Phase 1 Trial of CRISPR-Cas9 Gene Editing Targeting ANGPTL3

Authors, Journal, Affiliations, Type, DOI

• Authors: Luke J. Laffin, Stephen J. Nicholls (co-first authors); Russell S. Scott; Peter M. Clifton; John Baker; Ashish Sarraju; Shweta Singh; Qiuqing Wang; Kathy Wolski; Huansheng Xu; Jen Nielsen; Naimish Patel; Jason M. Duran; Steven E. Nissen (senior author)
• Affiliations: Cleveland Clinic (Dept of Cardiovascular Medicine; Coordinating Center for Clinical Research); Victorian Heart Institute, Monash University, Melbourne; New Zealand Clinical Research, Christchurch; Royal Adelaide Hospital; Aotearoa Clinical Trials, Auckland; CRISPR Therapeutics, Boston
• Journal: New England Journal of Medicine 2025;393:2119-30 (published online 8 November 2025)
• Type: Phase 1A, multicenter, open-label, single-ascending-dose first-in-human trial
• DOI: 10.1056/NEJMoa2511778
• Registration: ACTRN12623000809639 (Australia New Zealand Clinical Trials Registry)
• Funding: CRISPR Therapeutics (the sponsor also employs several authors)

Overview

First-in-human trial of CTX310, a lipid-nanoparticle–encapsulated CRISPR-Cas9 drug product designed to make a permanent loss-of-function edit in hepatic ANGPTL3 from a single intravenous infusion. Fifteen adults with hypercholesterolemia, hypertriglyceridemia, or mixed dyslipidemia refractory to maximally tolerated therapy received one dose across five ascending dose levels. There were no dose-limiting toxic effects and no CTX310-related serious adverse events; the commonest events were grade 2 infusion reactions (20%) and a single transient aminotransferase elevation that resolved by day 14. Lipid lowering was dose-dependent above 0.6 mg/kg, with the 0.8 mg/kg cohort reaching −48.9% LDL cholesterol and −55.2% triglycerides — a simultaneous reduction the authors note no existing therapy achieves. The trial is small, open-label and descriptive, and its 60-day efficacy follow-up is negligible against a permanently edited genome.

Keywords

ANGPTL3, angiopoietin-like protein 3, CRISPR-Cas9, in vivo gene editing, lipid nanoparticle, CTX310, hypercholesterolemia, hypertriglyceridemia, mixed dyslipidemia, familial hypercholesterolemia, LDL cholesterol, triglycerides, loss-of-function mutation, one-time therapy

Key Takeaways

Rationale and Target

• ANGPTL3 inhibits lipoprotein lipase and endothelial lipase; naturally occurring ANGPTL3 loss-of-function variants cause lifelong reductions in triglycerides and LDL cholesterol with reduced ASCVD risk and no apparent harmful effects
• ANGPTL3 is produced almost exclusively by the liver, making it an attractive target for hepatocyte-directed editing
• Existing ANGPTL3-directed approaches (monoclonal antibody, ASO, siRNA) all require long-term repeat administration; CRISPR-Cas9 editing offers durable modification after a single treatment
• The authors' stated unmet need: current oral therapies for hypertriglyceridemia have limited effectiveness and no treatment simultaneously and substantially lowers both LDL cholesterol and triglycerides

Drug Product (CTX310)

• Two components co-encapsulated in a lipid nanoparticle: mRNA encoding Streptococcus pyogenes Cas9, and a 100-mer single guide RNA targeting ANGPTL3
• The polyadenylated Cas9 mRNA contains methylated pseudouridine to reduce inflammatory response
• LNP comprises four components: an ionizable lipid, a polyethylene glycol lipid, 1,2-distearoyl-sn-glycero-3-phosphocholine, and cholesterol
• Hepatic uptake via apolipoprotein E–mediated LDL receptor uptake, scavenger receptor–mediated uptake, and potentially other receptors; endosomal escape delivers the payload to the cytoplasm with importin-mediated nuclear delivery
• Off-target editing was not observed in preclinical studies

Trial Design and Population

• Phase 1A, multicenter, open-label, single ascending dose; six sites in Australia, New Zealand and the United Kingdom; enrolment May 2024 – August 2025
• 27 persons screened, 15 enrolled and treated: 0.1 mg/kg (n=3), 0.3 (n=3), 0.6 (n=3), 0.7 (n=2), 0.8 (n=4). Doses calculated on total RNA administered, based on estimated lean body weight
• Eligibility: uncontrolled hypercholesterolemia (familial or non-familial), moderate-to-severe hypertriglyceridemia, or mixed dyslipidemia, refractory to maximally tolerated lipid-lowering therapy
• A 0.7 mg/kg cohort was added, and the 0.8 mg/kg cohort expanded, to better define the dose–response relationship
• Premedication with glucocorticoid and antihistamine; direct observation ≥24 h; daily safety assessments for 3 days; weekly follow-up for 30 days, then days 60, 90, 180, 270, 360
• Background lipid-lowering therapy was unchanged in all participants through 60 days, so the lipid changes are attributable to CTX310

Baseline Characteristics (Table 1, N=15)

• Median age 53 years (range 31–68); 87% male; BMI 31.1±4.9; 93% white
• Clinical ASCVD 40%; familial hypercholesterolemia 40% (1 homozygous, 5 heterozygous; 5 genetically confirmed)
• Severe hypertriglyceridemia 13%; mixed dyslipidemia 40%; non-familial hypercholesterolemia 7%; type 2 diabetes 33%
• Directly measured LDL cholesterol 154.6±79.2 mg/dL; triglycerides median 192.2 mg/dL (IQR 108.9–252.4); non-HDL cholesterol 203.2±73.1 mg/dL; apolipoprotein B 132.1±43.1 mg/dL; ANGPTL3 161.8±58.0 ng/mL
• Background therapy: statin 60%, ezetimibe 53%, PCSK9 monoclonal antibody 40%, fibrate 27%, icosapent ethyl 7%, apheresis 7%, evinacumab 7%; statin intolerance 27%

Primary End Point — Safety

• No dose-limiting toxic effects. No serious adverse events related to CTX310
• Dose-limiting toxicity was predefined as: aminotransferase elevation to CTCAE grade ≥3 persisting >14 days; bilirubin or INR to grade ≥3; any grade 3 laboratory abnormality persisting ≥7 days; or any grade 4 laboratory abnormality
• Any investigator-reported adverse event in 14 participants (93%): grade 1 in 4 (27%), grade 2 in 9 (60%), grade 3 in 0, grade 4 in 0, grade 5 in 1 (7%)
• Two serious adverse events (13%), neither attributed to CTX310: a spinal disk herniation requiring hospitalisation 7 months after the 0.3 mg/kg dose, and one death 179 days after the 0.1 mg/kg dose
• Adverse events of special interest in 4 participants (27%): allergic/localised reaction 1 (7%), infusion-related reaction 3 (20%), aminotransferase elevation 1 (7%)
• Infusion-related reactions (20%): grade 2 in all cases, occurring in two participants at 0.6 mg/kg and one at 0.8 mg/kg; symptoms began minutes after infusion start and consisted of back pain and nausea; managed by pausing the infusion with supportive care (antiemetics, repeat premedication, optionally acetaminophen); all resolved and every participant completed the infusion
• Allergic reaction (7%): rash on the upper torso one day after the 0.3 mg/kg dose, resolved the following day
• Aminotransferase elevation (7%): CTCAE grade 2, between 3 and 5 times baseline, in a participant who already had elevated aminotransferases at baseline (0.8 mg/kg dose, who also had an infusion reaction). Peaked on day 4, was less than twice baseline by day 7, and returned to baseline by day 14. No concomitant rise in bilirubin, alkaline phosphatase, or prothrombin time
• The authors note similar effects have been reported for other LNP-delivered therapies and consider the safety data acceptable to proceed to larger studies

Secondary Efficacy End Points

Percent change in ANGPTL3 was measured at 30 days; lipid biomarker changes are reported at 90 days for the 0.1, 0.3 and 0.6 mg/kg doses and at 60 days for the 0.7 and 0.8 mg/kg doses — the time points are therefore not directly comparable across rows.

• The two lowest doses produced no ANGPTL3 reduction — levels rose by roughly 9% at both 0.1 and 0.3 mg/kg, with LDL cholesterol and triglycerides also rising
• The highest dose (0.8 mg/kg, four participants, 60 days) produced a mean reduction of 48.9% in LDL cholesterol and 55.2% in triglycerides, with wide individual ranges (−86.6 to −19.5% for LDL-C; −83.7 to −37.9% for TG)
• The dose–response is not monotonic and ANGPTL3 knockdown does not track lipid lowering: 0.7 mg/kg achieved the deepest ANGPTL3 reduction (−79.7%) yet only −21.0% LDL-C and −19.2% TG, whereas 0.8 mg/kg achieved a shallower ANGPTL3 reduction (−73.2%) but roughly double the lipid effect. With n=2 and n=4 respectively this may be sampling variation, but the authors acknowledge that response is probably influenced by factors beyond administered dose — hepatic steatosis, inflammation, and pre-existing genetic and metabolic profile
• Wide within-cohort variability was observed at identical doses, which the authors attribute partly to the heterogeneity of the enrolled lipid disorders

Interpretation and Context

• The mechanism of action and the ANGPTL3 and lipid reductions through 60 days at the highest doses suggest that extensive hepatocyte gene editing was probably achieved
• Observed reductions are similar to those of evinacumab, the anti-ANGPTL3 monoclonal antibody approved for homozygous familial hypercholesterolemia
• Contrast with RNA-level ANGPTL3 inhibitors: vupanorsen (antisense oligonucleotide) development was discontinued for worsening hepatic steatosis and aminotransferase elevations, whereas zodasiran and solbinsiran (siRNA) are associated with reduced hepatic fat fraction
• The authors position one-time therapy as addressing the well-documented problem of waning adherence — up to half of patients discontinue statins and PCSK9 monoclonal antibodies within 12 months
• Regulatory framing: FDA guidance recommends follow-up for up to 15 years after genome-editing drug administration
• The authors explicitly temper enthusiasm: "as with all irreversible interventions, enthusiasm must be tempered by the need for rigorous, long-term follow-up to ensure both durability and safety"

Limitations of the document

• Small sample size (n=15) and a primary objective of safety rather than efficacy; the combination of small n and open-label design precludes formal statistical comparisons. All analyses were descriptive with no hypothesis testing and no imputation
• Heterogeneous population: participants had various lipid disorders with broad baseline ranges for LDL cholesterol and triglycerides, limiting assessment of efficacy in any specific population (e.g. homozygous FH, severe hypertriglyceridemia)
• Limited generalisability: restricted racial and ethnic diversity (93% white), limited geographic distribution (six sites in three countries, all Australia/NZ/UK), and few women enrolled (87% male)
• Short follow-up: 60-day minimum follow-up, with additional follow-up ongoing. This is negligible relative to potential lifetime exposure to an edited genome; late-emerging or low-incidence safety signals, including off-target editing consequences, cannot be excluded. The authors state ongoing surveillance is essential
• Efficacy time points are non-uniform: ANGPTL3 reported at 30 days; lipids at 90 days for the three lower doses and 60 days for the two highest — cross-dose comparisons of lipid effect are therefore not like-for-like
• Durability is not established. The premise is a permanent edit, but this report cannot demonstrate permanence; the separation of the 0.7 and 0.8 mg/kg cohorts (n=2, n=4) is far too small to define the dose–response relationship reliably
• Uncontrolled, no comparator arm: there is no placebo or active comparator (e.g. evinacumab, siRNA), so the comparative efficacy and safety of CTX310 remain unestablished
• Sponsor involvement: funded by CRISPR Therapeutics, which employs several authors; data were collected by PPD (a contract research organisation) and the sponsor performed the initial statistical analysis. Statisticians at the Cleveland Clinic Coordinating Center independently confirmed all results, and final content decisions were reserved for academic authors with no publication restrictions — but the sponsor reviewed the manuscript and suggested revisions
• The single death (179 days after the lowest dose) is reported as unrelated to CTX310, but with n=15 and one grade 5 event, the trial is far too small to exclude a low-frequency treatment-related risk
• Table 3 extraction caveat: the PDF conversion rendered Table 3 as fragmented column text. The per-dose LDL-C and TG values in this page were reconstructed from that table and cross-checked against the narrative — the 0.8 mg/kg row reproduces the discussion's stated −48.9% and −55.2% exactly, and every ANGPTL3 value matches the results narrative. The 0.7 mg/kg lipid values rest on the table alone and carry more uncertainty

Key Concepts Mentioned

• concepts/ANGPTL3-Inhibition — the target; this is the first clinical ANGPTL3 gene-editing trial, providing primary data for a therapy the wiki previously knew only via review articles
• concepts/Lipid-Gene-Therapy — in vivo CRISPR-Cas9 editing of a hepatic lipid gene via LNP; the ANGPTL3 arm of the clinical GET landscape
• concepts/Gene-Editing-Risk-Benefit-Framework — a worked example of the irreversibility trade-off: one-time durable therapy against 15-year regulatory follow-up
• concepts/CRISPR-Cas9-in-Channelopathies — same editing platform, different target organ and delivery route (LNP to liver, not AAV to myocardium)
• concepts/Familial-Hypercholesterolemia — 40% of the enrolled population; the eventual target indication
• concepts/Hypertriglyceridemia-Management — −55.2% triglycerides at the highest dose; addresses the limited effectiveness of existing oral TG therapies
• concepts/Dyslipidemia — context for where durable gene editing might sit relative to statins, ezetimibe, PCSK9 inhibitors and siRNA
• concepts/APOC3-Inhibition — parallel hepatically-targeted TG-lowering pathway

Key Entities Mentioned

• entities/CTX310 — the investigational drug product itself
• entities/Evinacumab — anti-ANGPTL3 monoclonal antibody; efficacy benchmarks directly comparable to CTX310 in this trial
• entities/CRISPR-Therapeutics — sponsor and developer of CTX310 (also CTX320 for Lp(a), CTX340 for AGT)
• concepts/Familial-Hypercholesterolemia — enrolled population and primary future indication

Wiki Pages Updated

• wiki/sources/CRISPR-Cas9-NEJM-2025.md — created (this page)
• wiki/entities/CTX310.md — created
• wiki/entities/CRISPR-Therapeutics.md — created
• wiki/concepts/ANGPTL3-Inhibition.md — CTX310 upgraded from secondary to primary source; dose–response table added; source_count 2→3
• wiki/concepts/Lipid-Gene-Therapy.md — CTX310 section rewritten with primary data; source_count 3→4
• wiki/concepts/Gene-Editing-Risk-Benefit-Framework.md — CTX310 added as a worked irreversibility case; source_count updated
• wiki/concepts/Hypertriglyceridemia-Management.md — simultaneous LDL-C/TG lowering added
• wiki/concepts/Familial-Hypercholesterolemia.md — primary CTX310 data added
• wiki/concepts/CRISPR-Cas9-in-Channelopathies.md — CTX310 primary citation added
• wiki/entities/Evinacumab.md — head-to-head efficacy comparison with CTX310
• wiki/sourceindex.md, wiki/wikiindex.md — index rows
• log.md — ingest entry