Evicenter
MarkdownMedical informationP&T meetings
Onvuzosiran (ADX-324)

Prophylaxis of hereditary angioedema attacks

Launch
Not announced
Last updated
119 sources

Section 2 of 6

Executive summary

6 evidence topics · 41 sources

Clinical benefits of Onvuzosiran

Burden of hereditary angioedema

Summary

Hereditary angioedema (HAE) is estimated to affect 1 in 50,000 people worldwide. A systematic literature review reported a pooled diagnosed annual prevalence of 1.70 per 100,000 (95% confidence interval, 1.58 to 1.83) across HAE types, and a US commercial claims algorithm estimated 2.43 per 100,000 (8694 people) in 2020. In one long-term cohort, symptoms began at a mean age of 11.2 years. Untreated patients may experience attacks every 7 to 14 days on average; abdominal involvement occurs in 70% to 90% of individuals, and laryngeal involvement, which accounts for less than 1% of attacks, occurs at least once in about half of patients. Reported mean or median delays from symptom onset to diagnosis ranged from 3.9 to 26 years.

In one cohort, 70 of 214 patients who had died asphyxiated during a laryngeal attack, 63 of them with undiagnosed HAE; in a second cohort, asphyxiation owing to laryngeal edema caused 39 of 75 deaths (52%). Without a diagnosis, the risk of death with laryngeal edema is estimated at 20% to 25%. In a patient registry, 36.9% of participants with HAE due to C1 inhibitor deficiency reported a family member who died from an HAE attack. Abdominal attacks may lead to “unnecessary surgeries or opiate dependence”, and another cohort reported higher odds of venous thromboembolic disease (odds ratio 4.20) and autoimmune disease (odds ratio 1.65) in HAE, with no increased risk of death (odds ratio 0.69).

In a survey of 260 participants, respondents reported a mean of 11.5 attacks (standard deviation 14.2) in the preceding 6 months, and 137 of 260 (52.7%) had an AE-QoL total score of 39 or higher. In a survey of 457 respondents, 42.5% had depression scores “indicative of depressive symptomatology”, and overall work impairment was 34%. In US commercial claims for 328 patients, mean annualized HAE-related healthcare costs per patient were $165,348 before and $515,333 after starting long-term prophylaxis, driven by prophylaxis pharmacy costs (mean $395,845 per patient per year). A modeled estimate of on-demand treatment alone placed direct costs at $363,795 and indirect costs at $52,576 per patient per year.

Onvuzosiran efficacy and safety

Summary: phase 1/2 pharmacodynamics, efficacy, and safety

Onvuzosiran (ADX-324) has been studied in one phase 1/2 trial (ADX-324-101, NCT05691361), a single-center study in Australia. Part A was a randomized, double-blind, placebo-controlled, single ascending dose study in healthy volunteers, with weight-based doses from 0.4 to 6 mg/kg and one cohort that received two 140 mg doses 3 months apart. Part B is an open-label phase 2a study in 3 participants with type I or type II HAE, dosed every 6 months: 1 participant received 122.5 mg and 2 received 245 mg. These doses are lower or less frequent than the phase 3 regimens of 300 mg every 6 months and 240 mg every 3 months.

In an interim analysis of Part A, mean nadir prekallikrein levels were 39% (n=6), 14% (n=6), and 20% (n=5) of baseline after single doses of 0.4, 1.0, and 2.0 mg/kg. The prospectus reports for the 6 mg/kg cohort a mean prekallikrein-kallikrein reduction of 94% (88% to 97%) at nadir and 84% (78% to 92%) at 6 months, with 3 of the 4 participants below the lower limit of quantification at nadir. For a single dose of 300 mg plus or minus 25 mg (n=5 in the AAAAI poster), the prospectus reports a mean reduction exceeding 90% (84% to 96%) at nadir and at least 80% (65% to 90%) through Day 169; the AAAAI poster and the February 2026 press release report 93% at nadir and 80% at 6 months. The sponsor reports a lag of about 22 days between the first dose and full inhibition, and states that reductions in the participants with HAE were similar to those in healthy volunteers at similar doses.

All 3 phase 2a participants were attack-free for 3 months after the first dose. The prospectus reports one participant attack-free for 18 consecutive months on every-6-month dosing after one mild attack on Day 153 (August 19, 2026 cutoff); the November 2026 presentation reports Patient 202 (245 mg every 6 months) attack-free for 21 consecutive months after one mild attack on Day 153 (November 5, 2026 cutoff). Neither document states whether the two reports describe the same participant. All phase 2a dosing was every 6 months; the November 2026 presentation describes the 3-month result as a “100% Q3M attack-free rate”. Sponsor documents state that no attacks occurred while the prekallikrein-kallikrein reduction exceeded 80%. No baseline attack rates, on-treatment attack rates, baseline characteristics, or quality-of-life results were identified.

In the Part A interim analysis, no serious adverse events were reported, all study drug-related adverse events were mild except one moderate event, and headache was the most frequent adverse event. The prospectus lists injection site reactions, headache, laboratory changes in blood creatine phosphokinase, and ALT/AST increases as the most common adverse events in Part A, and describes the laboratory changes as asymptomatic, transient, and resolved without treatment; it gives no counts for them. A workshop abstract names headache and injection-site reaction as the most common adverse events in Part A and gives one figure, 28.9%, without stating whether it applies to each event or to both together. No Part A participant had a serious adverse event or an adverse event leading to withdrawal. In Part B, no treatment-related serious adverse events were reported as of the August 5, 2026 safety cutoff, repeat dosing at 6-month intervals was not associated with a safety signal, and 1 of the 3 participants withdrew for administrative reasons.

The results are available only in congress abstracts, congress posters, a press release, a corporate presentation, and SEC filings. No peer-reviewed full article was identified, and the registry record shows “No Results Posted”.

Summary: differences between reporting documents

The documents report different numbers of Part A participants: 43 (32 onvuzosiran, 11 placebo) in 5 cohorts in the February 2026 press release, and 49 (38 onvuzosiran, 11 placebo) in the prospectus and the AAAAI poster; the EAACI poster also cites 49 healthy volunteers. Planned registry enrollment for both parts is 53. The number of Part A cohorts is given as seven in the prospectus, 6 dose-escalation cohorts plus 1 repeat-dose cohort in the workshop abstract, and 5 doses plus a repeat-dosing group in the EAACI abstract.

The prospectus attributes the 94% nadir and 84% six-month reductions to the 6 mg/kg cohort. The November 2026 presentation attributes the 94% and 84% means to 6 mg/kg on one page and the same means with their ranges to 300 mg plus or minus 25 mg on another, whereas the prospectus reports exceeding 90% and at least 80% for that fixed dose, and the AAAAI poster and February 2026 release report 93% and 80%.

The proposed suppression threshold for attack prevention is greater than 80% in the prospectus, the November 2026 presentation, and the AAAAI, EAACI, and ACAAI abstracts, and greater than 70% in the workshop abstract. The first Part B dose is written as 122 mg, 122.5 mg, or 2 mg/kg (approximately 125 mg). Serious adverse events are described as absent, absent when treatment-related, or absent when drug-related, depending on the document.

Summary: phase 3 trial status

STOP-HAE (NCT06960213) is a randomized, double-blind, placebo-controlled phase 3 trial that began in August 2025 and plans to enroll approximately 90 adults with type I or type II HAE at 70 sites (the EAACI 2026 poster reports at least 60 sites in 22 countries). Participants are randomized 1:1:1 to onvuzosiran 300 mg every 6 months, 240 mg every 3 months, or placebo. The primary endpoint is the time-normalized number of investigator-confirmed attacks per month from Day 22 to Week 25 for 300 mg every 6 months compared with placebo. No results are available; primary completion is estimated for 2027-11-30, and the sponsor expects topline data by year-end 2027. The sponsor's target profile is a greater than 90% reduction in monthly attacks with greater than 75% of patients attack-free. Participants who complete STOP-HAE may enter the extension study ADX-324-302 (NCT07428499), which began on 2026-04-02, gives onvuzosiran for up to 3 years to approximately 90 participants, has the incidence and severity of treatment-emergent adverse events through Month 36 as its primary endpoint, and has an estimated primary completion date of 2030-06-30.

Budget impact of Onvuzosiran

Summary

Onvuzosiran is investigational, and its sponsor states that it has “no products approved for commercial sale”. STOP-HAE topline data are expected by the end of 2027, and the sponsor anticipates a potential New Drug Application submission in 2028 if the data are positive. The sponsor is developing onvuzosiran as a subcutaneous treatment given at Q6M and Q3M intervals. No price has been published, and no budget impact model of onvuzosiran was identified. The published inputs a plan would use are listed below with the year of each source.

  • US prevalence: the 2018 ICER evidence report projected a US prevalence of 6,690 individuals with HAE type 1 or 2. A 2025 claims analysis estimated 9559, 9341, and 8694 people diagnosed with any type of HAE in 2018, 2019, and 2020 (2.67, 2.61, and 2.43 per 100,000), and 7602, 7275, and 6595 by expert physician analysis. A 2021 patient survey states that HAE caused by C1 inhibitor deficiency affects fewer than 8000 people in the United States. A 2026 systematic literature review reported a pooled diagnosed annual prevalence of 1.43 per 100,000 in North America. The sponsor cites a frequency of approximately 1 in 50,000 people worldwide (2026).
  • Fraction eligible for prophylaxis: the 2018 ICER report assumed, from expert opinion, that one-third of patients were eligible for prophylaxis, approximately 2,230 patients, with 446 patients initiating treatment each year. In the 2021 survey, 462 of 737 participants (62.7%) received prophylaxis. A 2026 claims analysis states that approximately 70% of US patients with HAE are treated with long-term prophylaxis.
  • Current prophylaxis mix: in the 2026 claims analysis of 328 patients, the first long-term prophylactic agent was lanadelumab in 138 (42%), subcutaneous C1 esterase inhibitor (Haegarda) in 97 (30%), intravenous C1 esterase inhibitor (Cinryze) in 54 (17%), and berotralstat in 39 (12%). The 2018 ICER budget impact analysis assumed a displaced mix of 49% Haegarda, 49% Cinryze, and 2% no long-term prophylaxis. Neither mix includes garadacimab or donidalorsen.
  • Annual comparator costs: 2025 Missouri Medicaid drug fact blasts list estimated annual wholesale acquisition costs (WAC) of $747,006 for donidalorsen every 4 weeks ($373,503 every 8 weeks), $685,200 for garadacimab ($742,300 in the first year of therapy), $685,171 for lanadelumab, $665,224 for Cinryze, $604,152 for Haegarda, and $579,885 for berotralstat. A 2026 Colorado WAC disclosure lists berotralstat 150 mg at $48,487.92 per carton of 28 capsules. The 2018 ICER analysis used an annual lanadelumab WAC of $565,557 and a net price of $423,344. Medicare Part D data for 2024 list 512 beneficiaries for lanadelumab, 221 for berotralstat, 187 for Haegarda, and 46 for Cinryze. Federal Supply Schedule and Big Four prices are listed for November 2026, and Medicare Part B payment limits for October 1, 2026 through December 31, 2026.
  • Cost offsets: in the 2026 claims analysis, mean annualized HAE-related healthcare costs per patient were $165,348 before and $515,333 after starting long-term prophylaxis, including mean long-term prophylaxis pharmacy costs of $395,845 per patient per year and a reduction in emergency room and inpatient costs of $8344 per patient per year.
  • Framework and threshold: the 2018 ICER framework defines potential budget impact as the differential health care cost of the new therapy, including drug costs, minus offsets from averted health care events, undiscounted over a five-year time horizon. Its threshold for 2018-19 was approximately $991 million per year for new drugs. Lanadelumab reached 13% of that threshold at WAC, with an average annual per-patient difference of $96,145 at WAC and -$45,729 at net price against the 49%/49%/2% mix.

The 2018 ICER inputs (prevalence, eligible fraction, displaced mix, lanadelumab price, and threshold) were derived when Cinryze, Haegarda, and lanadelumab were the three drugs approved for long-term prophylaxis of HAE-1/2.

Conclusions

Summary

Onvuzosiran (ADX-324) is an investigational small interfering RNA that targets prekallikrein, and its sponsor states that it has “no products approved for commercial sale”. The FDA granted orphan drug designation for the treatment of patients with HAE in October 2025 and Fast Track designation for prophylaxis to prevent HAE attacks in August 2026. The phase 3 STOP-HAE trial began in August 2025 and has an estimated primary completion date of 2027-11-30; the sponsor expects topline data by the end of 2027 and anticipates a New Drug Application in 2028 if the data are positive. No price has been published, and no cost-effectiveness or budget impact analysis of onvuzosiran was identified.

The clinical evidence comes from one phase 1/2 trial (NCT05691361). In healthy volunteers, a single dose of 300 mg plus or minus 25 mg (n=5) reduced prekallikrein-kallikrein by a mean of 93% at nadir and 80% at 6 months in the AAAAI poster and the February 2026 release; the prospectus reports more than 90% and at least 80% for the same dose group. The open-label phase 2a part has dosed 3 participants with HAE every 6 months (1 at 122.5 mg and 2 at 245 mg), doses lower or less frequent than the phase 3 regimens of 300 mg every 6 months and 240 mg every 3 months. All 3 participants were attack-free for 3 months after the first dose, and the November 2026 presentation reports that 1 participant (245 mg every 6 months) had one mild attack on Day 153 and then remained attack-free for 21 consecutive months as of the November 5, 2026 cutoff. Sponsor documents report that no attacks occurred while prekallikrein suppression exceeded 80%. No treatment-related serious adverse events were reported as of the August 5, 2026 safety cutoff; injection site reactions, headache, laboratory changes in blood creatine phosphokinase, and ALT/AST increases are listed as the most common adverse events, without counts.

This evidence has four limits. It rests on 3 participants with HAE in an uncontrolled study, with no published baseline or on-treatment attack rates, baseline characteristics, or quality-of-life results. It is reported only in congress abstracts and posters, a press release, a corporate presentation, and SEC filings; no peer-reviewed full article was identified, and the registry record shows “No Results Posted”. The documents disagree on the number of healthy volunteers (43 or 49), on the dose to which the 94% nadir reduction is attributed, and on the suppression threshold proposed for attack prevention (greater than 80% or greater than 70%). No head-to-head trial or indirect comparison that includes onvuzosiran was identified, and STOP-HAE compares it with placebo alone.

Questions a P&T committee would need answered when phase 3 results and a price are published:

  • The reduction in time-normalized attack rate from Day 22 to Week 25 with 300 mg every 6 months compared with placebo (the primary endpoint), the corresponding result for 240 mg every 3 months, and the proportion of participants attack-free with each regimen.
  • Protection during the first weeks of treatment: attacks before Day 22 are excluded from the primary endpoint, the sponsor reports a lag of about 22 days between the first dose and full inhibition, and STOP-HAE requires washout of previous long-term prophylaxis, so the trial does not test continuing an existing agent during that lag.
  • Durability and safety beyond 6 months, including the hepatic enzyme and creatine phosphokinase changes, which the extension study (onvuzosiran for up to 3 years; estimated primary completion 2030-06-30) is designed to address.
  • Applicability: STOP-HAE enrolls adults aged 18 years or older with type I or type II HAE and excludes clinically significant renal disease, so no data address children, HAE with normal C1 inhibitor, or renal impairment.
  • Whether onvuzosiran would be self-administered or given by a health care professional, which the sources do not state.
  • The annual cost relative to current agents, whose 2025 estimated annual wholesale acquisition costs range from $579,885 (berotralstat) to $747,006 (donidalorsen every 4 weeks), and the resulting budget impact.

Onvuzosiran appears in no guideline list of first-line prophylactic agents, which comprises subcutaneous plasma-derived C1 inhibitor, lanadelumab, berotralstat, garadacimab, and donidalorsen. The sponsor compares it with donidalorsen, which acts on the same target: in OASIS-HAE, donidalorsen 80 mg reduced trough prekallikrein by 73% every 4 weeks and 47% every 8 weeks, and the sponsor describes its attack-free rate as “only 35% (Q8W) to 53% (Q4W)”. The sponsor's target profile for onvuzosiran is a greater than 90% reduction in monthly attacks with greater than 75% of patients attack-free on Q3M or Q6M dosing, and the sponsor states that cross-study comparisons “should be interpreted with caution”. A network meta-analysis of approved agents, which does not include onvuzosiran, ranked garadacimab (SUCRA 94%) and lanadelumab every 2 weeks (SUCRA 83%) highest for attack-rate reduction. A second prekallikrein-directed siRNA, BW-20805, reported reductions in monthly attack rate of 83% to 96% across three regimens in a study of 25 participants with HAE.