We took benefit of the actual fact that HIV can be an evolving pathogen whose transmitting network could be reconstructed using hereditary sequence information to handle this shortcoming

We took benefit of the actual fact that HIV can be an evolving pathogen whose transmitting network could be reconstructed using hereditary sequence information to handle this shortcoming. NORTH PARK, California. We created and examined a network-based statistic for calculating treatment results using simulated Rabbit Polyclonal to NCOA7 scientific studies on our inferred transmitting network. We explored the statistical power of the network-based statistic against typical efficacy measures and discover that when upcoming transmitting is decreased, the prospect of improved statistical power could be understood. Furthermore, our simulations demonstrate the fact that network statistic can detect community-level results NU6300 (electronic.g., decrease in onward transmitting) of HIV treatment within a scientific trial establishing. This research demonstrates the utility of the network-based statistical metric when investigating HIV treatment options as a method to reduce onward transmission in a clinical trial setting. == Introduction == Randomized trials of preventive measures against HIV, such as male circumcision[1][3], vaginal microbicide gel[4], pre-exposure prophylaxis (PrEP)[5],[6], and vaccination[7], have demonstrated modest but potentially important benefits. In studies to date, such interventions have had estimated efficacy levels of 3050% in preventing individual infections. What remains to be quantified, however, is the potential such interventions have for larger community-level effects (i.e. reduced infectiousness leading to lower disease burden). It is unknown whether any current intervention strategy could meet the public health goal of bringing an epidemic under control. Prevention and treatment efforts [e.g. imperfect vaccination, PrEP followed by antiretroviral treatment (ART), or test-and-treat] could reduce onward transmission by decreasing the viral load of infected individuals[8]. Even interventions that were ineffective or not intended to prevent infections (e.g. test-and-treat[9]) may prove useful in lowering HIV’s capacity for NU6300 future infection by reducing viral load[10],[11]; such effects would likely be of great benefit to the susceptible population as a whole. It is noteworthy that single or two-drug ART combinations are sufficient for substantial reductions in maternal-child transmission, even though such therapies are not adequate for complete suppression of HIV[12]. Current clinical trials are not designed to detect a decrease in onward transmission as an effect of an intervention at the community level. Although group-level randomized trials have been undertaken[13], such studies are logistically complex and not always practical. This manuscript addresses the question of how to use putative transmission network information inferred from individual-level randomized studies to learn about the community-level impact of prevention strategies. HIV prevention trials face a combination of formidable obstacles: relatively low efficacy (around 30%) and a low incidence of HIV infections (1% per year) in most populations. This combination often results in relatively low statistical power (<80%) to detect efficacious intervention, even for very large numbers of trial participants (10000 per arm). The recent CAPRISA microbicide trial in South Africa is an exception, which was able to conduct the study in a population with an unusually high incidence of HIV (around 10%)[4]. Statistical tests used to detect the efficacy of prevention and intervention treatments take into account the time of infection of study subjects but not the evolutionary transmission history of the virus. HIV is a measurably evolving pathogen, and changes in its genetic sequence over time have been successfully used to reconstruct recent individual-to-individual transmission histories[14][20]. These NU6300 transmission histories are traditionally represented as phylogenetic trees, but they can also be depicted as (incomplete) transmission networks. These networks are comprised of nodes, representing HIV-infected individuals, which are connected by edges if the genetic similarity or phylogenetic relatedness of the viruses is sufficiently high[17],[21]. Individual nodes in a transmission network can be described using a variety of statistics, one of NU6300 which is degree: the number of edges connecting to a given node. Here we propose a statistical metric that accounts for the evolutionary relatedness of the virus and address two questions: (i) can transmission networks provide a basis for developing NU6300 more powerful statistical metrics to measure prevention effeciveness, and (ii) can these transmission networks be used to detect decreases in viral transmission from study participants to others in their sexual network? Through simulations on a transmission network inferred from men.