Efficacy of praziquantel has been maintained over four decades (from 1977 to 2018): A systematic review and meta-analysis of factors influence its efficacy

Abstract

Background: The antihelminthic drug praziquantel has been used as the drug of choice for treating schistosome infection for more than 40 years. Although some epidemiological studies have reported low praziquantel efficacy in cure rate (CR) and/or egg reduction rate (ERR), there is no consistent robust evidence of the development of schistosome resistance to praziquantel (PZQ). There is need to determine factors that lead to variable treatment CR and/or ERR. Therefore, we conducted a systematic review and meta-analysis to review CR and ERR as well as identify their predictors.

Methodology/principal findings: In this systematic review and meta-analysis, a literature review was conducted using Biosis Citation Index, Data Citation Index, MEDLINE, and Web of Science Core Collection all of which were provided through Web of Science. Alongside these, EMBASE, and CAB abstracts were searched to identify relevant articles. Random effect meta-regression models were used to identify the factors that influence CR and/or ERR by considering differences in host characteristics and drug dose. In total, 12,127 potential articles were screened and 146 eligible articles (published from 1979 to 2020) were identified and included for the meta-analysis. We found that there has been no significant reduction in CR or ERR over the study period. The results showed more variability in CR, compared with ERR which was more consistent and remained high. The results showed a positive effect of "PZQ treatment dose" with the current recommended dose of 40 mg/kg body weight achieving 57% to 88% CR depending on schistosome species, age of participants, and number of parasitological samples used for diagnosis, and ERR of 95%.

Conclusions/significance: Based on a review of over 40 years of research there is no evidence to support concerns about schistosomes developing resistance to PZQ. These results indicate that PZQ remains effective in treating schistosomiasis.

Fig 1. A systematic review flow diagram.

Diagram of the number of articles identified and examined at each stage of the review. A total of 146 articles published from 1979 to 2020 met all inclusion criteria and were included for the meta-analysis.

Fig 2. The distribution of reported parasitological cure rates and egg reduction rates following praziquantel treatment for (A) cure rate, 297 observations from 132 articles and (B) egg reduction rate, 240 observations from 108 articles.

Fig 3. Fitted graphs for predictors identified from a random effects meta-regression model.

Identified predictors’ effects on cure rate (graph A, B, C and D) and egg reduction rate (graph E): (A) praziquantel treatment dose over the range 10–60 mg/kg body weight on cure rate (B) the number of parasitological samples used for a diagnosis over the range 1–12 samples on cure rate (C) age category child (0 to 19 years old) and adults (20 years old or older) (D) parasite species S. mansoni and S. haematobium (E) praziquantel treatment dose over the range 10–60 mg/kg body weight on egg reduction rate. Scatter plot graphs A, B, and E: Data points indicate reported cure rate/egg reduction rate for each observation. Negative fraction indicates that praziquantel treatment was associated with increase of schistosome infection intensity. Lines are fitted graphs generated from random effects meta-regression. Scatter plot graph A and B: Dashed lines denote the highest level of cure rate over range that could be achieved among adults with S. mansoni infection treated with 60 mg/kg body weight praziquantel (for graph B) and one parasitological sample used for a diagnostic (for graph A). Heavy lines denote the cure rate over the reported dose/number of samples using mode of other predictors, i.e., children infected with S. mansoni and treated with 40 mg/kg body weight praziquantel treatment dose, diagnosed by one parasitological sample. Box plot C and D: Boxes denote range of possible maximum and minimum cure rate (%) estimated by models with heavy lines denoting model estimated cure rate using mode of predictors. Dashed lines denote the range of raw reported cure rate by each observation.

Fig 4

Scatter graphs of the reported cure rate (graph A) and egg reduction rate (graph B) by treatment year. Each plot represents the reported cure rate/egg reduction rage of S. haematobium in blue and S. mansoni in orange. The size of each plot represents the number of participants of each observation. Lines are fitted graphs generated from random-effects meta-regression models. Heavy lines denote the cure rate or egg reduction rate of S. mansoni infection. Dashed lines denote the cure rate and egg reduction rate of S. haematobium infection. For graph A: fitted lines denote the cure rate over the reported treatment years that could be achieved among children treated with 40 mg/kg bodyweight praziquantel and one parasitological sample used for a diagnostic. Similarly, for graph B, fitted lines denote the egg reduction rate with 40 mg/kg body weight praziquantel treatment.

Fig 5. Scatter graph of the reported egg reduction rate in percentage vs pre-treatment infection intensity (eggs/10ml urine for S. haematobium in blue circle and eggs/g of feces for S. mansoni in orange diamond).

The fitted line is from a random effects meta-regression model which confirmed statistically significant association between pre-treatment infection intensity and egg reduction rate [F = 11.257 (1, 118), p = 0.001]. For this model both schistosome species and type of pre-treatment infection intensity mean (arithmetic and geometric mean) were considered. A total of 134 observations from 70 articles were used for this analysis.

Fig 6. Scatter graph of the reported egg reduction rate in percentage (%) vs cure rate in observations that used 40 mg/kg body weight praziquantel treatment dose (163 observations from 87 articles).

The fitted line is from a random effects meta-regression model which confirmed statistically significant association between cure rate and egg reduction rate [F = 20.097 (1, 147), p<0.001].