Analysis of pharmaceutical compounds in cerebrospinal fluid (CSF) may present challenges due to the combination of the low protein content in this matrix and relatively low drug concentrations, often corresponding to free drug concentrations in plasma, typically found in CSF. A 30% loss of AMG 579 was observed during preparation of quality control samples and further investigation determined that this loss was likely due to binding to collection tubes. This observation also highlighted the possibility of additional losses of AMG 579 that could occur during collection of clinical samples, such as binding to catheters used in the collection of CSF.
Loss of AMG 579 in QC samples was reduced from 30% to 5% when the volume of CSF stored in 1.5 mL vials was increased from 0.06 mL to 1 mL. Modest but unavoidable losses of about 20% of AMG 579 were also found following perfusion through both silicone and polypropylene (Pharmed® BPT) collection catheters. Silicone tubing was used for CSF collection based on clinical site preference. An LC–MS/MS method was validated to quantify AMG 579 in human CSF to support clinical testing. The original range of the assay was 1–1000 ng/mL but the LLOQ was subsequently lowered to 0.1 ng/mL to better meet project requirements. Interday bias (% RE) and precision (% CV) were −4.2% and 12.3% at the LLOQ, and less than ±0.9% and 8.3% for higher concentrations, respectively. The compound was stable in human CSF for at least 5 h at room temperature, 55 days at −70 °C (−60 to −80 °C range), and through three freeze–thaw cycles.
Careful selection of assay conditions and materials minimized losses of the compound during sample collection and storage. While these losses could not be entirely eliminated, practical sample collection and storage conditions were established to allow for analysis of AMG 579 in human clinical trials.