Citations

AbstractData-driven machine-learning methods on neuroimaging (e.g. MRI) are of great interest for the investigation and classification of neurological diseases. However, traditional machine learning requires domain knowledge to delineate the brain regions first, followed by feature extraction from the regions. Compared with this semi-automated approach, recently developed deep learning methods have advantages since they do not require such prior knowledge; instead, deep learning methods can automatically find features that differentiate MRIs from different cohorts. In the present study, we developed a deep learning-based classification pipeline distinguishing brain MRIs of individuals with one of three types of headaches [migraine (n = 95), acute post-traumatic headache (n = 48) and persistent post-traumatic headache (n = 49)] from those of healthy controls (n = 532) and identified the brain regions that most contributed to each classification task. Our pipeline included: (i) data preprocessing; (ii) binary classification of healthy controls versus headache type using a 3D ResNet-18; and (iii) biomarker extraction from the trained 3D ResNet-18. During the classification at the second step of our pipeline, we resolved two common issues in deep learning methods, limited training data and imbalanced samples from different categories, by incorporating a large public data set and resampling among the headache cohorts. Our method achieved the following classification accuracies when tested on independent test sets: (i) migraine versus healthy controls—75% accuracy, 66.7% sensitivity and 83.3% specificity; (2) acute post-traumatic headache versus healthy controls—75% accuracy, 66.7% sensitivity and 83.3% specificity; and (3) persistent post-traumatic headache versus healthy controls—91.7% accuracy, 100% sensitivity and 83.3% specificity. The most significant biomarkers identified by the classifier for migraine were caudate, caudal anterior cingulate, superior frontal, thalamus and ventral diencephalon. For acute post-traumatic headache, lateral occipital, cuneus, lingual, pericalcarine and superior parietal regions were identified as most significant biomarkers. Finally, for persistent post-traumatic headache, the most significant biomarkers were cerebellum, middle temporal, inferior temporal, inferior parietal and superior parietal. In conclusion, our study shows that the deep learning methods can automatically detect aberrations in the brain regions associated with different headache types. It does not require any human knowledge as input which significantly reduces human effort. It uncovers the great potential of deep learning methods for classification and automatic extraction of brain imaging–based biomarkers for these headache types.

Abstract Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare hematologic malignancy with historically poor outcomes and no worldwide consensus treatment approach. Unique among most hematologic malignancies for its frequent cutaneous involvement, BPDCN can also invade other extramedullary compartments, including the central nervous system. Generally affecting older adults, many patients are unfit to receive intensive chemotherapy, and although hematopoietic stem cell transplantation is preferred for younger, fit individuals, not all are eligible. One recent therapeutic breakthrough is that all BPDCNs express CD123 (IL3Rα) and that this accessible surface marker can be pharmacologically targeted. The first-in-class agent for BPDCN, tagraxofusp, which targets CD123, was approved in December 2018 in the United States for patients with BPDCN aged ≥2 years. Despite favorable response rates in the frontline setting, many patients still relapse in the setting of monotherapy, and outcomes in patients with relapsed/refractory BPDCN remain dismal. Therefore, novel approaches targeting both CD123 and other targets are actively being investigated. To begin to formally address the state of the field, we formed a new collaborative initiative, the North American BPDCN Consortium (NABC). This group of experts, which includes a multidisciplinary panel of hematologists/oncologists, hematopoietic stem cell transplant physicians, pathologists, dermatologists, and pediatric oncologists, was tasked with defining the current standard of care in the field and identifying the most important research questions and future directions in BPDCN. The position findings of the NABC’s inaugural meetings are presented herein.

AbstractTransthyretin‐mediated amyloidosis (ATTR) is a rare, under‐recognized, progressively debilitating, fatal disease caused by the aggregation and extracellular deposition of amyloid transthyretin (TTR) fibrils in multiple organs and tissues throughout the body. TTR is predominantly synthesized by the liver and normally circulates as a homotetramer, while misfolded monomers aggregate to form amyloid fibrils. One strategy to treat ATTR amyloidosis is to reduce the amount of TTR produced by the liver using drugs that directly target the TTR mRNA or gene. This narrative review focuses on how TTR gene silencing tools act to reduce TTR production, describing strategies for improved targeted delivery of these agents to hepatocytes where TTR is preferentially expressed. Antisense oligonucleotides (ASOs) and small interfering RNAs (siRNAs), termed RNA silencers, cause selective degradation of TTR mRNA, while a TTR gene editing tool reduces TTR expression by introducing nonsense mutations into the TTR gene. Two strategies to facilitate tissue‐specific delivery of these nucleic acid‐based drugs employ endogenous receptors expressed by hepatocytes. Lipid nanoparticles (LNPs) that recruit apolipoprotein E support low‐density lipoprotein receptor‐mediated uptake of unconjugated siRNA and are now used for CRISPR gene editing tools. Additionally, conjugating N‐acetylgalactosamine (GalNAc) moieties to ASOs or siRNAs facilitates receptor‐mediated uptake by the asialoglycoprotein receptor. In summary, ATTR is a progressive disease with various clinical manifestations due to TTR aggregation, deposition, and amyloid formation. Receptor‐targeted ligands (eg, GalNAc) and nanoparticle encapsulation (eg, LNPs) are technologies to deliver ASOs, siRNAs, and gene editing tools to hepatocytes, the primary location of TTR synthesis.

Background. Nocardia is an environmental pathogen with a predilection for causing opportunistic infections in immunocompromised patients, including solid organ transplant (SOT) recipients. Although risk factors have been identified for developing nocardiosis in this population, little is known regarding clinical factors resulting in poor outcomes. We evaluated a cohort of SOT recipients with nocardiosis for associations with 12-month mortality. Methods. We performed a multicenter retrospective cohort study of adult SOT recipients diagnosed with culture-confirmed nocardiosis from 2000 to 2020. Patients were followed for 12 months after diagnosis, unless abbreviated by mortality. Multivariable Cox regression was performed to analyze associations with 12-month mortality. A subgroup analysis of patients with disseminated nocardiosis was performed to analyze treatment variables. Results. A total of 125 SOT recipients met inclusion criteria; 12-month mortality was 16.8%. Liver transplantation (hazard ratio [HR] 3.52; 95% confidence interval [CI] 1.27-9.76) and time from symptom onset to presentation (HR 0.92/d; 95% CI 0.86-0.99) were independently associated with 12-month mortality, whereas disseminated infection was not (HR 1.23; 95% CI 0.49-3.13). No treatment-specific factors were significantly associated with mortality in 33 patients with disseminated nocardiosis, although survivors had a higher rate of linezolid use. Conclusions. This study identified 2 independent associations with 12-month mortality, representing demographics and infection severity. Disseminated infection was not independently associated with poor outcomes, and specific sites of infection may be more important than dissemination itself. No treatment-specific factors were associated with mortality, though this analysis was likely underpowered. Further study of treatment strategies based on specific Nocardia syndromes is warranted.

ObjectiveSystemic lupus erythematosus (SLE) is marked by immune dysregulation linked to varied clinical disease activity. Using a unique longitudinal cohort of SLE patients, this study sought to identify optimal immune mediators informing an empirically refined flare risk index (FRI) reflecting altered immunity prior to clinical disease flare.MethodsThirty‐seven SLE‐associated plasma mediators were evaluated by microfluidic immunoassay in 46 samples obtained in SLE patients with an imminent clinical disease flare (preflare) and 53 samples obtained in SLE patients without a flare over a corresponding period (pre‐nonflare). SLE patients were selected from a unique longitudinal cohort of 106 patients with classified SLE (meeting the American College of Rheumatology 1997 revised criteria for SLE or the Systemic Lupus International Collaborating Clinics 2012 revised criteria for SLE). Autoantibody specificities, hybrid SLE Disease Activity Index (hSLEDAI) scores, clinical features, and medication usage were also compared at preflare (mean ± SD 111 ± 47 days prior to flare) versus pre‐nonflare (99 ± 21 days prior to nonflare) time points. Variable importance was determined by random forest analysis with logistic regression subsequently applied to determine the optimal number and type of analytes informing a refined FRI.ResultsPreflare versus pre‐nonflare differences were not associated with demographics, autoantibody specificities, hSLEDAI scores, clinical features, nor medication usage. Forward selection and backward elimination of mediators ranked by variable importance resulted in 17 plasma mediator candidates differentiating preflare from pre‐nonflare visits. A final combination of 11 mediators best informed a newly refined FRI, which achieved a maximum sensitivity of 97% and maximum specificity of 98% after applying decision curve analysis to define low, medium, and high FRI scores.ConclusionWe verified altered immune mediators associated with imminent disease flare, and a subset of these mediators improved the FRI to identify SLE patients at risk of imminent flare. This molecularly informed, proactive management approach could be critical in prospective clinical trials and the clinical management of lupus.

PURPOSE Binary cardiac response assessment using cardiac biomarkers is prognostic in light chain amyloidosis. Previous studies suggested four-level cardiac responses using N-terminal prohormone of brain natiuretic peptide improves prognostic prediction. This study was designed to validate graded cardiac response criteria using N-terminal prohormone of brain natiuretic peptide/brain natiuretic peptide. PATIENTS AND METHODS This retrospective, multicenter study included patients with light chain amyloidosis who achieved at least a hematologic partial response (PR) and were evaluable for cardiac response. Four response criteria were tested on the basis of natriuretic peptide response depth: cardiac complete response (CarCR), cardiac very good partial response (CarVGPR), cardiac PR (CarPR), and cardiac no response (CarNR). Response was classified as best response and at fixed time points (6, 12, and 24 months from therapy initiation). The study primary outcome was overall survival. RESULTS 651 patients were included. Best CarCR, CarVGPR, CarPR, and CarNR were achieved in 16%, 26.4%, 22.9%, and 34.7% of patients, respectively. Patients in cardiac stage II were more likely to achieve CarCR than patients in cardiac stage IIIA and IIIB (22% v 13.5% v 3.2%; P < .001). A deeper cardiac response was associated with a longer survival (5-year overall survival 93%, 79%, 65%, and 33% for CarCR, CarVGPR, CarPR, and CarNR, respectively; P < .001). Fixed time-point analyses and time-varying covariates Cox regression analysis, to minimize survivorship bias, affirmed the independent survival advantage of deeper cardiac responses. Four-level response performed better than two-level response as early as 12 months from therapy initiation. CONCLUSION Graded cardiac response criteria allow better assessment of cardiac improvement compared with the traditional binary response system. The study re-emphasizes the importance of early diagnosis, which increases the likelihood of deep cardiac responses.

This study examined the effects of different types of masks (no mask, surgical mask (SM), and N95-mask) on physiological and perceptual responses during 30-min of self-paced cycle ergometer exercise. This study was a prospective randomly assigned experimental design. Outcomes included workload (Watts), oxygen saturation (SpO2), end-tidal carbon dioxide (PetCO2), heart rate (HR), respiratory rate (RR), rating of perceived exertion (RPE), and rating of perceived dyspnea (RPD). Volunteers (54–83 years (n = 19)) completed two familiarization sessions and three testing sessions on an air braked cycle ergometer. No significant difference was found for condition x time for any of the dependent variables. RPE, RPD, and PetCO2 were significantly higher with an N95-mask vs. no mask (NM) ((p = 0.012), (p = 0.002), (p < 0.001)). HR was significantly higher with the SM compared to the NM condition (p = 0.027) (NM 107.18 ± 9.96) (SM 112.34 ± 10.28), but no significant difference was found when comparing the SM to the N95 condition or when comparing the N95condition to the NM condition. Watts increased across time in each condition (p = 0.003). Initially RR increased during the first 3 min of exercise (p < 0.001) with an overall gradual increase noted across time regardless of mask condition (p < 0.001). SpO2 significantly decreased across time but remained within normal limits (>95%). No significant difference was found in Watts, RR, or SpO2 regardless of mask condition. Overall, the N95mask was associated with increased RPE, RPD, and PetCO2 levels. This suggests trapping of CO2 inside the mask leading to increased RPE and RPD.

Abstract Background Antispike monoclonal antibodies are recommended for early treatment of high-risk persons with mild to moderate coronavirus disease 2019 (COVID-19). However, clinical outcomes of their use during the severe acute respiratory syndrome coronavirus 2 Omicron wave are limited. Methods This is a descriptive retrospective study of high-risk adult patients who received treatment with sotrovimab (January 1–March 20, 2022) or bebtelovimab (March 21–April 30, 2022). The primary outcome was the proportion of patients who progressed to severe outcome within 30 days after receiving antispike-neutralizing monoclonal antibody infusion. Results A total of 3872 high-risk patients (median age, 62.7 years; 41.1% male) with mild to moderate COVID-19 received sotrovimab (n = 2182) or bebtelovimab (n = 1690). Among sotrovimab-treated patients, the most common comorbidities were an immunosuppressed condition (46.7%), hypertension (38.2%), and diabetes (21.2%). The rates of severe outcome, intensive care unit (ICU) admission, and mortality were 2.2%, 1.0%, and 0.4%, respectively, after sotrovimab infusion. Among bebtelovimab-treated patients, the most common comorbidities were hypertension (42.7%), diabetes (17.1%), and an immunosuppressed condition (17.0%). The rates of severe disease, ICU admission, and mortality were 1.3%, 0.5%, and 0.2%, respectively, after bebtelovimab infusion. Older age, immunosuppressed status, and several comorbidities were associated with severe disease progression, while COVID-19 vaccination was associated with lower risk. No anaphylaxis was reported during monoclonal antibody infusion. Conclusions This real-world analysis of a large cohort of high-risk patients demonstrates low rates of severe disease after treatment with sotrovimab during the era dominated by Omicron B.1.1.529 and after treatment with bebtelovimab during the era dominated by BA.2 and Omicron subvariants.

Abstract The effectiveness of bebtelovimab in real-world settings has not been assessed. In this retrospective cohort study of 3607 high-risk patients, bebtelovimab was used more commonly than nirmatrelvir-ritonavir for treatment of coronavirus disease 2019 (COVID-19) among older patients, immunosuppressed patients, and those with multiple comorbid conditions. Despite its use in patients with multiple comorbid conditions, the rate of progression to severe disease after bebtelovimab (1.4% [95% confidence interval, 1.2%–1.7%]) was not significantly different from that for nirmatrelvir-ritonavir treatment (1.2% [.8%–1.5%]). Our findings support the emergency use authorization of bebtelovimab for treatment of COVID-19 during the Omicron epoch dominated by BA.2 and subvariants.