Gene editing with the CRISPR-Cas9 system moved from laboratories into clinical practice in 2023. That year, the U.S. Food and Drug Administration (FDA) approved for the first time a medicinal product based on CRISPR genome editing, for selected patients with sickle cell disease. An expanded approval followed, and in 2024 a conditional marketing authorisation in the European Union. This article summarises what these treatments can do, whom they are intended for, what risks they carry, and why their approval does not concern germline editing.
How CRISPR-Cas9 editing works
The CRISPR-Cas9 system can be described in simplified terms as a molecular tool that can find a short stretch of DNA and cut both of its strands. The cell then repairs the break with its own mechanisms, which may lead to the disruption of a regulatory segment or to a change in the function of the target gene. In the clinical applications approved through 2024, editing is not performed directly in the patient’s body but outside the body (ex vivo): physicians collect the patient’s haematopoietic stem cells, which are modified in the laboratory and, after checks, returned by infusion. This procedure reduces some of the uncertainties of direct administration into the body, but it requires prior chemotherapeutic conditioning (myeloablation), which itself carries serious risks. The advantage is that the modified cells can be tested before they are returned to the patient; the disadvantages are complexity, the length of hospitalisation, and the fact that only specialised centres can deliver the treatment. Data on long-term safety remain limited, and patients are followed over the long term.
Casgevy: the first approved CRISPR treatment
The product Casgevy (active substance exagamglogene autotemcel, abbreviated exa-cel) was developed by Vertex Pharmaceuticals and CRISPR Therapeutics. The FDA approved it on 8 December 2023 for patients aged 12 and older with sickle cell disease and recurrent vaso-occlusive crises, making it the first CRISPR-based medicine approved in the United States. The approval rested on data from single-arm clinical trials, that is, trials without a control group, which is common for rare and serious diseases but also limits the strength of the evidence. The detailed definition of the indication is given on the FDA product page. In January 2024, the FDA extended the approval to transfusion-dependent beta thalassaemia in patients aged 12 and older.
European approval and conditional marketing authorisation
The European Commission granted Casgevy a conditional marketing authorisation effective from 9 February 2024 for both sickle cell disease and transfusion-dependent beta thalassaemia in patients aged 12 and older for whom haematopoietic stem cell transplantation is appropriate and no suitable related donor is available. Conditional authorisation means that, in the view of the European Medicines Agency (EMA), the benefit of immediate availability outweighs the risk arising from less complete data than is normally required, and the marketing authorisation holder must subsequently submit results from ongoing studies — the product is therefore approved “on condition” that further evidence is provided. The product overview also stipulates that treatment may only take place at an approved centre under the supervision of a physician experienced in stem cell transplantation, and that safety is also monitored through a registry with a planned follow-up of 15 years.
Why sickle cell disease and beta thalassaemia
Sickle cell disease and beta thalassaemia are both inherited disorders of haemoglobin, the protein that carries oxygen in red blood cells. In sickle cell disease, a point mutation causes red blood cells to deform into a sickle shape, leading to painful vessel blockages and organ damage. In severe beta thalassaemia, the body does not produce enough functional haemoglobin chains, and patients depend on regular transfusions with subsequent chelation therapy against iron overload. Both diseases share the fact that they are caused by defective haemoglobin production in blood-forming cells, which makes them logical first targets for a therapy that modifies the patient’s own haematopoietic cells. Casgevy does not repair the mutation itself: the edit targets the erythroid-specific enhancer of the BCL11A gene, thereby restoring production of foetal haemoglobin, which may partly take over the function of the defective adult haemoglobin. The choice of this indirect target has its reason: in adulthood, the BCL11A gene helps suppress foetal haemoglobin, and its targeted downregulation specifically in the red blood cell lineage may increase the share of foetal haemoglobin without intervening in other tissues where BCL11A acts. Published interim results suggest that most followed-up patients achieved a longer period without severe crises or without the need for transfusions after treatment, but the data come from small cohorts with relatively short follow-up and must be interpreted cautiously (study in sickle cell disease, study in beta thalassaemia). A key unanswered question remains the durability of the effect over decades: no one can yet say with certainty whether the modified stem cells will keep producing enough foetal haemoglobin for the patient’s entire life, or whether their function will weaken over time.
Lyfgenia: a gene therapy that is not CRISPR
On the same day, 8 December 2023, the FDA also approved Lyfgenia (lovotibeglogene autotemcel from bluebird bio) for patients aged 12 and older with sickle cell disease and a history of vaso-occlusive events. Unlike Casgevy, it is not CRISPR editing but lentiviral gene therapy: a functional copy of the gene for a modified beta-globin is added to the patient’s stem cells. The distinction matters because both procedures have a different mechanism and a different risk profile. Lyfgenia carries in its package insert a boxed warning about the risk of haematologic malignancy, that is, a blood cancer, and patients are to be monitored long term after treatment. Details are given in the package insert on the FDA website. Lyfgenia had not received authorisation in the EU through 2024; its story illustrates that even approved one-time therapies remain burdened with serious safety questions.
Risks: off-target effects and what is not yet known
Every intervention in the genome carries a risk of unintended changes. They are technically called off-target effects: the tool may cut DNA at sites similar to the target segment, which may theoretically impair the function of another gene. Manufacturers and regulators therefore require analyses of potential off-target sites before approval, but according to review papers these methods capture only some of the possible events, and their clinical significance is difficult to assess (2025 review). Further risks include failure of the modified cells to engraft, cytopenias including a drop in blood platelets with a risk of bleeding, infertility after myeloablative conditioning, and as yet unknown long-term consequences, including a theoretical risk of cancers. The conditioning chemotherapy with busulfan itself, without which the modified cells would not engraft in the bone marrow, carries a risk of serious infections and organ toxicity, so the overall balance of the treatment also includes harm caused by the procedural steps, not only by the editing itself. That is precisely why both the EMA and the FDA require many years of follow-up of treated patients in registries. It must be stated openly that the datasets are small, follow-up is short, and the benefit-risk balance may change as more data accumulate.
Price and availability: a one-time treatment with a multimillion price tag
The approved gene therapies for haemoglobinopathies are among the most expensive medicines in the world. According to manufacturers’ announcements from December 2023, the U.S. list price of Casgevy was $2.2 million and of Lyfgenia $3.1 million for a single administration, while actual payments by insurers may differ and European prices are subject to national negotiations. Manufacturers argue that a one-time payment may replace decades of costly chronic care, while payers point to the limited data on durability of effect and to the budgetary burden. To the price of the product itself must be added the costs of cell collection, hospitalisation, myeloablative conditioning, and long-term follow-up, so the real costs of treatment are even higher. As a result, treatment remains available only at selected centres and for a narrow group of patients, which raises the question of fair access, especially in countries with a high incidence of sickle cell disease and limited resources. This text gives no treatment or reimbursement recommendations; decisions always belong to the treating physician and to regulators.
Somatic versus germline: where the boundary lies
The key ethical dividing line runs between somatic editing and germline editing. The approved therapies, including Casgevy, modify only the blood-forming cells of a specific patient; the change is not inherited and is not passed on to offspring. Editing embryos or germ cells, by contrast, would be heritable and would affect all future generations, which is why the expert community and regulators consider it unacceptable outside strictly defined research. A cautionary precedent is the case of the Chinese biophysicist He Jiankui, who in November 2018 announced the birth of twins from embryos edited with CRISPR with the aim of disabling the CCR5 gene. He carried out the procedure without a proper ethical and regulatory framework, with problematic informed consent and with a risk of mosaicism and off-target changes. On 30 December 2019, a court in Shenzhen sentenced him to three years in prison and a fine for illegal medical practice; his collaborators received lighter sentences. China subsequently tightened its legislation, and the World Health Organization called for a registry of human genome editing research. The case still serves as an argument for why the clinical success of somatic therapies must not open the path to heritable interventions without societal consensus and robust safety data.
Sources
- FDA approves first gene therapies for sickle cell disease (Casgevy and Lyfgenia) – https://www.fda.gov/news-events/press-announcements/fda-approves-first-gene-therapies-treat-patients-sickle-cell-disease
- Casgevy – product page on the FDA website – https://www.fda.gov/vaccines-blood-biologics/casgevy
- Vertex press release on FDA approval of Casgevy (2023) – https://news.vrtx.com/news-releases/news-release-details/vertex-and-crispr-therapeutics-announce-us-fda-approval
- Casgevy – product page on the EMA website – https://www.ema.europa.eu/en/medicines/human/EPAR/casgevy
- Casgevy – EMA medicine overview (PDF) – https://www.ema.europa.eu/en/documents/overview/casgevy-epar-medicine-overview_en.pdf
- Exagamglogene autotemcel in transfusion-dependent beta thalassaemia (NEJM, 2024) – https://www.nejm.org/doi/full/10.1056/NEJMoa2309673
- Exagamglogene autotemcel in severe sickle cell disease (NEJM, 2024) – https://www.nejm.org/doi/full/10.1056/NEJMoa2309676
- Lyfgenia – product page on the FDA website – https://www.fda.gov/vaccines-blood-biologics/lyfgenia
- Lyfgenia – FDA package insert (PDF) – https://www.fda.gov/media/174610/download
- Measurement and clinical interpretation of CRISPR off-target effects (Nature Genetics, 2025) – https://www.nature.com/articles/s41588-025-02428-3
- Summary of the European Commission approval of exa-cel (2024) – https://www.cgtlive.com/view/european-commission-approves-exa-cel-scd-tdt
- The He Jiankui case – summary and sentencing (2018–2019) – https://en.wikipedia.org/wiki/He_Jiankui_affair
Content Transparency and AI Assistance
How this article was created:
This article was written by Hermes Agent — an open-source self-improving AI agent (Nous Research) — in collaboration with our editors. The editors set the topic, research direction, and primary sources; Hermes prepared the research, structure, and draft. Every key claim passed iterative fact-checking against two independent sources plus blind review audits.
Editorial review and fact-checking:
- ✓ The text was editorially reviewed
- ✓ Fact-checking: All key claims and data were verified against two independent sources
- ✓ Fact corrections and enhancement: Our editors corrected inaccuracies and added subject matter expertise
- ✓ Multi-round audit: The article passed repeated blind audits including a final shared review round
AI model limitations (important disclaimer):
Language models can generate plausible-sounding but inaccurate or misleading information (known as “hallucinations”). We therefore strongly recommend:
- Verifying critical facts in the primary sources (official documentation, peer-reviewed research, subject matter authorities)
- Not relying on AI-generated content as your sole information source for decision-making
- Applying critical thinking when reading
Spotted a mistake?
Let us know in the comments under the article — we check every report.
Technical details:
- Agent: Hermes Agent (Nous Research, open-source)
- Model: Muse Spark 1.3 Contributor (Meta)
- Method: Self-improving editorial skill — research, drafting, multi-round blind audit, human review
- Run: Self-hosted Hermes Agent instance run by our editors, connected to the model provider’s API (not a local model)
Be First to Comment