PAGE. Abstracts of the Annual Meeting of the Population Approach Group in Europe.
PAGE 28 (2019) Abstr 9158 [www.page-meeting.org/?abstract=9158]
Click to open
Oral: Methodology - New Modelling Approaches
Mohammed H. Cherkaoui (1), Nicole Andenmatten (1), Rolf Fendel (2), Lydia Burgert (3,4), Chiara Fornari (5), Michael Gabel (6), Oluwaseun F. Egbelowo (7), John Ward (8), Joerg Moehrle (1), Nathalie Gobeau (1)
(1) Medicines for Malaria Venture, Geneva, Switzerland, (2) University of Tübingen, Tübingen, Germany, (3) Swiss Tropical & Public Health, Basel, Switzerland, (4) University of Basel, Basel, Switzerland, (5) AstraZeneca, Cambridge, UK, (6) University of Heidelberg, Heidelberg, Germany, (7) University of Witwatersrand, Johannesburg, South Africa, (8) University of Loughborough, Loughborough, UK
Introduction: Chemoprotective drugs are developed to protect people from being infected by plasmodium, and in particular by P. falciparum, the most prevalent and deadly malaria strain in Africa. MMV’s stretched goal is to develop a treatment able to protect people for one month after a single dose. It is thus important to understand how a drug kills parasites not only in the blood stage, which causes symptoms, but also in the liver stage, where the infection starts whilst remaining asymptomatic.
Whilst PKPD models have been developed to understand the blood-stage activity of antimalarial candidates to cure patients, little modelling has been proposed for liver-stage activity. What makes modelling liver-stage challenging is that the liver-stage parasites, unlike blood-stage parasites, cannot be counted.
Objective: Develop a PKPD model to describe the drug killing effect on both the liver and blood-stage parasites. The example of DSM265, a plasmodial dihydroorotate dehydrogenase (DHODH) inhibitor, is chosen to illustrate the approach.
Methods: A mathematical model was developed and consists of two ordinary differential equations which describes the dynamic of the liver and blood-stage parasites, respectively. Each equation includes a net growth rate and a drug killing rate specific to each stage. One term accounted for the release of the parasites from liver to blood-stage. Since liver-stage parasitemia cannot be monitored, the activity on liver-stage had to be deconvoluted from the knowledge of the blood-stage activity. Therefore, the estimation of the PKPD parameters was conducted in four steps; (i) parasite growth in the blood, (ii) parasite growth in the liver, (iii) drug activity in the blood and (iv) drug activity in the liver.
Two studies in which volunteers were injected infected red blood cells then administered DSM265 at 150mg and 400mg, respectively, were used to determine (iii) . Then two studies, in which volunteers were administered a dose of 400 mg DSM265 administered 1, 3 or 7 days prior to infecting them with an i.v. injection of 3200 sporozoites or 5 mosquito bites, were used to deduce (i), (ii) and (iv) . In these studies, the parasites invade first the liver, then the red blood cells. In all four Volunteer Infected Studies (VIS), PK concentrations and blood-stage parasitemia were measured in each individual.
All parameter estimations were conducted with Monolix (v2018R2), an NLME modelling software. When the number of subjects was limited, some of the inter-individual variability parameters were fixed. First, the blood-stage parameters were estimated with the blood-stage VIS data; then they were fixed, and the liver-stage parameters were estimated with the liver-stage VIS data.
Finally, a sensitivity analysis was conducted to identify which parameters are key in protecting people with DSM265. Moreover, to validate the model, simulations of the liver-stage VIS were conducted to compare the predicted fraction of subjects with breakthrough with the observations.
Results: The growth of the parasite in blood was better described by a cyclic model, with a growth rate estimated at 0.064 hr-1 and a period at 44.9 hours. For the liver-stage parasites, the growth rate was calculated to be 0.063 hr-1 and the fraction of viable sporozoites estimated to be 0.2% corresponding to 6 sporozoites invading hepatocytes. In comparison, it was estimated that about 35% of the 50-100 sporozoites injected after a mosquito bite reached the blood stream and could potentially invade hepatocytes . The blood-activity analysis led to the estimation of a Minimum Inhibiting Concentration (MIC) of 1180ng/mL against blood infection, whereas the liver-activity analysis led to a MIC against liver infection of 2440ng/mL. The sensitivity analysis showed that the chemo- protectiveness of DSM265 is sensible to its liver-activity, and to a lesser extent to its blood-activity. Finally, the simulations showed that the chemoprotection PKPD model could reproduce the liver-stage VIS results, despite the limited number of volunteers.
Conclusion: In conclusion, combining liver and blood-stage VIS made it possible to develop a promising PKPD model that can describe the activity of an antimalarial drug on both liver and blood-stages. Nevertheless, further analyses and more studies are needed to validate the model. Hopefully, this will help select better dosing regimen for chemoprotection to be tested in phase II studies.