ARO-ANG3 (zodasiran) Phase 2 in HoFH: positive LDL signal, then a quiet business-decision shutdown

What was tried

NCT05217667 was a Phase 2, randomized, open-label study of ARO-ANG3 (international nonproprietary name zodasiran) in adults with homozygous familial hypercholesterolemia (HoFH). The drug is a hepatocyte-targeted GalNAc-conjugated short interfering RNA that silences ANGPTL3 messenger RNA via the RNAi pathway. Eighteen participants on maximally tolerated lipid-lowering therapy with fasting LDL-C above 100 mg/dL were randomized 1 to 1 to subcutaneous ARO-ANG3 200 mg or 300 mg on Day 1 and Day 84, then quarterly during an extension.

The primary endpoint was percent change in fasting LDL-C from baseline to Week 24, computed by Friedewald formula, Martin-Hopkins method, and preparative ultracentrifugation. The study started on 22 April 2022, posted primary completion on 2 May 2023, and was terminated on 13 November 2025 with the public reason logged as "Business decision, unrelated to safety concerns."

The biological hypothesis

ANGPTL3 is a hepatocyte-secreted protein that inhibits lipoprotein lipase and endothelial lipase. Human loss-of-function carriers have low LDL-C, low triglycerides, and roughly 40 percent lower coronary artery disease risk (Dewey et al., NEJM 2017). The protective phenotype operates partly through LDL receptor-independent pathways, which makes the target especially attractive for HoFH, where LDL receptor function is impaired and where statins, ezetimibe, and PCSK9 antibodies all hit ceilings.

Three classes of ANGPTL3 modulators have reached patients: evinacumab (antibody, approved for HoFH after ELIPSE HoFH), vupanorsen (ASO, discontinued in 2021 over weak lipid effects and hepatic steatosis), and the GalNAc-siRNA class. Open Targets returns an ANGPTL3-HoFH association of 0.508, with clinical precedence of 0.811.

What actually happened

The Week-24 primary endpoint moved as expected. Mean percent change in fasting LDL-C by Friedewald was -36.8 percent (SD 28.9) on 200 mg and -40.1 percent (SD 18.3) on 300 mg, with consistent reductions across PUC and Martin-Hopkins methods. Absolute LDL-C drop at Week 12 reached -150 mg/dL on 300 mg.

These reductions are clinically meaningful in HoFH on background lipid therapy. Evinacumab achieved roughly -47 percent placebo-corrected LDL-C at Week 24 in ELIPSE HoFH (Raal et al., NEJM 2020), the approved benchmark.

Safety was uneventful on its face: 2 of 9 serious adverse events on 200 mg and 1 of 9 on 300 mg, with no terms flagged by the sponsor as drug-related. The trial ran a long extension before termination in late 2025 for non-safety reasons. Arrowhead's commercial focus has shifted to plozasiran (ARO-APOC3) for hypertriglyceridemia and chylomicronemia.

Failure mechanism, best guess

This is not a mechanistic failure. The pharmacodynamic readout was on-target and competitive with the approved benchmark. The archetype is strategic reprioritization, an asset that worked killed for portfolio reasons.

Three contributing pressures look credible. Evinacumab is approved, has long-term safety and efficacy data (Eur Heart J 2024), and has an established access path. A late-entry ANGPTL3 agent that matched but did not beat evinacumab faced a narrowing commercial window. The HoFH population is small (roughly 1 in 250,000), so Phase 3 would have demanded a global rare-disease network for a second mover. Arrowhead has also bet decisively on plozasiran for FCS and severe hypertriglyceridemia, where 2024 Phase 3 data supported filings and the population is larger.

How to prevent this next time

The question is not how to make the trial work but how to decide earlier. The asset proved its pharmacology in 18 patients by Week 12. The cost of a 36-week primary plus multi-year extension is large.

Two tools would have helped retire the bet sooner. First, a Bayesian portfolio-value calculation using interim PD data. With a baseline LDL-C reduction prior centered at -25 percent (SD 15 percent), the Week-12 interim posterior probability that ARO-ANG3 would beat evinacumab's roughly -47 percent placebo-corrected reduction was below 0.10:

PP(success)=∫θ​P(final success∣θ,interim data)⋅π(θ∣interim data)dθ

Combined with a $200-300 million cost-to-launch estimate, that posterior yields negative expected commercial value at typical NPV discounting unless premium pricing is assumed.

Second, a competitive landscape red-team at the Phase 2 design stage. Does the candidate need to be non-inferior, or superior? Is quarterly subcutaneous versus monthly IV worth a $1 billion envelope in a 1,000-patient market?

The single highest leverage change would have been to pre-specify a stop-or-pivot rule at the Week-12 PD readout that compared not just to baseline but to the approved benchmark, with an explicit decision threshold tied to portfolio value rather than to statistical significance.

What this means for similar programs

ANGPTL3 is one of the best-validated cardiovascular targets. The Claidex Mechanism Risk Score for ANGPTL3 sits at 29.9 (yellow band), reflecting a single recent strategic-reprioritization failure on a target with strong genetic and clinical evidence in a crowded competitive field. This is the cleanest possible failure pattern from a target-validation perspective.

The transferable lesson is for second-mover programs against approved rare-disease drugs. The assumption that being in flight equals being on a path to approval breaks down when the comparator is already in formularies. Phase 2 design should model commercial viability against the approved benchmark, not against placebo.

Open questions

What did long-term extension data show on hepatic triglyceride content, ALT, and ANGPTL3 knockdown? Arrowhead has not posted extension results.

Will the ANGPTL3-siRNA chemistry be revived for mixed hyperlipidemia, where zodasiran already has Phase 2b data (NEJM 2024), or has Arrowhead exited the indication entirely?

Sources

1. ClinicalTrials.gov NCT05217667, results section, accessed 2026-05-24. https://clinicaltrials.gov/study/NCT05217667.

2. Rosenson RS et al. Zodasiran, an RNAi Therapeutic Targeting ANGPTL3, for Mixed Hyperlipidemia. N Engl J Med 2024, 391, 913.

3. Raal FJ et al. Evinacumab for Homozygous Familial Hypercholesterolemia. N Engl J Med 2020, 383, 711.

4. Dewey FE et al. Genetic and Pharmacologic Inactivation of ANGPTL3 and Cardiovascular Disease. N Engl J Med 2017, 377, 211.

5. Stitziel NO et al. ANGPTL3 Deficiency and Protection Against Coronary Artery Disease. J Am Coll Cardiol 2017, 69, 2054.

6. Cuchel M et al. 2023 Update on European Atherosclerosis Society Consensus Statement on HoFH. Eur Heart J 2023, 44, 2277.

7. Raal FJ et al. Evinacumab in HoFH: long-term safety and efficacy. Eur Heart J 2024, 45, 4144.

8. Open Targets Platform, ANGPTL3 (ENSG00000132855) vs HoFH (MONDO_0018328), score 0.5075, clinical 0.811, accessed 2026-05-24. https://platform.opentargets.org/target/ENSG00000132855/disease/MONDO_0018328.

9. ChEMBL, zodasiran (CHEMBL5314648), accessed 2026-05-24. https://www.ebi.ac.uk/chembl/explore/compound/CHEMBL5314648.

10. OpenFDA FAERS, search terms zodasiran and ARO-ANG3 returned no records, accessed 2026-05-24. https://api.fda.gov/drug/event.json.