(W0930-10-55) Development of a Sensitive, Drug Tolerant Total Antibody Assay for the Detection of Antibodies against a Novel Capsid AAVPRV in Human Serum and CSF

Purpose: Due to the risk of immunogenicity, the characterization of pre-existing and treatment emergent anti-AAV immune responses are pivotal to the development of rAAV based gene therapies. To measure possible immune responses in clinical candidates, it is necessary to develop highly sensitive assays that can measure anti-drug antibodies (ADAs) in biofluids such as serum and CSF. Here, we describe the development of ADA assays for disease-modifying gene therapies for neurodegenerative disorders that utilize AAVPRV, an in-house generated AAV capsid, for transgene delivery to the CNS.

Methods: The AAVPRV ADA Assay uses the MSD platform for the highly sensitive detection of the antibodies. Briefly, unlabeled AAVPRV was coated to an assay plate; after a wash step, Positive Controls, Negative Controls, and Samples were diluted to the Minimum Required Dilution (MRD) and added to the coated plate. The plate was incubated overnight, washed, and then incubated with an acidic solution to break the formed virus-antibody complexes. The ADAs in the acidic solution were then transferred to an MSD SECTOR plate containing a neutralizing solution and incubated for two hours. After another wash step, ruthenium-labeled AAVPRV for tier 1, or ruthenium-labeled AAVPRV in excess unlabeled AAVPRV for tier 2, was then added to the plate and incubated for 1 hour. Upon completion, the plate was once more washed and 2x MSD Read Buffer is added to each well. The plate was then read on the MSD instrument and measured for electrochemiluminescence. A descriptive image summarizing this protocol can be found in Figure 1. A tiered assay approach to identify the presence, specificity, and titer of ADA responses has been adopted as a gold standard by industry leaders, the FDA, and the EMA. Specifically, tier 1 serves as a high-sensitivity preliminary screen used to define samples as ADA negative or screening assay positive. Tier 2 confirms the presence of ADAs in samples that screened positive, via the addition of excess drug into the detection solution; excess drug bind and removes ADAs from the assay, thus an inhibition of signal generated would be observed.

Results: 50 individual healthy human serum samples were evaluated for ADAs in tier 1 and 2 assays. Statistical analysis was used to remove outliers and determine cut points. We found that 54% (26/48) of samples had pre-existing antibodies to AAVPRV. In healthy CSF samples, 2% (1/47) of samples had pre-existing antibodies to AAVPRV. Figure 2 and 3 show the strong correlation of tier 1 and tier 2 signals for the individuals tested in serum and CSF. The high correlation between tier 1 and tier 2 in both serum and CSF suggests the highly specific nature of the assays. Negative Samples for each biofluid were used to create pools for further assay development. The negative serum or CSF pools were spiked with positive control antibodies and serially diluted, creating sensitivity curves used to determine the sensitivity of the assay in each biofluid across 5 independent runs for serum and 4 independent runs for CSF. The screening sensitivity for each assay was determined to be 4.08 ng/mL and 2.95 ng/mL, respectively (Tables 1A, 1B, 2A and 2B). To determine drug tolerance of the assay, excess drug at differing concentrations was incubated in 3 levels of positive control and negative control, then evaluated with the tier 1 assay. The lowest tested concentration of antibody screened positive in the presence of 5.55 x 10¹⁰ vg/mL of drug in serum, and 5.55 x 10¹¹ vg/mL of drug in CSF (Figures 4 and 5). To further characterize the robustness and reproducibility of results within the assay, we measured key performance benchmarks, including inter- and intra-assay precision, titer precision, selectivity, and prozone effect. All met internal acceptance criteria.

Conclusion: We have successfully developed a highly sensitive and drug tolerant assay to detect ADAs against the AAVPRV vector in human biofluids. ADA levels measured by this assay can serve as a clinical trial endpoint to assess immunogenicity and thus provide valuable insight into the effects of ADAs on drug efficacy and any adverse events that may arise.

References: Simmons E, Wen Y, Li J, Qian YW, Wong LC, Konrad RJ, Bivi N. A sensitive and drug tolerant assay for detecting anti-AAV9 antibodies using affinity capture elution. J Immunol Methods. 2023 Jan;512:113397. doi: 10.1016/j.jim.2022.113397. Epub 2022 Dec 5. PMID: 36481208.

Figure 1: Summary of assay protocol and tiered approach.
Figure 2: Correlation between tier 1 and tier 2 signals in human serum.
Figure 3: Correlation between tier 1 and tier 2 signals in human CSF.

Tables summarizing development runs to determine assay sensitivity.

Figures 4 and 5 exhibit the drug tolerance of the assay in both tested biofluids.