User engagement analysis of visual vs text-based summaries of new medical research in the general medical community (Preprint)

BACKGROUND In the digital age of publication and era of social media, there is increased need for precise and concise reporting of medical manuscripts to educate students, physicians, and other health professionals. Both visual abstracts (VAs) and textual reports (TRs) have been utilized online, but there is a paucity of study about whether VAs really boost user engagement through visual representation of key study findings. Furthermore, prior studies have focused on specialty journals content as opposed to the general medical community. OBJECTIVE The current study was designed to analyze the effectiveness of standardized VAs across the general medical community. METHODS In this prospective case-control study, we used Twitter to publish 18 pairs of VAs and TRs from an open-access, physician-run medical news organization with healthcare professional readership across all medical specialties in order to investigate the effect of VAs on user engagement. Each VA/TR pair, which generally covered new research applicable to the general internist, was published on the same day, covered the same content, and had the same title on Twitter; posts for TRs had a nonspecific control image and link to a text abstract, while VA posts had a partial preview and link to the full VA. The outcomes studied were views, engagement rate, and percent change in engagement rate. RESULTS Results showed that while there was no difference in number of views on Twitter (P=.83), VAs had significantly higher engagement rates (P=.002), with an average fold change of 2.75 (95% CI 1.83 to 3.67). CONCLUSIONS Our results show that Twitter posts containing previews and links to VAs had more user engagement than posts with no visual content and links to TRs, which suggests that VAs are more effective tools for promoting engagement with medical content.

The Impact of Anthelmintic Treatment on Human Gut Microbiota Based on Cross-Sectional and Pre- and Postdeworming Comparisons in Western Kenya

ABSTRACT Murine studies suggest that the presence of some species of intestinal helminths is associated with changes in host microbiota composition and diversity. However, studies in humans have produced varied conclusions, and the impact appears to vary widely depending on the helminth species present. To demonstrate how molecular approaches to the human gut microbiome can provide insights into the complex interplay among disparate organisms, DNA was extracted from cryopreserved stools collected from residents of 5 rural Kenyan villages prior to and 3 weeks and 3 months following albendazole (ALB) therapy. Samples were analyzed by quantitative PCR (qPCR) for the presence of 8 species of intestinal parasites and by MiSeq 16S rRNA gene sequencing. Based on pretreatment results, the presence of neither Ascaris lumbricoides nor Necator americanus infection significantly altered the overall diversity of the microbiota in comparison with age-matched controls. Following ALB therapy and clearance of soil-transmitted helminths (STH), there were significant increases in the proportion of the microbiota made up by Clostridiales (P = 0.0002; average fold change, 0.57) and reductions in the proportion made up by Enterobacteriales (P = 0.0004; average fold change, −0.58). There was a significant posttreatment decrease in Chao1 richness, even among individuals who were uninfected pretreatment, suggesting that antimicrobial effects must be considered in any posttreatment setting. Nevertheless, the helminth-associated changes in Clostridiales and Enterobacteriales suggest that clearance of STH, and of N. americanus in particular, alters the gut microbiota. IMPORTANCE The gut microbiome is an important factor in human health. It is affected by what we eat, what medicines we take, and what infections we acquire. In turn, it affects the way we absorb nutrients and whether we have excessive intestinal inflammation. Intestinal worms may have an important impact on the composition of the gut microbiome. Without a complete understanding of the impact of mass deworming programs on the microbiome, it is impossible to accurately calculate the cost-effectiveness of such public health interventions and to guard against any possible deleterious side effects. Our research examines this question in a “real-world” setting, using a longitudinal cohort, in which individuals with and without worm infections are treated with deworming medication and followed up at both three weeks and three months posttreatment. We quantify the impact of roundworms and hookworms on gut microbial composition, suggesting that the impact is small, but that treatment of hookworm infection results in significant changes. This work points to the need for follow-up studies to further examine the impact of hookworm on the gut microbiota and determine the health consequences of the observed changes.

Abstract 177: Exploratory Results From the B01-02 Trial: Administration of MultiStem® Results in Decreased Circulating CD3+ Cells and Lower Levels of Inflammatory Cytokines

Introduction: B01-02 was a double-blind, placebo-controlled study of ischemic stroke patients (NIHSS 8-20, inclusive) treated within 24-48 hours of symptom onset with MultiStem, an adult, adherent, stem cell product. In pre-clinical models of stroke, MultiStem appears to confer improved neurological benefit through multiple mechanisms, including down-regulation of the peripheral immune response. We sought to determine whether this observation translated to acute ischemic stroke patients treated with MultiStem. Methods: Patients were randomized 1:1 and received IV infusion of 1.2 billion cells or placebo. After randomization, patients had blood drawn to determine baseline levels of circulating levels of CD3+ cells or specific inflammatory cytokines. Blood was drawn again at days 2, 7 and 30 post-treatment to determine the average fold change from baseline in each patient at each time point. Results: 126 patients formed the Intention-To-Treat (ITT) population, 65 receiving MultiStem, 61 placebo. 38 MultiStem patients and 40 placebo patient samples were analyzed for CD3+ cells, while 60 MultiStem patients and 55 placebo patients samples were analyzed for serum cytokine levels. CD3+ cell levels were significantly decreased in the MultiStem treatment group compared to placebo (p=0.001) at Day 2, but not later. IL-6 (p=.031) and IL-12 (p=.035) were both significantly downregulated in the MultiStem treatment group at Day 7. IL-1β (p=.065), TNF-α (p=.068) and IFN-γ (p=.100) also trended towards being significantly reduced at Day 7. Conclusions: IV administration of MultiStem within 24-48 hours of an ischemic stroke results in decreased circulating CD3+ cells, while reducing inflammatory cytokines in the blood during the first 7 days after onset of symptoms. These results support the hypothesis that MultiStem treatment limits the activation of the innate immune system’s response to stroke, which may provide improved neurological and recovery outcomes. Additional studies are required to confirm this observation and optimize the time of administration for MultiStem mediated benefit

Microrna-150 Regulates STAT5b Levels in Juvenile Myelomonocytic Leukemia (JMML)

Introduction JMML is a rare leukemia characterized by aberrant myeloid proliferation and hypersensitivity to GM-CSF. Mutually exclusive mutations in PTPN11, N/K-RAS, CBL, or NF1 are found in ~90% of patients. These mutations cause the disease at least in part by activating STAT5 through phosphorylation and promoting cell growth. MicroRNAs (miRs) are small (~22 nucleotides) noncoding RNAs, regulating gene expression and often deregulated in leukemia. We investigated whether miRs are deregulated in JMML and whether miRs target critical proteins involved in the disease pathogenesis. Patients and Methods MiRs expression profile of 40 bone marrow (BM) samples from untreated JMML patients and 8 BM samples from healthy controls was performed using the Ncounter Human v2 miRNA Expression Assay (nanostring). MiR150-5p and STAT5b mRNA expression were assessed using quantitative (q)RT-PCR. Selected miRs and target expression were measured in BM and spleen from a JMML (PTPN11) murine model (D61Y mutation). We overexpressed miR-150-5p in AML cell lines (K562, OCI-AML-3, KG1a) using a miR-150-5p precursor plasmid and STAT5b mRNA was assessed by qRT-PCR. STAT5 and phospho-STAT5 protein levels were assessed by Western Blot in miR-150-5p transfected AML cell lines, human JMML and mice BM cells. STAT5b 3'UTR sequence was cloned in a Firefly/Renilla Luc construct and co-transfected with miR-150-5p miRNA mimic into 293T cell line. Results We found that 25 miRs were differentially expressed in whole BM cells from JMML patients respect to healthy controls (Table 1). MiR-150-5p was the most downregulated miR in JMML. We focused on miR-150-5p, since it has been described to be downregulated in AML cases and is predicted to target STAT5b, a critical gene in JMML biology. We validated that miR-150-5p was down-regulated in JMML cases respect to controls performing qRT-PCR on 38 BM samples from JMML patients. Likewise, miR-150-5p was downregulated in BM and spleen samples from PTPN11 mutated mice respect to controls (0.35 and 0.27 Average Fold Change decrease respectively). STAT5 protein, a predicted target for miR-150-5p, was highly expressed in the JMML patients and mice BM samples respect to their controls (4.3 and 1.3 Average Fold Change increase respectively). MiR-150-5p overexpression in K562, OCI-AML-3 and KG1a cell lines led to decrease of STAT5 protein levels and phosphorilation at 48 hours. Direct interaction of STAT5b 3'UTR with miR-150-5p was demonstrated by luciferase assay (~50% Luc activity inhibition, P<0.001). Last, overexpression of miR-150-5p in primary JMML samples using a GFP tagged lentivirus significantly decreased cell growth respect to controls (empty vector) in response to GM-CSF (P<0.01). Discussion We showed that miR-150-5p is downregulated in JMML samples and in JMML animal models. Functionally, miR-150-5p directly inhibits the translation of STAT5b mRNA resulting also in a decrease of phosphorylation of STAT5 total protein. These findings identify an alternative mechanism that supports STAT5 deregulation in JMML and that could be therapeutically targeted. Table 1. Deregulated miRNAs in JMML Gene Accession # P value Fold-Change (Log2) miR-150-5p MIMAT0000451 0,001 -2,382 let-7g-5p MIMAT0000414 0,002 -1,660 miR-1260a MIMAT0005911 0,009 -1,592 let-7a-5p MIMAT0000062 0,010 -1,574 miR-4454 MIMAT0018976 0,021 -1,399 miR-148a-3p MIMAT0000243 0,030 -1,211 miR-146b-5p MIMAT0002809 0,009 -1,084 miR-342-3p MIMAT0000753 0,010 -1,075 let-7f-5p MIMAT0000067 0,021 -1,022 miR-26a-5p MIMAT0000082 0,034 -1,008 let-7d-5p MIMAT0000065 0,038 -1,005 miR-30b-5p MIMAT0000420 0,019 -0,972 miR-29b-3p MIMAT0000100 0,044 -0,956 miR-29a-3p MIMAT0000086 0,024 -0,761 miR-338-3p MIMAT0000763 0,042 0,654 miR-23a-3p MIMAT0000078 0,040 0,689 miR-222-3p MIMAT0000279 0,018 0,756 miR-548ai MIMAT0018989 0,009 0,993 miR-494 MIMAT0002816 0,023 1,021 miR-320e MIMAT0015072 0,007 1,175 miR-224-5p MIMAT0000281 0,024 1,227 miR-4508 MIMAT0019045 0,016 1,369 miR-575 MIMAT0003240 0,014 1,429 miR-3195 MIMAT0015079 0,014 1,432 miR-630 MIMAT0003299 0,001 2,272 Figure 1. miR-150-5p Relative Expression for each subset of JMML patients with different mutational profiles and Healthy Controls Figure 1. miR-150-5p Relative Expression for each subset of JMML patients with different mutational profiles and Healthy Controls Disclosures No relevant conflicts of interest to declare.

Ex Vivo Expansion of Donor-Derived Anti-Tumor T Cells for CLL-Specific Adoptive Immunotherapy

Abstract 978 Although chronic lymphocytic leukemia (CLL) is sensitive to graft-versus-leukemia (GVL) effects, strategies to enhance donor-derived tumor immunity are needed to prevent relapse and improve outcomes after allogeneic hematopoietic stem cell transplantation (HSCT). Post-transplant infusion of mature donor T cells specific for recipient CLL cells could provide an effective treatment approach personalized to the individual tumor. However, unmanipulated CLL cells are weak antigen presenting cells (APCs), expressing low levels of costimulatory molecules, and therefore only poorly stimulate the expansion of tumor-reactive T cells. To overcome this barrier, we evaluated a novel formulation of human recombinant CD40L, a molecule known to enhance the immunostimulatory capacity of normal and malignant B cells. This formulation of CD40L (designated CD40L-Tri) was designed with the extracellular domain of CD40L connected by a long flexible linker to a leucine zipper for trimerization and an octahistidine motif for purification. We compared the immunostimulatory activity of CD40L-Tri with a murine fibroblast cell line that was engineered to express human CD40L (tCD40L/NIH3T3). In 3 of 3 cases, CD40L-Tri (at 0.5, 1, and 2 mg/ml) significantly expanded normal CD19+ B cells over 14 days by an average fold change of 21.5, 27.0 and 29.5, respectively (all p<0.05). We further observed that exposure of normal CD19+ B cells to CD40L-Tri (at 2mg/ml, n=3) resulted in significantly increased expression of the costimulatory molecules CD80, CD83 and CD86 at 48 hours by an average fold change of 28.7, 24, and 169.9, respectively(all p<0.05), which was comparable to the average fold change of tCD40L/NIH3T3 (24.7, 21.8, and 144.9, respectively, all p<0.05). In three separate experiments, thymidine incorporation assays revealed that exposure of normal B cells to CD40L-Tri consistently generated APCs with high capacity to stimulate allogeneic CD4+ and CD8+ T cells, at comparable levels to tCD40L/NIH3T3 cells. Consistent with these results, CD40L-Tri-activated B cells could be used to present pulsed peptide to specifically expand T cells specific for the influenza peptide M1 from a normal HLA-A2+ volunteer. Together, these studies confirmed that CD40L-Tri has potent immune-stimulatory effects on B cells, and can be utilized to expand human T cells without the risk of expansion of xenoantigen-specific T cells incurred through the use of conventional tCD40L/NIH3T3 cells. To evaluate the ability of CD40L-Tri to facilitate the expansion of CLL-specific T cells, we examined T cell responses against 4 cryopreserved CLL tumors for which cryopreserved HLA-matched normal donor cells were available. Donor CD8+ T cells were subjected to four weekly in vitro stimulations with CD40L-Tri-activated recipient CLL-B cells in the presence of IL-7, IL-12, and IL-15. In 3 of 4 cases, donor CD8+ T cells with specificity against recipient tumor were expanded. These CTL were not reactive with recipient-derived fibroblasts or PHA blasts, suggesting that T cell reactivity was tumor- rather than allo-specific. CD8+ CLL-reactive T cells secreted IFNg by Elispot and were cytotoxic to recipient CLL cells. Reactivity was blocked in the presence of MHC class I blocking antibody (W6/32). Cloning of CLL-reactive T cells from 2 patients resulted in the isolation of up to 8 tumor-reactive T cell clones each, that were confirmed to have tumor-specific and MHC class I-restricted reactivity. These T cell clones were also reactive against other CLL cells that shared at least one HLA molecule with original donor/recipient, suggesting that these CTL clones recognize common CLL antigens. Our results demonstrate that the combination of CD40L-Tri-activation of CLL cells and supportive cytokines can reliably generate CLL-specific T cells from HLA-matched donors. Our ongoing studies focus on the identification of CLL antigens recognized by these T cells, which may serve to identify useful immunogens for CLL immunotherapy. These studies suggest a potentially effective strategy for adoptive T cell therapy to enhance GVL after allogeneic HSCT in patients with CLL and possibly other mature B cell malignancies. Disclosures: No relevant conflicts of interest to declare.

Meta-Analysis and Meta-Review of Thyroid Cancer Gene Expression Profiling Studies Identifies Important Diagnostic Biomarkers

Purpose An estimated 4% to 7% of the population will develop a clinically significant thyroid nodule during their lifetime. In many cases, preoperative diagnoses by needle biopsy are inconclusive. Thus, there is a clear need for improved diagnostic tests to distinguish malignant from benign thyroid tumors. The recent development of high-throughput molecular analytic techniques should allow the rapid evaluation of new diagnostic markers. However, researchers are faced with an overwhelming number of potential markers from numerous thyroid cancer expression profiling studies. Materials and Methods To address this challenge, we have carried out a comprehensive meta-review of thyroid cancer biomarkers from 21 published studies. A gene ranking system that considers the number of comparisons in agreement, total number of samples, average fold-change and direction of change was devised. Results We have observed that genes are consistently reported by multiple studies at a highly significant rate (P < .05). Comparison with a meta-analysis of studies reprocessed from raw data showed strong concordance with our method. Conclusion Our approach represents a useful method for identifying consistent gene expression markers when raw data are unavailable. A review of the top 12 candidates revealed well known thyroid cancer markers such as MET, TFF3, SERPINA1, TIMP1, FN1, and TPO as well as relatively novel or uncharacterized genes such as TGFA, QPCT, CRABP1, FCGBP, EPS8 and PROS1. These candidates should help to develop a panel of markers with sufficient sensitivity and specificity for the diagnosis of thyroid tumors in a clinical setting.