To the best of our knowledge, this is the first study assessing a possible association between neutropenia and ACT treatment in pregnant women infected with malaria. We report that although AL has good efficacy in eliminating existing parasites, its prophylactic efficacy declines in pregnant women with neutropenia compared to pregnant women without neutropenia. However, our data did not reveal any link between neutropenia and reduced prophylactic effectiveness of neither DP nor PA following the treatment of malaria infection in pregnant women.
At least 63% of the pregnant women in this analysis had neutropenia before initiation of malaria treatment. This finding is consistent with our observations of children in Mali, Burkina and Guinea with malaria [9]. Previous studies have shown that neutropenia is quite common in the African general population [16, 17] and this study confirms this finding in the Malian pregnant population with malaria.
The findings in this study that show that the initial density of P. falciparum parasites is lower in pregnant women with neutropenia is in line with previous findings. A study in children in Mali, Burkina-Faso and Guinea showed that parasitemia in patients with neutropenia was lower compared to non-neutropenic children [9] and another study carried out along Africa by Olliaro et al., in 2011 showed a positive association between neutrophil levels and parasite density [18]. If the neutropenia is the cause of higher parasite density or the effect of the parasite density remains unclear but the association between the two was also described by Bostrom et al. They showed that the number of neutrophils in peripheral blood was reduced in pregnant women infected with P. falciparum compared to pregnant women not infected with malaria [19].
Next to the association between neutropenia and parasitemia this study also found an association between neutropenia and ACT treatment. Pregnant women with neutropenia at enrolment had a higher rate of clinical and parasitological clearance compared to women who were non neutropenic. In addition, it was shown that of the three ACTs studied, AL is the least protective against P. falciparum reappearance in neutropenic pregnant women. However, our data in children show that in addition to AL, DP and PA also have reduced prophylactic efficacy in cases of neutropenia [9].
A study conducted by the West African Network for Clinical Trials of Antimalarial Drugs (WANECAM) in Mali, Burkina-Faso and Guinea [20] revealed that for the treatment of episodes of uncomplicated P. falciparum malaria, parasite clearance time was slower with AL than with PA, and similarly slower with AL than with DP. Further analysis showed a higher risk of reinfection with P. falciparum during a 42-day follow-up with AL compared with PA and DP [20].
Although at Day 28, AL was less protective against new infections in pregnant women with neutropenia, we did not find statistically significant association between the re-emergence of parasites during a 63-day follow-up period and the drug used, or the level of urbanization of the pregnant woman’s residence according to neutrophil baseline status. This leads us to speculate that neutrophils, generally dying within 8 to 12 hours of entering the peripheral bloodstream [19], are more useful for predicting a short-term re-emergence (not exceeding 28 days) of malaria parasites after treatment.
The association between neutrophil counts and ACT efficacy was seen only when the pretreatment levels were assessed. The effect was not seen at neutrophil counts at day 7. What was seen is an increase of women that could be classified as neutropenic at day 7. This is in line with previous findings that show that several antimalarial drugs including ACTs can cause secondary neutropenia [21–23]. This may explain the lack of association found in this study as the group of women with normal or increased neutrophil levels was too small to detect statistically significant differences between of ACPR and with therapeutic failure (ETF, LPF or LCF), whatever the treatment arm.
One of the limitations of this study is that we do not know the neutrophil status of pregnant women before they are infected with P. falciparum. Another limitation of this study is the under-representation of pregnant women without neutropenia compared to those with neutropenia. A study with a larger number of women without neutropenia could reinforce our results.
In conclusion, this study has for the first time shown that lower pre-treatment neutrophil levels in peripheral blood tend to have a deleterious effect on the effectiveness of AL in pregnant women, while malaria infection leads to an increase in the neutrophil levels in pregnant women. As a result, treating malaria infection in neutropenic pregnant women could compromise the prophylactic effectiveness of AL in vulnerable populations such as pregnant women in areas where malaria transmission is permanent or long-standing.
Thus, if the pregnant woman infected with malaria is known to be neutropenic, PA or DP would be good alternatives for their treatment.