The Lineage-Defining Transcription Factors SOX2 and NKX2-1 Determine Lung Cancer Cell Fate and Shape the Tumor Immune Microenvironment.

The major types of non-small-cell lung cancer (NSCLC)-squamous cell carcinoma and adenocarcinoma-have distinct immune microenvironments. We developed a genetic model of squamous NSCLC on the basis of overexpression of the transcription factor Sox2, which specifies lung basal cell fate, and loss of the tumor suppressor Lkb1 (SL mice). SL tumors recapitulated gene-expression and immune-infiltrate features of human squamous NSCLC; such features included enrichment of tumor-associated neutrophils (TANs) and decreased expression of NKX2-1, a transcriptional regulator that specifies alveolar cell fate. In Kras-driven adenocarcinomas, mis-expression of Sox2 or loss of Nkx2-1 led to TAN recruitment. TAN recruitment involved SOX2-mediated production of the chemokine CXCL5. Deletion of Nkx2-1 in SL mice (SNL) revealed that NKX2-1 suppresses SOX2-driven squamous tumorigenesis by repressing adeno-to-squamous transdifferentiation. Depletion of TANs in SNL mice reduced squamous tumors, suggesting that TANs foster squamous cell fate. Thus, lineage-defining transcription factors determine the tumor immune microenvironment, which in turn might impact the nature of the tumor.

CXCL8/CXCR2 signaling mediates bone marrow fibrosis and is a therapeutic target in myelofibrosis.

Proinflammatory signaling is a hallmark feature of human cancer, including in myeloproliferative neoplasms (MPNs), most notably myelofibrosis (MF). Dysregulated inflammatory signaling contributes to fibrotic progression in MF; however, the individual cytokine mediators elicited by malignant MPN cells to promote collagen-producing fibrosis and disease evolution are yet to be fully elucidated. Previously, we identified a critical role for combined constitutive JAK/STAT and aberrant NF-κB proinflammatory signaling in MF development. Using single-cell transcriptional and cytokine-secretion studies of primary cells from patients with MF and the human MPLW515L (hMPLW515L) murine model of MF, we extend our previous work and delineate the role of CXCL8/CXCR2 signaling in MF pathogenesis and bone marrow fibrosis progression. Hematopoietic stem/progenitor cells from patients with MF are enriched for a CXCL8/CXCR2 gene signature and display enhanced proliferation and fitness in response to an exogenous CXCL8 ligand in vitro. Genetic deletion of Cxcr2 in the hMPLW515L-adoptive transfer model abrogates fibrosis and extends overall survival, and pharmacologic inhibition of the CXCR1/2 pathway improves hematologic parameters, attenuates bone marrow fibrosis, and synergizes with JAK inhibitor therapy. Our mechanistic insights provide a rationale for therapeutic targeting of the CXCL8/CXCR2 pathway among patients with MF.

Deep learning applications in visual data for benign and malignant hematologic conditions: a systematic review and visual glossary.

Deep learning (DL) is a subdomain of artificial intelligence algorithms capable of automatically evaluating subtle graphical features to make highly accurate predictions, which was recently popularized in multiple imaging-related tasks. Because of its capabilities to analyze medical imaging such as radiology scans and digitized pathology specimens, DL has significant clinical potential as a diagnostic or prognostic tool. Coupled with rapidly increasing quantities of digital medical data, numerous novel research questions and clinical applications of DL within medicine have already been explored. Similarly, DL research and applications within hematology are rapidly emerging, although these are still largely in their infancy. Given the exponential rise of DL research for hematologic conditions, it is essential for the practising hematologist to be familiar with the broad concepts and pitfalls related to these new computational techniques. This narrative review provides a visual glossary for key deep learning principles, as well as a systematic review of published investigations within malignant and non-malignant hematologic conditions, organized by the different phases of clinical care. In order to assist the unfamiliar reader, this review highlights key portions of current literature and summarizes important considerations for the critical understanding of deep learning development and implementations in clinical practice.

COVID-19 Mortality Prediction From Deep Learning in a Large Multistate Electronic Health Record and Laboratory Information System Data Set: Algorithm Development and Validation.

BACKGROUND: COVID-19 is caused by the SARS-CoV-2 virus and has strikingly heterogeneous clinical manifestations, with most individuals contracting mild disease but a substantial minority experiencing fulminant cardiopulmonary symptoms or death. The clinical covariates and the laboratory tests performed on a patient provide robust statistics to guide clinical treatment. Deep learning approaches on a data set of this nature enable patient stratification and provide methods to guide clinical treatment. OBJECTIVE: Here, we report on the development and prospective validation of a state-of-the-art machine learning model to provide mortality prediction shortly after confirmation of SARS-CoV-2 infection in the Mayo Clinic patient population. METHODS: We retrospectively constructed one of the largest reported and most geographically diverse laboratory information system and electronic health record of COVID-19 data sets in the published literature, which included 11,807 patients residing in 41 states of the United States of America and treated at medical sites across 5 states in 3 time zones. Traditional machine learning models were evaluated independently as well as in a stacked learner approach by using AutoGluon, and various recurrent neural network architectures were considered. The traditional machine learning models were implemented using the AutoGluon-Tabular framework, whereas the recurrent neural networks utilized the TensorFlow Keras framework. We trained these models to operate solely using routine laboratory measurements and clinical covariates available within 72 hours of a patient's first positive COVID-19 nucleic acid test result. RESULTS: The GRU-D recurrent neural network achieved peak cross-validation performance with 0.938 (SE 0.004) as the area under the receiver operating characteristic (AUROC) curve. This model retained strong performance by reducing the follow-up time to 12 hours (0.916 [SE 0.005] AUROC), and the leave-one-out feature importance analysis indicated that the most independently valuable features were age, Charlson comorbidity index, minimum oxygen saturation, fibrinogen level, and serum iron level. In the prospective testing cohort, this model provided an AUROC of 0.901 and a statistically significant difference in survival (P<.001, hazard ratio for those predicted to survive, 95% CI 0.043-0.106). CONCLUSIONS: Our deep learning approach using GRU-D provides an alert system to flag mortality for COVID-19-positive patients by using clinical covariates and laboratory values within a 72-hour window after the first positive nucleic acid test result.

Ruxolitinib Therapy Followed by Reduced-Intensity Conditioning for Hematopoietic Cell Transplantation for Myelofibrosis: Myeloproliferative Disorders Research Consortium 114 Study.

We evaluated the feasibility of ruxolitinib therapy followed by a reduced-intensity conditioning (RIC) regimen for patients with myelofibrosis (MF) undergoing transplantation in a 2-stage Simon phase II trial. The aims were to decrease the incidence of graft failure (GF) and nonrelapse mortality (NRM) compared with data from the previous Myeloproliferative Disorders Research Consortium 101 Study. The plan was to enroll 11 patients each in related donor (RD) and unrelated donor (URD) arms, with trial termination if ≥3 failures (GF or death by day +100 post-transplant) occurred in the RD arm or ≥6 failures occurred in the URD. A total of 21 patients were enrolled, including 7 in the RD arm and 14 in the URD arm. The RD arm did not meet the predetermined criteria for proceeding to stage II. Although the URD arm met the criteria for stage II, the study was terminated owing to poor accrual and a significant number of failures. In all 19 transplant recipients, ruxolitinib was tapered successfully without significant side effects, and 9 patients (47%) had a significant decrease in symptom burden. The cumulative incidences of GF, NRM, acute graft-versus-host disease (GVHD), and chronic GVHD at 24 months were 16%, 28%, 64%, and 76%, respectively. On an intention-to-treat basis, the 2-year overall survival was 61% for the RD arm and 70% for the URD arm. Ruxolitinib can be integrated as pretransplantation treatment for patients with MF, and a tapering strategy before transplantation is safe, allowing patients to commence conditioning therapy with a reduced symptom burden. However, GF and NRM remain significant.

American Registry of Pathology Expert Opinions: Immunohistochemical evaluation of classic Hodgkin lymphoma.

The diagnosis of classic Hodgkin lymphoma requires immunohistochemical confirmation in most cases and one can argue for these studies as standard-of-care in the diagnostic workup. The authors propose a panel of studies for primary identification of CHL to include: CD3, CD20, CD15, CD30 and PAX5. When pattern discordances are identified, additional assessment is recommended. In the case of overexpression of B lineage markers by Hodgkin/Reed-Sternberg cells, or a differential diagnosis that includes large B-cell lymphoma or variants, additional markers recommended are: CD45, OCT2, BOB1, CD79a and MUM1/IRF4. If primary mediastinal large B cell lymphoma is considered in the differential diagnosis, suggested additional markers include: P63, CD23, CD45 and CD79a. When considering a differential diagnosis that includes anaplastic large cell lymphoma we suggest: ALK, CD45, pan T cell antigens (such as CD2, CD5, CD7, and CD43), and cytotoxic markers (granzyme, perforin, and TIA1). If peripheral T cell lymphoma or T cell lymphomas of follicular helper origin are considered in the differential diagnosis, the following panel is recommended: pan T cell antigens, CD4, CD8, one or more follicular dendritic cell markers, and assessment for Epstein-Barr virus (EBV) infection, preferably EBV encoded RNA (EBER) as assessed by in situ hybridization When the differential diagnosis includes nodular lymphocyte predominant Hodgkin lymphoma, recommended additional studies include OCT2, CD21 and/or CD23, PD1, and assessment for EBV infection. The authors recognize that these panels may not be adequate to completely characterize other lymphomas, but these panels will usually be sufficient to distinguish classic Hodgkin lymphoma from other lymphoma types.

Gestational Trophoblastic Disease: Clinical and Imaging Features.

Gestational trophoblastic disease (GTD) is a spectrum of both benign and malignant gestational tumors, including hydatidiform mole (complete and partial), invasive mole, choriocarcinoma, placental site trophoblastic tumor, and epithelioid trophoblastic tumor. The latter four entities are referred to as gestational trophoblastic neoplasia (GTN). These conditions are aggressive with a propensity to widely metastasize. GTN can result in significant morbidity and mortality if left untreated. Early diagnosis of GTD is essential for prompt and successful management while preserving fertility. Initial diagnosis of GTD is based on a multifactorial approach consisting of clinical features, serial quantitative human chorionic gonadotropin (ß-hCG) titers, and imaging findings. Ultrasonography (US) is the modality of choice for initial diagnosis of complete hydatidiform mole and can provide an invaluable means of local surveillance after treatment. The performance of US in diagnosing all molar pregnancies is surprisingly poor, predominantly due to the difficulty in differentiating partial hydatidiform mole from nonmolar abortion and retained products of conception. While GTN after a molar pregnancy is usually diagnosed with serial ß-hCG titers, imaging plays an important role in evaluation of local extent of disease and systemic surveillance. Imaging also plays a crucial role in detection and management of complications, such as uterine and pulmonary arteriovenous fistulas. Familiarity with the pathogenesis, classification, imaging features, and treatment of these tumors can aid in radiologic diagnosis and guide appropriate management. ©RSNA, 2017.

Clinical utility of next-generation sequencing in the diagnosis of hereditary haemolytic anaemias.

Hereditary haemolytic anaemias are genetically and phenotypically heterogeneous disorders characterized by increased red cell destruction, with consequences ranging from innocuous to severe life-threatening anaemia. Diagnostic laboratories endeavour to assist clinicians reach the exact patient diagnosis by using tests principally based on morphological and biochemical techniques. However, these routine studies may be inconclusive, particularly in newborn infants and when transfusions have recently been administered. Large numbers and size of the potentially involved genes also impose a practical challenge for molecular diagnosis using routine sequencing approaches. To overcome these diagnostic shortcomings, we have utilized next-generation sequencing to provide a high-throughput, highly sensitive assay. We developed a panel interrogating 28 genes encoding cytoskeletal proteins and enzymes with sequencing coverage of the coding regions, splice site junctions, deep intronic and regulatory regions. We then evaluated 19 samples, including infants with unexplained extreme hyperbilirubinaemia and patients with transfusion-dependent haemolytic anaemia. Where possible, inheritance patterns of pathogenic mutations were determined by sequencing of immediate relatives. We conclude that this next-generation sequencing panel could be a cost-effective approach to molecular diagnosis of hereditary haemolytic anaemia, especially when the family history is uninformative or when routine laboratory testing fails to identify the causative haemolytic process.

Bone marrow fibrosis in myelofibrosis: pathogenesis, prognosis and targeted strategies.

Bone marrow fibrosis is a central pathological feature and World Health Organization major diagnostic criterion of myelofibrosis. Although bone marrow fibrosis is seen in a variety of malignant and non-malignant disease states, the deposition of reticulin and collagen fibrosis in the bone marrow of patients with myelofibrosis is believed to be mediated by the myelofibrosis hematopoietic stem/progenitor cell, contributing to an impaired microenvironment favoring malignant over normal hematopoiesis. Increased expression of inflammatory cytokines, lysyl oxidase, transforming growth factor-ß, impaired megakaryocyte function, and aberrant JAK-STAT signaling have all been implicated in the pathogenesis of bone marrow fibrosis. A number of studies indicate that bone marrow fibrosis is an adverse prognostic variable in myeloproliferative neoplasms. However, modern myelofibrosis prognostication systems utilized in risk-adapted treatment approaches do not include bone marrow fibrosis as a prognostic variable. The specific effect on bone marrow fibrosis of JAK2 inhibition, and other rationally based therapies currently being evaluated in myelofibrosis, has yet to be fully elucidated. Hematopoietic stem cell transplantation remains the only curative therapeutic approach that reliably results in resolution of bone marrow fibrosis in patients with myelofibrosis. Here we review the pathogenesis, biological consequences, and prognostic impact of bone marrow fibrosis. We discuss the rationale of various anti-fibrogenic treatment strategies targeting the clonal hematopoietic stem/progenitor cell, aberrant signaling pathways, fibrogenic cytokines, and the tumor microenvironment.

Extracellular sodium and potassium levels modulate cardiac conduction in mice heterozygous null for the Connexin43 gene.

UNLABELLED: Several studies have disagreed on measurements of cardiac conduction velocity (CV) in mice with a heterozygous knockout of the connexin gene Gja1--a mutation that reduces the gap junction (GJ) protein, Connexin43 (Cx43), by 50 %. We noted that perfusate ionic composition varied between studies and hypothesized that extracellular ionic concentration modulates CV dependence on GJs. CV was measured by optically mapping wild-type (WT) and heterozygous null (HZ) hearts serially perfused with solutions previously associated with no change (Solution 1) or CV slowing (Solution 2). In WT hearts, CV was similar for Solutions 1 and 2. However, consistent with the hypothesis, Solution 2 in HZ hearts slowed transverse CV (CVT) relative to Solution 1. Previously, we showed CV slowing in a manner consistent with ephaptic conduction correlated with increased perinexal inter-membrane width (W P) at GJ edges. Thus, W P was measured following perfusion with systematically adjusted [Na(+)]o and [K(+)]o in Solutions 1 and 2. A wider W P was associated with reduced CVT in WT and HZ hearts, with the greatest effect in HZ hearts. Increasing [Na(+)]o increased CVT only in HZ hearts. Increasing [K(+)]o slowed CVT in both WT and HZ hearts with large W P but only in HZ hearts with narrow W P. CONCLUSION: When perinexi are wide, decreasing excitability by modulating [Na(+)]o and [K(+)]o increases CV sensitivity to reduced Cx43. By contrast, CV is less sensitive to Cx43 and ion composition when perinexi are narrow. These results are consistent with cardiac conduction dependence on both GJ and non-GJ (ephaptic) mechanisms.

Optimized immunohistochemical panel to differentiate myeloid sarcoma from blastic plasmacytoid dendritic cell neoplasm.

Myeloid sarcoma (MS) and blastic plasmacytoid dendritic cell neoplasm (BPDCN) can be difficult to distinguish morphologically, even with the use of extensive immunohistochemical studies. Three new research markers, myxovirus A (MxA), CLA/CD162, and CD303/BDCA-2, have been reported to be positive in BPDCN, but their clinical utility has never been tested. We compared these markers to other antibodies that have been used traditionally to distinguish MS from BPDCN to assess the utility of these newer antibodies in differential diagnosis. Formalin-fixed, paraffin-embedded tissue sections of 23 MS and 17 BPDCN cases were assessed using immunohistochemical analysis for CD4, CD14, CD33, CD43, CD56, CD68, CD123, CD163, myeloperoxidase, lysozyme, terminal deoxynucleotidyl transferase (TdT), T-cell leukemia 1 (TCL-1), MxA, cutaneous lymphocyte-associated antigen (CLA)/CD162, and blood dendritic cell antigen 2 (BDCA2)/CD303. We identified antibodies with a high predictive value of ≥ 90% and used these markers to develop an approach to classification using specific staining criteria. Diagnostic classification criteria were based on staining patterns of one or more of the seven markers. BPDCN was associated with positive staining for CD56, TdT, or TCL1, or negative staining for lysozyme. MS was associated with positive staining for lysozyme or myeloperoxidase, or negative staining for CD56, CD123, myxovirus, or TCL1. The immunohistochemical staining patterns observed using a panel that includes MPO, CD56, CD123, TCL1, TdT, and MxA, are predictive of MS or BPDCN. In this study, neither CD162 nor CD303 had good predictive value in distinguishing MS from BPDCN.

Side-to-side nerve bridges reduce muscle atrophy after peripheral nerve injury in a rodent model.

BACKGROUND: Peripheral nerve injury can result in muscle atrophy and long-term disability. We hypothesize that creating a side-to-side bridge to link an injured nerve with a healthy nerve will reduce muscle atrophy and improve muscle function. METHODS: Sprague-Dawley rats were divided into four groups (n = 7 per group). Group 1: transection only--a 10-mm gap was created in the proximal tibial nerve; group 2: transected plus repaired--the transected tibial nerve was repaired; group 3: transected plus repaired plus nerve bridge--transected nerve repaired with a distal nerve bridge between the tibial and peroneal nerves via epineurial windows; and group 4: transected plus nerve bridge--transected tibial nerve left unrepaired and distal bridge added. Gait was assessed every 2 wk. At 90 d the following measures were determined: gastrocnemius mass, muscle and nerve nuclear density, and axonal infiltration into the nerve bridge. RESULTS: Groups 3 and 4 had greater improvements in walking track recovery than groups 1 and 2. Group 3's gastrocnemius muscles exhibited the least amount of atrophy. Groups 1, 2, and 4 exhibited greater histologic appearance of muscle breakdown compared with group 3 and control muscle. Finally, most bridges in groups 3 and 4 had neuronal sprouting via the epineurial windows. CONCLUSIONS: Our study demonstrated reduced muscle atrophy with a side-to-side nerve bridge in the setting of peripheral nerve injury. These results support the application of novel side-to-side bridges in combination with traditional end-to-end neurorrhaphy to preserve muscle viability after peripheral nerve injuries.

Progression from Prediabetes to Diabetes in a Diverse U.S. Population: A Machine Learning Model.

Objective: To date, there are no widely implemented machine learning (ML) models that predict progression from prediabetes to diabetes. Addressing this knowledge gap would aid in identifying at-risk patients within this heterogeneous population who may benefit from targeted treatment and management in order to preserve glucose metabolism and prevent adverse outcomes. The objective of this study was to utilize readily available laboratory data to train and test the performance of ML-based predictive risk models for progression from prediabetes to diabetes. Methods: The study population was composed of laboratory information services data procured from a large U.S. outpatient laboratory network. The retrospective dataset was composed of 15,029 adults over a 5-year period with initial hemoglobin A1C (A1C) values between 5.0% and 6.4%. ML models were developed using random forest survival methods. The ground truth outcome was progression to A1C values indicative of diabetes (i.e., ≥6.5%) within 5 years. Results: The prediabetes risk classifier model accurately predicted A1C ≥6.5% within 5 years and achieved an area under the receiver-operator characteristic curve of 0.87. The most important predictors of progression from prediabetes to diabetes were initial A1C, initial serum glucose, A1C slope, serum glucose slope, initial HDL, HDL slope, age, and sex. Conclusions: Leveraging readily obtainable laboratory data, our ML risk classifier accurately predicts elevation in A1C associated with progression from prediabetes to diabetes. Although prospective studies are warranted, the results support the clinical utility of the model to improve timely recognition, risk stratification, and optimal management for patients with prediabetes.

CKD Progression Prediction in a Diverse US Population: A Machine-Learning Model.

Rationale & Objective: Chronic kidney disease (CKD) is a major cause of morbidity and mortality. To date, there are no widely used machine-learning models that can predict progressive CKD across the entire disease spectrum, including the earliest stages. The objective of this study was to use readily available demographic and laboratory data from Sonic Healthcare USA laboratories to train and test the performance of machine learning-based predictive risk models for CKD progression. Study Design: Retrospective observational study. Setting & Participants: The study population was composed of deidentified laboratory information services data procured from a large US outpatient laboratory network. The retrospective data set included 110,264 adult patients over a 5-year period with initial estimated glomerular filtration rate (eGFR) values between 15-89 mL/min/1.73 m2. Predictors: Patient demographic and laboratory characteristics. Outcomes: Accelerated (ie, >30%) eGFR decline associated with CKD progression within 5 years. Analytical Approach: Machine-learning models were developed using random forest survival methods, with laboratory-based risk factors analyzed as potential predictors of significant eGFR decline. Results: The 7-variable risk classifier model accurately predicted an eGFR decline of >30% within 5 years and achieved an area under the curve receiver-operator characteristic of 0.85. The most important predictor of progressive decline in kidney function was the eGFR slope. Other key contributors to the model included initial eGFR, urine albumin-creatinine ratio, serum albumin (initial and slope), age, and sex. Limitations: The cohort study did not evaluate the role of clinical variables (eg, blood pressure) on the performance of the model. Conclusions: Our progressive CKD classifier accurately predicts significant eGFR decline in patients with early, mid, and advanced disease using readily obtainable laboratory data. Although prospective studies are warranted, our results support the clinical utility of the model to improve timely recognition and optimal management for patients at risk for CKD progression. Plain-Language Summary: Defined by a significant decrease in estimated glomerular filtration rate (eGFR), chronic kidney disease (CKD) progression is strongly associated with kidney failure. However, to date, there are no broadly used resources that can predict this clinically significant event. Using machine-learning techniques on a diverse US population, this cohort study aimed to address this deficiency and found that a 5-year risk prediction model for CKD progression was accurate. The most important predictor of progressive decline in kidney function was the eGFR slope, followed by the urine albumin-creatinine ratio and serum albumin slope. Although further study is warranted, the results showed that a machine-learning model using readily obtainable laboratory information accurately predicts CKD progression, which may inform clinical diagnosis and management for this at-risk population.

MANIFEST: Pelabresib in Combination With Ruxolitinib for Janus Kinase Inhibitor Treatment-Naïve Myelofibrosis.

PURPOSE: Standard therapy for myelofibrosis comprises Janus kinase inhibitors (JAKis), yet spleen response rates of 30%-40%, high discontinuation rates, and a lack of disease modification highlight an unmet need. Pelabresib (CPI-0610) is an investigational, selective oral bromodomain and extraterminal domain inhibitor (BETi). METHODS: MANIFEST (ClinicalTrails.gov identifier: NCT02158858), a global, open-label, nonrandomized, multicohort, phase II study, includes a cohort of JAKi-naïve patients with myelofibrosis treated with pelabresib and ruxolitinib. The primary end point is a spleen volume reduction of ≥ 35% (SVR35) at 24 weeks. RESULTS: Eighty-four patients received ≥ 1 dose of pelabresib and ruxolitinib. The median age was 68 (range, 37-85) years; 24% of patients were intermediate-1 risk, 61% were intermediate-2 risk, and 16% were high risk as per the Dynamic International Prognostic Scoring System; 66% (55 of 84) of patients had a hemoglobin level of < 10 g/dL at baseline. At 24 weeks, 68% (57 of 84) achieved SVR35, and 56% (46 of 82) achieved a total symptom score reduction of ≥ 50% (TSS50). Additional benefits at week 24 included 36% (29 of 84) of patients with improved hemoglobin levels (mean, 1.3 g/dL; median, 0.8 g/dL), 28% (16 of 57) with ≥ 1 grade improvement in fibrosis, and 29.5% (13 of 44) with > 25% reduction in JAK2V617F-mutant allele fraction, which was associated with SVR35 response (P = .018, Fisher's exact test). At 48 weeks, 60% (47 of 79) of patients had SVR35 response. Grade 3 or 4 toxicities seen in ≥ 10% patients were thrombocytopenia (12%) and anemia (35%), leading to treatment discontinuation in three patients. 95% (80 of 84) of the study participants continued combination therapy beyond 24 weeks. CONCLUSION: The rational combination of the BETi pelabresib and ruxolitinib in JAKi-naïve patients with myelofibrosis was well tolerated and showed durable improvements in spleen and symptom burden, with associated biomarker findings of potential disease-modifying activity.

Computational pathology in 2030: a Delphi study forecasting the role of AI in pathology within the next decade.

BACKGROUND: Artificial intelligence (AI) is rapidly fuelling a fundamental transformation in the practice of pathology. However, clinical integration remains challenging, with no AI algorithms to date in routine adoption within typical anatomic pathology (AP) laboratories. This survey gathered current expert perspectives and expectations regarding the role of AI in AP from those with first-hand computational pathology and AI experience. METHODS: Perspectives were solicited using the Delphi method from 24 subject matter experts between December 2020 and February 2021 regarding the anticipated role of AI in pathology by the year 2030. The study consisted of three consecutive rounds: 1) an open-ended, free response questionnaire generating a list of survey items; 2) a Likert-scale survey scored by experts and analysed for consensus; and 3) a repeat survey of items not reaching consensus to obtain further expert consensus. FINDINGS: Consensus opinions were reached on 141 of 180 survey items (78.3%). Experts agreed that AI would be routinely and impactfully used within AP laboratory and pathologist clinical workflows by 2030. High consensus was reached on 100 items across nine categories encompassing the impact of AI on (1) pathology key performance indicators (KPIs) and (2) the pathology workforce and specific tasks performed by (3) pathologists and (4) AP lab technicians, as well as (5) specific AI applications and their likelihood of routine use by 2030, (6) AI's role in integrated diagnostics, (7) pathology tasks likely to be fully automated using AI, and (8) regulatory/legal and (9) ethical aspects of AI integration in pathology. INTERPRETATION: This systematic consensus study details the expected short-to-mid-term impact of AI on pathology practice. These findings provide timely and relevant information regarding future care delivery in pathology and raise key practical, ethical, and legal challenges that must be addressed prior to AI's successful clinical implementation. FUNDING: No specific funding was provided for this study.

A Phase Ib Trial of AVID200, a TGFß 1/3 Trap, in Patients with Myelofibrosis.

PURPOSE: Myelofibrosis (MF) is a clonal myeloproliferative neoplasm characterized by systemic symptoms, cytopenias, organomegaly, and bone marrow fibrosis. JAK2 inhibitors afford symptom and spleen burden reduction but do not alter the disease course and frequently lead to thrombocytopenia. TGFß, a pleiotropic cytokine elaborated by the MF clone, negatively regulates normal hematopoiesis, downregulates antitumor immunity, and promotes bone marrow fibrosis. Our group previously showed that AVID200, a potent and selective TGFß 1/3 trap, reduced TGFß1-induced proliferation of human mesenchymal stromal cells, phosphorylation of SMAD2, and collagen expression. Moreover, treatment of MF mononuclear cells with AVID200 led to increased numbers of progenitor cells (PC) with wild-type JAK2 rather than JAK2V617F. PATIENTS AND METHODS: We conducted an investigator-initiated, multicenter, phase Ib trial of AVID200 monotherapy in 21 patients with advanced MF. RESULTS: No dose-limiting toxicity was identified at the three dose levels tested, and grade 3/4 anemia and thrombocytopenia occurred in 28.6% and 19.0% of treated patients, respectively. After six cycles of therapy, two patients attained a clinical benefit by IWG-MRT criteria. Spleen and symptom benefits were observed across treatment cycles. Unlike other MF-directed therapies, increases in platelet counts were noted in 81% of treated patients with three patients achieving normalization. Treatment with AVID200 resulted in potent suppression of plasma TGFß1 levels and pSMAD2 in MF cells. CONCLUSIONS: AVID200 is a well-tolerated, rational, therapeutic agent for the treatment of patients with MF and should be evaluated further in patients with thrombocytopenic MF in combination with agents that target aberrant MF intracellular signaling pathways.

Interstitial volume modulates the conduction velocity-gap junction relationship.

Cardiac conduction through gap junctions is an important determinant of arrhythmia susceptibility. Yet, the relationship between degrees of G(j) uncoupling and conduction velocity (θ) remains controversial. Conflicting results in similar experiments are normally attributed to experimental differences. We hypothesized that interstitial volume modulates conduction velocity and its dependence on G(j). Interstitial volume (V(IS)) was quantified histologically from guinea pig right ventricle. Optical mapping was used to quantify conduction velocity and anisotropy (AR(θ)). Albumin (4 g/l) decreased histologically assessed V(IS), increased transverse θ by 71 ± 10%, and lowered AR(θ). Furthermore, albumin did not change isolated cell size. Conversely, mannitol increased V(IS), decreased transverse θ by 24 ± 4%, and increased AR(θ). Mannitol also decreased cell width by 12%. Furthermore, mannitol was associated with spontaneous ventricular tachycardias in three of eight animals relative to zero of 15 during control. The θ-G(j) relationship was assessed using the G(j) uncoupler carbenoxolone (CBX). Whereas 13 µM CBX did not significantly affect θ during control, it slowed transverse θ by 38 ± 9% during mannitol (edema). These data suggest changes in V(IS) modulate θ, AR(θ), and the θ-G(j) relationship and thereby alter arrhythmia susceptibility. Therefore, V(IS) may underlie arrhythmia susceptibility, particularly in diseases associated with gap junction remodeling.