Detailed patient-level longitudinal outcome reporting
Tables I–III and Fig. 1 permit direct inspection of genotype, run-in and treatment attack rates, exposure duration, discontinuation, and adverse events for every patient.
↳ Tables I–III; Fig. 1
Abstract
BACKGROUND: Hereditary angioedema (HAE) with normal C1 inhibitor (HAE-nC1INH), including HAE arising from F12 (HAE-FXII) or PLG (HAE-PLG) mutations, has a high unmet therapeutic need. OBJECTIVE: Our aim was to explore long-term prophylaxis with monthly subcutaneous garadacimab (anti-activated factor XII mAb) in patients with HAE-FXII or HAE-PLG. METHODS: Efficacy and safety were evaluated for HAE-FXII or HAE-PLG in open-label phase 2 (a 13-week treatment period [TP1] and subsequent ≥44-week extension period [TP2]; 600 mg of garadacimab during each period) and ongoing phase 3 open-label extension (OLE; garadacimab dose 200 mg) studies. RESULTS: Patients 1 and 2 (HAE-FXII) completed phase 2 and then continued into OLE (total exposure time 42.9 and 40.2 months, respectively); patient 3 (HAE-FXII) discontinued TP1 (because of lack of efficacy). Patients 4 to 6 (HAE-PLG) completed TP1 only. Patients 1 and 2 (HAE-FXII) had a reduction in monthly attack rate (AR) of 88% or more versus run-in during phase 2 (TP1 AR = 0.4 and 0.0 and TP2 AR= 0.1 and 0.2, respectively; run-in = 3.2 for both) and were attack-free during the OLE. The AR was increased for patients 4 and 5 (HAE-PLG) and reduced for patient 6 (HAE-PLG). During TP1, 4 of 6 patients (HAE-FXII [n = 2]; HAE-PLG [n = 2]) experienced treatment-emergent adverse events (TEAEs): patient 1 (HAE-FXII) experienced a garadacimab-related TEAE (mild injection-site reaction), and patient 3 (HAE-FXII) experienced an unrelated serious TEAE (severe HAE attack). Patients 1 and 2 (HAE-FXII) collectively experienced 13 mild or moderate TEAEs during TP2 and 2 mild TEAEs during the OLE. CONCLUSIONS: Garadacimab showed a favorable safety profile in all 6 patients with HAE-nC1INH. Garadacimab also demonstrated efficacy in 2 of the 3 patients with HAE-FXII; a reduction in AR was observed in 1 of the 3 patients with HAE-PLG.
Key Claims
Primary findingTentative
Garadacimab also demonstrated efficacy in 2 of the 3 patients with HAE-FXII
causal efficacy language rests on uncontrolled observations in two continuing patients
Abstract; Conclusions
SecondarySupported
Garadacimab showed a favorable safety profile in all 6 patients with HAE-nC1INH.
observed adverse events were mostly mild or moderate, but six patients cannot characterize uncommon harms
Abstract; Conclusions
SecondarySupported
A reduction in AR was observed in 1 of the 3 patients with HAE-PLG.
Table II confirms the reduction, although it fell below the protocol responder threshold
Abstract; Results; Conclusions
SecondarySupported
Patients 1 and 2 (HAE-FXII) were attack-free during the phase 3 OLE (100% reduction vs run-in for more than 18 months).
Table II directly records zero attacks during more than 18 months of extension follow-up
Results and Discussion; Table II
Derived from the full evaluation — not a separate score.
Strengths · Limitations
Strengths
Mixed findings reported without concealment
The report identifies one HAE-FXII discontinuation for lack of efficacy, increased attack rates in two HAE-PLG patients, and a subthreshold reduction in the third.
↳ Results and Discussion, p. 1757; Table II
Mechanistic counterevidence informs interpretation
The discussion uses FXIIa-independent bradykinin generation to explain why PLG-mutated patients may not respond and links this possibility to genotype-specific treatment decisions.
↳ Results and Discussion, p. 1757; references 19–20
Limitations
Uncontrolled design limits efficacy attribution
Treatment was open-label with no control group, and run-in periods were short. Reduced attacks in two continuing HAE-FXII patients therefore constitute a treatment signal rather than established efficacy.
↳ Methods; Table E1; Limitations, p. 1759
Extremely narrow and heterogeneous cohort
The cohort comprised six White women split equally between F12 and PLG mutations, while long-term follow-up was confined to two HAE-FXII patients.
↳ Tables I–II; Limitations, p. 1759
Clinical recommendation exceeds evidence maturity
The conclusion calls conditional garadacimab use reasonable despite mixed responses and the acknowledged absence of a control group.
↳ Conclusion and Limitations, p. 1759
Nabu Commentary
The report’s principal value is its prospective, genetically confirmed, patient-level record of a very rare population, including more than 40 months of follow-up in two HAE-FXII patients. Tables I–III and Fig. 1 make the heterogeneous outcomes directly inspectable, and the discussion addresses several biological and clinical confounders. The score is tempered by open-label treatment, short run-ins, no control group, and long-term continuation confined to the two apparent HAE-FXII responders. The patient-level fractions are communicated consistently, but “demonstrated efficacy” and the concluding conditional-use recommendation convey more certainty and actionability than this design supports.
Check another paper
Author Note / Field Feedback
Nabu’s assessment, alongside the field’s view.
Are you an author of this paper?
Dimension breakdown
Study Quality
Sound3.6
Confidence mediumContribution
3.6
The report adds prospective, genetically confirmed observations with more than 40 months of exposure in two HAE-FXII patients and openly reports divergent PLG outcomes. Its advance remains narrow because it comprises six uncontrolled cases across two biologically distinct subgroups.
Results and Discussion, pp. 1757–1759; Table II
“one of the first prospective clinical studies with long-term exposures (>40 months)”
Methodological Rigour
3.6
Patient-level genotype, attack-rate, exposure, safety, and progression data permit reconstruction of each course, while the discussion addresses several alternative explanations. Open-label treatment, short run-ins, no comparator, and long-term continuation only among two apparent responders materially limit efficacy identification.
Tables I–III; Fig. 1; Results and Discussion, pp. 1757–1759
“garadacimab was administered on an open-label basis across both studies”
Reporting
3.8
The report follows a coherent progression from rationale and methods to subgroup-specific outcomes, mechanistic interpretation, limitations, and conclusions. Fractions remain consistent throughout, although “demonstrated efficacy” conveys more certainty than an uncontrolled case series can provide.
Abstract; Key Messages; Conclusion, p. 1759
“demonstrated efficacy in 2 of the 3 patients with HAE-FXII”
Positioning
3.7
The discussion relates the findings to guideline uncertainty, earlier treatment reports, and evidence that PLG variants may generate bradykinin independently of FXIIa. It acknowledges sample, demographic, open-label, and control limitations, but the final conditional-use recommendation does not fully carry those limitations through.
Introduction, p. 1756; Results and Discussion, pp. 1757–1759
“Limitations of this assessment include the small patient population with no male or non-White patients.”
Reliability
Caveats4 of 4 checks
Internal CoherenceNoted
The patient counts, attack-rate directions, exposure durations, and safety totals are otherwise coherent. Table II contains malformed percentage values that conflict typographically, but not directionally, with the accompanying prose and raw rates.
Research ConductNoted
Written consent requirements, trial identifiers, sponsor involvement, and author conflicts are disclosed. The report does not include an explicit ethics-committee or IRB approval declaration or identifier.
Flags: 2 declared / 5 total
Reference IntegrityClean
24 of 24 checkable references verified
25 references in manuscript 1 are books, websites or datasets — counted, but not index-checkable
Post-Publication RecordClean
No retraction notice found in Retraction Watch.
Last checked: 18 Sep 2026
Sources: Retraction Watch ✓
Readiness
Where this paper’s evidence sits on the path from initial observation to real-world use.
Demonstrated
Garadacimab is situated in a clinical treatment setting and the report offers a preliminary genotype-specific signal. The uncontrolled exploratory analysis is not validation in a representative operational population and supports further study rather than deployment-ready practice.
Methods; Table E3; Limitations, p. 1759
“with no control group”
AI-generated, human-governed. Something look off? Contact us to request a review.