor-4- and p38 MAPK-dependent pathways to induce conformational changes in β2 integrins that allow these phagocytes to effectively accomplish their mission in the front line of the inflammatory response. Hence, ageing in the circulation might represent a critical process for neutrophils that enables these immune cells to properly unfold their functional properties for host defense.

Objective— Cell–matrix interactions are crucial for regulating cellular activities, such as migration. This is of special importance for morphogenic processes, such as angiogenesis (the development of new blood vessels). Most of our understanding of cell migration relies on 2-dimensional (2D) experiments. However, the awareness that 3D settings might elicit different results has increased. Knowledge about endothelial cell (EC) behavior in 3D environments and the influence of matrix composition on EC migration, in particular, is still limited. Approach and Results— We characterize the migration of single ECs through 2 structurally different hydrogels: spongy Matrigel and fibrillar collagen I. Our observations reveal an elongated migration phenotype in Matrigel and a rounded phenotype with pronounced cell blebs (blebs >2 µm) in collagen I, which have not previously been described in ECs. Directed migration seems to depend on Rac1 and Cdc42 in collagen, but not in Matrigel (shown using appropriate pharmacological inhibitors). By applying anti-integrin antibodies and supplementing laminin in collagen gels, we identify laminin as the main determinant of the elongated phenotype. Laminin seems to induce a morphological switch between modes of migration. As an in situ proof of principle, we performed live imaging of EC migration during vascular growth in a murine retina in the absence and presence of anti-integrin antibodies. Conclusions— We show that, surprisingly, ECs can evade the pharmacological inhibition of central signaling pathways involved in migration (contractility, small GTPases, and proteolysis) by shifting gears between modes of migration. This finding indicates an unexpected contextual plasticity of EC behavior.

Summary Small molecules that interfere with microtubule dynamics, such as Taxol and the Vinca alkaloids, are widely used in cell biology research and as clinical anticancer drugs. However, their activity cannot be restricted to specific target cells, which also causes severe side effects in chemotherapy. Here, we introduce the photostatins, inhibitors that can be switched on and off in vivo by visible light, to optically control microtubule dynamics. Photostatins modulate microtubule dynamics with a subsecond response time and control mitosis in living organisms with single-cell spatial precision. In longer-term applications in cell culture, photostatins are up to 250 times more cytotoxic when switched on with blue light than when kept in the dark. Therefore, photostatins are both valuable tools for cell biology, and are promising as a new class of precision chemotherapeutics whose toxicity may be spatiotemporally constrained using light.

Determination of blood flow velocity and related hemodynamic parameters is an important aspect of physiological studies which in many settings requires fluorescent labeling. Here we show that Third Harmonic Generation (THG) microscopy is a suitable tool for label-free intravital investigations of the microcirculation in widely-used physiological model systems. THG microscopy is a non-fluorescent multi-photon scanning technique combining the advantages of label-free imaging with restriction of signal generation to a focal spot. Blood flow was visualized and its velocity was measured in adult mouse cremaster muscle vessels, non-invasively in mouse ear vessels and in Xenopus tadpoles. In arterioles, THG line scanning allowed determination of the flow pulse velocity curve and hence the heart rate. By relocating the scan line we obtained velocity profiles through vessel diameters, allowing shear rate calculations. The cell free layer containing the glycocalyx was also visualized. Comparison of the current microscopic resolution with theoretical, diffraction limited resolution let us conclude that an about sixty-fold THG signal intensity increase may be possible with future improved optics, optimized for 1200–1300 nm excitation. THG microscopy is compatible with simultaneous two-photon excited fluorescence detection. It thus also provides the opportunity to determine important hemodynamic parameters in parallel to common fluorescent observations without additional label.