Expression of Concern: Antivirus effectiveness of ivermectin on dengue virus type 2 in Aedes albopictus (PLoS Neglected Tropical Diseases)

Abstract

After this article [1] was published, a concern was raised that for each experimental treatment (group), the triplicate data reported in Table 1 align quite closely with the group average, with low deviation and no outliers. In response, the authors noted that that they had repeatedly performed preliminary experiments to optimize conditions that yielded consistent results. The authors provided PCR data (S1 File) and analysis results (S2 File) to support Table 1. In reviewing this issue, PLOS Neglected Tropical Diseases obtained input from multiple members of our Editorial Board. All consulted editors questioned the feasibility of the Table 1 results. They advised that the table and supporting dataset report infection rates that are more consistent within and across cohorts for each dose than are expected; while the results are theoretically possible, the probability of obtaining the reported values is exceedingly low for this assay and biological system. It was also raised that based on published studies [2–5], ivermectin may inhibit replication but is not expected to remove all viral RNA present in the system prior to treatment; this has implications for the ‘Positive mosquito (n)’ (and corresponding negative mosquito) data reported in Table 1. In response to this, the corresponding author commented that per [2–5] DENV-2 virus nucleic acid will be present for a long time after death, but this cannot be considered as the degradation rate of nucleic acid in live mosquitoes as used in [1] because in living organisms active enzymes metabolize viral nucleic acid [6–9]. Overall, the consulted editors advised that there are concerns about the feasibility, integrity, and reliability of the Table 1 results which were not resolved by the data and comments received in post-publication discussions. Therefore, the PLOS Neglected Tropical Diseases Editors issue this Expression of Concern. In the post-publication assessment, it was also raised that detection of viral RNA is not sufficient to demonstrate whether actual virus is present. This study [1] did not include assays to detect live virus, and so results and conclusions about virus positivity, virus loads, and infection rates should be interpreted as referring specifically to the presence or absence of viral RNA which may not accurately indicate the presence or absence of virus. The authors acknowledged that this was a shortcoming of the study but commented that detection of viral RNA has been used as an indicator of viral replication in other published studies, hence in their view it was a valid indicator of the antiviral effect of ivermectin. There was an error in [1], in the sentence, “Adult mosquitoes aged between three and five days post emergence from larvae were used as experiment objects.” In this sentence, ‘larvae’ should be replaced with ‘pupae’.