Background: Malaria in pregnancy is an important public health problem in sub Saharan Africa. It is known to be the most common and preventable cause of harmful birth outcomes in malaria endemic areas. It is therefore important for a pregnant woman to be treated with safe and effective antimalarial medication. Drug safety in pregnancy is of a greater concern due to limited safety data available in this vulnerable group. This is because pregnant women are not involved in clinical trials related to drug development process due to safety reasons and hence, most of these medicines come to market with limit information available about their safety in pregnancy. Hence, establishing a drug safety monitoring mechanism would be important to generate safety data when a given medicine is already in the market, especially medications against tropical diseases. Pregnant women are at increased risk of malaria infection and illness than non-pregnant individuals due to physiological, hormonal and immunological changes that occur in their body after conception. The changes are also responsible for various therapeutic challenges that face this vulnerable group. This explains the presence of significant alteration of antimalarial pharmacokinetic (PK) properties in pregnancy and hence lead to a reduced drug blood concentration, which will ultimately lower antimalarial cure rate. Another factor that affects antimalarial effectiveness in pregnancy is parasite resistance against sulfadoxine-pyrimethamine (SP), a drug that is used for intermittent preventive treatment of malaria in pregnancy (IPTp). The objectives of the thesis were to assess the magnitude of drugs exposure during pregnancy in relation to pregnancy outcomes, to describe the feasibility of establishing active pharmacovigilance system in developing countries using Health Demographic Surveillance System (HDSS) platform, to determine safety of artemether-lumefantrine (AL) exposure in first trimester of pregnancy, to evaluate pharmacokinetics and pharmacodynamics properties of artemether-lumefantrine in pregnant and nonpregnant women, and to determine the effectiveness of IPTp-SP in prevention of placental malaria, maternal anaemia and low birth weight in areas with different malaria transmission intensity. 

 Method: Three different study designs were used independently to respond to different specific objectives of this thesis; (i) a longitudinal follow up study was conducted to generate artemether/lumefantrine (AL) safety data in first trimester secondary to its inadvertent exposure in two Health Demographic Surveillance System (HDSS) areas in Tanzania. Pregnant women with gestational age ≤ 20 weeks were enrolled and followed up on monthly bases until delivery. Drugs exposures during the entire pregnancy period were also recorded. The latter was used to document the feasibility of establishing active pharmacovigilance system using HDSS platform in one of the studied HDSS area. (ii) To determine AL PK, a prospective study involving pregnant in second and third trimester and nonpregnant women, both with uncomplicated P falciparum malaria. Plasma samples were collected at predefined dates for bioassay to determine drug level. Participants were followed up on pre-defined schedule visits until day 42. Inter- and intra-individual variability was assessed and covariated effects quantified using a nonlinear mixed-effect modeling approach (NONMEM®). (iii) Another prospective study enrolling pregnant women to assess the effectiveness of IPTp in two areas with different malaria transmission intensity. Pregnant women were recruited in the labor ward and structured questionnaire was used for interview. Placental parasitaemia was screened by using both light microscope and realtime quantitative PCR.

FindingsPharmacovigilance system91% (994 of 1089) of pregnant women who were piloted to assess feasibility of establishing active PVsystem completed the follow up until delivery. 98% of pregnant women reported to have taken at least one medication during pregnancy, mainly drugs provided in the antenatal program. Other mostreported drugs were analgesics (24%), antibiotics (17%) and antimalarials (15%), excluding IPTp. Ironand folate supplementations were associated with decreased risk of miscarriage/stillbirth (OR 0.1; 0.08 – 0.3). AL safety82% (1783 of 2167) of pregnant women who used and not used antimalarial drugs in first trimesterwere followed until delivery and recorded their pregnancy outcome. 319 (17.9%) used antimalarialdrugs in first trimester and AL was the most frequent antimalarial used [53.9% (172 of 319)]. Otherswere 24.4 % quinine, 20.7% SP and 3.4% amodiaquine. Quinine exposure in first trimester wasassociated with increased risk of miscarriage/stillbirth (OR 2.5; 1.3 – 5.1) and premature birth (OR 2.6;1.3 – 5.3). AL, SP and amodiaquine exposure were found not to be harmful.PK analysis33 pregnant women and 22 non-pregnant women with malaria were treated with AL (80/480mg) twicedaily for 3 days. Lumefantrine (LF) bioavailability and metabolism rate into desmethyl-lumefantrinewere respectively 34% lower and 78% higher in pregnant than in non-pregnant patients. Overall PCRuncorrected therapeutic failure was 18% in pregnant and 5% in non-pregnant women (OR 4.0; p value 0.22). A higher median day 7 LF concentration was associated with adequate clinical and parasitological response.Effectiveness of IPTp350 pregnant women were recruited and screened for placental parasitaemia (175 each from high andlow malaria transmission areas). Prevalence of placenta parasitaemia was 16.6% in high transmissionarea and 2.3% in low transmission area. One or more doses of IPTp in high transmission area had 80%impact against placental malaria (OR 0.2; CI 0.06 – 0.7; p=0.015) and 60% in low transmission (OR 0.4; CI0.04 – 4.5; p=0.478). Primigravida and residing in high transmission area were significant risk factors for placental malaria (OR 2.4; CI 1.1 – 5.0) and (OR 9.4; CI 3.2 – 27.7), respectively. The numbers needed to treat (NNT) was 4 (CI 2 – 4) women in high transmission area and 33 (CI 20 – 50) low transmission area to prevent one placental malaria. IPTp use was not statistically significant associated with decreased risk of maternal anaemia or low birth weight, regardless are of transmission intensity.Conclusion:Overall medicine use in pregnancy period is very high, including AL exposure in first trimester albeit this drug is not the first line treatment for malaria in early pregnancy. AL use in first trimester was safer as opposed to quinine, the first line drug which was associated with adverse pregnancy outcomes. We therefore recommend to consider other options than quinine for standard antimalarial drug in first trimester, and AL could be the best one. HDSS platforms represent a reliable and feasible support to build on a pharmacovigilance system to assess safety of drugs in pregnancy since it has proved to be feasible. We recommend that pharmaceutical companies and other global financial bodies should invest more on the establishment of active pharmacovigilance system in pregnancy in tropical developing countries. The latter will boost safety data pool of newly marketed medicines and anti-infective agents for treating different illnesses in pregnancy. LF bioavailability is significantly lowered in pregnant women due to altered PK properties as opposed to non-pregnant women in the same area. This may be responsible for therapeutic failure among pregnant women secondary to the observed low post-treatment prophylaxis. We recommend to evaluate a modified treatment regimen of malaria in pregnancy.