Search

Machine learning applied to digital pathology has been increasingly used to assess kidney function and diagnose the underlying cause of chronic kidney disease (CKD). We developed a novel computational framework, clustering-based spatial analysis (CluSA), that leverages unsupervised learning to learn spatial relationships between local visual patterns in kidney tissue. This framework minimizes the need for time-consuming and impractical expert annotations. 107,471 histopathology images obtained from 172 biopsy cores were used in the clustering and in the deep learning model. To incorporate spatial information over the clustered image patterns on the biopsy sample, we spatially encoded clustered patterns with colors and performed spatial analysis through graph neural network. A random forest classifier with various groups of features were used to predict CKD. For predicting eGFR at the biopsy, we achieved a sensitivity of 0.97, specificity of 0.90, and accuracy of 0.95. AUC was 0.96. For predicting eGFR changes in one-year, we achieved a sensitivity of 0.83, specificity of 0.85, and accuracy of 0.84. AUC was 0.85. This study presents the first spatial analysis based on unsupervised machine learning algorithms. Without expert annotation, CluSA framework can not only accurately classify and predict the degree of kidney function at the biopsy and in one year, but also identify novel predictors of kidney function and renal prognosis., (© 2023. Springer Nature Limited.)

As global vaccination campaigns against SARS-CoV-2 proceed, there is particular interest in the longevity of immune protection, especially with regard to increasingly infectious virus variants. Neutralizing antibodies (Nabs) targeting the receptor binding domain (RBD) of SARS-CoV-2 are promising correlates of protective immunity and have been successfully used for prevention and therapy. As SARS-CoV-2 variants of concern (VOCs) are known to affect binding to the ACE2 receptor and by extension neutralizing activity, we developed a bead-based multiplex ACE2-RBD inhibition assay (RBDCoV-ACE2) as a highly scalable, time-, cost-, and material-saving alternative to infectious live-virus neutralization tests. By mimicking the interaction between ACE2 and the RBD, this serological multiplex assay allows the simultaneous analysis of ACE2 binding inhibition to the RBDs of all SARS-CoV-2 VOCs and variants of interest (VOIs) in a single well. Following validation against a classical virus neutralization test and comparison of performance against a commercially available assay, we analyzed 266 serum samples from 168 COVID-19 patients of varying severity. ACE2 binding inhibition was reduced for ten out of eleven variants examined compared to wild-type, especially for those displaying the E484K mutation such as VOCs beta and gamma. ACE2 binding inhibition, while highly individualistic, positively correlated with IgG levels. ACE2 binding inhibition also correlated with disease severity up to WHO grade 7, after which it reduced., (© 2022. The Author(s).)

Pathologists use visual classification to assess patient kidney biopsy samples when diagnosing the underlying cause of kidney disease. However, the assessment is qualitative, or semi-quantitative at best, and reproducibility is challenging. To discover previously unknown features which predict patient outcomes and overcome substantial interobserver variability, we developed an unsupervised bag-of-words model. Our study applied to the C-PROBE cohort of patients with chronic kidney disease (CKD). 107,471 histopathology images were obtained from 161 biopsy cores and identified important morphological features in biopsy tissue that are highly predictive of the presence of CKD both at the time of biopsy and in one year. To evaluate the performance of our model, we estimated the AUC and its 95% confidence interval. We show that this method is reliable and reproducible and can achieve 0.93 AUC at predicting glomerular filtration rate at the time of biopsy as well as predicting a loss of function at one year. Additionally, with this method, we ranked the identified morphological features according to their importance as diagnostic markers for chronic kidney disease. In this study, we have demonstrated the feasibility of using an unsupervised machine learning method without human input in order to predict the level of kidney function in CKD. The results from our study indicate that the visual dictionary, or visual image pattern, obtained from unsupervised machine learning can predict outcomes using machine-derived values that correspond to both known and unknown clinically relevant features., (© 2022. The Author(s).)

Increasing evidence suggests that single in kidney states (e.g., kidney transplantation and living donation) progressive glomerulosclerosis limits kidney lifespan. Modeling shows that post-nephrectomy compensatory glomerular volume (GV) increase drives podocyte depletion and hypertrophic stress resulting in proteinuria and glomerulosclerosis, implying that GV increase could serve as a therapeutic target to prevent progression. In this report we examine how Angiotensin Converting Enzyme inhibition (ACEi), started before uninephrectomy can reduce compensatory GV increase in wild-type Fischer344 rats. An unbiased computer-assisted method was used for morphometric analysis. Urine Insulin-like growth factor-1 (IGF-1), the major diver of body and kidney growth, was used as a readout. In long-term (40-week) studies of uni-nephrectomized versus sham-nephrectomized rats a 2.2-fold increase in GV was associated with reduced podocyte density, increased proteinuria and glomerulosclerosis. Compensatory GV increase was largely prevented by ACEi started a week before but not after uni-nephrectomy with no measurable impact on long-term eGFR. Similarly, in short-term (14-day) studies, ACEi started a week before uni-nephrectomy reduced both GV increase and urine IGF-1 excretion. Thus, timing of ACEi in relation to uni-nephrectomy had significant impact on post-nephrectomy "compensatory" glomerular growth and outcomes that could potentially be used to improve kidney transplantation and live kidney donation outcomes., (© 2021. The Author(s).)

Fast and reliable detection of patients with severe and heterogeneous illnesses is a major goal of precision medicine 1,2 . Patients with leukaemia can be identified using machine learning on the basis of their blood transcriptomes 3 . However, there is an increasing divide between what is technically possible and what is allowed, because of privacy legislation 4,5 . Here, to facilitate the integration of any medical data from any data owner worldwide without violating privacy laws, we introduce Swarm Learning-a decentralized machine-learning approach that unites edge computing, blockchain-based peer-to-peer networking and coordination while maintaining confidentiality without the need for a central coordinator, thereby going beyond federated learning. To illustrate the feasibility of using Swarm Learning to develop disease classifiers using distributed data, we chose four use cases of heterogeneous diseases (COVID-19, tuberculosis, leukaemia and lung pathologies). With more than 16,400 blood transcriptomes derived from 127 clinical studies with non-uniform distributions of cases and controls and substantial study biases, as well as more than 95,000 chest X-ray images, we show that Swarm Learning classifiers outperform those developed at individual sites. In addition, Swarm Learning completely fulfils local confidentiality regulations by design. We believe that this approach will notably accelerate the introduction of precision medicine.

Hepatocellular carcinoma (HCC) can have viral or non-viral causes 1-5 . Non-alcoholic steatohepatitis (NASH) is an important driver of HCC. Immunotherapy has been approved for treating HCC, but biomarker-based stratification of patients for optimal response to therapy is an unmet need 6,7 . Here we report the progressive accumulation of exhausted, unconventionally activated CD8 + PD1 + T cells in NASH-affected livers. In preclinical models of NASH-induced HCC, therapeutic immunotherapy targeted at programmed death-1 (PD1) expanded activated CD8 + PD1 + T cells within tumours but did not lead to tumour regression, which indicates that tumour immune surveillance was impaired. When given prophylactically, anti-PD1 treatment led to an increase in the incidence of NASH-HCC and in the number and size of tumour nodules, which correlated with increased hepatic CD8 + PD1 + CXCR6 + , TOX + , and TNF + T cells. The increase in HCC triggered by anti-PD1 treatment was prevented by depletion of CD8 + T cells or TNF neutralization, suggesting that CD8 + T cells help to induce NASH-HCC, rather than invigorating or executing immune surveillance. We found similar phenotypic and functional profiles in hepatic CD8 + PD1 + T cells from humans with NAFLD or NASH. A meta-analysis of three randomized phase III clinical trials that tested inhibitors of PDL1 (programmed death-ligand 1) or PD1 in more than 1,600 patients with advanced HCC revealed that immune therapy did not improve survival in patients with non-viral HCC. In two additional cohorts, patients with NASH-driven HCC who received anti-PD1 or anti-PDL1 treatment showed reduced overall survival compared to patients with other aetiologies. Collectively, these data show that non-viral HCC, and particularly NASH-HCC, might be less responsive to immunotherapy, probably owing to NASH-related aberrant T cell activation causing tissue damage that leads to impaired immune surveillance. Our data provide a rationale for stratification of patients with HCC according to underlying aetiology in studies of immunotherapy as a primary or adjuvant treatment.

The success of molecular therapies targeting specific metabolic pathways in cancer is often limited by the plasticity and adaptability of metabolic networks. Here we show that pharmacologically induced lipotoxicity represents a promising therapeutic strategy for the treatment of hepatocellular carcinoma (HCC). LXRα-induced liponeogenesis and Raf-1 inhibition are synthetic lethal in HCC owing to a toxic accumulation of saturated fatty acids. Raf-1 was found to bind and activate SCD1, and conformation-changing DFG-out Raf inhibitors could disrupt this interaction, thereby blocking fatty acid desaturation and inducing lethal lipotoxicity. Studies in genetically engineered and nonalcoholic steatohepatitis-induced HCC mouse models and xenograft models of human HCC revealed that therapies comprising LXR agonists and Raf inhibitors were well tolerated and capable of overcoming therapy resistance in HCC. Conceptually, our study suggests pharmacologically induced lipotoxicity as a new mode for metabolic targeting of liver cancer., (© 2021. The Author(s), under exclusive licence to Springer Nature America, Inc. part of Springer Nature.)

We extend an agent-based multiscale model of vascular tumour growth and angiogenesis to describe transarterial chemoembolisation (TACE) therapies. The model accounts for tumour and normal cells that are both nested in a vascular system that changes its structure according to tumour-related growth factors. Oxygen promotes nutrients to the tissue and determines cell proliferation or death rates. Within the extended model TACE is included as a two-step process: First, the purely mechanical influence of the embolisation therapy is modelled by a local occlusion of the tumour vasculature. There we distinguish between partial and complete responders, where parts of the vascular system are occluded for the first and the whole tumour vasculature is destroyed for the latter. In the second part of the model, drug eluding beads (DEBs) carrying the chemotherapeutic drug doxorubicin are located at destroyed vascular locations, releasing the drug over a certain time-window. Simulation results are parameterised to qualitatively reproduce clinical observations. Patients that undergo a TACE-treatment are categorised in partial and complete responders one day after the treatment. Another 90 days later reoccurance or complete response are detected by volume perfusion computer tomography (VPCT). Our simulations reveal that directly after a TACE- treatment an unstable tumour state can be observed, where regrowth and total tumour death have the same likeliness. It is argued that this short time-window is favorable for another therapeutical intervention with a less radical therapy. This procedure can shift the outcome to more effectiveness. Simulation results with an oxygen therapy within the unstable time-window demonstrate a potentially positive manipulated outcome. Finally, we conclude that our TACE model can motivate new therapeutical strategies and help clinicians analyse the intertwined relations and cross-links in tumours.

Due to late diagnosis and a pronounced chemoresistance, most patients with hepatocellular carcinoma (HCC) have an overall poor prognosis. Measles vaccine viruses (MeV) have been shown to possess anti-tumor properties and their efficacy has been enhanced by arming with suicide genes. To test armed MeV for the treatment of HCC, we equipped it with the suicide gene Super-cytosine deaminase (SCD) and tested the efficacy in cell culture and in a mouse xenograft model of human HCC. Prodrug conversion was investigated in cell culture and quantified by high-performance liquid chromatography. We observed a strong oncolytic activity of MeV-SCD against human HCC in vitro and in vivo. The prodrug was efficiently converted in infected cells leading to a significant enhancement of the cytotoxic effect. Treatment of HCC xenografts with MeV caused long-term virus replication in tumor tissue. We show that the suicide gene therapy induces an apoptosis-like cell death but is not dependent on intact apoptosis pathways. These results demonstrate that MeV-based suicide gene therapy is a promising novel therapy regimen for HCC overcoming resistance towards conventional therapy. The independence from apoptosis raises hopes for the treatment of patients whose tumor cells exert defects in this cell death mechanism.

The misuse of somatic gene therapy for the purpose of enhancing athletic performance is perceived as a coming threat to the world of sports and categorized as 'gene doping'. This article describes a direct detection approach for gene doping that gives a clear yes-or-no answer based on the presence or absence of transgenic DNA in peripheral blood samples. By exploiting a priming strategy to specifically amplify intronless DNA sequences, we developed PCR protocols allowing the detection of very small amounts of transgenic DNA in genomic DNA samples to screen for six prime candidate genes. Our detection strategy was verified in a mouse model, giving positive signals from minute amounts (20 μl) of blood samples for up to 56 days following intramuscular adeno-associated virus-mediated gene transfer, one of the most likely candidate vector systems to be misused for gene doping. To make our detection strategy amenable for routine testing, we implemented a robust sample preparation and processing protocol that allows cost-efficient analysis of small human blood volumes (200 μl) with high specificity and reproducibility. The practicability and reliability of our detection strategy was validated by a screening approach including 327 blood samples taken from professional and recreational athletes under field conditions.

Current concerns over insertional mutagenesis by retroviral vectors mitigate investigations into alternative, potentially persistent gene therapy vector systems not dependent on genomic integration, such as Sendai virus vectors (SeVV). Prenatal gene therapy requires efficient gene delivery to several tissues, which may not be achievable by somatic gene transfer to the adult. Initially, to test the potential and tropism of the SeVV for gene delivery to fetal tissues, first-generation (replication- and propagation-competent) recombinant SeVV, expressing beta-galactosidase was introduced into late gestation immunocompetent mice via the amniotic and peritoneal cavities and the yolk sac vessels. At 2 days, this resulted in very high levels of expression particularly in the airway epithelium, mesothelium and vascular endothelium, respectively. However, as expected, substantial vector toxicity was observed. The efficiency of gene transfer and the level of gene expression were then examined using a second-generation SeVV. The second generation was developed to be still capable of cytoplasmic RNA replication and therefore high-level gene expression, but incapable of vector spread due to lack of the gene for viral F-protein. Vector was introduced into the fetal amniotic and peritoneal cavities, intravascularly, intramuscularly and intraspinally; at 2 days, expression was observed in the airway epithelia, peritoneal mesothelia, unidentified cells in the gut wall, locally at the site of muscle injection and in the dorsal root ganglia, respectively. Mortality was dramatically diminished compared with the first-generation vector.

Heparin-induced priapism constitutes a special form of pharmaco-induced prolonged erection, but the pathophysiological principles are not yet definitely clear. Heparin-induced antiplatelet-antibodies may lead to the aggregation of thrombocytes and thus alter the penile blood flow leading to low-flow priapism. Alternatively, this condition may be explained by initial high-flow priapism that later turns into ischemic priapism. The question remains whether hemorrhage with subsequent organisation of the hematoma and late fibrosis constitutes a pathogenetic factor. Besides this pathogenetic discussion, this paper presents the differential diagnosis of priapism as well as diagnostic and therapeutic procedures.

The low transduction efficiency of viral and nonviral vectors is a major limitation in tumour gene therapy. The HSV-1 tegument protein VP22 has been shown to exhibit a novel intercellular transport property. VP22 wild-type as well as VP22 fusion proteins efficiently spread from the original expressing cell to numerous neighbouring cells, so that protein transport by VP22 chimaeric polypeptides into the surrounding cells offers a possible compensation for the inadequate gene transfer efficiencies. To improve the therapeutic efficacy of the E. coli cytosine deaminase (CD) suicide gene we made use of the VP22 transport property in CD transducing adenoviral (Ad) vectors. C- and N-terminal fusions of CD linked in-frame with VP22 were generated and cloned into recombinant adenoviral vectors. Following in vitro transduction immunofluorescence analysis of Ad-transduced producer cells coplated with naive cells confirmed that the characteristic foci pattern of central producer and adjoining neighbour cells displaying nuclear staining was retained. After transduction of rat hepatoma cells with adenoviral vectors and subsequent incubation with the prodrug 5-FC, we observed enhanced cell cytotoxicity when comparing the CD-VP22 fusion (Ad-CD-VP22) with Ad-vectors expressing the CD gene only (Ad-CD). Thereby employment of Ad-vectors encoding VP22 fusion proteins opens up new possibilities to potentiate the efficiency of suicide gene therapy for the treatment of solid tumours.

It is now generally accepted that transforming growth factor-beta (TGF-beta) has an important role in the pathogenesis of both acute and chronic forms of renal disease. Although TGF-beta's potent fibrogenic activity is considered a major factor in chronic progression of renal disease, this cytokine participates in the control of several fundamental cellular responses in the kidney including inflammation, programmed cell death, cell growth, cell differentiation, and cellular hypertrophy. Recent identification of Smad proteins as intracellular mediators of TGF-beta signaling has provided important insights into mechanisms that may determine the specificity of TGF-beta action in different renal and inflammatory cells. Thus, Smads are characterized by an astonishingly complex array of molecular and functional interactions with other signaling pathways. These emerging patterns of signaling cross talk involving Smad proteins suggest a dynamic profile of positive or negative transmodulation of TGF-beta signaling, depending on the cellular context. Understanding the interplay between these signaling cascades is an important field of investigation that will ultimately reveal new targets for precise and selective modulation of TGF-beta's diverse actions in renal diseases.