SURVIVAL ANALYSIS OF PEDIATRIC WILMS TUMOR BASED ON RISK STRATIFICATION

Objective: This study aims to determine prognostic factors of WT patient in Adam Malik Hospital, Medan. Material & Methods: at Adam Malik Hospital, Medan. Univariate and multivariate Cox regression analyses were performed to determine independent prognostic factors for WT. The primary endpoint of this study were patients’ overall survival (OS) obtained by performing Kaplan-Meier analysis on significant variables. Results: From the univariate Cox regression analysis, gender was found to be the sole significant factor (HR = 0.218, p = 0.005) with males have a higher hazard ratio. The multivariate Cox regression analysis yielded age of diagnosis (HR = 13.860, p = 0.014) and incomplete tumor removals (HR = 0.056, p = 0.008). Kaplan-Meier analysis were performed on three significant variables mentioned before. Only gender yielded a significant Mantel-Cox log-rank score (p = 0.002) with male patients were found to have better survivability (which median survival 476 days compared to females’ 11 days). The survival of the boys was 45.45% while all of the girls did not survive until the cut-off. Conclusion: Three prognostic factors, including children’s gender, age of diagnosis, and tumor removal status, were confirmed to be prognostic factors for the overall survival of children with WT. Further studies covering broader demographic areas were suggested to confirm significant results.

Predictors of colorectal cancer survival using cox regression and random survival forests models based on gene expression data

Understanding and identifying the markers and clinical information that are associated with colorectal cancer (CRC) patient survival is needed for early detection and diagnosis. In this work, we aimed to build a simple model using Cox proportional hazards (PH) and random survival forest (RSF) and find a robust signature for predicting CRC overall survival. We used stepwise regression to develop Cox PH model to analyse 54 common differentially expressed genes from three mutations. RSF is applied using log-rank and log-rank-score based on 5000 survival trees, and therefore, variables important obtained to find the genes that are most influential for CRC survival. We compared the predictive performance of the Cox PH model and RSF for early CRC detection and diagnosis. The results indicate that SLC9A8, IER5, ARSJ, ANKRD27, and PIPOX genes were significantly associated with the CRC overall survival. In addition, age, sex, and stages are also affecting the CRC overall survival. The RSF model using log-rank is better than log-rank-score, while log-rank-score needed more trees to stabilize. Overall, the imputation of missing values enhanced the model’s predictive performance. In addition, Cox PH predictive performance was better than RSF.

Abstract 13607: Correlation in Hospital Survival Following Out-of-Hospital Cardiac Arrest With Variation in Post-Arrest Inpatient Intervention Rates in Michigan

Introduction: Wide variations in rates of survival to hospital discharge exist for survivors of out-of-hospital cardiac arrest (OHCA). The potential influence of variation in post-OHCA hospital care has not been adequately explored. We hypothesized that variation of in hospital survival rates may be influenced by variation of in-hospital care in Michigan. Methods: We performed a secondary analysis of a statewide cardiac arrest database constructed from two probabilistically-linked cardiac arrest registries [Cardiac Arrest Registry to Enhance Survival (CARES) and Michigan Inpatient Database (MIDB)] from 2014 - 2017. A novel composite rank score was created to characterize post-arrest in-hospital care, incorporating four specific interventions: left heart catheterization within 24 hours (LHC), emergent mechanical circulatory support (EMCS), targeted temperature management (TTM), and do-not-resuscitate order placed within 72 hours of arrival (DNR). The highest score (1 of 38) was given to the hospital with highest procedure rate (LHC, TTM, LHC) and the lowest rate of early DNR. Spearman’s correlation coefficients assessed the relationship between the equal weight composite rank score and rate of hospital survivors. Results: We included 3,644 patients admitted to 38 hospitals who treated >30 OHCA patients during the study period. Patient mean age was 62.4 years, and 59.3% were male. Survival, rank scores and correlation coefficients are listed below: We observed four-fold variation in survival for all patients and witnessed arrest, with a non-significant correlation with care provision. However, we identified a sixteen-fold variation in survival among unwitnessed arrests, which was significantly correlated with a higher rank of care provided. Conclusions: In Michigan, the greatest variation in survival was identified among unwitnessed arrests. This variation was robustly associated with a composite rank of in-hospital post-arrest interventions.

The Clonotypic BCR IG of CLL Patients Contain Predicted T-Cell Class I Epitopes with Shared Structural Properties

Targeted therapies have revolutionized the treatment of chronic lymphocytic leukemia (CLL) with remarkable overall response rates. Against that, however, CLL remains incurable, indicating a need for novel strategies towards disease control and eventual eradication, including reinvigoration of anti-tumor immune responses. T cells in CLL display an oligoclonal profile and appear selected by restricted antigens, with recent evidence suggesting that the selecting epitopes may lie within the clonotypic B-cell receptor immunoglobulins (BcR IG). Should this prove to be the case, such neoepitopes could be exploited as idiotypic targets for cellular therapy or for peptide vaccine design, aiming to augment response to current treatments. Here we performed ad hoc prediction of putative T-cell class I neoepitopes contained within the clonotypic BcR IGs of CLL patients, with an intended bias towards major stereotyped CLL subsets. We selected 27 patients to represent the full spectrum of CLL: (i) with mutated IGHV genes (M-CLL, n=5), (ii) with unmutated IGHV genes (U-CLL, n=5), (iii) assigned to major stereotyped subsets (subset #1, n=7; subset #2, n=5; subset #4, n=5). RT-PCR was performed for the heavy (H) and the light (K/L) IG chains using subgroup-specific Leader primers for the IGHV/IGK/LV gene and universal primers annealing to the constant domain (IGHG, IGHM, IGKC, IGLC), in order to produce the full-length V-(D)-J gene rearrangement sequence, plus the start of the constant domain. PCR products were subjected to direct double-strand Sanger sequencing with a quality-optimized protocol. The amino acid sequences were subsequently parsed in peptides and subjected to NetMHCpan. The rank score was calculated, considering the 4-digit HLA-A and -B typing for each individual patient. High- and medium-binding peptides (rank score <2%) were selected. Exact matches to germline and/or proteome databases were excluded. Overall, 1,007 predicted neoepitopes were identified. All patients had predicted CD8 + T-cell epitopes within the clonotypic BcR IG, either in the heavy chain (26/27 pts, n=632 epitopes) or the light chain (26/27 pts, n=375 epitopes). The majority of the peptides resulted from somatic hypermutations (SHMs) across the IGHV gene outside the complementarity-determining region 3 (CDR3; n=538, 53.4%). With the exception of few peptides located within the FR4 region (n=11, 0.1%), the remaining (n=458, 45.5%) involved (at least part of) the CDR3, which is particularly relevant given its small length (9-27 aa) within the full sequence (331-660 aa). There was no statistically significant difference in the rank score of peptides involving the CDR3 vs. all others. Peptide clustering assigned most of the predicted neoepitopes (970/1007, 96.3%) in 54 clusters of similar length and amino acid composition. Also, it revealed similar or identical predicted neoepitopes among different patients (30 clusters of two, 10 clusters of three, 8 clusters of four, 4 clusters of five, 1 cluster of six and 1 cluster of eight). Importantly, these clusters involved: (i) shared CDR3 patterns in patients assigned to the same stereotyped subset, but also (ii) subset-specific recurrent SHMs across the rearranged IGHV gene, e.g. G-to-E SHM at position 28 in the VH CDR1 of subset #4, a recurrent SHM in this subset. Also of note, the two most highly populated clusters involved peptides within the VL CDR3, and were biased towards specific subsets; the cluster of eight patients contained 4 patients assigned to subset #1 and the cluster of six patients included 4 patients assigned to subset #2. In conclusion, in silico prediction identified a significant number of putative T-cell class I neoepitopes contained within the clonotypic BcR IG of CLL patients. The majority of these neoepitopes can be assigned to clusters based on amino acid similarity and are shared among different patients. Many of them culminate from subset-specific ('stereotyped') CDR3 patterns or recurrent SHMs, suggesting that the targeted SHM which shapes the CLL BcR IG repertoire may produce immunogenic CD8 + T-cell epitopes. Their actual immunogenicity has to be tested in ex vivo studies, currently underway by our group. Disclosures Anagnostopoulos: Abbvie: Other: clinical trials; Sanofi: Other: clinical trials ; Ocopeptides: Other: clinical trials ; GSK: Other: clinical trials; Incyte: Other: clinical trials ; Takeda: Other: clinical trials ; Amgen: Other: clinical trials ; Janssen: Other: clinical trials; novartis: Other: clinical trials; Celgene: Other: clinical trials; Roche: Other: clinical trials; Astellas: Other: clinical trials . Chatzidimitriou: Abbvie: Honoraria, Research Funding; Janssen: Honoraria, Research Funding.

Structure-Based Virtual Screening, Molecular Docking, MolecularDynamics Simulation of VEGF Inhibitors for the Clinical Treatment of Ovarian Cancer

VEGF and its receptor play an important role both in physiologic and pathologic angiogenesis, which is identified in ovarian cancer progression and metastasis development. The aim of the present investigation is to identify a potential VEGF inhibitor which is playing a crucial role in stimulating the immunosuppressive microenvironment in tumour cells of ovary and to examine for an effectiveness of identified inhibitor for treatment of ovarian cancer using various In silico approaches. 12 established VEGF inhibitors were collected from various literature. The compound AEE788 displays the great affinity towards the target protein as a result of docking study. AEE78 was further used for structure base virtual screening in order to obtain more structurally similar compound with high affinity. Among the 80 Virtual screened compounds, CID 88265020, explicates much better affinity than established compound AEE788. Based on Molecular Dynamics Simulation, pharmacophore and comparative toxicity analysis of both the best established compound and the best virtual screened compound displayed a trivial variation in associated properties. The virtual screened compound CID 88265020 have the high affinity with the lowest re-rank score, and holds a huge potential to inhibit the VGFR and can be implemented for prospective of future investigations in Ovarian Cancer.

ABCDE: Approximating Betweenness-Centrality ranking with progressive-DropEdge

Betweenness-centrality is a popular measure in network analysis that aims to describe the importance of nodes in a graph. It accounts for the fraction of shortest paths passing through that node and is a key measure in many applications including community detection and network dismantling. The computation of betweenness-centrality for each node in a graph requires an excessive amount of computing power, especially for large graphs. On the other hand, in many applications, the main interest lies in finding the top-k most important nodes in the graph. Therefore, several approximation algorithms were proposed to solve the problem faster. Some recent approaches propose to use shallow graph convolutional networks to approximate the top-k nodes with the highest betweenness-centrality scores. This work presents a deep graph convolutional neural network that outputs a rank score for each node in a given graph. With careful optimization and regularization tricks, including an extended version of DropEdge which is named Progressive-DropEdge, the system achieves better results than the current approaches. Experiments on both real-world and synthetic datasets show that the presented algorithm is an order of magnitude faster in inference and requires several times fewer resources and time to train.

Powerful and robust non-parametric association testing for microbiome data via a zero-inflated quantile approach (ZINQ)

Background Identification of bacterial taxa associated with diseases, exposures, and other variables of interest offers a more comprehensive understanding of the role of microbes in many conditions. However, despite considerable research in statistical methods for association testing with microbiome data, approaches that are generally applicable remain elusive. Classical tests often do not accommodate the realities of microbiome data, leading to power loss. Approaches tailored for microbiome data depend highly upon the normalization strategies used to handle differential read depth and other data characteristics, and they often have unacceptably high false positive rates, generally due to unsatisfied distributional assumptions. On the other hand, many non-parametric tests suffer from loss of power and may also present difficulties in adjusting for potential covariates. Most extant approaches also fail in the presence of heterogeneous effects. The field needs new non-parametric approaches that are tailored to microbiome data, robust to distributional assumptions, and powerful under heterogeneous effects, while permitting adjustment for covariates. Methods As an alternative to existing approaches, we propose a zero-inflated quantile approach (ZINQ), which uses a two-part quantile regression model to accommodate the zero inflation in microbiome data. For a given taxon, ZINQ consists of a valid test in logistic regression to model the zero counts, followed by a series of quantile rank-score based tests on multiple quantiles of the non-zero part with adjustment for the zero inflation. As a regression and quantile-based approach, the method is non-parametric and robust to irregular distributions, while providing an allowance for covariate adjustment. Since no distributional assumptions are made, ZINQ can be applied to data that has been processed under any normalization strategy. Results Thorough simulations based on real data across a range of scenarios and application to real data sets show that ZINQ often has equivalent or higher power compared to existing tests even as it offers better control of false positives. Conclusions We present ZINQ, a quantile-based association test between microbiota and dichotomous or quantitative clinical variables, providing a powerful and robust alternative for the current microbiome differential abundance analysis.

Pharmacokinetic predictions and docking studies of substituted aryl amine-based triazolopyrimidine designed inhibitors of Plasmodium falciparum dihydroorotate dehydrogenase (PfDHODH)

Background The sixteen (16) designed data set of substituted aryl amine-based triazolopyrimidine were docked against Plasmodium falciparum dihydroorotate dehydrogenase (PfDHODH) employing Molegro Virtual Docker (MVD) software and their pharmacokinetic property determined through SwissADME predictor. Results The docking studies shows compound D16, 5-((6-methoxy-5-methyl-[1,2,4]triazolo[1,5-a]pyrimidin-7-yl)amino)benzo[b]thiophen-4-ol to be the most interactive and stable derivative (re-rank score = − 114.205 kcal/mol) resulting from the hydrophobic as well as hydrogen interactions. The hydrogen interaction produced one hydrogen bond with the active residues LEU359 (H∙∙H∙∙O) at a bond distances of 2.2874 Å. All the designed derivatives were found to pass the Lipinski rule of five tests, supporting the drug-likeliness of the designed compounds. Conclusion The ADME analysis revealed a perfect concurrence with the Lipinski Ro5, where the derivatives were found to possess good pharmacokinetic properties such as molar refractivity (MR), number of rotatable bonds (nRotb), log of skin permeability (log Kp), blood-brain barrier (BBB). These results could a deciding factor for the optimization of novel antimalarial compounds.

Web service composition ranking based on QoS and social network analysis

Web service composition refers to the aggregation of web services for producing composite solutions in order to satisfy user requirements which can't be satisfied by atomic services. It is an essential challenge to find the most reliable and trustable complex services in each composition process. Many of current composition approaches use QoS (Quality of Service) values to select among different composition candidates. However, QoS values can't be trusted all the time since some service providers may promote their services by publishing wrong QoS values. Social network analysis techniques such as PageRank have been used successfully in finding the trustable and authoritative web resources. We believe that these techniques can also be used to improve the service composition process. We have developed a modified PageRank algorithm called Service Rank in order to find the importance level of each service in a composition based on its connectivity and invocation history. This can be accomplished by assigning higher weights to links which have more number of invocations, more up-to-date invocation time and contract signing time, and longer contract durations. Eventually the Service Rank score will be combined with the QoS score for composition ranking. Preliminary results from our experiments have proved the effectiveness of this method. As a consequence users can be more satisfied with the service composition result.