(T1230-02-11) Sensitive Quantitation of Antisense Oligonucleotides Using Capitainer^® qDBS Microsampling Device Coupled with Hybridization LC-MS/MS

Purpose: Novel finger-stick derived capillary blood collection, such as that supported by Capitainer’s quantitative dried blood spot (qDBS) microsampling device, has recently emerged as a viable alternative to traditional venipuncture. Not only less invasive, microsampling facilitates a patient-centric approach in allowing remote/ at-home collection of small volume blood samples (typically < 20 mL). However, there is limited data detailing the applicability of dried blood as a matrix suitable for the analysis of antisense oligonucleotides (ASOs), in part due to the lack of selectivity in traditional chemical extraction methods, but also as a result of compromised sensitivity from low sample volume coupled with poor extractability. In the current research, we present a sensitive and selective workflow that combines microsample homogenization with a hybridization LC-MS/MS approach for the simultaneous quantitation of a representative 21-mer ASO, fomivirsen (FME), and its 5' n-1 metabolite ranging from 0.50 – 500 ng/mL collected using the 10 µL Capitainer qDBS device. Additionally, we establish a bioanalytical procedure for FME and the n-1 metabolite in human plasma, critical for establishing concordance with blood.

Methods: Human blood samples fortified with FME and n-1 metabolite were absorbed onto the 10 µL Ahlstrom 222 pre-cut cellulose substrate of the Capitainer qDBS device and allowed to air-dry in the presence of desiccant for a minimum of 24 hours. Dried blood qDBS microsamples were carefully detached and placed in a 96-well plate, to which a 5/32” stainless steel grinding bead was added, followed by adding the n+2 analogue of FME as an internal standard (IS). Dried blood microsamples were homogenized in the presence of surfactant, followed by affinity purification of the ASO-Probe complex using functionalized magnetic beads. After multiple wash steps, the targeted ASO was released from the capture probe via thermal denaturation. The resulting extracts were stored at 4ºC for LC-MS/MS analysis. In the human plasma method, 100 µL of matrix was spiked with FME and n-1 metabolite at concentrations ranging from 0.050 – 50 ng/mL and extracted using the identical affinity purification process applied to dried blood microsamples. Liquid chromatographic separation was performed on an Acquity Premier BEH C18 column (50 x 2.1mm, 1.7 µm) with mobile phase containing ion-pairing reagents. FME, n-1 metabolite, and n+2 internal standard were detected using a SCIEX 6500+ QQQ operated in negative ESI/MRM mode

Results: Typically, lower LC flow rates generate smaller and more highly charged droplets in ESI. This results in high ionization efficiency, particularly for multiply (negatively) charged ASOs. However, utilizing a low flow rate (≤ 0.3 mL/min) may exacerbate issues related to pre-column void volume, resulting in drawbacks such as peak broadening, diminished separation resolution, and compromised sample throughput. To address these challenges, a post-injection divert valve was integrated as a strategic solution (Figure 1). This valve effectively eliminates the invalid volume after sample injection, resulting in sharper peak profiles, in turn leading to improved separation resolution and signal-to-noise ratio (Figure 2). Under optimized LC-MS conditions, the simultaneous determination of FME and its key 5' n-1 metabolite were achieved with LLOQs of 0.50 ng/mL (10 µL blood microsample) and 50 pg/mL (100 µL plasma). Sample preparation leveraging complete microsample homogenization overcomes the potential for reduced extractability of ASOs with increasing blood hematocrit. Additionally, detergent was introduced in the desorption solution. This facilitates buffer permeation into the cellulose substrate and aids in ASO solubility, enabling near-quantitative recovery. However, treatment with detergent results in the complete desorption of all blood components from the substrate, and it was therefore necessary to use a highly selective hybridization approach to purification. In this manner, a biotinylated capture probe is designed to complement the FME sequence, allowing formation of a distinct ASO-Probe-Bead complex, which was also applicable to the 5' n-1 metabolite and n+2 analog. Calibration curves were established for both FME and n-1 metabolite in blood microsamples (0.50 – 500 ng/mL) and plasma (0.050 – 50 ng/mL) with precision and accuracy assessments performed at four QC concentration levels meeting all acceptance criteria. Specificity and lack of matrix effect bias were confirmed from individual blood/plasma donors, maintaining consistent quantitation for each of FME and n-1 metabolite. Furthermore, the impact of blood hematocrit was evaluated at low and high QC concentration with results confirming suitability up to 50% HCT (Table 1).

Conclusion: A sensitive and selective bioanalytical method for the determination of FME and n-1 metabolite was successfully developed for an analytical range of 0.50 – 500 ng/ml when extracted from a 10 µL dried blood microsample collected using the Capitainer qDBS and analyzed by hybridization LC-MS/MS. The method was linear, precise and accurate with all evaluated bioanalytical assessments meeting acceptance criteria. Further, the same workflow was established for plasma samples, offering a valuable tool for bridging PK data between matrices, encouraging the future use of blood microsampling for ASOs in decentralized clinical trials.

Figure 1. LC-MS configuration to bypass the autosampler during sample analysis.

Figure 2. Representative chromatograms of FME, metabolites and analogs without (A) and with (B) post-injection divert valve at a flow rate of 0.3 mL/min.

Table 1. Impact of Blood Hematocrits on the Quantitation of FME and 5' n-1 Metabolite