Retrospective Analysis Using Pharmacokinetic/Pharmacodynamic Modeling and Simulation Offers Improvements in Efficiency of the Design of Volunteer Infection Studies for Antimalarial Drug Development

Volunteer infection studies using the induced blood stage malaria (IBSM) model have been shown to facilitate antimalarial drug development. Such studies have traditionally been undertaken in single‐dose cohorts, as many as necessary to obtain the dose‐response relationship. To enhance ethical and logistic aspects of such studies, and to reduce the number of cohorts needed to establish the dose‐response relationship, we undertook a retrospective in silico analysis of previously accrued data to improve study design. A pharmacokinetic (PK)/pharmacodynamic (PD) model was developed from initial fictive‐cohort data for OZ439 (mixing the data of the three single‐dose cohorts as: n = 2 on 100 mg, 2 on 200 mg, and 4 on 500 mg). A three‐compartment model described OZ439 PKs. Net growth of parasites was modeled using a Gompertz function and drug‐induced parasite death using a Hill function. Parameter estimates for the PK and PD models were comparable for the multidose single‐cohort vs. the pooled analysis of all cohorts. Simulations based on the multidose single‐cohort design described the complete data from the original IBSM study. The novel design allows for the ascertainment of the PK/PD relationship early in the study, providing a basis for rational dose selection for subsequent cohorts and studies.

Study Highlights

• WHAT IS THE CURRENT KNOWLEDGE ON THE TOPIC?

☑ Volunteer infection studies are routinely used in antimalarial drug development to generate early pharmacokinetic/pharmacodynamic data for compounds.

• WHAT QUESTION DID THIS STUDY ADDRESS?

☑ Can in silico analyses be used to suggest improvements to volunteer infection study designs?

• WHAT DOES THIS STUDY ADD TO OUR KNOWLEDGE?

☑ Multiple dose adaptive trial designs can potentially reduce the number of cohorts needed to establish the dose‐response relationship in volunteer infection studies.

• HOW MIGHT THIS CHANGE CLINICAL PHARMACOLOGY OR TRANSLATIONAL SCIENCE?

☑ Real time data analyses can be used to recommend doses for adaptive volunteer infection studies.

Volunteer infection studies using the induced blood stage malaria (IBSM) model have been recognized as a valuable system for defining the key pharmacokinetic (PK) and pharmacodynamic (PD) relationships for dose selection in antimalarial drug development. ¹ , ² , ³ , ⁴ , ⁵ , ⁶ , ⁷ In such studies, healthy volunteers are inoculated intravenously with a given quantity (with small variability) of Plasmodium‐infected red cells. Parasitemia is then followed by quantitative polymerase chain reaction until a prespecified treatment threshold is reached when the test drug is administered. Parasite and drug concentrations are then measured. These studies are conducted prior to phase II dose‐response (D‐R) trials and can be included in an integrated first‐in‐human study protocol, or after completion of the first‐in‐human PK and safety study. IBSM studies have been typically designed as flexible multiple cohort studies where each volunteer of one cohort receives a single dose of the same amount of drug (“single dose per cohort”). ² , ³ , ⁴ , ⁵ After each cohort, a decision is made to stop or to add a cohort to test a lower or higher dose based on the response observed in the previous cohorts.For the multiple single‐dose‐per‐cohort design, the starting dose is typically selected based on safety and PK information from a phase I single ascending dose (SAD) study and, more recently, on preclinical data from a severe combined immunodeficient mouse model, with the dose selected on the basis of being best able to inform the D‐R relationship, rather than aiming for cure. This approach, where a single dose is tested in all subjects of the initial cohort, risks missing the dose likely to be most informative for defining the PK/PD relationship.An alternative approach is to spread a range of doses across a smaller number of subjects within the initial cohort and use PK/PD models developed based on data from this cohort to support dose selections of subsequent cohorts and studies. Using data from a previous study, ² we undertook an in silico investigation of such an adaptive study design, aiming to reduce the number of subjects exposed to inefficacious doses, and to establish a D‐R relationship. This multiple‐dose‐groups‐per‐cohort design, referred to as the “2‐2‐4” design, is contrasted with the already implemented study design depicted in Figure  1 .Open in a separate windowFigure 1Comparison of standard and adaptive designs of IBSM studies. A/B/C, dose levels to be selected during the progress of the study based on pharmacokinetic/pharmacodynamic results of the initial cohort; CHMI, controlled human malaria infection; D‐R, dose‐response; IBSM, induced blood stage malaria infection; n, number of subjects at each dose.The objectives of this retrospective analysis were to: (i) compare PK/PD parameter estimates from the initial cohort of the 2‐2‐4 study design with the prior results from the data of the full study and (ii) propose a preliminary workflow to establish D‐R early in an IBSM study, and use modeling and simulation (M&S) to support dose selections for subsequent cohorts and later phase clinical trials.     

Clinical Practice Guidance: Surveillance for phaeochromocytoma and paraganglioma in paediatric succinate dehydrogenase gene mutation carriers

The succinate dehydrogenase (SDH) enzyme complex functions as a key enzyme coupling the oxidation of succinate to fumarate in the citric acid cycle. Inactivation of this enzyme complex results in the cellular accumulation of the oncometabolite succinate, which is postulated to be a key driver in tumorigenesis. Succinate accumulation inhibits 2‐oxoglutarate‐dependent dioxygenases, including DNA and histone demethylase enzymes and hypoxic gene response regulators. Biallelic inactivation (typically resulting from one inherited and one somatic event) at one of the four genes encoding the SDH complex (SDHA/B/C/D) is the most common cause for SDH deficient (dSDH) tumours. Germline mutations in the SDHx genes predispose to a spectrum of tumours including phaeochromocytoma and paraganglioma (PPGL), wild type gastrointestinal stromal tumours (wtGIST) and, less commonly, renal cell carcinoma and pituitary tumours. Furthermore, mutations in the SDHx genes, particularly SDHB, predispose to a higher risk of malignant PPGL, which is associated with a 5‐year mortality of 50%. There is general agreement that biochemical and imaging surveillance should be offered to asymptomatic carriers of SDHx gene mutations in the expectation that this will reduce the morbidity and mortality associated with dSDH tumours. However, there is no consensus on when and how surveillance should be performed in children and young adults. Here, we address the question: “What age should clinical, biochemical and radiological surveillance for PPGL be initiated in paediatric SDHx mutation carriers?”.     

Examination of potential novel biochemical factors in relation to prostate cancer incidence and mortality in UK Biobank

Background Although prostate cancer is a leading cause of cancer death, its aetiology is not well understood. We aimed to identify novel biochemical factors for prostate cancer incidence and mortality in UK Biobank.Methods A range of cardiovascular, bone, joint, diabetes, renal and liver-related biomarkers were measured in baseline blood samples collected from up to 211,754 men at recruitment and in a subsample 5 years later. Participants were followed-up via linkage to health administrative datasets to identify prostate cancer cases. Hazard ratios (HRs) and 95% confidence intervals were calculated using multivariable-adjusted Cox regression corrected for regression dilution bias. Multiple testing was accounted for by using a false discovery rate controlling procedure.Results After an average follow-up of 6.9 years, 5763 prostate cancer cases and 331 prostate cancer deaths were ascertained. Prostate cancer incidence was positively associated with circulating vitamin D, urea and phosphate concentrations and inversely associated with glucose, total protein and aspartate aminotransferase. Phosphate and cystatin-C were the only biomarkers positively and inversely, respectively, associated with risk in analyses excluding the first 4 years of follow-up. There was little evidence of associations with prostate cancer death.Conclusion We found novel associations of several biomarkers with prostate cancer incidence. Future research will examine associations by tumour characteristics.Subject terms: Predictive markers, Prostate cancer, Risk factors