Background Deep learning has been widely used for glaucoma diagnosis. However, there is no clinically validated algorithm for glaucoma incidence and progression prediction. This study aims to develop a clinically feasible deep-learning system for predicting and stratifying the risk of glaucoma onset and progression based on color fundus photographs (CFPs), with clinical validation of performance in external population cohorts.Methods We established data sets of CFPs and visual fields collected from longitudinal cohorts. The mean follow-up duration was 3 to 5 years across the data sets. Artificial intelligence (AI) models were developed to predict future glaucoma incidence and progression based on the CFPs of 17,497 eyes in 9346 patients. The area under the receiver operating characteristic (AUROC) curve, sensitivity, and specificity of the AI models were calculated with reference to the labels provided by experienced ophthalmologists. Incidence and progression of glaucoma were determined based on longitudinal CFP images or visual fields, respectively.Results The AI model to predict glaucoma incidence achieved an AUROC of 0.90 (0.81–0.99) in the validation set and demonstrated good generalizability, with AUROCs of 0.89 (0.83–0.95) and 0.88 (0.79–0.97) in external test sets 1 and 2, respectively. The AI model to predict glaucoma progression achieved an AUROC of 0.91 (0.88–0.94) in the validation set, and also demonstrated outstanding predictive performance with AUROCs of 0.87 (0.81–0.92) and 0.88 (0.83–0.94) in external test sets 1 and 2, respectively.Conclusion Our study demonstrates the feasibility of deep-learning algorithms in the early detection and prediction of glaucoma progression.FUNDING National Natural Science Foundation of China (NSFC); the High-level Hospital Construction Project, Zhongshan Ophthalmic Center, Sun Yat-sen University; the Science and Technology Program of Guangzhou, China (2021), the Science and Technology Development Fund (FDCT) of Macau, and FDCT-NSFC.
Fei Li, Yuandong Su, Fengbin Lin, Zhihuan Li, Yunhe Song, Sheng Nie, Jie Xu, Linjiang Chen, Shiyan Chen, Hao Li, Kanmin Xue, Huixin Che, Zhengui Chen, Bin Yang, Huiying Zhang, Ming Ge, Weihui Zhong, Chunman Yang, Lina Chen, Fanyin Wang, Yunqin Jia, Wanlin Li, Yuqing Wu, Yingjie Li, Yuanxu Gao, Yong Zhou, Kang Zhang, Xiulan Zhang
BACKGROUND. In human lupus nephritis (LN), tubulointerstitial inflammation (TII) on biopsy predicts progression to end-stage renal disease (ESRD). However, only about half of patients with moderate/severe TII develop ESRD. We hypothesized that this heterogeneity in outcome reflects different underlying inflammatory states. Therefore, we interrogated renal biopsies from LN longitudinal and cross-sectional cohorts. METHODS. Data was acquired using conventional and highly multiplexed confocal microscopy. To accurately segment cells across whole biopsies, and to understand their spatial relationships, we developed computational pipelines by training and implementing several deep learning models and other computer vision techniques. RESULTS. High B cell densities were associated with protection from ESRD. In contrast, CD8, γδ and other CD4-CD8- T cells, were associated with both acute renal failure and progression to ESRD. B cells were often organized into large periglomerular neighborhoods with T follicular helper cells while CD4- T cells formed small neighborhoods in the tubulointerstitium whose frequency predicted progression to ESRD. CONCLUSIONS. These data reveal that specific in situ inflammatory states are associated with refractory and progressive renal disease. FUNDING. These studies were funded by the NIH Autoimmunity Centers of Excellence (AI082724), Department of Defense (LRI180083) and Alliance for Lupus Research, NIH S10-OD025081, S10-RR021039, and P30-CA14599 awards.
Rebecca Abraham, Madeleine S. Durkee, Junting Ai, Margaret Veselits, Gabriel Casella, Yuta Asano, Anthony Chang, Kichul Ko, Charles Oshinsky, Emily Peninger, Maryellen L. Giger, Marcus R. Clark
BACKGROUND. Patients undergoing immune-modifying therapies demonstrate a reduced humoral response after COVID-19 vaccination, but we lack a proper evaluation of the impact of such therapies on vaccine-induced T cell responses. METHODS. We longitudinally characterized humoral and Spike-specific T cell responses in inflammatory bowel disease (IBD) patients who are on antimetabolite therapy (azathioprine or methotrexate), TNF inhibitors and/or other biologic treatment (anti-integrin or anti-p40) for up to 6 months after completing two-dose COVID-19 mRNA vaccination. RESULTS. We demonstrated that a Spike-specific T cell response is not only induced in treated IBD patients at levels similar to healthy individuals, but also sustained at higher magnitude for up to 6 months after vaccination, particularly in those treated with TNF inhibitor therapy. Furthermore, the Spike-specific T cell response in these patients is mainly preserved against mutations present in SARS-CoV-2 B.1.1.529 (Omicron) and characterized by a Th1/IL-10 cytokine profile. CONCLUSION. Despite the humoral response defects, patients under immune-modifying therapies demonstrated a favorable profile of vaccine-induced T cell responses that might still provide a layer of COVID-19 protection. FUNDING. This study was funded by the National Centre for Infectious Diseases NCID Catalyst Grant (FY2021ES) and the National Research Fund Competitive Research Programme (NRF-CRP25-2020-0003). The funders played no role in the design, conduct, or reporting of this study.
Martin Qui, Nina Le Bert, Webber Pak Wo Chan, Malcolm Tan, Shou Kit Hang, Smrithi Hariharaputran, Jean Xiang Ying Sim, Jenny Guek Hong Low, Weiling Ng, Wei Yee Wan, Tiing Leong Ang, Antonio Bertoletti, Ennaliza Salazar
BACKGROUND Hyaluronan (HA), an extracellular matrix glycosaminoglycan, has been implicated in the pathophysiology of COVID-19 infection, pulmonary hypertension, pulmonary fibrosis, and other diseases, but is not targeted by any approved drugs. We asked whether hymecromone (4-methylumbelliferone [4-MU]), an oral drug approved in Europe for biliary spasm treatment that also inhibits HA in vitro and in animal models, could be repurposed as an inhibitor of HA synthesis in humans.METHODS We conducted an open-label, single-center, dose-response study of hymecromone in healthy adults. Subjects received hymecromone at 1200 (n = 8), 2400 (n = 9), or 3600 (n = 9) mg/d divided into 3 doses daily, administered orally for 4 days. We assessed safety and tolerability of hymecromone and analyzed HA, 4-MU, and 4-methylumbelliferyl glucuronide (4-MUG; the main metabolite of 4-MU) concentrations in sputum and serum.RESULTS Hymecromone was well tolerated up to doses of 3600 mg/d. Both sputum and serum drug concentrations increased in a dose-dependent manner, indicating that higher doses lead to greater exposures. Across all dose arms combined, we observed a significant decrease in sputum HA from baseline after 4 days of treatment. We also observed a decrease in serum HA. Additionally, higher baseline sputum HA levels were associated with a greater decrease in sputum HA.CONCLUSION After 4 days of exposure to oral hymecromone, healthy human subjects experienced a significant reduction in sputum HA levels, indicating this oral therapy may have potential in pulmonary diseases where HA is implicated in pathogenesis.TRIAL REGISTRATION ClinicalTrials.gov NCT02780752.FUNDING Stanford Medicine Catalyst, Stanford SPARK, Stanford Innovative Medicines Accelerator program, NIH training grants 5T32AI052073-14 and T32HL129970.
Joelle I. Rosser, Nadine Nagy, Riya Goel, Gernot Kaber, Sally Demirdjian, Jamie Saxena, Jennifer B. Bollyky, Adam R. Frymoyer, Ana E. Pacheco-Navarro, Elizabeth B. Burgener, Jayakumar Rajadas, Zhe Wang, Olga Arbach, Colleen E. Dunn, Anissa Kalinowski, Carlos E. Milla, Paul L. Bollyky
BACKGROUND. Neutralizing antibodies are considered a key correlate of protection by current SARS-CoV-2 vaccines. The manner in which human infections respond to therapeutic SARS-CoV-2 antibodies, including convalescent plasma therapy (CPT), remains to be fully elucidated. METHODS. Here, we conducted a proof-of-principle study of CPT based on a phase I trial in thirty hospitalized COVID-19 patients with a median interval between the onset of symptoms and the first transfusion of 9 days (IQR, 7-11.8 days). A comprehensive longitudinal monitoring of the virologic, serologic, and disease status of recipients allowed deciphering of parameters on which plasma therapy efficacy depends. RESULTS. In the context of this trial CPT was safe as evidenced by the absence of transfusion related adverse events and a low mortality (3.3%). Treatment with highly neutralizing plasma was significantly associated with faster virus clearance, as demonstrated by Kaplan-Meier analysis (p= 0.034) and confirmed in a parametric survival model including viral load and comorbidity (adjusted hazard ratio (HR)= 3.0 [95% confidence interval (CI) 1.1;8.1], p= 0.026). The onset of endogenous neutralization had a noticeable effect on viral clearance but, importantly, even after adjusting for their pre-transfusion endogenous neutralization status recipients benefitted from plasma therapy with high neutralizing antibodies (HR= 3.5 [95% CI 1.1;11], p= 0.034). CONCLUSION. In summary, our data demonstrate a clear impact of exogenous antibody therapy on the rapid clearance of viremia before and after onset of the endogenous neutralizing response and more broadly point beyond antibody-based interventions to critical laboratory parameters for improved evaluation of current and future SARS-CoV-2 therapies. TRIAL REGISTRATION. ClinicalTrials.gov NCT04869072 FUNDING. This study was funded via an “Innovation-Pool” project by the University Hospital Zurich, the “Swiss Red Cross “Glückskette” Corona Funding”, Pandemiefonds of the UZH Foundation and the Clinical Research Priority Program ‘Comprehensive Genomic Pathogen Detection’ of the University of Zurich.
Maddalena Marconato, Irene A. Abela, Anthony Hauser, Magdalena Schwarzmüller, Rheliana Katzensteiner, Dominique L. Braun, Selina Epp, Annette Audigé, Jacqueline Weber, Peter Rusert, Emèry Schindler, Chloé Pasin, Emily West, Jürg Böni, Verena Kufner, Michael Huber, Maryam Zaheri, Stefan Schmutz, Beat M. Frey, Roger D. Kouyos, Huldrych F. Günthard, Markus G. Manz, Alexandra Trkola
BACKGROUND. Cytomegalovirus (CMV) is the most common intrauterine infection, leading to infant brain damage. Prognostic assessment of CMV-infected fetuses has remained an ongoing challenge in prenatal care, in the absence of established prenatal biomarkers of congenital CMV (cCMV) infection severity. We aimed to identify prognostic biomarkers of cCMV-related fetal brain injury. METHODS. Global proteome analysis was performed in mid-gestation amniotic fluid samples, comparing fetuses with severe cCMV to asymptomatic CMV-infected fetuses. The levels of selected differentially-excreted proteins were further determined by specific immunoassays. RESULTS. Employing unbiased proteome analysis in a discovery cohort, we identified amniotic fluid proteins related to inflammation and neurological disease pathways, which demonstrated distinct abundance in fetuses with severe cCMV. Amniotic fluid levels of two of these proteins - the immunomodulatory proteins chemerin and galectin-3-binding-protein - were highly predictive of the severity of cCMV in an independent validation cohort, differentiating between fetuses with severe (N=17) and asymptomatic (N=26) cCMV, with 100-93.8% positive predictive value, and 92.9-92.6% negative predictive value (for chemerin - galectin-3-binding-protein, respectively). CONCLUSION. Analysis of chemerin and galectin-3-binding-protein in mid-gestation amniotic fluids could be employed in the clinical setting to profoundly improve the prognostic assessment of CMV-infected fetuses. TRIAL REGISTRATION. NA FUNDING. Israel Science Foundation; Research Fund - Hadassah Medical Organization.
Olesya Vorontsov, Lorinne Levitt, Daniele Lilleri, Gilad W. Vainer, Orit Caplan, Licita Schreiber, Alessia Arossa, Arsenio Spinillo, Milena Furione, Or Alfi, Esther Oiknine-Djian, Meital Kupervaser, Yuval Nevo, Sharona Elgavish, Moran Yassour, Maurizio Zavattoni, Tali Bdolah-Abram, Fausto Baldanti, Miriam Geal-Dor, Zichria Zakay-Rones, Nili Yanai, Simcha Yagel, Amos Panet, Dana G. Wolf
BACKGROUND. It is unclear whether the level of serum hepatitis B virus (HBV) DNA at baseline impacts the on-treatment risk of hepatocellular carcinoma (HCC) in HBeAg positive, non-cirrhotic patients with chronic hepatitis B (CHB). METHODS. We conducted a multicenter cohort study including 2,073 entecavir- or tenofovir-treated, HBeAg-positive, non-cirrhotic, adult CHB patients with baseline HBV DNA levels ≥5.00 log10 IU/mL at three centers in Korea between January 2007 and December 2016. We evaluated the on-treatment incidence rate of HCC by baseline HBV DNA levels. RESULTS. During a median 5.7 years of continuous antiviral treatment, 47 patients developed HCC (0.39 per 100 person-years). By Kaplan–Meier analysis, HCC risk was the lowest in those with baseline HBV DNA levels ≥8.00 log10 IU/mL, increased incrementally with decreasing viral load, and the highest with HBV DNA levels 5.00–5.99 log10 IU/mL (P<0.001). By multivariable analysis, baseline HBV DNA level was an independent factor that was inversely associated with HCC risk. Compared with HBV DNA ≥8.00 log10 IU/mL, the adjusted hazard ratios for HCC risk with HBV DNA 7.00–7.99 log10 IU/mL, 6.00–6.99 log10 IU/mL, and 5.00–5.99 log10 IU/mL were 2.48 (P=0.03), 3.69 (P=0.002), and 6.10 (P<0.001), respectively. CONCLUSION. On-treatment HCC risk increased incrementally with decreasing baseline HBV DNA levels in the range of ≥5.00 log10 IU/mL in HBeAg-positive, non-cirrhotic, adult patients with CHB. Early initiation of antiviral treatment with a high viral load (≥8.00 log10 IU/mL) may maintain the lowest risk of HCC in those patients. FUNDING. Korean Government.
Won-Mook Choi, Gi-Ae Kim, Jonggi Choi, Seungbong Han, Young-Suk Lim
BACKGROUND. Tuberous Sclerosis Complex (TSC) is a neurogenetic syndrome due to loss-of-function mutations in TSC2 or TSC1, characterized by tumors at multiple body sites, including facial angiofibroma (FAF). Here, an ultrasensitive assessment of the extent and range of UV-induced mutations in TSC facial skin was performed. METHODS. A Multiplex High-sensitivity PCR Assay (MHPA) was developed, enabling mutation detection at extremely low (<0.1%) variant allele frequencies (VAF). RESULTS. MHPA assays were developed for both TSC2 and TP53, and applied to 81 samples, including 66 skin biopsies. UV-induced second hit mutation causing inactivation of TSC2 was pervasive in TSC facial skin with an average of 4.8 mutations per 2 mm biopsy at median VAF 0.08%, generating >150,000 incipient facial tumors (subclinical ‘micro-FAFs’) in the average TSC subject. The MHPA analysis also led to the identification of a refined UV-related indel signature and a recurrent complex mutation pattern, consisting of both a single or dinucleotide variant, and a 1-9 nt deletion, in cis. CONCLUSION. TSC facial skin can be viewed as harboring a patchwork of clonal fibroblast proliferations (micro-FAF) with indolent growth, a small proportion of which develop into clinically observable FAF. Our observations also expand the spectrum of UV-related mutation signatures. FUNDING. This work was supported by the TSC Alliance, Engles Family Fund for Research in TSC and LAM, and National Institutes of Health, National Heart, Lung, and Blood Institute [U01HL131022-04; Intramural Research Program].
Katarzyna Klonowska, Joannes M. Grevelink, Krinio Giannikou, Barbara A. Ogorek, Zachary T. Herbert, Aaron R. Thorner, Thomas N. Darling, Joel Moss, David J. Kwiatkowski
BACKGROUND. Responses to conventional donor lymphocyte infusion (DLI) for post-allogeneic hematopoietic cell transplantation (HCT) relapse are typically poor. Natural killer (NK) cell-based therapy is a promising modality to treat post-HCT relapse. METHODS. We initiated this ongoing phase I trial of adoptively transferred cytokine induced memory-like (CIML) NK cells in patients with myeloid malignancies relapsed after haploidentical HCT. All patients received a donor-derived NK cell dose of 5–10 million cells/kg after lymphodepleting chemotherapy, followed by systemic IL-2 for 7 doses. High resolution profiling with mass cytometry and single cell RNA sequencing characterized the expanding and persistent NK cell subpopulations in a longitudinal manner after infusion. RESULTS. In the first 6 enrolled patients on the trial, infusion of CIML NK cells led to a rapid 10 to 50-fold in vivo expansion that was sustained over months. The infusion was well-tolerated, with fever and pancytopenia as the most common adverse events. Expansion of NK cells was distinct from IL-2 effects on endogenous post-HCT NK cells, and not dependent on CMV viremia. Immunophenotypic and transcriptional profiling revealed a dynamic evolution of the activated CIML NK cell phenotype, superimposed on the natural variation in donor NK cell repertoires. CONCLUSION. Given their rapid expansion and long-term persistence in an immune compatible environment, CIML NK cells serve as a promising platform for the treatment of post-transplant relapse of myeloid disease. Further characterization of their unique in vivo biology and interaction with both T cells and tumor targets will lead to improvements in cell-based immunotherapies. TRIAL REGISTRATION. NCT04024761 FUNDING. Supported by Dunkin Donuts Breakthrough Award, the NIH/National Cancer Institute R21 CA245413, the Leukemia and Lymphoma Society Scholar and TRP awards.
Roman M. Shapiro, Grace C. Birch, Guangan Hu, Juliana Vergara Cadavid, Sarah Nikiforow, Joanna Baginska, Alaa K. Ali, Mubin Tarannum, Michal Sheffer, Yasmin Z. Abdulhamid, Benedetta Rambaldi, Yohei Arihara, Carol Reynolds, Max S. Halpern, Scott J. Rodig, Nicole Cullen, Jacquelyn O. Wolff, Kathleen L. Pfaff, Andrew A. Lane, R. Coleman Lindsley, Corey S. Cutler, Joseph H. Antin, Vincent T. Ho, John Koreth, Mahasweta Gooptu, Haesook T. Kim, Karl-Johan Malmberg, Catherine J. Wu, Jianzhu Chen, Robert J. Soiffer, Jerome Ritz, Rizwan Romee
BACKGROUND. Myotonic dystrophy type 1 (DM1) is a complex life-limiting neuromuscular disorder characterized by severe skeletal muscle atrophy, weakness, and cardio-respiratory defects. Exercised DM1 mice exhibit numerous physiological benefits that are underpinned by reduced CUG foci and improved alternative splicing. However, the efficacy of physical activity in patients is unknown. METHODS. Eleven genetically diagnosed DM1 patients were recruited to examine the extent to which 12-weeks of cycling can recuperate clinical, and physiological metrics. Furthermore, we studied the underlying molecular mechanisms through which exercise elicits benefits in skeletal muscle of DM1 patients. RESULTS. DM1 was associated with impaired muscle function, fitness, and lung capacity. Cycling evoked several clinical, physical, and metabolic advantages in DM1 patients. We highlight that exercise-induced molecular and cellular alterations in patients do not conform with previously published data in murine models and propose a significant role of mitochondrial function in DM1 pathology. Lastly, we discovered a subset of small nucleolar RNAs (snoRNAs) that correlated to indicators of disease severity. CONCLUSION. With no available cures, our data supports the efficacy of exercise as a primary intervention to partially mitigate the clinical progression of DM1. Additionally, we provide evidence for the involvement of snoRNAs and other noncoding RNAs in DM1 pathophysiology. TRIAL REGISTRATION. This trial was approved by the HiREB committee (#7901) and registered under ClinicalTrials.gov (NCT04187482). FUNDING. This work was primarily supported by Neil and Leanne Petroff. This study was also supported by a Canadian Institutes of Health Research Foundation Grant to MAT (#143325).
Andrew I. Mikhail, Peter L. Nagy, Katherine Manta, Nicholas Rouse, Alexander Manta, Sean Y. Ng, Michael F. Nagy, Paul Smith, Jian-Qiang Lu, Joshua P. Nederveen, Vladimir Ljubicic, Mark A. Tarnopolsky
No posts were found with this tag.