P562: capillary blood sampling allows feasible method for venetoclax concentration measurement in an academic multicenter clinical trial context

Background: BCL-2 inhibitor venetoclax has revolutionized the therapy of acute myeloid leukemia (AML) patients unfit for chemotherapy. Venetoclax is cleared from the body mainly by hepatic metabolism (CYP3A4) and its urinary excretion is minimal. Wide variability in venetoclax pharmacokinetics exists between patients. Nevertheless, in previous studies the clinical factors including age, gender, weight, kidney or liver function did not correlate with venetoclax plasma concentration. Drug concentration measurement from plasma is challenging in academic multicenter clinical trial context due to the sample preservation requirements. Only limited data exists on using commercially available devices for capillary blood sampling for drug concentration. Aims: We collected capillary blood for venetoclax trough concentration (Cmin) analysis within our VenEx trial (NCT04267081). For validation, we collected plasma blood samples from a subset of the patients. Our aim was to show the capillary blood method for venetoclax concentration measurements is reliable in an academic multicenter clinical trial context. Methods: VenEx - a phase 2 prospective multicenter trial conducted by Finnish Leukemia Group - recruited 104 participants with de novo or relapsed/ refractory (R/R) AML unfit for chemotherapy. Informed consent was obtained from all participants. The first three 28-day cycles of the therapy consisted of azacytidine (75mg/m2 for 7 days) and venetoclax (400mg daily for 21-28 days). Venetoclax was started with a ramp-up of 100-200-400mg during the first three days. We collected fingertip capillary blood with VAMS, Mitra® device (Neoteryx) (N=148) for Cmin measurements in C1d15 and C2d15. To validate the use of capillary concentrations, we collected parallel blood plasma samples from 36 patients in C1d15. Capillary samples were stored at room temperature until evaluated. The analysis of the extracted samples was performed using an LC-MS-MS system, comprising a liquid chromatography equipment (Shimadzu Nexera X2, UHPLC system) and an API3000 mass spectrometer (Sciex) with a turbo ion spray source. We also collected DNA for analysis of the CYP3A4*2 (c.666T>C), CYP3A4*22 (c.522-191G>A), and CYP3A5*3 (c.219-237A>G) alleles. Results: The C1d15 capillary blood Cmin correlated excellently (R2 0.835, p<0.001) with the parallel plasma concentration (fig 1A) demonstrating the feasibility of the assay. In the whole population, median capillary blood Cmin was 720 (58-12600) ng/L in C1d15 and 846 (134-9790) ng/L in C2d15. There was no significant correlation between age (R2 0,017, p = 0.115) or weight (R2 0,003, p = 0.486) and Cmin on d15. The mean Cmin at d15 did not differ significantly between genders (two-tailed T-test; C1d15 p = 0.593, C2d15 p= 0.471). The median Cmin in C1d15 was similar between the patients with no serious adverse events (SAE) and those with one or more SAE during C1 (fig 1B). Interestingly, one patient in the cohort was homozygous for the reduced function CYP3A4*22 allele and a had a very high Cmin of 12 600 ng/L in C1d15. However, in this cohort, carriers of the CYP3A4*22 allele did not have a significantly higher Cmin, compared to noncarriers. Summary/Conclusion We demonstrated the capillary venetoclax concentration method is reliable and allows feasible pharmacokinetic sampling and concentration measurement in a multicenter academic trial context. In concordance with the previous studies the venetoclax Cmin did not correlate with the patient demographics nor affected on the occurrence of SAE. Our finding of exceptionally high venetoclax concentration in a patient homozygous for CYP3A4*22 suggests that in rare circumstances, CYP3A4 variants may have a relevant effect on venetoclax pharmacokinetics.Keywords: Acute myeloid leukemia, Pharmacogenetics, Pharmacokinetic, Venetoclax