{ "items": [ "\n\n
BACKGROUND:HIV-infected individuals on antiretroviral therapy (ART) require treatment with artemisinin-based combination therapy (ACT) when infected with malaria. Artemether-lumefantrine (AL) is the most commonly used ACT for treatment of falciparum malaria in Africa but there is limited evidence on the safety and efficacy of AL in HIV-infected individuals on ART, among whom drug-drug interactions are expected. Day-42 adequate clinical and parasitological response (ACPR) and incidence of adverse events was assessed in HIV-infected individuals on efavirenz-based ART with uncomplicated falciparum malaria treated with AL. METHODS:A prospective, open label, non-randomized, interventional clinical trial was conducted at St Paul's Hospital in northern Zambia, involving 152 patients aged 15-65\u00a0years with uncomplicated falciparum malaria, who were on efavirenz-based ART. They received a 3-day directly observed standard treatment of AL and were followed up until day 63. Day-42 polymerase chain reaction (PCR)-corrected ACPRs (95% confidence interval [CI]) were calculated for the intention-to-treat population. RESULTS:Enrolled patients had a baseline geometric mean (95% CI) parasite density of 1108 (841-1463) parasites/\u00b5L; 16.4% (25/152) of the participants had a recurrent malaria episode by day 42. However, PCR data was available for 17 out of the 25 patients who had malaria recurrence. Among all the 17 patients, PCR findings demonstrated malaria re-infection, making the PCR-adjusted day-42 ACPR 100% in the 144 patients who could be evaluated. Even when eight patients with missing PCR data were considered very conservatively as failures, the day-42 ACPR was over 94%. None of the participants, disease or treatment characteristics, including day-7 lumefantrine concentrations, predicted the risk of malaria recurrence by day 42. AL was well tolerated following administration. There were only two cases of grade 3 neutropaenia and one serious adverse event of lobar pneumonia, none of which was judged as probably related to intake of AL. CONCLUSIONS:AL was well tolerated and efficacious in treating uncomplicated falciparum malaria in HIV co-infected adults on efavirenz-based ART. However, a higher than anticipated proportion of participants experienced malaria re-infection, which highlights the need for additional malaria prevention measures in this sub-population after treatment with AL. Trial registration Pan African Clinical Trials Registry (PACTR): PACTR201311000659400. Registered on 4 October 2013. https://pactr.samrc.ac.za/Search.aspx.
\n \n\n \n \nMOTIVATION:Artificial intelligence, trained via machine learning (e.g. neural nets, random forests) or computational statistical algorithms (e.g. support vector machines, ridge regression), holds much promise for the improvement of small molecule drug discovery. However, small molecule structure-activity data are high dimensional with low signal-to-noise ratios and proper validation of predictive methods is difficult. It is poorly understood which, if any, of the currently available machine learning algorithms will best predict new candidate drugs. METHODS:The quantile-activity bootstrap is proposed as a new model validation framework using quantile splits on the activity distribution function to construct training and testing sets. In addition, we propose two novel rank-based loss functions which penalize only the out-of-sample predicted ranks of high activity molecules. The combination of these methods was used to assess the performance of neural nets, random forests, support vector machines (regression) and ridge regression applied to 25 diverse high quality structure-activity datasets publicly available on ChEMBL. RESULTS:Model validation based on random partitioning of available data favours models that overfit and 'memorize' the training set, namely random forests and deep neural nets. Partitioning based on quantiles of the activity distribution correctly penalizes extrapolation of models onto structurally different molecules outside of the training data. Simpler, traditional statistical methods such as ridge regression can outperform state-of-the-art machine learning methods in this setting. In addition, our new rank-based loss functions give considerably different results from mean squared error highlighting the necessity to define model optimality with respect to the decision task at hand. AVAILABILITY:All software and data are available as Jupyter notebooks found at https://github.com/owatson/QuantileBootstrap. SUPPLEMENTARY INFORMATION:Supplementary data are available at Bioinformatics online.
\n \n\n \n \nBACKGROUND:Plasmodium ovale curtisi and wallikeri are perceived as relapsing malarial parasites. Contrary to Plasmodium vivax, direct evidence for this hypothesis is scarce. The aim of this prospective study was to characterize the reappearance patterns of ovale parasites. METHODS:P. ovale spp. infected patients were treated with artemether-lumefantrine and followed biweekly for up to 1 year for the detection of reappearing parasitemia. Molecular analysis of reappearing isolates was performed to identify homologous isolates by genotyping and to define cases of relapse following predefined criteria. RESULTS:At inclusion, 26 participants were positive for P. ovale curtisi and/or P. ovale wallikeri. The median duration of follow-up was 35 weeks. Reappearance of the same P. ovale species was observed in 46% of participants; 61% of P. ovale curtisi and 19% of P. ovale wallikeri infection-free intervals were estimated to end with reappearance by week 32. Based on the predefined criteria, 23% of participants were identified with 1 or 2 relapses, all induced by P. ovale curtisi. CONCLUSION:These findings are in line with the currently accepted relapse theory inasmuch as the reappearance of P. ovale curtisi strains following initial blood clearance was conclusively demonstrated. Interestingly, no relapse of P. ovale wallikeri was observed.
\n \n\n \n \n