PodoSighter: A Cloud-Based Tool for Label-Free Podocyte Detection in Kidney Whole-Slide Images.

BACKGROUND: Podocyte depletion precedes progressive glomerular damage in several kidney diseases. However, the current standard of visual detection and quantification of podocyte nuclei from brightfield microscopy images is laborious and imprecise. METHODS: We have developed PodoSighter, an online cloud-based tool, to automatically identify and quantify podocyte nuclei from giga-pixel brightfield whole-slide images (WSIs) using deep learning. Ground-truth to train the tool used immunohistochemically or immunofluorescence-labeled images from a multi-institutional cohort of 122 histologic sections from mouse, rat, and human kidneys. To demonstrate the generalizability of our tool in investigating podocyte loss in clinically relevant samples, we tested it in rodent models of glomerular diseases, including diabetic kidney disease, crescentic GN, and dose-dependent direct podocyte toxicity and depletion, and in human biopsies from steroid-resistant nephrotic syndrome and from human autopsy tissues. RESULTS: The optimal model yielded high sensitivity/specificity of 0.80/0.80, 0.81/0.86, and 0.80/0.91, in mouse, rat, and human images, respectively, from periodic acid-Schiff-stained WSIs. Furthermore, the podocyte nuclear morphometrics extracted using PodoSighter were informative in identifying diseased glomeruli. We have made PodoSighter freely available to the general public as turnkey plugins in a cloud-based web application for end users. CONCLUSIONS: Our study demonstrates an automated computational approach to detect and quantify podocyte nuclei in standard histologically stained WSIs, facilitating podocyte research, and enabling possible future clinical applications.

Automated Computational Detection of Interstitial Fibrosis, Tubular Atrophy, and Glomerulosclerosis.

BACKGROUND: Interstitial fibrosis, tubular atrophy (IFTA), and glomerulosclerosis are indicators of irrecoverable kidney injury. Modern machine learning (ML) tools have enabled robust, automated identification of image structures that can be comparable with analysis by human experts. ML algorithms were developed and tested for the ability to replicate the detection and quantification of IFTA and glomerulosclerosis that renal pathologists perform. METHODS: A renal pathologist annotated renal biopsy specimens from 116 whole-slide images (WSIs) for IFTA and glomerulosclerosis. A total of 79 WSIs were used for training different configurations of a convolutional neural network (CNN), and 17 and 20 WSIs were used as internal and external testing cases, respectively. The best model was compared against the input of four renal pathologists on 20 new testing slides. Further, for 87 testing biopsy specimens, IFTA and glomerulosclerosis measurements made by pathologists and the CNN were correlated to patient outcome using classic statistical tools. RESULTS: The best average performance across all image classes came from a DeepLab version 2 network trained at 40× magnification. IFTA and glomerulosclerosis percentages derived from this CNN achieved high levels of agreement with four renal pathologists. The pathologist- and CNN-based analyses of IFTA and glomerulosclerosis showed statistically significant and equivalent correlation with all patient-outcome variables. CONCLUSIONS: ML algorithms can be trained to replicate the IFTA and glomerulosclerosis assessment performed by renal pathologists. This suggests computational methods may be able to provide a standardized approach to evaluate the extent of chronic kidney injury in situations in which renal-pathologist time is restricted or unavailable.

The mutational landscapes of genetic and chemical models of Kras-driven lung cancer.

Next-generation sequencing of human tumours has refined our understanding of the mutational processes operative in cancer initiation and progression, yet major questions remain regarding the factors that induce driver mutations and the processes that shape mutation selection during tumorigenesis. Here we performed whole-exome sequencing on adenomas from three mouse models of non-small-cell lung cancer, which were induced either by exposure to carcinogens (methyl-nitrosourea (MNU) and urethane) or by genetic activation of Kras (Kras(LA2)). Although the MNU-induced tumours carried exactly the same initiating mutation in Kras as seen in the Kras(LA2) model (G12D), MNU tumours had an average of 192 non-synonymous, somatic single-nucleotide variants, compared with only six in tumours from the Kras(LA2) model. By contrast, the Kras(LA2) tumours exhibited a significantly higher level of aneuploidy and copy number alterations compared with the carcinogen-induced tumours, suggesting that carcinogen-induced and genetically engineered models lead to tumour development through different routes. The wild-type allele of Kras has been shown to act as a tumour suppressor in mouse models of non-small-cell lung cancer. We demonstrate that urethane-induced tumours from wild-type mice carry mostly (94%) Kras Q61R mutations, whereas those from Kras heterozygous animals carry mostly (92%) Kras Q61L mutations, indicating a major role for germline Kras status in mutation selection during initiation. The exome-wide mutation spectra in carcinogen-induced tumours overwhelmingly display signatures of the initiating carcinogen, while adenocarcinomas acquire additional C > T mutations at CpG sites. These data provide a basis for understanding results from human tumour genome sequencing, which has identified two broad categories of tumours based on the relative frequency of single-nucleotide variations and copy number alterations, and underline the importance of carcinogen models for understanding the complex mutation spectra seen in human cancers.

Inflammation and Hras signaling control epithelial-mesenchymal transition during skin tumor progression.

Epithelial-mesenchymal transition (EMT) is thought to be an important, possibly essential, component of the process of tumor dissemination and metastasis. About 20%-30% of Hras mutant mouse skin carcinomas induced by chemical initiation/promotion protocols have undergone EMT. Reduced exposure to TPA-induced chronic inflammation causes a dramatic reduction in classical papillomas and squamous cell carcinomas (SCCs), but the mice still develop highly invasive carcinomas with EMT properties, reduced levels of Hras and Egfr signaling, and frequent Ink4/Arf deletions. Deletion of Hras from the mouse germline also leads to a strong reduction in squamous tumor development, but tumors now acquire activating Kras mutations and exhibit more aggressive metastatic properties. We propose that invasive carcinomas can arise by different genetic and biological routes dependent on exposure to chronic inflammation and possibly from different target cell populations within the skin. Our data have implications for the use of inhibitors of inflammation or of Ras/Egfr pathway signaling for prevention or treatment of invasive cancers.

Sonographically Diagnosed Urothelial Thickening in Kidney Allografts: A Noninvasive and Clinically Highly Relevant Marker for the Detection of Acute Rejection.

OBJECTIVE. The objective of our study was to investigate the significance of sonographic features in assessing for acute kidney allograft rejection in the modern era. MATERIALS AND METHODS. In this retrospective study, 107 adult patients with a kidney allograft biopsy performed between 2015 and 2018 and diagnostic ultrasound performed within 2 weeks of the biopsy were included. Acute rejection was diagnosed on the basis of biopsy tissue sample results using the Banff criteria. The following ultrasound features were assessed: perfusion, cortical echogenicity, corticomedullary differentiation, urothelial thickening, change in renal length, renal artery velocity, and intraparenchymal arterial resistive index. Subjective measures of perfusion, echogenicity, corticomedullary differentiation, and urothelial thickening were assessed independently and in consensus by three abdominal radiologists; multirater kappa values were calculated for interobserver variability. The Wilcoxon rank sum test and chi-square test were used to evaluate the differences between two groups (rejection vs no rejection) and the sonographic features. Sensitivity, specificity, positive predictive value, and negative predictive value (NPV) were calculated for sonographic features that are associated with acute rejection. RESULTS. Of the sonographic features, only the presence of urothelial thickening was significantly associated with acute rejection (p < 0.001) and had substantial agreement (κ = 0.61) among readers. Urothelial thickening was highly sensitive (96%; 95% CI, 79-100%) with a high NPV (98%; 95% CI, 86-100%). CONCLUSION. Urothelial thickening on ultrasound is a highly sensitive finding for acute kidney rejection with a high NPV and thus may play a role in sonographic prebiopsy screening. Other historically associated sonographic features seem to play little, if any, role in the screening and assessment for kidney allograft rejection in the modern era.

Tonsillar Kaposi sarcoma in a renal transplant patient.

Kaposi sarcoma (KS) is a vascular neoplasm caused by human herpesvirus-8 (HHV-8) infection. KS is most often seen in individuals with acquired immunodeficiency syndrome but can occur in patients who are on immunosuppressive therapy. While the skin and oral mucosa are the typical sites for KS, lesions of the tonsil are quite rare with only a few reported cases. Here, we present a case of tonsillar KS occurring in a renal transplant patient. He presented with dysphagia, odynophagia, and weight loss. Oral examination revealed tonsillar hypertrophy with purple discoloration. Imaging revealed diffuse enlargement of Waldeyer's ring with enlarged right cervical lymph nodes, worrisome for post-transplant lymphoproliferative disorder. Microscopic examination of the tonsillectomy specimen showed a vascular proliferation positive for HHV-8, consistent with KS. The patient was subsequently treated with immunosuppression reduction and the addition of sirolimus, which resulted in complete resolution of oropharyngeal and cervical lesions.

Computational Segmentation and Classification of Diabetic Glomerulosclerosis.

BACKGROUND: Pathologists use visual classification of glomerular lesions to assess samples from patients with diabetic nephropathy (DN). The results may vary among pathologists. Digital algorithms may reduce this variability and provide more consistent image structure interpretation. METHODS: We developed a digital pipeline to classify renal biopsies from patients with DN. We combined traditional image analysis with modern machine learning to efficiently capture important structures, minimize manual effort and supervision, and enforce biologic prior information onto our model. To computationally quantify glomerular structure despite its complexity, we simplified it to three components consisting of nuclei, capillary lumina and Bowman spaces; and Periodic Acid-Schiff positive structures. We detected glomerular boundaries and nuclei from whole slide images using convolutional neural networks, and the remaining glomerular structures using an unsupervised technique developed expressly for this purpose. We defined a set of digital features which quantify the structural progression of DN, and a recurrent network architecture which processes these features into a classification. RESULTS: Our digital classification agreed with a senior pathologist whose classifications were used as ground truth with moderate Cohen's kappa κ = 0.55 and 95% confidence interval [0.50, 0.60]. Two other renal pathologists agreed with the digital classification with κ1 = 0.68, 95% interval [0.50, 0.86] and κ2 = 0.48, 95% interval [0.32, 0.64]. Our results suggest computational approaches are comparable to human visual classification methods, and can offer improved precision in clinical decision workflows. We detected glomerular boundaries from whole slide images with 0.93±0.04 balanced accuracy, glomerular nuclei with 0.94 sensitivity and 0.93 specificity, and glomerular structural components with 0.95 sensitivity and 0.99 specificity. CONCLUSIONS: Computationally derived, histologic image features hold significant diagnostic information that may augment clinical diagnostics.

Diverse tumour susceptibility in Collaborative Cross mice: identification of a new mouse model for human gastric tumourigenesis.

OBJECTIVE: The Collaborative Cross (CC) is a mouse population model with diverse and reproducible genetic backgrounds used to identify novel disease models and genes that contribute to human disease. Since spontaneous tumour susceptibility in CC mice remains unexplored, we assessed tumour incidence and spectrum. DESIGN: We monitored 293 mice from 18 CC strains for tumour development. Genetic association analysis and RNA sequencing were used to identify susceptibility loci and candidate genes. We analysed genomes of patients with gastric cancer to evaluate the relevance of genes identified in the CC mouse model and measured the expression levels of ISG15 by immunohistochemical staining using a gastric adenocarcinoma tissue microarray. Association of gene expression with overall survival (OS) was assessed by Kaplan-Meier analysis. RESULTS: CC mice displayed a wide range in the incidence and types of spontaneous tumours. More than 40% of CC036 mice developed gastric tumours within 1 year. Genetic association analysis identified Nfκb1 as a candidate susceptibility gene, while RNA sequencing analysis of non-tumour gastric tissues from CC036 mice showed significantly higher expression of inflammatory response genes. In human gastric cancers, the majority of human orthologues of the 166 mouse genes were preferentially altered by amplification or deletion and were significantly associated with OS. Higher expression of the CC036 inflammatory response gene signature is associated with poor OS. Finally, ISG15 protein is elevated in gastric adenocarcinomas and correlated with shortened patient OS. CONCLUSIONS: CC strains exhibit tremendous variation in tumour susceptibility, and we present CC036 as a spontaneous laboratory mouse model for studying human gastric tumourigenesis.

No difference in 4-nitroquinoline induced tumorigenesis between germ-free and colonized mice.

Variations in oral bacterial communities have been linked to oral cancer suggesting that the oral microbiome is an etiological factor that can influence oral cancer development. The 4-nitroquinoline 1-oxide (4-NQO)-induced murine oral and esophageal cancer model is frequently used to assess the effects of preventive and/or therapeutic agents. We used this model to assess the impact of the microbiome on tumorigenesis using axenic (germ-free) and conventionally housed mice. Increased toxicity was observed in germ-free mice, however, no difference in tumor incidence, multiplicity, and size was observed. Transcriptional profiling of liver tissue from germ-free and conventionally housed mice identified 254 differentially expressed genes including ten cytochrome p450 enzymes, the largest family of phase-1 drug metabolizing enzymes in the liver. Gene ontology revealed that differentially expressed genes were enriched for liver steatosis, inflammation, and oxidative stress in livers of germ-free mice. Our observations emphasize the importance of the microbiome in mediating chemical toxicity at least in part by altering host gene expression. Studies on the role of the microbiome in chemical-induced cancer using germ-free animal models should consider the potential difference in dose due to the microbiome-mediated changes in host metabolizing capacity, which might influence the ability to draw conclusions especially for tumor induction models that are dose dependent.

Identification of high-risk human papillomavirus and Rb/E2F pathway genomic alterations in mutually exclusive subsets of colorectal neuroendocrine carcinoma.

Colorectal neuroendocrine carcinomas, both small cell and large cell types, are highly aggressive tumors with poor prognosis compared with colorectal adenocarcinoma. The molecular drivers of neuroendocrine carcinoma are best defined in small cell lung cancer, which shows near-universal genomic alterations in TP53 and RB1. The genetics of colorectal neuroendocrine carcinoma remain poorly understood; recent studies demonstrated infrequent RB1 alterations and genetics closely resembling colorectal adenocarcinoma. To better define the molecular pathogenesis of colorectal neuroendocrine carcinoma, we performed capture-based next-generation sequencing on 25 cases and evaluated for expression of p53, Rb, p16, and high-risk human papillomavirus (HR-HPV) subtypes using immunohistochemistry, in situ hybridization, and polymerase chain reaction. Rb/E2F pathway dysregulation was identified in nearly all cases (23/25, 92%) and occurred via three distinct mechanisms. First, RB1 genomic alteration was present in 56% (14/25) of cases and was accompanied by Rb protein loss, high p16 expression, and absence of HR-HPV; these cases also had frequent genomic alterations in TP53, the PI3K/Ras and Wnt pathways, as well as in DNA repair genes, with 4/14 cases being hypermutated. Second, 16% (4/25) of cases, all left-sided, had TP53 alteration without RB1 alteration; half of these harbored high-level amplifications in CCNE1 and MYC or MYCN and arose in patients with ulcerative colitis. Finally, 28% (7/25) of cases, all rectal or anal, lacked genomic alterations in RB1 or TP53 but were positive for HR-HPV. Our data demonstrate that Rb/E2F pathway dysregulation is essential in the pathogenesis of colorectal neuroendocrine carcinoma, akin to neuroendocrine carcinomas in other anatomic sites. Moreover, colorectal neuroendocrine carcinomas stratify into three distinct molecular subgroups, which can be differentiated based on Rb protein and HR-HPV status. HR-HPV infection represents a distinct mechanism for Rb and p53 inactivation in cases lacking genomic alterations in either gene. Differential treatment strategies for hypermutated and HPV-driven cases could improve patient outcomes.

Cocaine/levamisole-associated autoimmune syndrome: a disease of neutrophil-mediated autoimmunity.

PURPOSE OF REVIEW: Levamisole was previously used for its immunomodulatory properties to treat rheumatoid arthritis and some cancers. However, because of serious side-effects, it was taken off the market in the United States. Recently, levamisole has reemerged as a popular cocaine adulterant. Some individuals who consume levamisole-adulterated cocaine can develop a life-threatening autoimmune syndrome. In this review, the medical consequences of levamisole exposure and postulated mechanisms by which levamisole induces these adverse effects are discussed. RECENT FINDINGS: Although agranulocytosis and cutaneous vasculitis are the major findings in patients who develop cocaine/levamisole-associated autoimmune syndrome (CLAAS), more recent experience indicates that other organ systems can be involved as well. Current studies point to neutrophil activation and neutrophil extracellular trap formation with subsequent antineutrophil cytoplasmic antibody-mediated tissue injury as a possible mechanism of CLAAS. SUMMARY: In the past decade, the detrimental effects of levamisole have reemerged because of its popularity as a cocaine adulterant. Although infrequent, some individuals develop a systemic autoimmune syndrome characterized by immune-mediated agranulocytosis and antineutrophil cytoplasmic antibody-mediated vasculitis. Mechanistically, neutrophil antigens appear to be a major player in inducing CLAAS. Prompt cessation of levamisole exposure is key to treatment, although relapses are frequent because of the addictive effects of cocaine and the high prevalence of levamisole within the cocaine supply.

Arginine reprogramming in ADPKD results in arginine-dependent cystogenesis.

Research into metabolic reprogramming in cancer has become commonplace, yet this area of research has only recently come of age in nephrology. In light of the parallels between cancer and autosomal dominant polycystic kidney disease (ADPKD), the latter is currently being studied as a metabolic disease. In clear cell renal cell carcinoma (RCC), which is now considered a metabolic disease, we and others have shown derangements in the enzyme arginosuccinate synthase 1 (ASS1), resulting in RCC cells becoming auxotrophic for arginine and leading to a new therapeutic paradigm involving reducing extracellular arginine. Based on our earlier finding that glutamine pathways are reprogrammed in ARPKD, and given the connection between arginine and glutamine synthetic pathways via citrulline, we investigated the possibility of arginine reprogramming in ADPKD. We now show that, in a remarkable parallel to RCC, ASS1 expression is reduced in murine and human ADPKD, and arginine depletion results in a dose-dependent compensatory increase in ASS1 levels as well as decreased cystogenesis in vitro and ex vivo with minimal toxicity to normal cells. Nontargeted metabolomics analysis of mouse kidney cell lines grown in arginine-deficient versus arginine-replete media suggests arginine-dependent alterations in the glutamine and proline pathways. Thus, depletion of this conditionally essential amino acid by dietary or pharmacological means, such as with arginine-degrading enzymes, may be a novel treatment for this disease.

Short-term early exposure to thirdhand cigarette smoke increases lung cancer incidence in mice.

Exposure to thirdhand smoke (THS) is a recently described health concern that arises in many indoor environments. However, the carcinogenic potential of THS, a critical consideration in risk assessment, remains untested. Here we investigated the effects of short-term early exposure to THS on lung carcinogenesis in A/J mice. Forty weeks after THS exposure from 4 to 7 weeks of age, the mice had increased incidence of lung adenocarcinoma, tumor size and, multiplicity, compared with controls. In vitro studies using cultured human lung cancer cells showed that THS exposure induced DNA double-strand breaks and increased cell proliferation and colony formation. RNA sequencing analysis revealed that THS exposure induced endoplasmic reticulum stress and activated p53 signaling. Activation of the p53 pathway was confirmed by an increase in its targets p21 and BAX. These data indicate that early exposure to THS is associated with increased lung cancer risk.

Hepatic steatosis after pediatric liver transplant.

Hepatic steatosis develops after liver transplantation (LT) in 30% of adults, and nonalcoholic fatty liver disease (NAFLD) is the most common chronic liver disease in nontransplanted children. However, posttransplant steatosis has been minimally studied in pediatric LT recipients. We explored the prevalence, persistence, and association with chronic liver damage of hepatic steatosis in these children. In this single-center study of pediatric patients transplanted 1988-2015 (n = 318), 31% of those with any posttransplant biopsy (n = 271) had ≥ 1 biopsy with steatosis. Median time from transplant to first biopsy with steatosis was 0.8 months (interquartile range [IQR], 0.3-6.5 months) and to last biopsy with steatosis was 5.5 months (IQR, 1.0-24.5 months); 85% of patients with steatosis also had for-cause biopsies without steatosis. All available for-cause biopsies were re-evaluated (n = 104). Of 9 biopsies that could be interpreted as nonalcoholic steatohepatitis (NASH)/borderline NASH, with steatosis plus inflammation or ballooning, 8 also had features of cholestasis or rejection. Among 70 patients with surveillance biopsies 3.6-20.0 years after transplant, only 1 overweight adolescent had a biopsy with NAFLD (grade 1 steatosis, mild inflammation, no ballooning or fibrosis)-despite a 30% prevalence of overweight/obesity in the cohort and 27% with steatosis on previous for-cause biopsy. Steatosis on preceding for-cause biopsy was not associated with portal (P = 0.49) or perivenular fibrosis (P = 0.85) on surveillance biopsy. Hepatic steatosis commonly develops early after transplant in children and adolescents, but it rarely persists. Biopsies that did have steatosis with NASH characteristics were all for-cause, mostly in patients with NAFLD risk factors and/or confounding causes of liver damage. Prospective studies that follow children into adulthood will be needed to evaluate if and when hepatic steatosis presents a longterm risk for pediatric LT recipients. Liver Transplantation 23 957-967 2017 AASLD.

The Trp53 delta proline (Trp53ΔP) mouse exhibits increased genome instability and susceptibility to radiation-induced, but not spontaneous, tumor development.

The tumor suppressor TP53 can initiate a plethora of anti-proliferative effects to maintain genomic integrity under conditions of genotoxic stress. The N-terminal proline-rich domain (PRD) of TP53 is important in the regulation of TP53 activity and stability. A common polymorphism at codon 72 in this region has been associated with altered cancer risk in humans. The Trp53ΔP mouse, which carries a germline homozygous deletion of a region of the PRD, does not develop spontaneous tumors in a mixed 129/Sv and C57BL/6 genetic background, but is highly susceptible to a broad range of tumor types following total body exposure to 4 Gy gamma (γ) radiation. This contrasts with the tumor spectrum in Trp53 null (-/-) mice, which mainly develop thymic lymphomas and osteosarcomas. Analysis of genomic instability in tissues and cells from Trp53ΔP mice demonstrated elevated basal levels of aneuploidy, but this is not sufficient to drive spontaneous tumorigenesis, which requires an additional DNA damage stimulus. Levels of genomic instability did not increase significantly in Trp53ΔP mice following irradiation exposure, suggesting that other radiation effects including tissue inflammation, altered metabolism or autophagy, may play an important role. The Trp53ΔP mouse is a novel model to dissect the mechanisms of tumor development induced by radiation exposure. © 2015 Wiley Periodicals, Inc.

Characteristics of AA amyloidosis patients in San Francisco.

BACKGROUND: AA amyloidosis due to subcutaneous injection of drugs of abuse has been described in the USA, but all the existing literature is from more than 20 years ago. There is more recent literature from Europe. We have observed a high incidence of AA amyloidosis in the county hospital in San Francisco. DESIGN: Here, we describe 24 patients who had kidney biopsy-proven AA amyloidosis from our hospital from 1998 to 2013. All the patients were thought to have AA amyloidosis from skin popping of illicit drugs after having exhausted the intravenous route. These patients with biopsy-proven AA amyloidosis were analysed further. RESULTS: All patients were found to have hepatitis C infection, hypertension was not common, most had advanced kidney failure, and acidosis was common as was tubulointerstitial involvement on the kidney biopsy. Other organ involvement included hepatomegaly and splenomegaly in a number of patients; direct myocardial involvement was not seen, but pulmonary hypertension, history of deep vein thrombosis and pulmonary embolism were common. The prognosis of these patients was poor. The mortality rate approached 50% 1 year after biopsy, and most of the patient needed dialysis shortly after diagnosis. Cessation of drug use seemed beneficial but rarely achievable. CONCLUSION: AA amyloidosis from skin popping is common in San Francisco. Most patients with renal involvement end up on dialysis, and mortality rates are exceedingly high.

Lupus-like membranous nephropathy: Is it lupus or not?

BACKGROUND: Membranous glomerulonephritis is typically classified as idiopathic or secondary to systemic lupus erythematosus (SLE), hepatitis B, drugs, toxins, other infections, or malignancy. Not infrequently in some patients without a definite diagnosis of SLE, pathologic features of secondary membranous nephropathy are seen e.g., mesangial and/or subendothelial deposits, tubuloreticular inclusions, and full house immunofluorescence. In these patients, there is uncertainty about the etiology, response to therapy, and prognosis of membranous GN. METHODS: We retrospectively reviewed the charts of 98 patients with membranous GN at San Francisco General Hospital and John Stroger Hospital of Cook County over a 10-year period. Data were collected and analyzed using SPSS.18. RESULTS: Thirty-nine (40 %) had idiopathic membranous GN (Group 1), thirty-six (37 %) had lupus membranous GN (Group 2) and twenty-three (23 %) had some pathological features of secondary membranous GN, but no definite etiology of membranous GN (Group 3). At baseline (at time of renal biopsy) and after mean follow-up of 3.5 years, the average serum creatinine (in mg/dL) in Group 1 was (1.6 ± 1.0 versus 1.6 ± 1.7), Group 2 was (1.8 ± 2.5 versus 1.2 ± 0.9) and Group 3 was (1.1 ± 0.4 versus 1.27 ± 0.83), respectively. For the same time points, the average urine protein to creatinine ratio (g/g) in Group 1 was (9.8 ± 7.1 versus 5.7 ± 6.7), Group 2 was (4.2 ± 3.9 versus 1.7 ± 2.2), and Group 3 was (7.4 ± 5.7 versus 3.1 ± 3.8). In addition, during the follow-up period, eleven of 39 (28 %) in Group 1, two of 36 (6 %) in Group 2, and three of 23 (13 %) in Group 3 progressed to end-stage renal disease and were started on dialysis. CONCLUSIONS: It appears that patients with lupus membranous GN have better renal prognosis than patients with idiopathic membranous GN. The renal prognosis for patients with pathological features of lupus membranous but no diagnosis of systemic lupus (lupus-like membranous GN) falls in between. Further studies are needed to determine if Group 3 patients can (a) definitively be classified as true idiopathic membranous GN or lupus membranous GN or (b) they have a separate disease from either M-type phospholipase A2 receptor membranous nephropathy or systemic lupus-induced membranous nephropathy.

Monoclonal light chain deposits within the stomach manifesting as immunotactoid gastropathy.

Immunotactoid deposits are defined by their ultrastructural appearance and are characterized by microtubular or cylindrical structures typically measuring greater than 30 nm in diameter. Although a rare entity, immunotactoid deposition most often manifests as immunotactoid glomerulopathy and is associated with underlying lymphoplasmacytic disorders. Corneal immunotactoid deposition known as immunotactoid keratopathy has also been reported in patients with paraproteinemia. Here, we describe the first reported case of immunotactoid deposition in the stomach. The deposits were composed solely of kappa immunoglobulin light chains without significant lambda light chain or immunoglobulin heavy chain components. The patient displayed no renal signs or symptoms, and additional thorough clinical examination failed to detect any evidence of a paraproteinemia or plasma cell dyscrasia. Thus, the gastric immunotactoid deposits in this case appear to be an isolated finding of light chain deposition, of which the significance and etiology are unclear.

Sequential mutations in Notch1, Fbxw7, and Tp53 in radiation-induced mouse thymic lymphomas.

T-cell acute lymphoblastic lymphomas commonly demonstrate activating Notch1 mutations as well as mutations or deletions in Fbxw7. However, because Fbxw7 targets Notch1 for degradation, genetic alterations in these genes are expected to be mutually exclusive events in lymphomagenesis. Previously, by using a radiation-induced Tp53-deficient mouse model for T-cell acute lymphoblastic lymphoma, we reported that loss of heterozygosity at the Fbxw7 locus occurs frequently in a Tp53-dependent manner. In the current study, we show that these thymic lymphomas also commonly exhibit activating Notch1 mutations in the proline-glutamic acid-serine-threonine (PEST) domain. Moreover, concurrent activating Notch1 PEST domain mutations and single-copy deletions at the Fbxw7 locus occur with high frequency in the same individual tumors, indicating that these changes are not mutually exclusive events. We further demonstrate that although Notch1 PEST domain mutations are independent of Tp53 status, they are completely abolished in mice with germline Fbxw7 haploinsufficiency. Therefore, Notch1 PEST domain mutations only occur when Fbxw7 expression levels are intact. These data suggest a temporal sequence of mutational events involving these important cancer-related genes, with Notch1 PEST domain mutations occurring first, followed by Fbxw7 deletion, and eventually by complete loss of Tp53.

exKidneyBERT: a language model for kidney transplant pathology reports and the crucial role of extended vocabularies.

Background: Pathology reports contain key information about the patient's diagnosis as well as important gross and microscopic findings. These information-rich clinical reports offer an invaluable resource for clinical studies, but data extraction and analysis from such unstructured texts is often manual and tedious. While neural information retrieval systems (typically implemented as deep learning methods for natural language processing) are automatic and flexible, they typically require a large domain-specific text corpus for training, making them infeasible for many medical subdomains. Thus, an automated data extraction method for pathology reports that does not require a large training corpus would be of significant value and utility. Objective: To develop a language model-based neural information retrieval system that can be trained on small datasets and validate it by training it on renal transplant-pathology reports to extract relevant information for two predefined questions: (1) "What kind of rejection does the patient show?"; (2) "What is the grade of interstitial fibrosis and tubular atrophy (IFTA)?" Methods: Kidney BERT was developed by pre-training Clinical BERT on 3.4K renal transplant pathology reports and 1.5M words. Then, exKidneyBERT was developed by extending Clinical BERT's tokenizer with six technical keywords and repeating the pre-training procedure. This extended the model's vocabulary. All three models were fine-tuned with information retrieval heads. Results: The model with extended vocabulary, exKidneyBERT, outperformed Clinical BERT and Kidney BERT in both questions. For rejection, exKidneyBERT achieved an 83.3% overlap ratio for antibody-mediated rejection (ABMR) and 79.2% for T-cell mediated rejection (TCMR). For IFTA, exKidneyBERT had a 95.8% exact match rate. Conclusion: ExKidneyBERT is a high-performing model for extracting information from renal pathology reports. Additional pre-training of BERT language models on specialized small domains does not necessarily improve performance. Extending the BERT tokenizer's vocabulary library is essential for specialized domains to improve performance, especially when pre-training on small corpora.