Gene Editing Technology: How CRISPR Could Change Medicine

01 / The story

How gene editing works, in plain language

CRISPR works like a search-and-edit tool for DNA. A guide RNA leads an enzyme (usually Cas9) to a matching DNA sequence. Conventional CRISPR cuts the DNA there, and the cell's own repair processes produce an edit. Importantly, a treatment does not have to fix the disease-causing mutation itself—some work by disabling a regulatory switch instead. [1, 4]

Two newer tools refine the idea. Base editing chemically changes individual DNA letters, usually without cutting both strands. Prime editing uses a targeting system plus an RNA-encoded template to write in a specified change. Both can reduce certain risks, but neither eliminates unintended edits. [7, 8]

There is also a logistical split. Ex vivo editing removes cells from the patient, edits them in a lab, tests them and returns them. In vivo editing delivers the editing machinery directly into the body—simpler in theory, but the machinery must reach enough of the right cells without affecting the wrong ones. Fat-based nanoparticles work reasonably well for the liver; reaching many other tissues remains hard. [1, 4, 6, 8]

02 / The story

What is actually approved: Casgevy

Casgevy (exagamglogene autotemcel) is the first approved CRISPR-based medicine. It edits a regulatory region controlling a gene called BCL11A in a patient's own blood-forming stem cells, which raises production of fetal hemoglobin. It does not directly repair the original hemoglobin mutation—it works around it. [1, 2]

The UK's MHRA authorized it on November 16, 2023 for eligible patients aged 12 and older with sickle cell disease or transfusion-dependent beta thalassemia. The FDA approved it for sickle cell disease with recurrent vaso-occlusive crises on December 8, 2023, and for transfusion-dependent beta thalassemia on January 16, 2024. [1, 2, 3]

One detail worth updating: the FDA's Casgevy page, dated July 2, 2026, now lists an indication for patients aged two years and older. If you see the original '12 and older' threshold quoted as the current U.S. indication, that is out of date—check the current package insert. [2]

In the FDA's initial sickle-cell evaluation, 29 of 31 evaluable patients stayed free of severe vaso-occlusive crises for at least 12 consecutive months. That is an evaluable subgroup, not every enrolled patient, and it is not evidence of lifetime freedom from symptoms or reversal of existing organ damage. [1]

Also approved alongside Casgevy for sickle cell disease was Lyfgenia, which uses a different gene-therapy approach entirely. Not every gene therapy is CRISPR. [1]

03 / The story

The treatment is not a simple injection

Casgevy's pathway involves collecting stem cells, editing them in a specialized facility, then conditioning—the patient receives chemotherapy to make room in the bone marrow—before the edited cells are reinfused. Conditioning carries serious risks: very low blood counts, infection and fertility concerns. [1, 2]

This matters for how we weigh benefits. Some of the burden belongs to the transplant-style workflow, not the editor itself. It also explains why treatment is limited to specialized centers and why long-term follow-up is essential. [1, 2]

04 / The story

Recent developments, in proportion

On September 8, 2026, Intellia announced that the FDA accepted and granted priority review to its application for lonvoguran ziclumeran, an in-vivo CRISPR therapy for hereditary angioedema, with a target action date of March 10, 2027. This is a substantial regulatory milestone—but it is not approval, and the source is a company announcement. [5]

On September 7, 2026, Cell Stem Cell published a four-patient report on base editing for sickle cell disease and transfusion-dependent thalassemia across three trials. The authors reported favorable outcomes and no off-target mutations detected by the tests used. 'None detected' is not the same as 'none exist,' and four patients cannot establish population-level safety. [9]

On September 30, 2026, Science Translational Medicine reported prime editing of patient-derived blood stem cells for GATA2 deficiency, with up to 70% correction of the studied variant. This is preclinical cell-based work—no patients were treated. The editing percentage is not a clinical cure rate. [10]

Earlier milestones matter too. A December 2025 NEJM paper reported the first human use of prime editing: two participants with chronic granulomatous disease showed restored enzyme activity within a month, persisting through six and four months of follow-up. Early proof of concept, not established efficacy. [7] And a May 2025 NEJM report described a custom in-vivo base-editing therapy made for one infant with CPS1 deficiency, who tolerated more dietary protein over seven weeks without serious adverse events. An exceptional single-patient result with short follow-up—not a ready-made treatment model. [6]

05 / The story

Risk, cost and access

Editing risk is broader than 'off-target cuts.' It includes unintended changes at the target site, large deletions or rearrangements, bystander edits from base editors, delivery-related toxicity and immune responses. The specific risk depends on the product, target, cell type and delivery method. Greater molecular precision is not the same as proven clinical safety. In April 2026 the FDA issued draft guidance on sequencing-based safety assessment for genome-editing products—draft and nonbinding, but a sign of where oversight is heading. [4, 8, 12]

A January 2025 Connecticut government presentation records a U.S. list price of $2.2 million for Casgevy. That is a dated list price, not a verified current net price, and it excludes the full hospital pathway. Specialized centers, eligibility rules, manufacturing logistics, travel and fertility preservation all create barriers beyond the drug payment. [1, 13]

On financing, CMS documents updated in September 2026 describe 32 states plus DC and Puerto Rico participating in its cell and gene therapy access model, covering 84% of Medicaid beneficiaries with sickle cell disease. That is progress on payment structure—not proof that 84% have received therapy or that access barriers are gone. [11]

06 / The story

Somatic versus heritable: the line that matters

Every treatment discussed here edits the patient's own body cells—blood stem cells or liver cells. None is intended to alter reproductive cells, so the changes are not passed to future children. [14]

Editing embryos or reproductive cells for reproduction is a separate category with distinct risks: inherited changes, consent from future generations who cannot give it, and governance questions no single country can settle alone. The World Health Organization recommends robust oversight and international coordination; those are recommendations, not a single worldwide law, and national rules differ. [14]

07 / The story

What to watch next

The field's near-term questions are concrete. Will the FDA approve the first in-vivo CRISPR therapy when it rules on lonvo-z in 2027? Will base and prime editing results hold up in larger groups with longer follow-up? Will durability data from the earliest Casgevy patients stay strong? And will access models actually move patients through treatment, not just payment agreements? [5, 7, 11]

CRISPR's medical era has genuinely begun. But the distance between 'approved for specific blood disorders at specialized centers' and 'routine genetic medicine' is measured in delivery problems, durability data and dollars—not in headlines.