2.1 Trial Design

This randomized, open-label, three-arm, parallel-treatment study assessed the safety, PK, and immunogenicity of a single AZD7442 IM dose in healthy eligible adults when administered as a co-formulated product compared with separately formulated component mAbs, and when derived from clonal cell line material compared with cell pool material. The study was approved by institutional review boards and conducted across 14 US sites. All participants provided written informed consent prior to enrollment. The trial consisted of a screening period of ≤ 28 days, after which randomization and dosing occurred on Day 1, followed by a follow-up period of 1 year.

Participants were randomized on Day 1 in a 1:1:1 ratio to one of three treatments to receive a single IM AZD7442 300 mg dose (150 mg each of tixagevimab and cilgavimab) via the lateral thigh, the gluteal dorsal, or the gluteal ventral, and stratified according to body weight (< 70 kg, ≤ 70 to < 80 kg, and ≥ 80 kg) and COVID-19 vaccination status (unvaccinated or vaccinated against COVID-19). Treatment A was a single AZD7442 dose of co-formulated tixagevimab and cilgavimab derived from clonal cell line material (the planned subsequent commercial product). Treatment B was AZD7442 administered as consecutive doses of separately formulated tixagevimab and cilgavimab derived from clonal cell line material (the initial commercial product during the pandemic). Treatment C was AZD7442 as consecutive doses of separately formulated tixagevimab and cilgavimab derived from cell pool material (the initial clinical trial formulation). When administered separately, tixagevimab was administered first, immediately followed by cilgavimab on the opposite site of the same injection region.

Participants abstained from alcohol and strenuous activity for 72 h and caffeine-containing foods/beverages for 24 h prior to study Day 1 and every outpatient visit. On study Day 1 in the clinical unit participants received a standard diet and rested for 1 h after dosing.

2.2 Participants

Eligible healthy participants were aged ≥18 years weighing 50–110 kg and screening BMI of 18–30 kg/m2, and either a documented negative SARS-CoV-2 reverse transcriptase polymerase chain reaction test within 3 days before Day 1 or a negative rapid SARS-CoV-2 antigen test on Day 1 (pre-dose). Participants could be unvaccinated or vaccinated with ≥1 locally approved/authorized COVID-19 vaccine doses, with vaccinated participants not in receipt of the COVID-19 vaccine ≤14 days of Day 1. These criteria allowed a sensitive, controlled and ethical way to detect true formulation differences in exposure with lower variability and risk.

Ineligible participants were those with a known history of allergy or reaction to tixagevimab or cilgavimab, or previous hypersensitivity or severe adverse reaction following IM administration of a mAb, and those presenting with a significant infection or acute illness (including fever >37.8°C) ≤1 day before randomization. Also excluded were individuals with immunodeficiency due to illness or drugs, a history of SARS or Middle East Respiratory Syndrome infection, clinical signs/symptoms consistent with COVID-19 within 4 weeks of screening, history of malignancy or clinically significant bleeding disorder or of significant bleeding/bruising following IM injections or venipuncture, active hepatitis B or C infection or positive test for hepatitis C or hepatitis B surface antigen at screening, receipt of a mAb within 6 months or five antibody half-lives (whichever is longer) prior to screening, prior receipt of any mAb indicated for prevention or treatment of SARS-CoV-2 infection or COVID-19, or any other significant disease that may greatly increase the participant risk in the study. As the study did not provide prospects of clinical benefits, which is required for studying in pediatric participants, healthy individuals aged 12 to < 18 years were not enrolled.

2.3 Blood Sampling and Bioanalysis

Blood samples for serum mAb concentration measurement were collected before the dose on study Day 1, and on study Days 2, 5, 8, 15, 22, 31, 61, 91, 181, 271, and 361 (all cohorts). Blood samples for ADA assessment were collected before the dose on study Day 1, and on study Days 31, 91, 181, and 361 (all cohorts). Serum concentrations of tixagevimab and cilgavimab as well as ADAs were analyzed as reported previously. Briefly, bioanalytical analyses of tixagevimab and cilgavimab in serum were performed by PPD Laboratories (Richmond, VA, USA), using immunocapture to SARS-CoV-2 receptor-binding domain followed by denaturation and detection of AZD7442 specific peptides by liquid chromatography–mass spectrometry [4, 8]. The lower level of quantification for both tixagevimab and cilgavimab was 0.30 µg/mL. Serum AZD7442 concentration was the sum of serum concentrations of tixagevimab and cilgavimab. The PK parameters of AZD7442, tixagevimab and cilgavimab were derived using non-compartmental methods in Phoenix® WinNonlin® (Certara, Version 8.1 or higher).

Serum samples for ADA detection were determined using a three-tier testing scheme [5]. Briefly, samples were screened for ADA against tixagevimab and cilgavimab separately using an electro-chemiluminescence (ECL) solution-phase bridging method, validated by PPD Laboratories. Diluted samples were incubated in a solution phase with biotinylated tixagevimab or cilgavimab and ruthenium ([sulfo-TAG]-labeled tixagevimab or cilgavimab). Samples were reported screen positive for ADA in the screening assay if the mean ECL value was at or above the ECL value of the plate-specific cut point factor. Screen positive samples were retested in a confirmation assay, where samples were analyzed both in the absence and presence of excess drug to determine if the sample’s positive response is specific to tixagevimab or cilgavimab. Titers were measured in confirmed positive samples and were reported as the reciprocal of the highest twofold dilution that measured positive in the assay, before returning a negative response. Titers for negative samples were reported as < 80 for tixagevimab or < 40 for cilgavimab; the minimum required dilution of the assay was 1:80 and 1:40, respectively.

Treatment-emergent (TE)-ADA to AZD7442 was defined as either treatment-induced ADA positive (ADA positive post-baseline but not detected at baseline for either or both of tixagevimab and cilgavimab), or treatment-boosted ADA positive (baseline ADA titer that was boosted by ≥ 4-fold following drug administration for either or both of tixagevimab and cilgavimab).

2.4 Statistical Analyses2.4.1 Sample Size Considerations

Preliminary analysis of PK data from the first-in-human study indicated that after an IM dose of 300 mg AZD7442 (two consecutive doses of tixagevimab and cilgavimab), the coefficient of variation (CV%) for Cmax of the individual mAbs, tixagevimab and cilgavimab were 35.6 and 38.5%, respectively, while the CV% for AUCs of both individual mAbs were < 36.4%. Therefore, assuming a mean ratio of one between comparators, 69 participants per treatment were expected to provide at least 95% power to demonstrate comparability (90% confidence intervals [CIs] contained within 0.8000 and 1.2500).

2.4.2 Bioequivalence Evaluation

Statistical comparisons were performed based on linear mixed effects analysis of variance models with the natural logarithms of primary PK parameters (Cmax, AUClast, AUCinf, and pAUCs) for tixagevimab, cilgavimab and AZD7442 as response variables. Treatment, injection site, COVID-19 vaccination status and continuous baseline body weight on the log scale were included as fixed effects, and participant as a random effect. Sex was not included in the model, as previous population PK analysis found no significant effect of sex on the PK of tixagevimab and cilgavimab [7].

One statistical model was fit for each PK parameter. No adjustments for multiple comparisons or multiplicity were made. Transformed back from the logarithmic scale, the geometric least square mean (and their two-sided 95% CIs) of each treatment was calculated and the geometric mean ratio (GMR) as well as their 90% CIs for each pairwise comparison was estimated. Bioequivalence was demonstrated if the GMRs (and their 90% CIs) fell within a pre‑defined limit of 0.8000–1.2500.

Demographics, baseline characteristics, safety and immunogenicity variables were summarized using descriptive statistics.

All analyses were conducted with SAS, Version 9.4 (SAS Institute Inc., Cary, NC, USA).

2.4.3 Correlation Analysis

Pairwise associations among PK parameters were assessed using Pearson correlation coefficients computed on the natural log–transformed values of each parameter for each drug component, by treatment group.

2.5 Safety Assessments

Safety was assessed throughout the study, including adverse events (AEs), serious adverse events (SAEs), adverse events of special interest (AESIs) and stratified per COVID-19 vaccination status at randomization. Adverse events of special interest were defined as anaphylaxis and other serious hypersensitivity reactions including immune complex disease, injection-site reactions and cardiac ischemia, cardiac failure, and thrombotic events. Adverse events of special interest were coded using the Medical Dictionary for Regulatory Activities (MedDRA; version 26.0).