Alexander Gutfraind

Efficacy and Optimization of Palivizumab Injection Regimens Against Respiratory Syncytial Virus Infection

IMPORTANCE Infection with the respiratory syncytial virus (RSV) is the leading cause of hospitalizations in children, accounting for more than 90,000 hospitalizations every year in the United States. For children who are at risk for severe RSV infections, the American Academy of Pediatrics recommends immunoprophylaxis with a series of up to 5 injections of the antibody palivizumab administered monthly, beginning on November 1 of each year. However, many practitioners initiate injections at the onset of RSV season as indicated by local surveillance. OBJECTIVES To evaluate the effectiveness of current regimens for palivizumab injections across different cities and to design an optimized regimen. DESIGN, SETTING, AND PARTICIPANTS We performed a mathematical modeling study of the risk for hospitalization due to RSV infection. The model accounted for the pharmacokinetics of the antibody, the timing of the injections, and seasonal patterns of RSV, including geographic and year-to-year variability. We used the model to estimate the efficacy of current regimens, including the American Academy of Pediatrics recommendation, and to design a more effective injection regimen, the optimized fixed start (OFS), which uses city-specific initiation dates. Participants were the approximately 700,000 individuals who had specimens tested for RSV by National Respiratory and Enteric Virus Surveillance System laboratories in 18 US cities from July 1, 1994, through June 30, 2011 (a total of 725,741 tests). INTERVENTIONS Different palivizumab injection regimens. MAIN OUTCOMES AND MEASURES The primary outcome measure was reduction in hospitalizations due to RSV infections. The secondary measures were cost (number of palivizumab doses) and duration of protection (in days). RESULTS The American Academy of Pediatrics-recommended 5-injection regimen is expected to reduce hospitalization risk by a median of 2.7% (range, -2.2% to 6.1%) compared with the conventional regimen based on RSV surveillance. The 5-injection OFS regimen is expected to further reduce risk by a median of 6.8% (range, 4.9% to 14.8%), and the 4-injection OFS regimen is expected to achieve efficacy comparable to that of the conventional 5-injection regimen while reducing costs by 20%. CONCLUSIONS AND RELEVANCE Modified palivizumab regimens can improve protection for children at risk for severe outcomes of RSV infection and thereby lower rates of hospitalization due to RSV.

Evaluating Large-scale Blood Transfusion Therapy for the Current Ebola Epidemic in Liberia

BACKGROUND To combat the 2014-2015 Ebola virus disease (EVD) epidemic in West Africa, the World Health Organization urged the rapid evaluation of convalescent whole blood (CWB) and plasma (CP) transfusion therapy. However, the feasibility and likely impacts of broad implementation of transfusions are yet unknown. METHODS We extended an Ebola virus transmission model published by the Centers for Disease Control and Prevention to include hospital-based convalescent donations and transfusions. Using recent epidemiological estimates for EVD in Liberia and assuming that convalescent transfusions reduce the case-fatality rate to 12.5% (range, 7.5%-17.5%), we projected the impacts of a countrywide ramp-up of transfusion therapy. RESULTS Under the 10% case-hospitalization rate estimated for Liberia in September 2014, large-scale CP therapy is expected to save 3586 lives by October 2015 (3.1% mortality reduction; 95% confidence interval [CI], .52%-4.5%). Under a higher 30% hospitalization rate, CP transfusions are expected to save 151 lives (0.9% of the total; 95% CI, .21%-11%). CONCLUSIONS Transfusion therapy for EVD is a low-cost measure that can potentially save many lives in West Africa but will not measurably influence the prevalence. Under all scenarios considered, CP transfusions are predicted to achieve greater reductions in mortality than CWB.

Agent-Based Model Forecasts Aging of the Population of People Who Inject Drugs in Metropolitan Chicago and Changing Prevalence of Hepatitis C Infections

People who inject drugs (PWID) are at high risk for blood-borne pathogens transmitted during the sharing of contaminated injection equipment, particularly hepatitis C virus (HCV). HCV prevalence is influenced by a complex interplay of drug-use behaviors, social networks, and geography, as well as the availability of interventions, such as needle exchange programs. To adequately address this complexity in HCV epidemic forecasting, we have developed a computational model, the Agent-based Pathogen Kinetics model (APK). APK simulates the PWID population in metropolitan Chicago, including the social interactions that result in HCV infection. We used multiple empirical data sources on Chicago PWID to build a spatial distribution of an in silico PWID population and modeled networks among the PWID by considering the geography of the city and its suburbs. APK was validated against 2012 empirical data (the latest available) and shown to agree with network and epidemiological surveys to within 1%. For the period 2010–2020, APK forecasts a decline in HCV prevalence of 0.8% per year from 44(±2)% to 36(±5)%, although some sub-populations would continue to have relatively high prevalence, including Non-Hispanic Blacks, 48(±5)%. The rate of decline will be lowest in Non-Hispanic Whites and we find, in a reversal of historical trends, that incidence among non-Hispanic Whites would exceed incidence among Non-Hispanic Blacks (0.66 per 100 per years vs 0.17 per 100 person years). APK also forecasts an increase in PWID mean age from 35(±1) to 40(±2) with a corresponding increase from 59(±2)% to 80(±6)% in the proportion of the population >30 years old. Our studies highlight the importance of analyzing subpopulations in disease predictions, the utility of computer simulation for analyzing demographic and health trends among PWID and serve as a tool for guiding intervention and prevention strategies in Chicago, and other major cities.

Generating Scaled Replicas of Real-World Complex Networks

Research on generative models plays a central role in the emerging field of network science, studying how statistical patterns found in real networks can be generated by formal rules. During the last two decades, a variety of models has been proposed with an ultimate goal of achieving comprehensive realism for the generated networks. In this study, we (a) introduce a new generator, termed ReCoN; (b) explore how models can be fitted to an original network to produce a structurally similar replica, and (c) aim for producing much larger networks than the original exemplar. In a comparative experimental study, we find ReCoN often superior to many other stateof- the-art network generation methods. Our design yields a scalable and effective tool for replicating a given network while preserving important properties at both microand macroscopic scales and (optionally) scaling the replica by orders of magnitude in size. We recommend ReCoN as a general practical method for creating realistic test data for the engineering of computational methods on networks, verification, and simulation studies. We provide scalable open-source implementations of most studied methods, including ReCoN.

Modelling the neighbour aid phenomenon for installing costly complex networks

Complex networks are frequently used to model natural, social and engineered systems. A lot of studies looked at the problem of modelling the topology of a network or designing it. Less is known, even in a stylized form, about how one might optimally install a given network topology, assuming that installing a node aids the installation of neighbouring nodes. This problem arises in education where it is desired to teach mutually reinforcing words, as well as in engineering whenever damaged infrastructure networks are to be recovered. We introduce a new discrete optimization problem where the goal is to minimize the total cost of installing a given network. This cost is determined by the structure of the network and the sequence with which the nodes are installed. Namely, the cost of installing a node is a function of the number of its neighbours that have been installed before it. We analyse the common case where the cost function is decreasing and convex, and provide bounds on the cost of the optimal solution. We also show that all sequences have the same cost when the cost function is linear and give results on the cost of a random solution for several models of random graphs. Examining the computational complexity, we show that the problem is NP-hard when the cost function is arbitrary. Finally, we provide an integer programme, a dynamic programming algorithm, and greedy heuristics which give high quality solutions.

Optimizing Distribution of Pandemic Influenza Antiviral Drugs

Effective distribution of these drugs will reduce illness and death in underinsured populations.

High-Risk Geographic Mobility Patterns among Young Urban and Suburban Persons who Inject Drugs and their Injection Network Members

Young people in the USA who inject drugs, particularly those at a risk of residence instability, experience the highest incidence of hepatitis C (HCV) infections. This study examined associations between geographic mobility patterns and sociodemographic, behavioral, and social network characteristics of 164 young (ages 18–30) persons who inject drugs (PWID). We identified a potential bridge sub-population who reported residence in both urban and suburban areas in the past year (crossover transients) and higher-risk behaviors (receptive syringe sharing, multiple sex partners) compared to their residentially localized counterparts. Because they link suburban and urban networks, crossover transients may facilitate transmission of HIV and HCV between higher and lower prevalence areas. Interventions should address risk associated with residential instability, particularly among PWID who travel between urban and suburban areas.

Modeling of patient virus titers suggests that availability of a vaccine could reduce hepatitis C virus transmission among injecting drug users

A future vaccine could reduce hepatitis C virus transmission among those sharing syringes, even in the presence of postexposure viral replication in vaccinees. Hampering hepatitis C virus transmission No hepatitis C virus (HCV) vaccine is currently available, and evidence from studies in nonhuman primates suggests that any future human HCV vaccine would be unlikely to induce complete immunity against the virus. Major et al. examined whether lowered HCV titers potentially resulting from an imperfect vaccine might still stem HCV transmission in people who inject drugs. The authors measured the HCV RNA from infected human plasma retained in contaminated needles and syringes. Their mathematical model combining these measurements with published HCV viral kinetics data suggested that a partially effective vaccine could reduce the HCV transmission risk among individuals who share contaminated needles and syringes. The major route of hepatitis C virus (HCV) transmission in the United States is injection drug use. We hypothesized that if an HCV vaccine were available, vaccination could affect HCV transmission among people who inject drugs by reducing HCV titers after viral exposure without necessarily achieving sterilizing immunity. To investigate this possibility, we developed a mathematical model to determine transmission probabilities relative to the HCV RNA titers of needle/syringe-sharing donors. We simulated sharing of two types of syringes fitted with needles that retain either large or small amounts of fluid after expulsion. Using previously published viral kinetics data from both naïve subjects infected with HCV and reinfected individuals who had previously cleared an HCV infection, we estimated transmission risk between pairs of serodiscordant injecting drug users, accounting for syringe type, rinsing, and sharing frequency. We calculated that the risk of HCV transmission through syringe sharing increased ~10-fold as viral titers (log10 IU/ml) increased ~25-fold. Cumulative analyses showed that, assuming sharing episodes every 7 days, the mean transmission risk over the first 6 months was >90% between two people sharing syringes when one had an HCV RNA titer >5 log10 IU/ml. For those with preexisting immunity that rapidly controlled HCV, the cumulative risk decreased to 1 to 25% depending on HCV titer and syringe type. Our modeling approach demonstrates that, even with transient viral replication after exposure during injection drug use, HCV transmission among people sharing syringes could be reduced through vaccination if an HCV vaccine were available.

Generating realistic scaled complex networks

Research on generative models plays a central role in the emerging field of network science, studying how statistical patterns found in real networks could be generated by formal rules. Output from these generative models is then the basis for designing and evaluating computational methods on networks including verification and simulation studies. During the last two decades, a variety of models has been proposed with an ultimate goal of achieving comprehensive realism for the generated networks. In this study, we (a) introduce a new generator, termed ReCoN; (b) explore how ReCoN and some existing models can be fitted to an original network to produce a structurally similar replica, (c) use ReCoN to produce networks much larger than the original exemplar, and finally (d) discuss open problems and promising research directions. In a comparative experimental study, we find that ReCoN is often superior to many other state-of-the-art network generation methods. We argue that ReCoN is a scalable and effective tool for modeling a given network while preserving important properties at both micro- and macroscopic scales, and for scaling the exemplar data by orders of magnitude in size.

A graph database framework for covert network analysis: An application to the Islamic State network in Europe

Abstract This paper proposes a new framework, based on graph database theory, for encoding complex data on covert networks, mapping their structure, and conducting a sensitivity analysis. The framework is then applied to reconstruct the terrorist network of the 2015–2016 attacks in Paris and Brussels, and related plots in Europe by the Islamic State group. The resulting network was found to be qualitatively different from the ideologically-related Al-Qaeda network, having a lower secrecy and a lower mean degree, under different network-generating assumptions.