In December 2023, the FDA approved Casgevy (exagamglogene autotemcel), the first CRISPR-based gene therapy, for sickle cell disease. It was a landmark: the technology that Jennifer Doudna and Emmanuelle Charpentier had first described in 2012 had reached patients in just over a decade. But Casgevy, for all its significance, used the original, bluntest form of CRISPR — Cas9 nuclease, which cuts both strands of DNA at a target site and relies on the cell's own (imprecise) repair machinery to fix the break.
The next generation of gene editors — base editors and prime editors — don't cut the double helix at all. And their clinical results are beginning to arrive.
The Precision Ladder
Understanding the clinical significance of base editing and prime editing requires understanding what they fix about first-generation CRISPR.
CRISPR-Cas9 creates a double-strand break (DSB) at a specified genomic location. The cell then repairs the break through one of two pathways: non-homologous end joining (NHEJ), which is error-prone and typically disrupts the target gene, or homology-directed repair (HDR), which can insert a desired sequence but works inefficiently in most cell types. Cas9 is excellent for gene knockout (disruption) but unreliable for precise correction.
Base editors , developed by David Liu's laboratory at the Broad Institute (first published in 2016), fuse a catalytically impaired Cas9 to a deaminase enzyme that chemically converts one DNA base to another without cutting the double helix. Cytosine base editors (CBEs) convert C·G to T·A; adenine base editors (ABEs) convert A·T to G·C. Together, they can correct approximately 30% of known pathogenic point mutations. Because no DSB is created, base editing avoids the insertions, deletions, and chromosomal rearrangements that can accompany Cas9 cutting.
Prime editors , also from Liu's lab (published in 2019), use a modified Cas9 fused to a reverse transcriptase, guided by a "prime editing guide RNA" (pegRNA) that contains both the targeting sequence and a template for the desired edit. Prime editors can make any single-base change, plus small insertions and deletions, without double-strand breaks or donor DNA templates. They are the most precise and versatile editors yet developed — capable of correcting an estimated 89% of known pathogenic mutations.
David Liu was awarded the 2025 Breakthrough Prize in Life Sciences for developing both technologies.
Verve Therapeutics: Base Editing for Heart Disease
The most clinically advanced base editing programme is Verve Therapeutics' VERVE-102, targeting PCSK9 in the liver to reduce LDL cholesterol. The approach uses an adenine base editor delivered via lipid nanoparticles — a one-time intravenous infusion intended to permanently lower cholesterol without the need for ongoing statin therapy or PCSK9 antibody injections.
Preclinical data in non-human primates showed durable reductions in PCSK9 protein and LDL cholesterol lasting over a year after a single dose. The Phase I trial, initiated in 2024, is evaluating safety and LDL reduction in patients with heterozygous familial hypercholesterolaemia — a genetic condition affecting roughly 1 in 250 people that dramatically increases cardiovascular risk.
Verve's second programme, targeting ANGPTL3 (CTX310, developed with CRISPR Therapeutics), addresses triglyceride levels through a similar base editing approach. Preclinical primate studies showed sustained reductions in the ANGPTL3 protein and triglycerides after a single treatment.
The significance of these programmes extends beyond cardiology. They represent the first systematic application of base editing to common diseases affecting millions of people, as opposed to the rare monogenic conditions that have dominated gene therapy development. If successful, they would demonstrate that gene editing can be a mainstream therapeutic modality — not just a last resort for otherwise untreatable genetic disorders.
Prime Medicine: The First Human Data
In May 2025, Prime Medicine reported what may be the most significant milestone in gene editing since Casgevy's approval: the first-ever clinical data from a prime editing therapeutic.
PM359 was administered to a patient with chronic granulomatous disease (CGD), a rare primary immunodeficiency caused by mutations in genes encoding components of the NADPH oxidase complex. Patients with CGD cannot generate the reactive oxygen species that immune cells use to kill bacteria and fungi, leaving them vulnerable to severe, recurrent infections.
A single dose of PM359 — involving ex vivo prime editing of the patient's own haematopoietic stem cells, followed by reinfusion — restored NADPH oxidase activity. The company reported that the treatment showed both safety and efficacy, with functional correction of the underlying enzymatic defect.
This is a proof of concept for the entire prime editing modality. CGD was chosen deliberately: it has a clear molecular defect, a measurable functional endpoint (oxidase activity), and a well-understood natural history. A positive result in CGD demonstrates that prime editing can make precise corrections in human stem cells, that those corrections persist after transplantation, and that the edited cells function normally in the body.
The Expanding Clinical Landscape
As of February 2026, CRISPR Medicine News monitors approximately 250 clinical trials involving gene-editing therapeutic candidates, with more than 150 currently active. The field has expanded dramatically beyond the initial sickle cell disease and beta-thalassaemia programmes.
Oncology represents the largest single category, with CRISPR-edited CAR-T cells in trials for multiple haematological malignancies and solid tumours. The editing typically knocks out genes that limit T-cell persistence or function (PD-1, TGF-β receptor, TCR) or inserts synthetic receptors that redirect the cells against tumour antigens.
In vivo editing — delivering the editor directly into the patient's body rather than editing cells outside the body — is the frontier. Intellia Therapeutics' NTLA-2001 demonstrated the first successful in vivo CRISPR editing in humans, targeting TTR in the liver for transthyretin amyloidosis. Phase I data showed dose-dependent reductions in serum TTR protein, with the highest dose group achieving an 87% mean reduction — comparable to the efficacy of chronic RNA interference therapies but from a single infusion.
Neurological applications are emerging. Base editing and prime editing offer the precision needed to correct mutations in the central nervous system, where the consequences of off-target editing would be particularly severe. Programmes targeting Huntington's disease, Rett syndrome, and various neurometabolic disorders are in preclinical development.
Safety and the Off-Target Question
The central safety concern for all gene editing approaches is off-target editing — unintended modifications at genomic locations similar to the target site. For Cas9, which creates double-strand breaks, off-target cuts can cause insertions, deletions, or chromosomal translocations. For base editors and prime editors, which do not cut both DNA strands, the off-target risk profile is different and generally considered lower, but not zero.
Extensive preclinical characterisation using whole-genome sequencing, unbiased off-target detection methods (GUIDE-seq, CIRCLE-seq, DISCOVER-seq), and long-term animal studies has been required for each clinical programme. The FDA has issued guidance documents specifying the analytical methods expected for characterising off-target editing in investigational gene therapies.
To date, no clinical trial of any CRISPR-based therapy has reported a serious adverse event attributable to off-target editing. However, the follow-up periods remain relatively short (1–3 years for most programmes), and the theoretical risk of rare off-target events contributing to cancer or other delayed toxicities cannot be fully excluded until longer-term data accumulate.
The Delivery Challenge
The biggest bottleneck for gene editing therapeutics is not the editor itself — it is delivery. Getting the editing machinery into the right cells, in sufficient quantities, without triggering immune responses, remains the field's most significant engineering challenge.
For ex vivo approaches (editing cells outside the body), delivery is relatively straightforward: electroporation of ribonucleoprotein complexes into isolated cells works efficiently. But ex vivo editing is limited to cell types that can be extracted, edited, and returned — primarily blood and immune cells.
For in vivo approaches, lipid nanoparticles (LNPs) are the current standard, but they traffic predominantly to the liver. Reaching other tissues — muscle, brain, lung, kidney — requires novel delivery vehicles: engineered adeno-associated viruses (AAVs), virus-like particles, exosomes, or synthetic nanoparticles with tissue-specific targeting ligands. Each delivery system adds its own safety, manufacturing, and regulatory complexity.
The 2026 Readouts to Watch
The next 12 months will bring clinical data that could reshape the field. Verve's Phase I base editing data for PCSK9 reduction will determine whether in vivo base editing is safe and effective in a common disease population. Prime Medicine's expanding CGD programme and additional prime editing candidates will establish whether the PM359 result is reproducible and durable. And the continued follow-up of Casgevy recipients — now approaching three years post-treatment — will provide the longest-term safety data for any CRISPR therapy.
Gene editing is no longer a laboratory curiosity or a regulatory experiment. It is a clinical modality with approved products, expanding trial programmes, and next-generation technologies entering humans. The question has shifted from "can we edit the human genome safely?" to "how many diseases can we treat this way?" The early data suggests the answer is: far more than anyone expected a decade ago.