Longitudinal changes in iron homeostasis in human experimental and clinical malaria

Abstract

Background: The interaction between iron status and malaria is incompletely understood. We evaluated longitudinal changes in iron homeostasis in volunteers enrolled in malaria volunteer infection studies (VIS) and in Malaysian patients with falciparum and vivax malaria.

Methods: We retrieved data and samples from 55 participants (19 female) enrolled in malaria VIS, and 171 patients (45 female) with malaria and 30 healthy controls (13 female) enrolled in clinical studies in Malaysia. Ferritin, hepcidin, erythropoietin, and soluble transferrin receptor (sTfR) were measured by ELISA.

Findings: In the VIS, participants' parasitaemia was correlated with baseline mean corpuscular volume (MCV), but not iron status (ferritin, hepcidin or sTfR). Ferritin, hepcidin and sTfR all increased during the VIS. Ferritin and hepcidin normalised by day 28, while sTfR remained elevated. In VIS participants, baseline ferritin was associated with post-treatment increases in liver transaminase levels. In Malaysian patients with malaria, hepcidin and ferritin were elevated on admission compared to healthy controls, while sTfR increased following admission. By day 28, hepcidin had normalised; however, ferritin and sTfR both remained elevated.

Interpretation: Our findings demonstrate that parasitaemia is associated with an individual's MCV rather than iron status. The persistent elevation in sTfR 4 weeks post-infection in both malaria VIS and clinical malaria may reflect a causal link between malaria and iron deficiency.

Funding: National Health and Medical Research Council (Program Grant 1037304, Project Grants 1045156 and 1156809; Investigator Grants 2016792 to BEB, 2016396 to JCM, 2017436 to MJG); US National Institute of Health (R01-AI116472-03); Malaysian Ministry of Health (BP00500420).

Keywords: Controlled human malaria infection; Hepcidin; Iron; Iron deficiency; Malaria; Volunteer infection studies.

Fig. 1

Longitudinal changes in the markers of iron metabolism in 55 volunteers experimentally infected with P. falciparum. Data points represent the means, and error bars the 95% confidence intervals. A repeated measures ANOVA was performed on log₁₀ transformed data which were back-transformed to original scale for presentation. Displayed P values have been adjusted using the Bonferroni correction to account for the 6 pairwise comparisons performed. Ferritin was adjusted as per the BRINDA project.^,

Fig. 2

Longitudinal changes in the markers of liver transaminases in 55 volunteers experimentally infected with P. falciparum. Data points represent the means, and error bars the 95% confidence intervals. A repeated measures ANOVA was performed on log₁₀ transformed data which were back-transformed to original scale for presentation. Displayed P values have been adjusted using the Bonferroni correction to account for the 6 pairwise comparisons performed.

Fig. 3

Longitudinal changes in the markers of iron metabolism in Malaysian patients with P. falciparum (n = 109) and P. vivax (n = 62) malaria. Data points and error bars represent marginal means and 95% confidence intervals, respectively. Data were analysed using a linear mixed effects model, as day 7 samples were available for only 33/109 patients with falciparum malaria, and 56/62 patients with vivax malaria. Non-normally distributed variables were log₁₀ transformed, and back-transformed for presentation. Displayed P values have been adjusted using the Bonferroni correction to account for the 6 pairwise comparisons performed. sTfR, soluble transferrin receptor.