CRSP CTX310: One Year of Durability for In Vivo Editing

One number buried in today’s ESC Congress presentation does more work than any headline. CRISPR Therapeutics reported that a single IV infusion of CTX310, its in vivo CRISPR-Cas9 therapy targeting ANGPTL3, kept a mean 79% reduction in ANGPTL3 (maximum 89%), a 48% cut in triglycerides (maximum 78%), and a 53% cut in LDL cholesterol (maximum 84%) sustained through one full year of follow-up. The efficacy numbers themselves are not news. They were already published in a Phase 1 paper in the New England Journal of Medicine last year. What is new, and what the field has been waiting for, is the durability. The one-year persistence answers the only question that ever mattered for in vivo editing: does the edit last?

CRSP trades at $57.72, a roughly $5.5 billion market cap against $2.36 billion in cash and marketable securities as of June 30, 2026. Back out the cash and the market is assigning roughly $3.2 billion of enterprise value to the entire pipeline: CTX310, the companion Lp(a) program CTX320, the CTX112 CAR-T franchise, and a slice of the CASGEVY royalty stream from Vertex. The read: that $3.2 billion is pricing the one-and-done lipid platform at close to zero, and the durability data is the first hard evidence that the platform is worth far more.

What the Data Actually Showed

CTX310 is an LNP-delivered CRISPR-Cas9 therapy that edits the ANGPTL3 gene in hepatocytes after a single intravenous infusion. That is in vivo CRISPR in its purest form: the edit happens inside the body, no cells removed, no transplant. The Phase 1 trial (NCT07491172) was open-label, dose-escalating from 0.1 to 0.8 mg/kg, and enrolled 15 patients across four lipid groups: homozygous familial hypercholesterolemia (HoFH), heterozygous familial hypercholesterolemia (HeFH), severe hypertriglyceridemia (sHTG), and mixed dyslipidemia.

Two things matter more than the raw numbers. First, the trial’s primary endpoint was safety and tolerability, not efficacy. All 15 participants completed at least a year of follow-up with no treatment-related serious adverse events and no grade 3 or higher transaminase elevations. That is the line that matters, because liver toxicity is the historical fear for any LNP-delivered, CRISPR-based liver program. Second, the durability held. The November 2025 readout showed a mean 73% ANGPTL3 reduction at the top dose. A year later the mean is 79%. The edit did not fade, drift, or lose potency. That persistence, published in NEJM as a correspondence, is the proof point that turns a gene-editing proof of concept into a potential one-time therapeutic.

The program now advances into a Phase 1b trial at a fixed 0.8 mg/kg flat dose, prioritizing sHTG and mixed dyslipidemia, with the next update expected in the second half of 2026. Approval is years away, through the standard Phase 2 and Phase 3 gauntlet, and CRISPR has been honest about that in its own framing.

Why Durability Is the Field’s Core Question

The reason this matters has nothing to do with the mechanism’s elegance and everything to do with a boring, unglamorous problem: people do not stay on lipid-lowering therapy. Statins are daily pills that roughly half of patients quietly abandon within a year. PCSK9 monoclonal antibodies require injections every two to four weeks, and even the siRNA Leqvio (inclisiran) still needs a shot every six months. The entire value proposition of a one-time edit is that it removes the adherence variable entirely. The Cleveland Clinic investigator who presented the data put it plainly: the biggest challenge in lipid management is staying on therapy, not finding a therapy that works.

That is the bull case in one sentence. A single infusion that produces a durable 53% LDL reduction competes with a lifetime of daily pills and intermittent injections, and it wins on the one axis patients consistently fail: persistence. The one-year data is the first real evidence that the edit is stable enough to make that promise credible.

The Competitive Picture: Same Target, Different Technology

CTX310 is not alone in the ANGPTL3 race, and the competitor list is how you size the opportunity.

Arrowhead’s zodasiran is the direct comp. It is an RNAi therapy targeting the exact same ANGPTL3 gene, dosed quarterly, and it is already in the Phase 3 YOSEMITE trial in HoFH, with enrollment complete and a readout expected in mid-2027. In Phase 2, zodasiran cut fasting LDL by a mean 35.7% at the 200 mg dose and 39.9% at 300 mg in HoFH patients whose baseline LDL averaged roughly 9.8 mmol/L. Arrowhead trades at a $12.5 billion market cap, more than twice CRISPR’s enterprise value, for a program that still requires a shot every three months forever. That is the comparison that frames the whole trade: if CTX310’s one-and-done edit is durable, it should ultimately be worth more than a quarterly siRNA that does the same thing, not less.

Verve’s VERVE-102 is the gene-editing comp, and it is no longer independent. Eli Lilly acquired Verve in a deal worth roughly $1.0 billion to $1.3 billion, closing in the third quarter of 2025. VERVE-102 uses base editing to knock out PCSK9 rather than CRISPR to knock out ANGPTL3, and its Heart-2 Phase 1b trial showed dose-dependent LDL cuts of 9% to 62% with durability out to 18 months. The distinction between the two editing tools matters, and it is worth reading the base editing versus prime editing breakdown to understand why a nuclease (CRISPR) and a base editor carry different safety and durability profiles. The Lilly deal is the cleanest signal that big pharma is already paying up for one-and-done cardiovascular editing. It also means the pure-play in vivo cardiovascular editing universe is now essentially CRISPR Therapeutics, which is exactly why a $5.5 billion valuation looks interesting.

Regeneron’s Evkeeza (evinacumab) is the approved ANGPTL3 antibody for HoFH, dosed by monthly IV infusion. It validates the target but carries the infusion burden the field is trying to escape. The siRNA and antibody approaches prove ANGPTL3 is a real, approvable target. CTX310 is betting the winner will be the one that does not need a recurring dose.

Valuation: The Market Is Pricing the Platform at Zero

CRISPR at $5.5 billion with $2.36 billion in cash is not a typical biotech balance sheet. More than 40% of the market cap is cash. The remaining roughly $3.2 billion of enterprise value has to cover CTX310, CTX320 (the Lp(a) program, the largest untapped lipid market), CTX112 CAR-T, and the CASGEVY economics with Vertex.

Compare that to Arrowhead at $12.5 billion for a portfolio where zodasiran and plozasiran are both still pre-approval and both require recurring dosing. Compare it to what Lilly paid for Verve: $1.3 billion for a Phase 1b PCSK9 base editor. On both comps, CRISPR’s in vivo lipid franchise is being valued at a discount to a recurring-dosing RNAi peer and at a fraction of what the mechanism could command if the one-and-done promise holds through Phase 2. This is the same second-wave argument laid out in the CRISPR stocks roundup: the first wave of gene editing was ex vivo and niche, and the second wave, in vivo editing for common diseases, is where the market-cap upside lives.

The reason the discount exists is legitimate. This is Phase 1a, open-label, fifteen patients, no efficacy endpoint, and no randomized comparison. Single-arm Phase 1 durability data, no matter how clean, does not guarantee Phase 3 success. And the nuclease approach CRISPR uses makes a double-strand break, which carries a different long-term off-target question than base editing. Those unknowns only larger, longer trials can resolve.

Risks

Arrowhead is ahead: zodasiran is in Phase 3 with a readout in mid-2027, and if it shows strong efficacy with a clean safety profile, it will be on the market years before CTX310 and will set the bar. The durability proof is still one year. Cardiovascular disease is a decades-long problem, and the field needs five-year and ten-year persistence data before “one and done” is more than marketing. Further out, CTX310’s next catalyst is a Phase 1b update in H2 2026, not a registrational readout. Between now and any approval there are multiple binary events, any one of which could reset the thesis. On top of that, the company is running multiple programs at once, including the CTX112 CAR-T effort, so capital allocation and the $600 million of convertible notes issued this year are real balance-sheet variables to watch.

The Verdict

The durability data is real, peer-reviewed, and answers the field’s core question. It is a validation of the platform, not yet a validation of the drug. At $57.72 and a $5.5 billion market cap with $2.36 billion in cash, the option is the buy, not the thesis. A small position, one to two percent of a biotech portfolio, sized for the binary reality that Phase 1b data and a Phase 3 launch are still years away. If the H2 2026 Phase 1b readout confirms durable sHTG lowering at the flat dose, the stock re-rates toward the Arrowhead comp and beyond, because the market will finally start pricing the one-and-done lipid franchise instead of zero. If the durability fades or a safety signal emerges, you lose the option premium and little else. That premium is worth paying for the only pure-play way to bet that a single infusion can replace a lifetime of pills.