From stratified to individualized precision dosing: Development of a liquid biopsy technique for measuring hepatic drug-metabolizing enzymes

Activity: Talk or presentationOral presentation


Background: Precision dosing aims to deliver the right drug dose for a specific patient based on individual characteristics in order to improve efficacy and reduce toxicity. ‘Liquid Biopsy’ is expected to define the metabolic capacity of an individual patient beyond the typical genotyping and hence facilitate precision dosing by linking the administered dose to expected clearance. Methods: Liver tissue (20-250 mg) and plasma (1-3 mL) samples from the same cancer patients (n=9) were analysed using proteomic (targeted and global) and transcriptomic (next generation sequencing, NGS) methods. Healthy controls (n=5) were used as baseline. In-house QconCAT [1] methodology was used for protein quantification on nanoHPLC-Orbitrap Elite system (Thermo Fisher) with an inclusion list focusing on drug-metabolizing enzymes. NGS followed Ampliseq workflow [2] performed at Life Technologies (Thermo Fisher Scientific, Texas) at a depth of 8 million reads/mL. Expression levels of drug-metabolising enzymes were normalized using a novel liver-specific shedding correction factor (LSCF), computed using a combination of 14 liver-specific marker genes measured in plasma to offset variability in the level of liver exosomal shedding. Results: Proteins (~2500) were measured, including key drug-metabolising enzymes (the targets), in non-cancerous liver samples from cancer patients and controls, alongside RNA transcripts (~21000) in matched plasma. Coverage of targets and markers was 80-100% in the protein data and 64-100% in RNA data. Exosome shedding was higher and more variable in cancer patients than healthy controls; LSCF(cancer)=26.08±19.63 reads per million (rpm), n=9; LSCF(healthy)=0.83±0.26 rpm, n=5; t-test p<0.01). Tissue protein and LSCF-corrected plasma RNA levels were assessed for correlation; major drug-metabolising enzymes were significantly correlated between plasma (RNA) and liver (protein); CYP3A4 (Pearson r=0.98, p<0.01, R2=0.95); CYP2C9 (r=0.76, p=0.03, R2=0.57); CYP1A2 (r=0.93, p=0.02, R2=0.86); CYP2A6 (r=0.98, p<0.01, R2=0.96) Conclusion: A liquid biopsy test based only on plasma samples to determine liver content of key enzymes was established; this was relevant to: (a) enzymes exclusively/predominantly expressed in liver and (b) predominantly shed into the blood. This technology should facilitate efforts towards precision dosing, as an essential element of precision medicine. [1] Achour B, Russell MR, Barber J, Rostami-Hodjegan A. Simultaneous quantification of the abundance of several cytochrome P450 and uridine 5'-diphospho-glucuronosyltransferase enzymes in human liver microsomes using multiplexed targeted proteomics. Drug Metab Dispos 2014; 42: 500-10. [2] Millat G, Chanavat V, Rousson R. Evaluation of a new NGS method based on a custom AmpliSeq library and Ion Torrent PGM sequencing for cardiomyopathies. Clinica Chimica Acta 2014; 433: 266-71.
17 Dec 2019

Event (Conference)

TitlePharmacology 2019 by British Pharmacological Society (BPS)
LocationEdinburgh International Conference Centre
CountryUnited Kingdom
Degree of recognitionInternational event