Faculty Bibliography

PURPOSE/UNASSIGNED:This phase Ib open-label, multicenter, platform study (NCT02646748) explored safety, tolerability, and preliminary activity of itacitinib (Janus kinase 1 inhibitor) or parsaclisib (phosphatidylinositol 3-kinase δ inhibitor) in combination with pembrolizumab [programmed death-1 (PD-1) inhibitor]. EXPERIMENTAL DESIGN/UNASSIGNED:Patients with advanced or metastatic solid tumors with disease progression following all available therapies were enrolled and received itacitinib (Part 1 initially 300 mg once daily) or parsaclisib (Part 1 initially 10 mg once daily; Part 2 all patients 0.3 mg once daily) plus pembrolizumab (200 mg every 3 weeks). RESULTS/UNASSIGNED:A total of 159 patients were enrolled in the study and treated with itacitinib (Part 1, n = 49) or parsaclisib (Part 1, n = 83; Part 2, n = 27) plus pembrolizumab. The maximum tolerated/pharmacologically active doses were itacitinib 300 mg once daily and parsaclisib 30 mg once daily. Most common itacitinib treatment-related adverse events (TRAE) were fatigue, nausea, and anemia. Most common parsaclisib TRAEs were fatigue, nausea, diarrhea, and pyrexia in Part 1, and fatigue, maculopapular rash, diarrhea, nausea, and pruritus in Part 2. In patients receiving itacitinib plus pembrolizumab, four (8.2%) achieved a partial response (PR) in Part 1. Among patients receiving parsaclisib plus pembrolizumab, 5 (6.0%) achieved a complete response and 9 (10.8%) a PR in Part 1; 5 of 27 (18.5%) patients in Part 2 achieved a PR. CONCLUSIONS/UNASSIGNED:Although combination of itacitinib or parsaclisib with pembrolizumab showed modest clinical activity in this study, the overall response rates observed did not support continued development in patients with solid tumors. SIGNIFICANCE/UNASSIGNED:PD-1 blockade combined with targeted therapies have demonstrated encouraging preclinical activity. In this phase I study, patients with advanced solid tumors treated with pembrolizumab (PD-1 inhibitor) and either itacitinib (JAK1 inhibitor) or parsaclisib (PI3Kδ inhibitor) experienced limited clinical activity beyond that expected with checkpoint inhibition alone and showed little effect on T-cell infiltration in the tumor. These results do not support continued development of these combinations.

Artificial Intelligence Based on Machine Learning in Pharmacovigilance: A Scoping Review, written by Benjamin Kompa, Joe B. Hakim, Anil Palepu, Kathryn Grace Kompa, Michael Smith, Paul A. Bain, Stephen Woloszynek, Jeffery L. Painter, Andrew Bate, Andrew L. Beam, was originally Online First without Open Access. After publication in volume 45, issue 5, page 477"“491 the author decided to opt for Open Choice and to make the article an Open Access publication. With the author(s)"™ decision to opt for Open Choice the copyright of the article changed on 27 January 2023 to © The Author(s) 2023 and the article is forthwith distributed under a Creative Commons Attribution NonCommercial 4.0 International License, which permits any non-commercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article"™s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article"™s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by- nc/4. 0/". The original article has been corrected.

STUDY DESIGN/METHODS:A retrospective review of operative patients at a single institution. OBJECTIVE:To validate a novel method of detecting pseudarthrosis on dynamic radiographs. SUMMARY OF BACKGROUND DATA/BACKGROUND:A common complication after anterior cervical discectomy and fusion is pseudarthrosis. A previously published method for detecting pseudarthrosis identifies a 1 mm difference in interspinous motion (ISM), which requires calibration of images and relies on anatomic landmarks difficult to visualize. An alternative is to use angles between spinous processes, which does not require calibration and relies on more visible landmarks. METHODS:ISM was measured on dynamic radiographs using the previously published linear method and new angular method. Angles were defined by lines from screw heads to dorsal points of spinous processes. Angular cutoff for fusion was calculated using a regression equation correlating linear and angular measures, based on the 1 mm linear cutoff. Pseudarthrosis was assessed with both cutoffs. Sensitivity, specificity, inter- and intra-reliability of angular and linear measures used post-operative CT as the reference. RESULTS:242 fused levels (81 allograft, 84 PEEK, 40 titanium, 37 standalone cages) were measured in 143 patients (mean age 52.0±11.5, 42%F). 36 patients (66 levels) had 1-year postoperative CTs; 13 patients (13 levels) had confirmed pseudarthrosis. Linear and angular measurements closely correlated (R=0.872), with 2.3° corresponding to 1 mm linear ISM. Potential pseudarthroses was found in 28.0% and 18.5% levels using linear and angular cutoffs, respectively. Linear cutoff had 85% sensitivity, 87% specificity; angular cutoff had 85% sensitivity, 96% specificity for detecting CT-validated pseudarthrosis. Interclass correlation coefficients were 0.974 and 0.986 (both P<0.001); intra-rater reliability averaged 0.953 and 0.974 (P<0.001 for all) for linear and angular methods, respectively. CONCLUSIONS:The angular measure for assessing potential pseudarthrosis is as sensitive as and more specific than published linear methods, has high inter-observer reliability, and can be used without image calibration.

INTRODUCTION:Artificial intelligence based on machine learning has made large advancements in many fields of science and medicine but its impact on pharmacovigilance is yet unclear. OBJECTIVE:The present study conducted a scoping review of the use of artificial intelligence based on machine learning to understand how it is used for pharmacovigilance tasks, characterize differences with other fields, and identify opportunities to improve pharmacovigilance through the use of machine learning. DESIGN:The PubMed, Embase, Web of Science, and IEEE Xplore databases were searched to identify articles pertaining to the use of machine learning in pharmacovigilance published from the year 2000 to September 2021. After manual screening of 7744 abstracts, a total of 393 papers met the inclusion criteria for further analysis. Extraction of key data on study design, data sources, sample size, and machine learning methodology was performed. Studies with the characteristics of good machine learning practice were defined and manual review focused on identifying studies that fulfilled these criteria and results that showed promise. RESULTS:The majority of studies (53%) were focused on detecting safety signals using traditional statistical methods. Of the studies that used more recent machine learning methods, 61% used off-the-shelf techniques with minor modifications. Temporal analysis revealed that newer methods such as deep learning have shown increased use in recent years. We found only 42 studies (10%) that reflect current best practices and trends in machine learning. In the subset of 154 papers that focused on data intake and ingestion, 30 (19%) were found to incorporate the same best practices. CONCLUSION:Advances from artificial intelligence have yet to fully penetrate pharmacovigilance, although recent studies show signs that this may be changing.

BACKGROUND CONTEXT/BACKGROUND:The transpsoas lateral lumbar interbody fusion (LLIF) technique is an effective alternative to traditional anterior and posterior approaches to the lumbar spine; however, nerve injuries are the most reported postoperative complication. Commonly used strategies to avoid nerve injury (eg, limiting retraction duration) have not been effective in detecting or preventing femoral nerve injuries. PURPOSE/OBJECTIVE:To evaluate the efficacy of emerging intraoperative femoral nerve monitoring techniques and the importance of employing prompt surgical countermeasures when degraded femoral nerve function is detected. STUDY DESIGN/SETTING/METHODS:We present the results from a retrospective analysis of a multi-center study conducted over the course of 3 years. PATIENT SAMPLE/METHODS:One hundred and seventy-two lateral lumbar interbody fusion procedures were reviewed. OUTCOME MEASURES/METHODS:Intraoperative femoral nerve monitoring data was correlated to immediate postoperative neurologic examinations. METHODS:Femoral nerve evoked potentials (FNEP) including saphenous nerve somatosensory evoked potentials (snSSEP) and motor evoked potentials with quadriceps recordings were used to detect evidence of degraded femoral nerve function during the time of surgical retraction. RESULTS:In 89% (n=153) of the surgeries, there were no surgeon alerts as the FNEP response amplitudes remained relatively unchanged throughout the surgery (negative group). The positive group included 11% of the cases (n=19) where the surgeon was alerted to a deterioration of the FNEP amplitudes during surgical retraction. Prompt surgical countermeasures to an FNEP alert included loosening, adjusting, or removing surgical retraction, and/or requesting an increase in blood pressure from the anesthesiologist. All the cases where prompt surgical countermeasures were employed resulted in recovery of the degraded FNEP amplitudes and no postoperative femoral nerve injuries. In two cases, the surgeons were given verbal alerts of degraded FNEPs but did not employ prompt surgical countermeasures. In both cases, the degraded FNEP amplitudes did not recover by the time of surgical closure, and both patients exhibited postoperative signs of sensorimotor femoral nerve injury including anterior thigh numbness and weakened knee extension. CONCLUSIONS:Multimodal femoral nerve monitoring can provide surgeons with a timely alert to hyperacute femoral nerve conduction failure, enabling prompt surgical countermeasures to be employed that can mitigate or avoid femoral nerve injury. Our data also suggests that the common strategy of limiting retraction duration may not be effective in preventing iatrogenic femoral nerve injuries.

Advances in imaging technologies, gene editing, and fluorescent molecule development have made real-time imaging of nucleic acids practical. Here, we detail methods for imaging the human telomerase RNA template, hTR via the use of three inserted MS2 stem loops and cognate MS2 coat protein (MCP) tagged with superfolder GFP or photoactivatable GFP. These technologies enable tracking of the dynamics of RNA species through Cajal bodies and offer insight into their residence time in Cajal bodies through photobleaching and photoactivation experiments. For complete details on the use and execution of this protocol, please refer to Laprade et al. (2020).

Aortic carboxypeptidase-like protein (ACLP) is a collagen-binding extracellular matrix (ECM) protein that has important roles in wound healing and fibrosis. ACLP contains thrombospondin repeats, a collagen-binding discoidin domain, and a catalytically inactive metallocarboxypeptidase domain. Recently, mutations in the ACLP-encoding gene, AE-binding protein 1 (AEBP1), have been discovered, leading to the identification of a new variant of Ehlers-Danlos syndrome (EDS) causing connective tissue disruptions in multiple organs. Currently, little is known about the mechanisms of ACLP secretion or the role of posttranslational modifications in these processes. We show here that the secreted form of ACLP contains N-linked glycosylation and that inhibition of glycosylation results in its intracellular retention. Using site-directed mutagenesis, we determined that glycosylation of Asn-471 and Asn-1030 is necessary for ACLP secretion and identified a specific N-terminal proteolytic ACLP fragment. To determine the contribution of secreted ACLP to ECM mechanical properties, we generated and mechanically tested wet-spun collagen ACLP composite fibers, finding that ACLP enhances the modulus (or stiffness), toughness, and tensile strength of the fibers. Some AEBP1 mutations were null alleles, whereas others resulted in expressed proteins. We tested the hypothesis that a recently discovered 40-amino-acid mutation and insertion in the ACLP discoidin domain regulates collagen binding and assembly. Interestingly, we found that this protein variant is retained intracellularly and induces endoplasmic reticulum (ER) stress identified with an XBP1-based ER stress reporter. Our findings highlight the importance of N-linked glycosylation of ACLP for its secretion and contribute to our understanding of ACLP-dependent disease pathologies.

BACKGROUND:Spinal epidural lipomatosis (SEL) is a rare condition defined as pathological overgrowth of the normally present epidural fat within the spinal canal. SEL is associated with Cushing disease, obesity and chronic corticosteroid therapy. Diabetes mellitus type 1 (DM1) has not known to be a risk factor for SEL. The neurological symptoms of SEL are attributed mainly to mechanical compression on the spinal cord and the cauda equina. METHODS:A retrospective chart review of patients evaluated at NYU Multiple Sclerosis Care Center identified three diabetic patients with progressive myelopathy associated with SEL. We report the clinical course, diagnostic workup and outcomes in these three patients with SEL-associated myelopathy. RESULTS:Three patients (2 females and 1 male) had long-standing DM1 and developed progressive myelopathy in their early 40's. All were found to have thoracic SEL (extensive extradural T1, T2 hyperintense signal; biopsy confirmed in one case) with associated extensive abnormal cord signal in lower cervical/upper thoracic spinal cord. A comprehensive evaluation for metabolic, infectious, autoimmune and vascular causes of myelopathy that included serologies, cerebrospinal fluid analyses, and spinal angiography did not reveal an alternative cause for myelopathy. One of the patients underwent a surgical decompression of SEL with subsequent clinical and radiologic improvement. CONCLUSIONS:Our case series suggest that patients with DM1 and myelopathy of unknown cause should be evaluated for SEL. Timely diagnosis and appropriate intervention may forestall progression of neurological disability and even result in neurologic improvement. SEL should be considered on the short list of diagnoses that cause potentially reversible progressive myelopathy.

This paper provides a synopsis of discussions related to biomedical engineering core curricula that occurred at the Fourth BME Education Summit held at Case Western Reserve University in Cleveland, Ohio in May 2019. This summit was organized by the Council of Chairs of Bioengineering and Biomedical Engineering, and participants included over 300 faculty members from 100+ accredited undergraduate programs. This discussion focused on six key questions: QI: Is there a core curriculum, and if so, what are its components? QII: How does our purported core curriculum prepare students for careers, particularly in industry? QIII: How does design distinguish BME/BIOE graduates from other engineers? QIV: What is the state of engineering analysis and systems-level modeling in BME/BIOE curricula? QV: What is the role of data science in BME/BIOE undergraduate education? QVI: What core experimental skills are required for BME/BIOE undergrads? s. Indeed, BME/BIOI core curricula exists and has matured to emphasize interdisciplinary topics such as physiology, instrumentation, mechanics, computer programming, and mathematical modeling. Departments demonstrate their own identities by highlighting discipline-specific sub-specialties. In addition to technical competence, Industry partners most highly value our students' capacity for problem solving and communication. As such, BME/BIOE curricula includes open-ended projects that address unmet patient and clinician needs as primary methods to prepare graduates for careers in industry. Culminating senior design experiences distinguish BME/BIOE graduates through their development of client-centered engineering solutions to healthcare problems. Finally, the overall BME/BIOE curriculum is not stagnant-it is clear that data science will become an ever-important element of our students' training and that new methods to enhance student engagement will be of pedagogical importance as we embark on the next decade.
Copyright