Base Editing vs Prime Editing: Post-CRISPR Gene Therapy

Everyone remembers the moment. December 2023: the FDA approved Casgevy, the first CRISPR-edited therapy in history. The stock of CRISPR Therapeutics jumped 20% in a day. The narrative was simple: gene editing works, and it is here.

Here is what nobody told you: Casgevy cuts both DNA strands. That double-strand break triggers cellular repair mechanisms that can delete chromosome arms, shuffle gene sequences, or silence tumor suppressors. The FDA knows this. Casgevy’s label requires 15 years of long-term follow-up for malignancy monitoring. At $2.2 million per patient and a manufacturing process that takes months per batch, fewer than 500 patients have started treatment in two and a half years.

CRISPR stocks are priced for the Casgevy ramp. That thesis is fully in the market. The next trade is what comes after the double-strand break: base editing and prime editing, two technologies that edit DNA without cutting it. Here are the five companies that matter, ranked by technology, clinical data, and timeline to an approval.

Base Editing: Beam Has the Lead

Beam Therapeutics (BEAM) closed at $27.48 with a $2.84 billion market cap. The company owns the foundational IP for base editing, a technology that fuses a modified Cas9 enzyme that nicks a single DNA strand to a deaminase enzyme that chemically converts one DNA letter to another. No double-strand break, no chromosomal chaos.

Beam has two clinical programs that together form the most advanced base editing pipeline in biotech. The first is ristoglogene autogetemcel, or risto-cel, an ex vivo base-edited cell therapy for sickle cell disease. In the Phase 1/2 BEACON trial, published in the New England Journal of Medicine, four patients with at least one month of follow-up achieved mean fetal hemoglobin above 60% with sickle hemoglobin suppressed below 40%. Total hemoglobin rose from a baseline of 9.3 g/dL to 17.9 g/dL at peak. No vaso-occlusive crises were reported post-treatment. One patient died of respiratory failure attributed to busulfan conditioning, not the edited cells. Dosing in the registrational cohort is now complete and Beam is enrolling the global registrational cohort designed to support an accelerated approval.

The second program, BEAM-302, is an in vivo liver-targeted base editing therapy for alpha-1 antitrypsin deficiency. A single lipid nanoparticle infusion corrects the PiZ mutation that causes toxic protein aggregation in the liver and emphysema in the lungs. Beam reported topline data from 29 patients in March 2026 showing durable correction, and in July 2026 dosed the first patient in the global registrational cohort. Updated data arrives at the European Respiratory Society Congress on September 8, 2026. BEAM-302 is the most advanced in vivo gene editing program in the clinic.

Prime Editing: The Search-and-Replace

Prime Medicine (PRME) closed at $3.08 with a market cap of $559 million. If base editing changes a single letter, prime editing can search for any DNA sequence and replace it with any other. The technology fuses a nickase Cas9 to a reverse transcriptase with a “prime editing guide RNA” that carries the edit template. It can perform all 12 base substitutions plus small insertions and deletions without a double-strand break.

PRME’s lead program, PM359, is an ex vivo prime-edited cell therapy for chronic granulomatous disease, a severe immunodeficiency where patients cannot fight bacterial and fungal infections. In December 2025, the NEJM published first-in-human data on two patients. Both achieved rapid neutrophil engraftment with 69% and 83% of neutrophils showing restored NADPH oxidase activity by Day 30, far exceeding the 20% threshold considered clinically meaningful. Both patients remained free of new CGD-related complications through six months of follow-up. The FDA has indicated clinical data may be sufficient to support accelerated approval.

The catch: CGD affects roughly 3,500 people in the US. The market is too small to sustain a public company. PRME’s new CEO, Allan Reine, paused active recruitment in the CGD trial after joining in May 2025 and pivoted the pipeline toward Wilson’s disease, AATD, and cystic fibrosis, indications where the patient populations are an order of magnitude larger. The pivot is rational but it resets the clinical clock. PRME is a platform technology with no near-term catalyst. The NEJM data proves prime editing works in humans; the question is whether the company can survive long enough to apply it to a commercial indication.

In Vivo: The Delivery Race

Two companies are competing to deliver gene editing inside the body rather than in a lab dish. In vivo editing eliminates apheresis, lab processing, and reinfusion: a single IV infusion does the job. Intellia Therapeutics (NTLA, $11.94, $1.67 billion market cap) uses CRISPR-Cas9 delivered by lipid nanoparticle. Verve Therapeutics (VERV, $11.13, $992 million market cap) uses base editing, also by LNP.

Intellia’s nex-z, also called NTLA-2001, is an in vivo CRISPR therapy for transthyretin amyloidosis with polyneuropathy. In Phase 1, a single dose produced a 92% mean serum TTR reduction sustained at 24 months with 72% of patients showing clinically meaningful neuropathy improvement. The Phase 3 MAGNITUDE-2 trial is enrolling and Intellia targets a BLA filing by 2028.

But the program hit a wall in early 2026 when a patient in the Phase 3 MAGNITUDE trial for cardiomyopathy experienced a grade 4 liver adverse event. Intellia paused enrollment in both Phase 3 trials. The FDA lifted the clinical hold after the company implemented risk mitigation strategies, but the liver signal is a read-through for the entire in vivo editing field: delivering editing machinery to the liver in a lipid nanoparticle is a trade-off between efficacy and hepatotoxicity. Intellia is the CRISPR company closest to an in vivo approval, but the safety overhang is real.

Verve’s story is similar but with a twist. VERVE-101, an in vivo base editor targeting PCSK9 to permanently lower LDL cholesterol, showed up to 73% LDL-C reduction in the Heart-1 trial. But the same trial produced liver laboratory abnormalities that led the company to pause enrollment and shift to VERVE-102, a second-generation construct. VERVE-102 has shown dose-dependent, sustained PCSK9 and LDL-C reductions in the ongoing Heart-2 trial with an improved safety profile, published in NEJM. The cardiovascular market is massive: 39 million Americans have elevated LDL-C. If VERVE-102 clears Phase 2 with clean liver data, Verve targets a market orders of magnitude larger than any rare disease gene editing program.

The CRISPR Cautionary Tale

Editas Medicine (EDIT) closed at $2.88 with a market cap of $442 million. The company was once valued above $4 billion. It has fallen 94% from its 2021 peak.

Editas uses CRISPR-Cas12a to edit the HBG1/2 promoter in patient stem cells ex vivo. The program, reni-cel, has produced real clinical data: 28 patients in the Phase 1/2/3 RUBY trial achieved mean total hemoglobin of 13.8 g/dL at Month 6 with fetal hemoglobin at 48.1%, well-tolerated with a safety profile consistent with busulfan conditioning.

So why is the stock at $2.88? Editas spent seven years and over $2 billion to produce a CRISPR therapy entering the same sickle cell market already occupied by Casgevy and Beam’s risto-cel. Being third to market in a disease affecting 100,000 US patients with a $2 million price tag is not a business. Editas is the cautionary tale: great science, wrong product strategy, stock destruction.

The Timeline Is the Trade

The investment case for gene editing 2.0 is a timeline bet. Base editing reaches the market first. Beam’s risto-cel could be approved by 2028 for sickle cell if the registrational cohort replicates the BEACON data. BEAM-302 for AATD could follow in 2029. That gives Beam two shots on goal within three years. At $2.84 billion, Beam trades at roughly the same market cap as Intellia at $1.67 billion despite having two programs in registrational cohorts to Intellia’s post-hold Phase 3 restart. The market is pricing Beam as if both programs carry 50% failure risk. The BEACON data implies a higher probability, particularly for risto-cel where the hemoglobin switch mechanism is validated by Casgevy and the base editing advantage is cleaner editing with less off-target risk.

Prime Medicine at $559 million is an option on a platform the NEJM just validated in humans. The market prices PM359’s CGD indication at approximately zero because the company deprioritized it. That may be correct for CGD, but it ignores the Wilson’s disease and AATD programs. The risk is that PRME runs out of cash before proving prime editing in a commercially relevant indication. The company burned $166 million in 2025. At $559 million, the market is saying “show me the next indication.”

Verve at $992 million is a pure bet on VERVE-102. The Heart-2 trial needs clean liver data and durable LDL-C reduction in a larger cohort. If it does, the addressable market justifies a valuation north of $5 billion. If it does not, the stock drops 50%. This is a binary catalyst in the next 12 months.

Intellia at $1.67 billion is the in vivo CRISPR leader with a cloud: MAGNITUDE-2 is enrolling again, but the liver AE changed the risk profile. The stock is down 85% from its 2021 peak because investors price the safety overhang. If nex-z delivers clean Phase 3 data, the re-rating potential is substantial. But the BLA timeline is 2028 at earliest, and the liver signal is not resolved by regulatory clearance alone. It is resolved by the next data cut.

The Risks

The single biggest risk for every company in this analysis is the same: delivery. Lipid nanoparticles accumulate in the liver. This is both a feature, for programs targeting liver-expressed genes like AATD and PCSK9, and a bug, for programs where liver toxicity limits dosing. Verve and Intellia both triggered liver safety signals because their LNPs hit the liver at doses high enough to edit the target gene. The line between therapeutic editing and hepatotoxicity is narrow, and it is the same line for every in vivo program.

For ex vivo programs, the risk is manufacturing and commercial execution. Casgevy proved that the science works but the patient journey is too long and too expensive for mass adoption. Beam’s risto-cel and Editas’s reni-cel face the same logistical hurdles. Beam has an advantage: base editing produces higher editing yields with fewer cell collection cycles, reducing the manufacturing burden. But it does not eliminate it.

The Verdict

Base editing is the technology to own for the next three years. Beam has two programs in registrational cohorts, NEJM-published data, and a market cap that does not reflect the probability of at least one approval by 2029. BEAM is a buy under $30 with a two-year hold. The downside is $15 if both programs fail; the upside is $80 if either succeeds. That is 5:1 risk-reward.

Prime Medicine is the long-dated option. Prime editing is the most versatile technology in gene editing: it can theoretically correct 89% of known disease-causing mutations. But PRME at $559 million is pricing in zero near-term revenue. If the Wilson’s disease program produces preclinical data by early 2027, the stock re-rates. If it does not, the cash runs out. This is a 1% position in a biotech portfolio, not a conviction holding.

Verve is the binary bet. VERVE-102 data in the next 12 months either validates in vivo base editing for a $5B+ indication or kills the thesis. Buy after the liver safety readout, not before.

Intellia is a hold. The CRISPR in vivo thesis is intact but the liver overhang caps the multiple. Wait for data.

Editas is a sell. The technology and clinical data are real, but being third to a market that cannot support three competitors at $2 million per patient is a structural problem, not a scientific one.