(M0930-02-08) Fit for Purpose Qualified Assays to Assess Exploratory Biomarkers in Human Aqueous Humor to Unveil Target Engagement and Pharmacological Activity in Complement Factor C5 Inhibition Clinical Studies

Purpose: Geographic Atrophy (GA) is an advanced form of dry age-related macular degeneration that results in retinal cell death and vision loss. Complement factors are dysregulated in GA withincreased levels of C5 and C3 present in aqueous humor (AH). Activation of the complement terminal pathway results in the breakdown of C5 into C5a and C5b. C5b allows the non-covalent association of complement factors C6 through C9 and the production of the pore-forming Membrane Attack Complex (MAC)/ or C5b-9 in the cell membrane that leads to cell lysis. A portion of the C5b-9 associates with proteins Vitronectin and Clusterin which prevents its insertion into the cell membrane. This complex exists as soluble C5b-9 (sC5b-9) and can be quantitated. Therapies targeting the complement pathway have been approved for slowing GA progression (e.g., SYFOVRE). ABBV-6628 is a Fab fragment that inhibits C5, designed to slow disease progression by disrupting the terminal complement pathway. Disrupting the terminal pathway and formation of the MAC/C5b-9, is hypothesized to block cell death and cellular dysfunction. As GA is a slowly progressing disease, biomarker assessment can provide an early snapshot into the effects of ABBV-6628 treatment. The first-in-human biomarker strategy aims to demonstrate ABBV-6628's target engagement of C5 with a reduction in freeC5 (fC5) and sC5b-9 in patient AH.The purpose of this study was to qualify Fit for Purpose (FFP) ligand binding assays to measure exploratory biomarkers, fC5 and sC5b9 in human AH as well as the comparison of sC5b-9 and fC5 levels from representative AMD or glaucomatous donor AH samples to non-diseased donor AH samples using these assays.

Methods: We established the FFP qualified sC5b-9 assay by leveraging a plasma based commercial sC5b-9 sandwich ELISA kit (Complement TCC ELISA RUO) from SVAR, Sweden. The kit was modified for use in human AH by adjusting the calibration range and minimum required dilution (MRD) recommended for the kit. The fC5 assay is an in-house developed sandwich ELISA. The presence of high levels of endogenous C5 in human AH posed a challenge in this assay. We also faced an additional challenge for both assays due to limited sample volume. We were able to overcome both these issues by using a higher sample MRD and surrogate matrix for sample dilution (Assay Diluent (AD) in the sC5b9 assay and General Serum Diluent 1 (SD1) from Immunochemistry Technologies in the fC5 assay respectively). Parallelism, stability, accuracy/precision, selectivity, sensitivity and MRD were assessed for both assays. The ELISA formats are illustrated in Figure 1. These assays were used to compare sC5b-9 and fC5 levels from representative AMD or glaucomatous donor AH samples to non-diseased donor AH samples (San Diego eye bank).

Results: The FFP qualified sC5B-9 assay has a sensitivity of 15.6 ng/ml in neat AH, while the fC5 assay has a sensitivity of 62.5ng/ml in neat AH and require AH sample volumes of 5µl and 50µL, respectively. The results of the MRD, sensitivity, precision, relative accuracy, dilutional linearity, parallelism, and sample stability assessment can be found in Table 1. The assays demonstrated robust performance and are fit for their intended purpose. The results from the comparison of sC5b-9 and fC5 levels from representative AMD or glaucomatous donor AH samples to non-diseased donor AH samples are shown in Figure 2. The results show a significant increase (p < 0.05) in both sC5b-9 and fC5 in AMD vs. non-diseased donors AH samples. No significant increase in either biomarker was seen in AH from glaucomatous donor samples.

Conclusion: We have successfully FFP qualified assays to measure sC5B-9 and fC5 in human AH. The use of surrogate matrix and sample dilution were sufficient to address the challenges of endogenous protein interference as well as limited sample volume. Assay sensitivity, precision, relative accuracy, dilutional linearity, parallelism, MRD, and sample stability were assessed as met the acceptance criteria. The data from analysing representative AMD AH samples presents additional proof of complement pathway involvement in GA and demonstrates that the sC5b-9 and fC5 assays demonstrate robust performance and are sensitive enough for their intended purpose. These assays will be used for exploratory biomarker assessment in the FIH ABBV-6628 clinical study.

References: 1. Bioanalysis Zone: 12/8/2018, Afshin Safavi (Bioagilytix) Exploratory biomarker testing: to qualify or validate the assay?
2. Bioanalysis (2014) 6(2), 185–198: Stevenson, et.al. Parallelism: Considerations for the development, validation and implementation of PK and biomarker ligand-binding assays.
3. Pharmaceutical Research, Volume 23, No. 2, February 2006: Lee et al. Fit-for-Purpose Method Development and Validation for Successful Biomarker Measurement.

Acknowledgements: The authors would like to thank Hannah Huh (AbbVie, Clinical Pharmacology), Matt Mason and Corine Ghosn (AbbVie, Research Pharmacology) for providing reagents and human AH samples.

Figure 1: Bioanalytical Assays for fC5 and sC5b-9 Biomarker Measurement

Table 1: Assay Performance of fC5 and sC5b9 Assays (Fit-for-Purpose Qualification)


Figure 2: Comparison of fC5 and sC5b-9 levels in AH from non-diseased and diseased donor eyes using qualified assays