Search

PURPOSE: Current guidelines recommend molecular genotyping for patients newly diagnosed with metastatic nonsquamous (mNSq) non–small-cell lung cancer (NSCLC). The association between availability of molecular genotyping before first line (1L) therapy and overall survival (OS) is not known. METHODS: We conducted a real-world cohort study using electronic health records in patients newly diagnosed with mNSq NSCLC. Cox proportional-hazards multivariable regression models were constructed to examine the association between OS and test result availability before 1L therapy, adjusting for covariates. Additional analyses were conducted to assess the consistency and strength of the relationship. Multivariable logistic regression models were used to examine the association between concurrent tissue and plasma testing (v tissue alone) and result availability. RESULTS: Three hundred twenty-six patients were included, 80% (261/326) with results available before 1L (available testing group), and 20% (65/326) without results available (unavailable testing group). With 14.2-month median follow-up, patients in the available testing group had significantly longer OS relative to the unavailable testing group (adjusted hazard ratio, 0.43; 95% CI, 0.30 to 0.62; P <.0001). The adjusted odds of availability of results before 1L therapy was higher with concurrent tissue and plasma testing (v tissue testing alone; adjusted odds ratio, 2.06; 95% CI, 1.09 to 3.90; P =.026). CONCLUSION: Among patients with mNSq NSCLC in a real-world cohort, availability of molecular genotyping results before 1L therapy was associated with significantly better OS. Concurrent tissue and plasma testing was associated with a higher odds of availability of results before 1L therapy. These findings warrant renewed attention to the completion of molecular genotyping before 1L therapy. Availability of NGS results before 1L therapy for patients with mNSCLC is associated with better overall survival. [ABSTRACT FROM AUTHOR]

Simple Summary: ARID1A abnormalities disturb gene coding processes and correlate with immunotherapy responsiveness. We report the first blood sample-based genomic sequencing of ARID1A in DNA shed from tumors into the circulation (known as cell-free DNA (cfDNA) from liquid biopsy). Altogether, of 62,851 cancer patients with ≥1 cfDNA alteration in the blood, 3137 (5%) had ≥1 deleterious ARID1A alteration (a frequency similar to the ~6% generally reported in tissue sequencing), suggesting this non-invasive test's value in detecting ARID1A. ARID1A alterations were most frequent in endometrial (21.3% of patients) and bladder cancer (12.9% of patients). As compared to blood samples without ARID1A aberrations, those with a functional (deleterious) ARID1A abnormality had more DNA alterations/sample (median, 6 versus 4; p < 0.0001) and more frequent co-alterations in one or more genes in key pathways promoting cancer development/progression, which may inform therapeutic strategies.ARID1A abnormalities disturb transcriptional processes regulated by chromatin remodeling and correlate with immunotherapy responsiveness. We report the first blood-based cell-free DNA (cfDNA) next-generation sequencing (NGS) ARID1A analysis. From November 2016 through August 2019, 71,301 patients with advanced solid tumors underwent clinical blood-derived cfDNA testing. Of these patients, 62,851 (88%) had ≥1 cfDNA alteration, and 3137 (of the 62,851) (5%) had ≥1 deleterious ARID1A alteration (a frequency similar to the ~6% generally reported in tissue NGS), suggesting this non-invasive test's value in interrogating ARID1A. ARID1A cfDNA alterations were most frequent in endometrial cancer, 21.3% of patients; bladder cancer, 12.9%; gastric cancer, 11%; cholangiocarcinoma, 10.9%; and hepatocellular carcinoma, 10.6%. Blood samples with a functional ARID1A abnormality had more alterations/sample (median, 6 versus 4; p < 0.0001) and more frequent co-alterations in ≥1 gene in key oncogenic pathways: signal transduction, RAS/RAF/MAPK, PI3K/Akt/mTor, and the cell cycle. Taken together, our data suggest that liquid (blood) biopsies identify ARID1A alterations at a frequency similar to that found in primary tumor material. Furthermore, co-alterations in key pathways, some of which are pharmacologically tractable, occurred more frequently in samples with functional (deleterious) ARID1A alterations than in those without such aberrations, which may inform therapeutic strategies. [ABSTRACT FROM AUTHOR]

Short text is ambiguous and often relies predominantly on the domain and context at hand in order to attain semantic relevance. Existing classification models perform poorly on short text due to data sparsity and inadequate context. Auxiliary context, which can often provide sufficient background regarding the domain, is typically available in several application scenarios. While some of the existing works aim to leverage real-world knowledge to enhance short-text representations, they fail to place appropriate emphasis on the auxiliary context. Such models do not harness the full potential of the available context in auxiliary sources. To address this challenge, we reformulate short-text classification as a dual channel self-supervised learning problem (that leverages auxiliary context) with a generation network and a corresponding prediction model. We propose a self-supervised framework, Pseudo-Auxiliary Context generation network for Short-text Modeling (PACS), to comprehensively leverage auxiliary context and it is jointly learned with a prediction network in an end-to-end manner. Our PACS model consists of two sub-networks: a Context Generation Network (CGN) that models the auxiliary context's distribution and a Prediction Network (PN) to map the short-text features and auxiliary context distribution to the final class label. Our experimental results on diverse datasets demonstrate that PACS outperforms formidable state-of-the-art baselines. We also demonstrate the performance of our model on cold-start scenarios (where contextual information is non-existent) during prediction. Furthermore, we perform interpretability and ablation studies to analyze various representational features captured by our model and the individual contribution of its modules to the overall performance of PACS, respectively. [ABSTRACT FROM AUTHOR]

Plasmablasts represent a specialized class of antibody-secreting effector B cells that transiently appear in blood circulation following infection or vaccination. The expansion of these cells generally tends to be massive in patients with systemic infections such as dengue or Ebola that cause hemorrhagic fever. To gain a detailed understanding of human plasmablast responses beyond antibody expression, here, we performed immunophenotyping and RNA sequencing (RNA-seq) analysis of the plasmablasts from dengue febrile children in India. We found that plasmablasts expressed several adhesion molecules and chemokines or chemokine receptors that are involved in endothelial interactions or homing to inflamed tissues, including skin, mucosa, and intestine, and upregulated the expression of several cytokine genes that are involved in leukocyte extravasation and angiogenesis. These plasmablasts also upregulated the expression of receptors for several B-cell prosurvival cytokines that are known to be induced robustly in systemic viral infections such as dengue, some of which generally tend to be relatively higher in patients manifesting hemorrhage and/or shock than in patients with mild febrile infection. These findings improve our understanding of human plasmablast responses during the acute febrile phase of systemic dengue infection. IMPORTANCE Dengue is globally spreading, with over 100 million clinical cases annually, with symptoms ranging from mild self-limiting febrile illness to more severe and sometimes life-threatening dengue hemorrhagic fever or shock, especially among children. The pathophysiology of dengue is complex and remains poorly understood despite many advances indicating a key role for antibody-dependent enhancement of infection. While serum antibodies have been extensively studied, the characteristics of the early cellular factories responsible for antibody production, i.e., plasmablasts, are only beginning to emerge. This study provides a comprehensive understanding of the transcriptional profiles of human plasmablasts from dengue patients. [ABSTRACT FROM AUTHOR]

• Routine molecular testing in stage I–III NSCLC has not been required to date. • However, targeted agents are undergoing trials in adjuvant and neoadjuvant settings. • Adjuvant osimertinib was recently approved for EGFR m (ex19del/L858R) resected NSCLC. • Molecular testing in stage I–III NSCLC will be necessary in the near future. • We discuss opportunities/challenges of integrating molecular testing into practice. Precision medicine in non-small cell lung cancer (NSCLC) is a rapidly evolving area, with the development of targeted therapies for advanced disease and concomitant molecular testing to inform clinical decision-making. In contrast, routine molecular testing in stage I–III disease has not been required, where standard of care comprises surgery with or without adjuvant or neoadjuvant chemotherapy, or concurrent chemoradiotherapy for unresectable stage III disease, without the integration of targeted therapy. However, the phase 3 ADAURA trial has recently shown that the epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI), osimertinib, reduces the risk of disease recurrence by 80% versus placebo in the adjuvant setting for patients with stage IB–IIIA EGFR mutation-positive NSCLC following complete tumor resection with or without adjuvant chemotherapy, according to physician and patient choice. Treatment with adjuvant osimertinib requires selection of patients based on the presence of an EGFR-TKI sensitizing mutation. Other targeted agents are currently being evaluated in the adjuvant and neoadjuvant settings. Approval of at least some of these other agents is highly likely in the coming years, bringing with it in parallel, a requirement for comprehensive molecular testing for stage I–III disease. In this review, we consider the implications of integrating molecular testing into practice when managing patients with stage I–III non-squamous NSCLC. We discuss best practices, approaches and challenges from pathology, surgical and oncology perspectives. [ABSTRACT FROM AUTHOR]

Background: To ensure safe delivery of oncologic care during the COVID-19 pandemic, telemedicine has been rapidly adopted. However, little data exist on the impact of telemedicine on quality and accessibility of oncologic care. This study assessed whether conducting an office visit for thoracic oncology patients via telemedicine affected time to treatment initiation and accessibility.Methods: This was a retrospective cohort study of patients with thoracic malignancies seen by a multidisciplinary team during the first surge of COVID-19 cases in Philadelphia (March 1 to June 30, 2020). Patients with an index visit for a new phase of care, defined as a new diagnosis, local recurrence, or newly discovered metastatic disease, were included.Results: 240 distinct patients with thoracic malignancies were seen: 132 patients (55.0%) were seen initially in-person vs 108 (45.0%) via telemedicine. The majority of visits were for a diagnosis of a new thoracic cancer (87.5%). Among newly diagnosed patients referred to the thoracic oncology team, the median time from referral to initial visit was significantly shorter amongst the patients seen via telemedicine vs. in-person (median 5.0 vs. 6.5 days, p < 0.001). Patients received surgery (32.5%), radiation (24.2%), or systemic therapy (30.4%). Time from initial visit to treatment initiation by modality did not differ by telemedicine vs in-person: surgery (22 vs 16 days, p = 0.47), radiation (27.5 vs 27.5 days, p = 0.86, systemic therapy (15 vs 13 days, p = 0.45).Conclusions: Rapid adoption of telemedicine allowed timely delivery of oncologic care during the initial surge of the COVID19 pandemic by a thoracic oncology multi-disciplinary clinic. [ABSTRACT FROM AUTHOR]

Network alignment aims at inferring a set of anchor links matching the shared entities between different information networks, which has become a prerequisite step for effective fusion of multiple information networks. In this paper, we will study the network alignment problem to fuse online social networks specifically. Social network alignment is extremely challenging to address due to several reasons, i.e., lack of training data, network heterogeneity and one-to-one constraint. Existing network alignment works usually require a large number of training instances, but such a demand can hardly be met in applications, as manual anchor link labeling is extremely expensive. Significantly different from other homogeneous network alignment works, information in online social networks is usually of heterogeneous categories, the incorporation of which in model building is not an easy task. Furthermore, the one-to-one cardinality constraint on anchor links renders their inference process intertwistingly correlated. To resolve these three challenges, a novel network alignment model, namely ActiveIter (Active Iterative Alignment), is introduced in this paper. The model ActiveIter defines a set of inter-network meta diagrams for anchor link feature extraction, adopts active learning for effective label query and uses greedy link selection for anchor link cardinality filtering. Extensive experiments were performed on a real-world aligned networks dataset, and the experimental results have demonstrated the effectiveness of ActiveIter compared with other state-of-the-art baseline methods. [ABSTRACT FROM AUTHOR]

The 2 articles by Youn et al and Weiss et al in this issue of Cancer confirm that there is no longer room for therapeutic nihilism in the management of older adults with non–small cell lung cancer. Fit elderly patients with preserved Eastern Cooperative Oncology Group performance status should be treated with standard‐of‐care therapies similar to their younger counterparts. [ABSTRACT FROM AUTHOR]

Neural networks have become very popular in recent years, because of the astonishing success of deep learning in various domains such as image and speech recognition. In many of these domains, specific architectures of neural networks, such as convolutional networks, seem to fit the particular structure of the problem domain very well and can therefore perform in an astonishingly effective way. However, the success of neural networks is not universal across all domains. Indeed, for learning problems without any special structure, or in cases where the data are somewhat limited, neural networks are known not to perform well with respect to traditional machine-learning methods such as random forests. In this article, we show that a carefully designed neural network with random forest structure can have better generalization ability. In fact, this architecture is more powerful than random forests, because the back-propagation algorithm reduces to a more powerful and generalized way of constructing a decision tree. Furthermore, the approach is efficient to train and requires a small constant factor of the number of training examples. This efficiency allows the training of multiple neural networks to improve the generalization accuracy. Experimental results on real-world benchmark datasets demonstrate the effectiveness of the proposed enhancements for classification and regression. [ABSTRACT FROM AUTHOR]

Purpose: TP53 mutation (MT) in epidermal growth factor receptor (EGFR) -MT non–small cell lung cancer (NSCLC) is associated with poor response to targeted therapy; however, its impact on survival is not clearly established. Patients and Methods: We performed an analysis of patients with stage IV EGFR MT NSCLC with available gene sequencing data. Associations between baseline characteristics; molecular profile, including TP53 MT; and survival outcomes were assessed. Results: We identified 131 consecutive patients with EGFR MT; 81 (62%) had a TP53 MT, and 55 (42%) had other coexisting oncogenic MTs. Emergent EGFR T790M MT was observed in 42 patients (32%). Overall survival (OS) was longer for younger patients (P =.003), never smokers (P =.002), those with Eastern Cooperative Oncology Group performance status 0 to 1 (P =.004), and emergent T790M MT (P =.018). TP53 MT (P =.021) and other coexisting oncogenic MTs (P = 0.011) were associated with inferior OS. In a multivariable regression analysis adjusted for age, smoking, Eastern Cooperative Oncology Group performance status, and the presence of TP53 MT (P =.063) and other coexisting MTs (P =.064) did not achieve statistical significance. Patients with EGFR T790M /TP53 double MT had worse OS compared with patients with T790M MT alone (46.4 months v 82.9 months). In our series, five patients transformed to small-cell lung cancer (5.6%). All had TP53 MT. In four patients, allelic fraction of TP53 MT increased at the time of transformation. Conclusion: The presence of TP53 and other coexisting MTs in EGFR MT NSCLC were associated with inferior OS, including patients with emergent T790M MT. An increase in TP53 mutation allelic fraction may potentially be a useful clinical predictor of small-cell transformation. [ABSTRACT FROM AUTHOR]

Background There are no biomarkers for assessment of disease burden or activity of therapy in SCLC. Patients and methods We conducted a prospective study enumerating serial CTCs in patients with newly diagnosed limited disease (LD) and extensive stage (ED) SCLC. CTCs demonstrating DNA damage and apoptosis based on γH2AX and M30 staining were also assessed. We correlated CTC number with disease stage, survival outcomes and tumor burden by RECIST. Results Between 03/2011-10/2013, 50 evaluable patients were enrolled (20 LD). Baseline CTC number was higher for ED (median CTC 71 vs. 1.5 for LD; p 0.0004 ). Patients with <5 CTC had longer PFS but not OS (11 vs. 6.7 months, p 0.0259 and 15.5 vs. 12.9 months, p 0.4357 ). A higher cutoff (CTC < 50 or CTC ≥ 50) was significantly correlated with both OS (20.2 vs. 11.8 months, p 0.0116 ) and PFS (10 vs. 4.8 months, p 0.0002 ). Patients with <5 CTC on day 1 of cycle 2 had longer PFS (10 vs. 3.17 months, p < 0.001) and OS (18 vs. 9 months, p 0.0001 ). Patients with an increase in γ2HAX-positive CTCs after chemotherapy had longer OS compared to patients without an increase (25.3 vs. 9 months, p 0.15 ). Conclusions This study demonstrates that CTCs at baseline and Cycle 2 of chemotherapy correlate with disease stage and survival in patients with SCLC, suggesting that CTCs may be used as a surrogate biomarker for clinical response. Confirmatory prospective clinical trials are needed before we can incorporate routine evaluation of CTCs into clinical practice. [ABSTRACT FROM AUTHOR]

Link prediction is one of the most fundamental problems in graph modeling and mining. It has been studied in a wide range of scenarios, from uncovering missing links between different entities in databases, to recommending relations between people in social networks. In this problem, we wish to predict unseen links in a growing target network by exploiting existing structures in source networks. Most of the existing methods often assume that abundant links are available in the target network to build a model for link prediction. However, in many scenarios, the target network may be too sparse to enable robust inference process, which makes link prediction challenging with the paucity of link data. On the other hand, in many cases, other (more densely linked) auxiliary networks can be available that contains similar link structure relevant to that in the target network. The linkage information in the existing networks can be used in conjunction with the node attribute information in both networks in order to make more accurate link recommendations. Thus, this paper proposes the use of learning methods to perform link inference by transferring the link information from the source network to the target network. We also note that the source network may contain the link information irrelevant to the target network. This leads to cross-network bias between the networks, which makes the link model built upon the source network misaligned with the link structure of the target network. Therefore, we re-sample the source network to rectify such cross-network bias by maximizing the cross-network relevance measured by the node attributes, as well as preserving as rich link information as possible to avoid the loss of source link structure caused by the re-sampling algorithm. The link model based on the re-sampled source network can make more accurate link predictions on the target network with aligned link structures across the networks. We present experimental results illustrating the effectiveness of the approach. [ABSTRACT FROM PUBLISHER]

In many text mining applications, side-information is available along with the text documents. Such side-information may be of different kinds, such as document provenance information, the links in the document, user-access behavior from web logs, or other non-textual attributes which are embedded into the text document. Such attributes may contain a tremendous amount of information for clustering purposes. However, the relative importance of this side-information may be difficult to estimate, especially when some of the information is noisy. In such cases, it can be risky to incorporate side-information into the mining process, because it can either improve the quality of the representation for the mining process, or can add noise to the process. Therefore, we need a principled way to perform the mining process, so as to maximize the advantages from using this side information. In this paper, we design an algorithm which combines classical partitioning algorithms with probabilistic models in order to create an effective clustering approach. We then show how to extend the approach to the classification problem. We present experimental results on a number of real data sets in order to illustrate the advantages of using such an approach. [ABSTRACT FROM PUBLISHER]

Network and linked data have become quite prevalent in recent years because of the ubiquity of the web and social media applications, which are inherently network oriented. Such networks are massive, dynamic, contain a lot of content, and may evolve over time. In this paper, we will study the problem of efficient dynamic link inference in temporal and heterogeneous information networks. The problem of efficiently performing dynamic link inference is extremely challenging in massive and heterogeneous information network because of the challenges associated with the dynamic nature of the network, and the different types of nodes and attributes in it. Both the topology and type information need to be used effectively for the link inference process. We propose an effective two-level scheme which makes efficient macro- and micro-decisions for combining structure and content in a dynamic and time-sensitive way. The time-sensitive nature of the links is leveraged in order to perform effective link prediction. We will also study how to apply the method to the problem of community prediction. We illustrate the effectiveness of our technique over a number of real data sets. Statistical Analysis and Data Mining 2013 DOI: 10.1002/sam.11198 [ABSTRACT FROM AUTHOR]

In recent years, a large amount of information has become available online in the form of web documents, social networks, or blogs. Such networks are large, heterogeneous, and often contain a huge number of links. This linkage structure encodes rich structural information about the topical behavior of the network. Such networks are often dynamic and evolve rapidly over time. Much of the work in the literature has focused on classification either with purely text behavior or with purely linkage behavior. Furthermore, the work in the literature is mostly designed for static networks. However, a given network may be quite diverse, and the use of either content or structure could be more or less effective in different parts of the network. In this paper, we examine the problem of node classification in dynamic information networks with both text content and links. Our techniques use a random walk approach in conjunction with the content of the network to facilitate an effective classification process. Our approach is dynamic, and can be applied to networks which are updated incrementally. Our results suggest that an approach based on both content and links is extremely robust and effective. We also present methods to perform supervised keyword-based clustering of nodes using this approach. We present experimental results illustrating the effectiveness and efficiency of our classification approach. We also show that the approach is able to find effective and coherent clusters. © 2012 Wiley Periodicals, Inc. Statistical Analysis and Data Mining 5: 16-34, 2012 [ABSTRACT FROM AUTHOR]

The U.S. Census Bureau's projection states that Hispanics form the largest minority group in the U.S. construction industry. A major challenge faced by American construction companies is the continually increasing number of fatalities among its Hispanic workers. According to the literature, illiteracy, the language barrier, and cultural differences are major causes of Hispanic worker fatalities. This study focuses on two out of these three aspects, i.e., language barriers and cultural differences. The study was undertaken to explore the reasons behind the lack of safety awareness of Hispanic workers, and to investigate the measures that construction companies take to deal with this problem. The literature review identified the main issues that need to be addressed by construction companies to ensure the safety of Hispanic workers. A survey was created, asking supervisors/directors of the top U.S. construction companies about their safety initiatives. As a result, it was found that construction companies address the problems of language and cultural barriers by implementing initiatives such as hands-on training, English as a Second Language courses, bilingual trainers, vocabulary cards, social gatherings, and common workshops. This study may help the construction industry identify areas requiring attention in order to improve the safety of its Hispanic employees. [ABSTRACT FROM AUTHOR]

In recent years, a number of indirect data collection methodologies have led to the proliferation of uncertain data. Such databases are much more complex because of the additional challenges of representing the probabilistic information. In this paper, we provide a survey of uncertain data mining and management applications. We will explore the various models utilized for uncertain data representation. In the field of uncertain data management, we will examine traditional database management methods such as join processing, query processing, selectivity estimation, OLAP queries, and indexing. In the field of uncertain data mining, we will examine traditional mining problems such as frequent pattern mining, outlier detection, classification, and clustering. We discuss different methodologies to process and mine uncertain data in a variety of forms. [ABSTRACT FROM AUTHOR]

The Class II human leukocyte antigen (HLA) DRB1 antigen DR4 is associated with autoimmune disorders and response to cyclosporine immunosuppression in T-lymphoproliferative disorders. We retrospectively reviewed the role of HLA DR4 on outcomes in 77 related and 22 unrelated consecutive first HLA-matched alloHSCT patients with lymphoid malignancies treated between 1992 and 2003. HLA DRB1 typing was determined by molecular (n = 69) or serologic (n = 30) methods. The proportion of patients with one HLA DR4 antigen was 18% (18/99). At a median follow-up of 5.6 years, there were no significant differences in aGvHD, cGvHD and OS between the HLA DR4 positive versus negative patients in any disease or donor subgroups. Nine of 18 (50%) DR4 positive patients and 20 of 81 (25%) DR4 negative patients had disease progression post HSCT (p = 0.033). Progression-free survival (PFS) at 3 years was 29% in the DR4 positive group and 70% in the DR4 negative group (p = 0.004). In univariate and multivariate analyses, DR4 positivity was the only significant factor associated with PFS (RR = 3.2, p = 0.007). Our results suggest that in addition to the known role of HLA DRB1 disparity in mediating histocompatibility, HLA DR4 is associated with inferior PFS in patients with lymphoid malignancy undergoing alloHSCT. [ABSTRACT FROM AUTHOR]

In recent years, privacy-preserving data mining has become an important problem because of the large amount of personal data which is tracked by many business applications. In many cases, users are unwilling to provide personal information unless the privacy of sensitive information is guaranteed. In this paper, we propose a new framework for privacy-preserving data mining of multidimensional data. Previous work for privacy-preserving data mining uses a perturbation approach which reconstructs data distributions in order to perform the mining. Such an approach treats each dimension independently and therefore ignores the correlations between the different dimensions. In addition, it requires the development of a new distribution-based algorithm for each data mining problem, since it does not use the multidimensional records, but uses aggregate distributions of the data as input. This leads to a fundamental re-design of data mining algorithms. In this paper, we will develop a new and flexible approach for privacy-preserving data mining that does not require new problem-specific algorithms, since it maps the original data set into a new anonymized data set. These anonymized data closely match the characteristics of the original data including the correlations among the different dimensions. We will show how to extend the method to the case of data streams. We present empirical results illustrating the effectiveness of the method. We also show the efficiency of the method for data streams. [ABSTRACT FROM AUTHOR]

High-dimensional data is a difficult case for most subspace-based classification methods because of the large number of combinations of dimensions, which have discriminatory power. This is because there are an exponential number of combinations of dimensions that could decide the correct class instance, and this combination could vary with data locality and test instance. Therefore, most summarized models such as decision trees and rule-based systems only aim to have a global summary of the data, which is used for classification. Because of this incompleteness, a particular classification model may be more or less suited to individual test instances. Furthermore, it may not provide sufficient insight into the most representative characteristics of a particular test instance. This is undesirable for many classification applications in which the diagnostic reasoning behind the classification of a test instance is as important as the classification process itself. In an interactive application, a user may find it more valuable to develop a diagnostic decision support method, which can reveal significant classification behaviors of exemplar records. Such an approach has the additional advantage of being able to optimize the decision process for the individual record in order to design more effective classification methods. In this paper, we propose the Subspace Decision Path (SD-Path) method, which provides the user with the ability to interactively explore a small number of nodes of a hierarchical decision process so that the most significant classification characteristics for a given test instance are revealed. In addition, the SD-Path method can provide enormous interpretability by constructing views of the data in which the different classes are clearly separated out. Even in difficult cases where the classification behavior of the test instance is ambiguous, the SD-Path method provides a diagnostic understanding of the characteristics, which results in this ambiguity. Therefore, this method combines the abilities of the human and the computer in creating an effective diagnostic tool for instance-centered high-dimensional classification. [ABSTRACT FROM AUTHOR]

Abstract: Forty-nine patients with intermediate- and high-risk aggressive non-Hodgkin lymphoma underwent autologous hematopoietic stem cell transplantation (autoHSCT) using the regimen of busulfan (Bu), cyclophosphamide (Cy), and etoposide (E) that was originally developed for allogeneic HSCT. Eighteen patients treated before 1999 received Cy 2.5 g/m2 on days −3 to −2 and E 1800 mg/m2 on day −3 after oral (PO) administration of Bu 1 mg/kg every 6 hours × 4 days for a total of 16 doses beginning on day −7. After April 1999, 31 patients similar in all pretransplantation risk assessments received the same regimen except that intravenous (IV) Bu was substituted for PO Bu and pharmacokinetic-directed (PKD) dosing was attempted to achieve an area under the concentration time curve of 1000-1500 μmol/min for each dose. Nonrelapse mortality was 28% for PO Bu patients versus 3% for the IV PKD group (P = .01, chi-square test). Actuarial 5-year overall survivals were 28% for patients who received the PO Bu regimen and 58% for patients who received the IV Bu regimen (P = .010, log-rank test), and progression-free survivals were 17% and 50%, respectively (P = .008, log-rank test). After substitution of PKD IV Bu in the BuCyE regimen, we observed lower nonrelapse mortality with increased overall and progression-free survivals in patients with intermediate- and high-risk aggressive non-Hodgkin lymphoma who underwent autoHSCT. [Copyright &y& Elsevier]

Current models of the classification problem do not effectively handle bursts of particular classes coming in at different times. In fact, the current model of the classification problem simply concentrates on methods for one-pass classification modeling of very large data sets. Our model for data stream classification views the data stream classification problem from the point of view of a dynamic approach in which simultaneous training and test streams are used for dynamic classification of data sets. This model reflects real-life situations effectively, since it is desirable to classify test streams in real time over an evolving training and test stream. The aim here is to create a classification system in which the training model can adapt quickly to the changes of the underlying data stream. In order to achieve this goal, we propose an on-demand classification process which can dynamically select the appropriate window of past training data to build the classifier. The empirical results indicate that the system maintains a high classification accuracy in an evolving data stream, while providing an efficient solution to the classification task. [ABSTRACT FROM AUTHOR]

In recent years, the progress in hardware technology has made it possible for organizations to store and record large streams of transactional data. This results in databases which grow without limit at a rapid rate. This data can often show important changes in trends over time. In such cases, it is useful to understand, visualize, and diagnose the evolution of these trends. In this paper, we introduce the concept of velocity density estimation, a technique used to understand, visualize, and determine trends in the evolution of fast data streams. We show how to use velocity density estimation in order to create both temporal velocity profiles and spatial velocity profiles at periodic instants in time. These profiles are then used in order to predict three kinds of data evolution: dissolution, coagulation, and shift. Methods are proposed to visualize the changing data trends in a single online scan of the data stream and a computational requirement which is linear in the number of data points. The visualization techniques can also be used to provide online animations which show the changes in the data characteristics while they occur. In addition, batch processing techniques are proposed in order to quantify the level of change across different combinations of dimensions. This quantification is then used in order to determine dimensional combinations with significant evolution. The techniques discussed in this paper can be easily extended to spatiotemporal data, changes in data snapshots at fixed instances in time, or any other data which has a temporal component during its evolution. [ABSTRACT FROM AUTHOR]

Data reduction can improve the storage, transfer time, and processing requirements of very large data sets. One of the challenges of designing effective data reduction techniques is to be able to preserve the ability to use the reduced format directly for a wide range of database and data mining applications. In this paper, we propose the novel idea of hierarchical subspace sampling in order to create a reduced representation of the data. The method is naturally able to estimate the local implicit dimensionalities of each point very effectively and, thereby, create a variable dimensionality reduced representation of the data. Such a technique is very adaptive about adjusting its representation depending upon the behavior of the immediate locality of a data point. An important property of the subspace sampling technique is that the overall efficiency of compression improves with increasing database size. Because of its sampling approach, the procedure is extremely fast and scales linearly both with data set size and dimensionality. We propose new and effective solutions to problems such as selectivity estimation and approximate nearest-neighbor search. These are achieved by utilizing the locality specific subspace characteristics of the data which are revealed by the subspace sampling technique. [ABSTRACT FROM AUTHOR]

Clustering is a central task in data mining applications such as customer segmentation. High-dimensional data has always been a challenge for clustering algorithms because of the inherent sparsity of the points. Therefore, techniques have recently been proposed to find clusters in hidden subspaces of the data. However, since the behavior of the data can vary considerably in different subspaces, it is often difficult to define the notion of a cluster with the use of simple mathematical formalizations. The widely used practice of treating clustering as the exact problem of optimizing an arbitrarily chosen objective function can often lead to misleading results. In fact, the proper clustering definition may vary not only with the application and data set but also with the perceptions of the end user. This makes it difficult to separate the definition of the clustering problem from the perception of an end-user. In this paper, we propose a system which performs high-dimensional clustering by cooperation between the human and the computer. The complex task of cluster creation is accomplished through a combination of human intuition and the computational support provided by the computer. The result is a system which leverages the best abilities of both the human and the computer for solving the clustering problem. [ABSTRACT FROM AUTHOR]

In this paper, we discuss the merits of building text categorization systems by using supervised clustering techniques. Traditional approaches for document classification on a predefined set of classes are often unable to provide sufficient accuracy because of the difficulty of fitting a manually categorized collection of documents in a given classification model. This is especially the case for heterogeneous collections of Web documents which have varying styles, vocabulary, and authorship. Hence, this paper investigates the use of clustering in order to create the set of categories and its use for classification of documents. Completely unsupervised clustering has the disadvantage that it has difficulty in isolating sufficiently fine-grained classes of documents relating to a coherent subject matter. In this paper, we use the information from a preexisting taxonomy in order to supervise the creation of a set of related clusters, though with some freedom in defining and creating the classes. We show that the advantage of using partially supervised clustering is that it is possible to have some control over the range of subjects that one would like the categorization system to address, but with a precise mathematical definition of how each category is defined. An extremely effective way then to categorize documents is to use this a priori knowledge of the definition of each category. We also discuss a new technique to help the classifier distinguish better among closely related clusters. [ABSTRACT FROM AUTHOR]

Incomplete data sets have become almost ubiquitous in a wide variety of application domains. Common examples can be found in climate and image data sets, sensor data sets, and medical data sets. The incompleteness in these data sets may arise from a number of factors: In some cases, it may simply be a reflection of certain measurements not being available at the time, in others, the information may be lost due to partial system failure, or it may simply be a result of users being unwilling to specify attributes due to privacy concerns. When a significant fraction of the entries are missing in all of the attributes, it becomes very difficult to perform any kind of reasonable extrapolation on the original data. For such cases, we introduce the novel idea of conceptual reconstruction in which we create effective conceptual representations on which the data mining algorithms can be directly applied. The attraction behind the idea of conceptual reconstruction is to use the correlation structure of the data in order to express it in terms of concepts rather than the original dimensions. As a result, the reconstruction procedure estimates only those conceptual aspects of the data which can be mined from the incomplete data set, rather than force errors created by extrapolation. We demonstrate the effectiveness of the approach on a variety of real data sets. [ABSTRACT FROM AUTHOR]

The results discussed in this paper are relevant to a large database consisting of consumer profile information together with behavioral (transaction) patterns. We introduce the concept of profile association rules, which discusses the problem of relating consumer buying behavior to profile information. The problem of online mining of profile association rules in this large database is discussed. We show how to use multidimensional indexing structures in order to actually perform the mining. The use of multidimensional indexing structures to perform profile mining provides considerable advantages in terms of the ability to perform very generic range-based online queries. [ABSTRACT FROM AUTHOR]

Redefines clustering for high-dimensional applications. Popularity of clustering problems in the database literature; Applications on customer segmentation, classification and trend analysis; Introduction of a general concept of projected clustering; Construction of clusters in arbitrarily aligned subspaces of lower dimensionality.

Focuses on the technique for discovering localized associations on the market basket data. Application of clustering on the data segments; Designation of algorithms on the customer pattern; Analysis of aggregate data on the marketing.

Focuses on the correlation between association rules and database of sales transactions. Determination of the buying patterns in consumer behavior; Description of the notion of a strongly-collective itemset; Development of an itemset generation algorithm.

Proposes the concept of intelligent crawling which learns characteristics of the linkage structure of the World Wide Web while performing the crawling. Overview of the statistical model which is used for the purpose of crawling; Discussion of the actual implementation and algorithmic details; Empirical results.

Discusses the problem of online mining of association rules in a large database of sales transactions. Description of the framework for organizing the primary item sets; Process for predicting the size of the output at a given level of user specified parameter; Efficiency in the generation of the rules; Redundancy in the rules generated from large interests; Algorithm for finding the primary itemsets.

Presents information on a study which introduced the notion of factored queue length and proposed a batching policy that scheduled the video with the maximum factored queue length (MFQL). Estimation of relative frequency to the different videos; Description of the results of simulation experiments comparing the various batching policies; Ideal policy for video-on-demand systems.

Introduction: It was hypothesized that use of proton beam therapy (PBT) in patients with locally advanced non–small cell lung cancer treated with concurrent chemoradiation and consolidative immune checkpoint inhibition is associated with fewer unplanned hospitalizations compared with intensity‐modulated radiotherapy (IMRT). Methods: Patients with locally advanced non–small cell lung cancer treated between October 2017 and December 2021 with concurrent chemoradiation with either IMRT or PBT ± consolidative immune checkpoint inhibition were retrospectively identified. Logistic regression was used to assess the association of radiation therapy technique with 90‐day hospitalization and grade 3 (G3+) lymphopenia. Competing risk regression was used to compare G3+ pneumonitis, G3+ esophagitis, and G3+ cardiac events. Kaplan–Meier method was used for progression‐free survival and overall survival. Inverse probability treatment weighting was applied to adjust for differences in PBT and IMRT groups. Results: Of 316 patients, 117 (37%) received PBT and 199 (63%) received IMRT. The PBT group was older (p <.001) and had higher Charlson Comorbidity Index scores (p =.02). The PBT group received a lower mean heart dose (p <.0001), left anterior descending artery V15 Gy (p =.001), mean lung dose (p =.008), and effective dose to immune circulating cells (p <.001). On inverse probability treatment weighting analysis, PBT was associated with fewer unplanned hospitalizations (adjusted odds ratio, 0.55; 95% CI, 0.38–0.81; p =.002) and less G3+ lymphopenia (adjusted odds ratio, 0.55; 95% CI, 0.37–0.81; p =.003). There was no difference in other G3+ toxicities, progression‐free survival, or overall survival. Conclusions: PBT is associated with fewer unplanned hospitalizations, lower effective dose to immune circulating cells and less G3+ lymphopenia compared with IMRT. Minimizing dose to lymphocytes may be warranted, but prospective data are needed. The use of proton beam therapy in locally advanced non–small cell lung cancer is associated with fewer 90‐day acute hospitalizations, lower effective dose to immune circulating cells, and less grade 3+ lymphopenia compared with intensity‐modulated radiation therapy. [ABSTRACT FROM AUTHOR]

Reirradiation (reRT) with proton beam therapy (PBT) may offer a chance of cure while minimizing toxicity for patients with isolated intrathoracic recurrences of non-small cell lung cancer (NSCLC). However, distant failure remains common, necessitating strategies to integrate more effective systemic therapy. This was a phase 2, single-arm trial (NCT03087760) of consolidation pembrolizumab after PBT reRT for locoregional recurrences of NSCLC. Four to 12 weeks after completion of 60 to 70 Gy PBT reRT, patients without progressive disease received pembrolizumab for up to 12 months. Primary endpoint was progression-free survival (PFS), measured from the start of reRT. Secondary endpoints were overall survival (OS) and National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0 toxicity. Between 2017 and 2021, 22 patients received PBT reRT. Median interval from prior radiation end to reRT start was 20 months. Most recurrences (91%) were centrally located. Most patients received concurrent chemotherapy (95%) and pencil beam scanning PBT (77%), and 36% had received prior durvalumab. Fifteen patients (68%) initiated consolidation pembrolizumab on trial and received a median of 3 cycles (range, 2-17). Pembrolizumab was discontinued most commonly due to toxicity (n = 5; 2 were pembrolizumab-related), disease progression (n = 4), and completion of 1 year (n = 3). Median follow-up was 38.7 months. Median PFS and OS were 8.8 months (95% CI, 4.2-23.7) and 22.8 months (95% CI, 6.9-not reached), respectively. There was only one isolated in-field failure after reRT. Grade ≥3 toxicities occurred in 10 patients (45%); 2 were pembrolizumab-related. There were 2 grade 5 toxicities, an aorto-esophageal fistula at 6.9 months and hemoptysis at 46.8 months, both probably from reRT. The trial closed early due to widespread adoption of immunotherapy off-protocol. In the first-ever prospective trial combining PBT reRT with consolidation immunotherapy, PFS was acceptable and OS favorable. Late grade 5 toxicity occurred in 2 of 22 patients. This approach may be considered in selected patients with isolated thoracic recurrences of NSCLC. [ABSTRACT FROM AUTHOR]

Discusses general main memory cache replacement policies designed specifically for use by World Wide Web caches. Theoretical optimization model for Web caching which generalizes least recently used (LRU) standard; Features of the variant of the size-adjusted LRU policy called the Pyramidal selection scheme; Performance of caching algorithms in event and trace-driven simulations.

The objective of this study was to describe the patterns of failure, frequency of low-volume relapse (LVR), and candidacy for ablative therapy at time of disease progression (PD) after chemoradiation and consolidative immunotherapy (CRT + ICI) in patients with stage III non-small cell lung cancer. We identified 229 consecutive patients with stage III non-small cell lung cancer treated with CRT + ICI between October 2017 and December 2021 at a single institution. PD was classified as isolated locoregional failure (LRF), isolated distant failure (DF), or synchronous LRF + DF. Any LRF was subclassified as in-field failure, marginal failure, or out-of-field failure. LVR was defined as 3 or fewer sites of PD in any number of organs. Ablative candidates were defined as having 5 or fewer sites of PD radiographically amenable to high-dose radiation or surgery. Time-to-event data were calculated using cumulative incidence analysis and Kaplan-Meier methods. Multivariable Cox modeling was used to examine the correlations between characteristics of relapse and postprogression survival. Of the 229 patients, 119 (52%) had PD. Of these 119 patients, 20 (21%) had isolated LRF, 28 (24%) had synchronous LRF + DF, and 71 (60%) had isolated DF. Of the 48 patients with any LRF, 28 (58%) had in-field failure, 10 (21%) marginal failure, and 10 (21%) out-of-field failure. The cumulative incidence of LRF and DF was 13% (95% CI, 9.2%-18%) and 32% (95% CI, 26%-38%) at 1 year and 19% (95% CI, 14%-24%) and 39% (95% CI, 33%-46%) at 2 years, respectively. Overall, 64 patients (54%) were considered to have LVR. At time of PD, 60 patients (50%) were eligible for ablative therapy. Patients with LVR had longer median survival versus with high-volume relapse (37.4 vs 15.2 months, P <.001). On multivariable analysis, LVR (hazard ratio, 0.32; 95% CI, 0.18-0.56; P <.001) was associated with improved postprogression survival. After CRT + ICI, approximately half of patients experience LVR at time of PD and are candidates for ablative therapies. Prospective trials are needed to validate the optimal treatment strategy for LVR. [ABSTRACT FROM AUTHOR]

We hypothesized that after adoption of immune checkpoint inhibitor (ICI) consolidation for patients with locally advanced non-small cell lung cancer (LA-NSCLC) receiving concurrent chemoradiation therapy (cCRT), rates of symptomatic pneumonitis would increase, thereby supporting efforts to reduce lung radiation dose. This single institution, multisite retrospective study included 783 patients with LA-NSCLC treated with definitive cCRT either before introduction of ICI consolidation (pre-ICI era cohort [January 2011-September 2017]; N = 448) or afterward (ICI era cohort [October 2017-December 2021]; N = 335). Primary endpoint was grade ≥2 pneumonitis (G2P) and secondary endpoint was grade ≥3 pneumonitis (G3P), per Common Terminology Criteria for Adverse Events v5.0. Pneumonitis was compared between pre-ICI era and ICI era cohorts using the cumulative incidence function and Gray's test. Inverse probability of treatment weighting (IPTW)-adjusted Fine-Gray models were generated. Logistic models were developed to predict the 1-year probability of G2P as a function of lung dosimetry. G2P was higher in the ICI era than in the pre-ICI era (1-year cumulative incidence 31.4% vs 20.1%; P <.001; IPTW-adjusted multivariable subdistribution hazard ratio, 2.03; 95% confidence interval, 1.53-2.70; P <.001). There was no significant interaction between ICI era treatment and either lung volume receiving ≥20 Gy (V20) or mean lung dose in Fine-Gray regression for G2P; however, the predicted probability of G2P was higher in the ICI era at clinically relevant values of lung V20 (≥24%) and mean lung dose (≥14 Gy). Cut-point analysis revealed a lung V20 threshold of 28% in the ICI era (1-year G2P rate 46.0% above vs 19.8% below; P <.001). Among patients receiving ICI consolidation, lung V5 was not associated with G2P. G3P was not higher in the ICI era (1-year cumulative incidence 7.5% vs 6.0%; P =.39; IPTW-adjusted multivariable subdistribution hazard ratio, 1.12; 95% confidence interval, 0.63-2.01; P =.70). In patients with LA-NSCLC treated with cCRT, the adoption of ICI consolidation was associated with an increase in G2P but not G3P. With ICI consolidation, stricter lung dose constraints may be warranted. [ABSTRACT FROM AUTHOR]

The therapeutic landscape for patients with advanced malignancies has changed dramatically over the last twenty years. The growing number of targeted therapies and immunotherapeutic options available have improved response rates and survival for a subset of patients, however determining which patients will experience clinical benefit from these therapies in order to avoid potential toxicities and reduce healthcare costs remains a clinical challenge. Cell-free circulating tumor DNA (ctDNA) is shed by tumor cells into systemic circulation and is already an integral part of routine clinical practice for the non-invasive tumor genotyping in advanced non-small cell lung cancer as well as other malignancies. The short half-life of ctDNA offers a unique opportunity to utilize early on-treatment changes in ctDNA for real-time assessment of therapeutic response and outcome, termed molecular response. Here, we provide a summary and review of the use of molecular response for the prediction of outcomes in patients with advanced cancer, including the current state of science, its application in clinic, and next steps for the development of this predictive tool. [ABSTRACT FROM AUTHOR]

CD8 T cells are important tools for protection against intracellularly replicating pathogens such as viruses. Previous studies showed that a discrete population of HLADR and CD38-expressing CD8 T cells expands massively during the acute febrile phase of human dengue virus infection--but very few of these cells secrete IFNγ upon in vitro stimulation with dengue peptides. To gain a better understanding of what other cytokines/chemokines do these massively expanding HLADR+CD38+ CD8 T cells express, we performed RNA seq of sorted HLADR+CD38+ CD8 T cell subsets after peptide stimulation. A majority of these peptide-stimulated HLADR+CD38+ CD8 T cells were CD69- IFNγ-, nearly a third were CD69+ IFNγ-, whereas very few (<10%) were CD69+ IFNγ+. The CD69- IFNγ- subset was enriched for the expression of key genes implicated in the negative regulation of T cell receptor (TCR) signaling and T-cell exhaustion, attraction of B cells and other lymphocytes, and cytokines related to Tc17/T-reg lineages or those that are implicated in immunosuppression/immunomodulatory and anti-inflammatory activities and angiogenesis. The CD69+ IFNγ- subset showed enriched transcription of key genes implicated in cytotoxic effector functions as well as costimulatory and signaling adaptors implicated in fine balancing of T cell receptor signaling. The CD69+ IFNγ+ subset largely shared the transcriptional profile with the CD69+ IFNγsubset--but with relatively more pronounced expression along with additional genes such as chemokines XCL1/XCL2. Our findings showing distinct functional subsets among these massively expanding CD8 T cells in dengue CD8 T cells warrant further studies to carefully examine the precise role of these T cell subsets in protection against dengue. [ABSTRACT FROM AUTHOR]

Purpose: This trial tested the safety and efficacy of a novel, deintensified radiation therapy (RT) approach after initial surgical resection for patients with human papilloma virus (HPV)-associated oropharyngeal squamous cell carcinoma (OPSCC).Methods and Materials: This single-arm phase 2 prospective clinical trial enrolled 60 patients with stage pT1-pT2 N1-3 HPV-associated OPSCC treated with transoral robotic surgery (TORS) and selective neck dissection at a single institution between May 2014 and September 2017. Patients had favorable features at the primary site (negative surgical margins ≥2 mm, no perineural invasion, and no lymphovascular invasion) but required adjuvant therapy based on lymph node involvement. Surgeries were all performed at a high-volume head and neck cancer center with expertise in TORS. Patients received postoperative RT to at-risk areas in the involved neck (60-66 Gy) and uninvolved neck (54 Gy). The resected primary site was treated as an active avoidance structure in the treatment planning of postoperative RT. Concurrent chemotherapy was administered for patients with extranodal extension.Results: Median follow-up of the 60 patients enrolled was 2.4 years (range, 8.5-53.8 months). A single patient recurred at the primary site, for 2-year local control of 98.3%. One patient (1.7%) developed a regional neck recurrence, and 2 patients (3.3%) developed distant metastases. Measured 2-year local recurrence-free survival was 97.9% (95% confidence interval, 86.1%-99.7%). Overall survival was 100% at the time of analysis. The mean radiation dose to the primary site was 36.9 Gy (standard deviation, 10.3 Gy). Two patients (3.3%) experienced late soft tissue necrosis in the primary site surgical bed that resolved within 2 months. Feeding tube dependence rates were 0% during RT, 3.3% temporarily during follow-up, and 0% at last follow-up.Conclusions: Deintensified postoperative RT that avoids the resected primary tumor site and targets only the at-risk neck after TORS for selected patients with HPV-associated OPSCC may be safe and is worthy of further study. [ABSTRACT FROM AUTHOR]

The article presents a case study of a 45-year-old man presented with a painless right neck mass in late 2013l. It notes that Computed tomography (CT) scan of the neck demonstrated a 3.2-centimeter mass in the right parapharyngeal space arising from the deep lobe of the right parotid gland. The article discusses targeting HER2/neu oncogene overexpression with ado-trastuzumab emtansine in the treatment of metastatic salivary sland neoplasms.

Clinical chorioamnionitis can result either from the ascending organisms from vagina after rupture of membranes or via the blood stream. This report describes a case of chorioamnionitis caused by Arcanobacterium haemolyticum, an unusual organism associated with this infection. A 19-year-old primigravida at 32 weeks of gestation presented with the complaints of fever, pain in the abdomen, and discharge per vaginum for the past 2 days. Watery, odorless, colorless, and discharge passing intermittently were noticed. A. haemolyticum was isolated from amniotic fluid. Upon induction, a healthy male baby weighing 1.9 kg was delivered. The patient was started on gentamicin and metronidazole for 8 days. Recovery was uneventful. The mother and baby were discharged on ninth day. Chorioamnionitis can result in significant maternal and fetal mortality and morbidity. Hence, it is important to ascertain the prompt diagnosis and treatment of suspected cases. Increase in awareness of clinicians and laboratories will reduce misdiagnosis and facilitate appropriate treatment. [ABSTRACT FROM AUTHOR]

Objectives: In previous clinical studies using a murine-derived CAR construct targeted to mesothelin (SS1 CAR), we observed limited persistence and efficacy. To improve the outcome, we created a fully human anti-mesothelin scFV fused to the intracellular signaling domains 4-1BB (CD137) and TCR zeta (M5 CAR). Pre-clinical evaluation of the M5 CAR revealed enhanced in vivo antitumor activity compared to the SS1 CAR. Methods: Fourteen patients with mesothelin-expressing tumors (ovarian, lung, malignant mesothelioma) were treated with the M5 CAR. Three patients received a single IV dose of 3 x 107 CART-meso cells/m2 (Cohort 1), three patients received a single IV dose of 3 x 107 CART-meso cells/m2 following lymphodepletion (Cohort 2), and two patients received a single IV dose of 3 x 108 CART-meso cells/m2 (Cohort 3). Due to toxicity, the last six patients received lymphodepletion and an initial IV dose of 3 x 107 CART-meso cells/m2 followed by up to two additional IV doses spaced 21-42 days apart (Cohort 4). Results: The infusions were well tolerated. M5 CART cells were detectable in blood one hour after infusion and expanded to a peak at Day 7-10. Higher levels were detected in patients treated with the higher dose CART cells and with lymphodepletion. M5 CART cells persisted up to 28-42 days in all subjects and at three and six months in two patients. M5 CART cells were trafficked to tumors in nine of the 14 patients. At 28-35 days after treatment, eight patients showed stable disease (8/14, 57%) per RECIST criteria. One patient had SD for three months; another had SD for nine months. Four subjects experienced cytokine release syndrome (CRS) (three grade 3 and one grade 4), and related AEs included hypoxia, hypotension, hyponatremia, fatigue, and febrile neutropenia. In Cohort 3, both subjects had acute respiratory SAEs. One patient experienced grade 3 SAEs (CRS, hypotension, hypoxia) but fully recovered. The other patient, with extensive pulmonary tumor burden and residual PD-1 blockade, suffered fatal respiratory failure. Due to these SAEs, the MTD was defined at 3 x 107 CART-meso cells/m2 and six subsequent patients had no pulmonary SAEs. Conclusions: This dose-escalation trial demonstrates the feasibility of treating multiple tumor types with M5 CART cells. The M5 CART cells engrafted and expanded in the blood of all subjects but persisted at relatively low blood levels and for short time periods. CART trafficking into tumors was identified in 9/14 patients. Short-term stable disease was noted in the majority of patients, but no clinical responses were seen. Pulmonary toxicity was seen when given at the 3 x 10 8 CART cells/m2 dose. These preliminary results may help guide future M5 CART clinical trials, such as loco-regional delivery or modifications to the CART cells to improve safety and enhance trafficking and persistence. [ABSTRACT FROM AUTHOR]

A network with $n$ nodes contains $O(n^2)$ possible links. Even for networks of modest size, it is often difficult to evaluate all pairwise possibilities for links in a meaningful way. Further, even though link prediction is closely related to missing value estimation problems, it is often difficult to use sophisticated models such as latent factor methods because of their computational complexity on large networks. Hence, most known link prediction methods are designed for evaluating the link propensity on a specified subset of links, rather than on the entire networks. In practice, however, it is essential to perform an exhaustive search over the entire networks. In this article, we propose an ensemble enabled approach to scaling up link prediction, by decomposing traditional link prediction problems into subproblems of smaller size. These subproblems are each solved with latent factor models, which can be effectively implemented on networks of modest size. By incorporating with the characteristics of link prediction, the ensemble approach further reduces the sizes of subproblems without sacrificing its prediction accuracy. The ensemble enabled approach has several advantages in terms of performance, and our experimental results demonstrate the effectiveness and scalability of our approach. [ABSTRACT FROM AUTHOR]