Assembling the evidence…
Garadacimab is a novel, fully human, anti-activated factor XII monoclonal antibody approved for long-term prophylaxis of patients with hereditary angioedema. This open-label, parallel-group, Phase 1, single-center, bridging study in healthy adults (18-55 years of age) characterized the pharmacokinetics and safety of a single 200 mg subcutaneous injection of garadacimab administered via autoinjector/pre-filled pen (AI/PFP) compared with the pre-filled syringe (PFS) used in previous studies. The aim of the study was to bridge the understanding of the PK and safety of garadacimab between PFS and AI/PFP modes of administration. Participants (N = 132) were stratified by body weight, randomized evenly in six groups by device (AI/PFP or PFS) and injection site (abdomen, thigh, or upper arm). The primary endpoint comprised pharmacokinetic parameter comparison between devices. Safety/tolerability were secondary endpoints. The geometric mean ratio (GMR) for Cmax and AUC0-inf comparing administration by PFS and AI/PFP was close to 1 with 90% confidence intervals within a range of 0.8-1.25, meeting bioequivalence criteria; GMR (90%) was 0.92 (0.81, 1.05) for Cmax and 0.96 (0.87, 1.07) for AUC0-inf. No participants in this study had anti-drug antibodies against garadacimab. Treatment-related emergent adverse events were reported in 9/66 (13.6%) participants in the PFS group and 11/66 (16.7%) participants in the AI/PFP group. Garadacimab 200 mg administered as a single subcutaneous dose via AI/PFP had a consistent safety and tolerability profile to that administered via PFS. These findings support administration of garadacimab via AI/PFP, providing at-home convenience for patients and physicians.
The geometric mean ratio (GMR) for Cmax and AUC0-inf comparing administration by PFS and AI/PFP was close to 1 with 90% confidence intervals within a range of 0.8-1.25, meeting bioequivalence criteria
accurate numerical intervals paired with formal bioequivalence wording that exceeds the stated study design
Garadacimab 200 mg administered as a single subcutaneous dose via AI/PFP had a consistent safety and tolerability profile to that administered via PFS
systematic single-dose safety reporting, tempered by open-label assessment and more AI/PFP injection-site events
No participants in this study had anti-drug antibodies against garadacimab
directly observed in 132 participants after one dose, limiting longer-term immunogenicity inference
garadacimab PK parameters were consistent across different sites of administration (abdomen, thigh, or upper arm)
small descriptive injection-site subgroups without formal statistical comparisons
Derived from the full evaluation — not a separate score.
Strengths
Participants were stratified by weight and randomized through an interactive response technology system, with balanced reported characteristics across device groups. The objective PK endpoints and analysis populations were defined in advance.
↳ Methods, Study Design and Participants; Statistical Methods; Results, Table 1
The paper reports device-specific Cmax and AUC0–inf values plus GMRs and 90% confidence intervals, allowing readers to inspect the central device comparison directly.
↳ Results, Table 2 and Figure 3b–c
Adverse events, device-related events, severe hypersensitivity, laboratory findings, and anti-drug antibodies are reported, including the higher injection-site event incidence with AI/PFP. The Discussion also identifies the healthy-volunteer and non-crossover limitations.
↳ Results, Safety, Table 3 and Tables S7–S8; Discussion, Limitations
Limitations
The Abstract says the confidence intervals met bioequivalence criteria, while the Discussion states that the study was not formally designed to assess bioequivalence. This distinction affects interpretation of the headline PK result.
↳ Abstract; Discussion, opening paragraph
All injections were administered by trained personnel at the study site, and the paper acknowledges that convenience and preference were not assessed. The Conclusions therefore extend beyond the outcomes observed.
↳ Methods, Study Design and Participants; Discussion, Limitations; Conclusions
The study enrolled healthy adults at one center for a single dose rather than patients with HAE using the device over time. Prior population-PK evidence supports exposure extrapolation but does not establish usability, adherence, or longer-term tolerability.
↳ Methods, Study Design and Participants; Discussion, Limitations
The central PK comparison is supported by randomized allocation, balanced device groups, and directly reported GMRs with 90% confidence intervals. The contribution remains incremental because it addresses delivery-device bridging for an established drug rather than efficacy or disease mechanisms. Open-label assessment modestly limits the secondary tolerability comparison, particularly for subjective injection-site outcomes. The strongest presentational issues are the formal bioequivalence wording and the claim of at-home convenience despite site-based administration and no convenience or preference assessment. The regulatory pathway is credible, but transfer to long-term self-use by patients with HAE remains indirect.
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Sound3.6
Confidence highThe study resolves a genuine device-bridging uncertainty at the intended 200 mg dose, providing practical evidence for product development. Its advance is narrow because it concerns delivery of an already-characterized drug rather than efficacy, disease biology, or patient usability.
The aim of the study was to bridge the understanding of the PK and safety of garadacimab between PFS and AI/PFP modes of administration.
Stratified IRT randomization, balanced device groups, defined analysis sets, precision-based sample sizing, and systematic harms collection fit the primary PK question. Open-label assessment leaves subjective local tolerability outcomes vulnerable to expectation or assessor effects, although the objective PK endpoint is not similarly affected.
Eligible participants were stratified by body weight...and then randomized...via an interactive response technology system.
The manuscript follows a coherent progression from the bridging objective to PK and safety results. Precision is reduced by describing the numerical intervals as “meeting bioequivalence criteria” despite acknowledging that formal bioequivalence was not the study design, and by presenting unmeasured at-home convenience as an outcome.
Although this study was not formally designed to assess bioequivalence
Prior garadacimab trials and broader autoinjector literature establish the study's development context. The Discussion acknowledges the parallel-group and healthy-volunteer limitations and traces PK extrapolation to HAE through prior population-PK evidence.
population PK analysis demonstrated that garadacimab PK is similar between healthy individuals and patients with HAE
Lower confidence on Contribution — domain match limited.
Caveats4 of 4 checks
The reported participant counts, PK estimates, and adverse-event totals reconcile across the manuscript. The conclusion nevertheless extends from device PK and safety findings to at-home convenience, which the study did not measure.
Ethics approval, written consent, prospective registration, sponsor funding, author conflicts, and a conditional IPD-sharing process are declared without an identified contradiction.
Flags: 4 declared / 5 total
35 of 35 checkable references verified
36 references in manuscript 1 are books, websites or datasets — counted, but not index-checkable
No retraction notice found in Retraction Watch.
Sources: Retraction Watch ✓
Where this paper’s evidence sits on the path from initial observation to real-world use.
The comparative PK evidence can directly inform device development and a regulatory bridging pathway for an existing treatment program. It does not establish operational self-administration, patient preference, adherence, or long-term device safety.
does not allow for within-subject comparisons, including assessments of convenience and preference
AI-generated, human-governed. Something look off? Contact us to request a review.