On September 13, 1999, Jesse Gelsinger drove from Tucson, Arizona, to the University of Pennsylvania's Institute for Human Gene Therapy in Philadelphia. He was eighteen years old, healthy enough to live a near-normal life, and eager to participate in a clinical trial he hoped might eventually help infants with his condition — ornithine transcarbamylase (OTC) deficiency, a metabolic disorder affecting the liver's ability to process ammonia. Jesse had a mild form of the disease, managed through diet and medication. The infants his participation might help were far sicker.
Three days after receiving an infusion of a high-dose adenoviral vector carrying the corrective gene, Jesse Gelsinger was dead. The cause was a massive systemic inflammatory response — his immune system had attacked the viral vector and, in doing so, destroyed him. He was the first person to die directly as a result of gene therapy. Subsequent investigation revealed that the University of Pennsylvania team had concealed from regulators the occurrence of serious adverse events in previous subjects. The principal investigator, James Wilson, had a financial stake in a company developing the technology. Seventeen monkeys had died in preclinical studies — information that had not been fully disclosed to the FDA.
Jesse Gelsinger's death halted the field. Clinical trials were suspended or placed under intense regulatory scrutiny. Congressional hearings were held. The promise that had made gene therapy one of the most anticipated fields in medicine — the idea that hereditary disease could be cured by correcting the underlying genetic error — suddenly seemed both distant and dangerous.
Fast-forward to December 2023. Victoria Gray, a woman from Mississippi who had spent her life in debilitating pain from sickle cell disease, was told that her treatment with Casgevy — the first approved therapy using CRISPR gene editing — had succeeded. Her bone marrow was producing fetal hemoglobin that compensated for her defective adult hemoglobin. She had been transfusion-free for years. At the American Society of Hematology annual meeting, her physician described her case as one of medicine's defining moments. The same month, Casgevy received FDA approval, making it the first CRISPR-based therapy approved for clinical use anywhere in the world.
"It's hard to convey how dark it was after Jesse Gelsinger died. It felt like the end. And now here we are." — Katherine High, Remarks at the American Society of Gene and Cell Therapy (2019)
Key Definitions
Gene therapy: A medical approach that treats or prevents disease by delivering functional genetic material into cells to compensate for, replace, or edit defective genes.
Viral vector: A virus that has been genetically modified to remove pathogenic genes and carry a therapeutic payload — a functional gene or gene-editing components — into target cells. The virus retains its cell-entry machinery but cannot replicate or cause disease.
Adeno-associated virus (AAV): The most widely used viral vector in current gene therapy. A small, naturally occurring virus that does not cause human disease, capable of infecting a wide range of tissues, and relatively well-tolerated by the immune system. Its limited cargo capacity (approximately 4.7 kilobases) is the main constraint.
Lentiviral vector: A vector derived from retroviruses (including HIV) that can integrate stably into the host cell's genome. Useful for therapies requiring long-term gene expression in dividing cells; integration carries a theoretical risk of insertional mutagenesis.
CRISPR-Cas9: A gene editing tool derived from a bacterial immune system. A guide RNA directs the Cas9 protein to a specific DNA sequence, where it makes a targeted double-strand cut. The cell's repair machinery then either disrupts the gene (NHEJ) or uses a provided template to correct it (HDR).
Somatic gene therapy: Gene therapy that modifies cells in the patient's body; changes are not heritable and affect only the treated individual.
Germline gene therapy: Modification of embryonic or reproductive cells, which would result in heritable changes passed to future generations. Currently prohibited for reproductive purposes in most jurisdictions.
Ex vivo gene therapy: Cells are removed from the patient, modified in the laboratory, and reinfused. Provides greater control over the editing process and quality assessment before cells are returned.
In vivo gene therapy: Genetic material is delivered directly into the patient's body, with the target cells receiving the therapy in their natural location.
Insertional mutagenesis: The risk that a viral vector integrating into the genome does so at a location that disrupts a tumor suppressor gene or activates an oncogene, potentially contributing to cancer.
The Science: How Genes Cause Disease
To understand gene therapy, it helps to understand how defective genes cause disease in the first place.
Most of the body's functions are carried out by proteins — enzymes that catalyze biochemical reactions, structural proteins that give cells their shape, receptor proteins that receive signals, transport proteins that carry molecules across membranes. Proteins are produced from instructions encoded in DNA. Mutations — changes in the DNA sequence — can disrupt this process in several ways: they can prevent a protein from being produced at all, produce a truncated or unstable protein, produce a protein with altered function, or produce a protein that actively causes harm.
Hereditary diseases are those caused by mutations that are present in every cell of the body because they were present in the fertilized egg. For single-gene (monogenic) disorders, the disease is caused by mutations in one specific gene. These are the diseases most amenable to gene therapy: if you can deliver a functional copy of the gene to the relevant cells, you can restore the missing function. More than 7,000 rare diseases are caused by single-gene mutations, and most currently have no approved treatment.
The cells that need to be targeted vary by disease. For spinal muscular atrophy, the motor neurons of the spinal cord need the SMN1 gene product. For hemophilia B, the liver cells need Factor IX. For sickle cell disease, the blood-forming stem cells in the bone marrow need correction of the hemoglobin gene. Reaching the right cells is one of the central engineering challenges of gene therapy.
Delivery: The Challenge at the Heart of the Field
A functional gene is a sequence of DNA — a large, charged molecule that cannot simply be injected into the bloodstream and expected to find its way to the nucleus of the right cell. Cells have elaborate defenses against foreign nucleic acids, which they appropriately treat as potential viral threats. Crossing the cell membrane, escaping endosomes, reaching the nucleus, and achieving stable expression of the therapeutic gene are each formidable engineering problems.
Viruses have solved these problems through billions of years of evolution. They enter cells via surface receptor interactions, escape endosomal degradation, and deliver their genetic material to the right intracellular location. Gene therapists exploit this toolkit by creating viral vectors: viruses from which the pathogenic genes have been removed and in which the therapeutic gene of interest has been inserted. The resulting vector is a delivery vehicle that can enter cells and deliver its cargo without causing disease.
The choice of vector depends on the biology of the target tissue and the requirements of the therapy. AAV vectors are the workhorses of the current generation of approved gene therapies. Different AAV serotypes — AAV1 through AAV9 and beyond — have different natural tropisms for different tissues. AAV9 crosses the blood-brain barrier and can reach motor neurons when delivered intravenously, which is why it is used for SMA. AAV2 has tropism for retinal cells, making it suitable for ocular gene therapy. AAV5 has hepatocyte tropism, used for hemophilia gene therapy. AAV does not integrate into the genome in most non-dividing cell types; it persists as an episome — a stable DNA circle separate from the chromosomes. This means there is no integration risk, but it also means the therapy is diluted as cells divide. For tissues with low cell turnover, like neurons and photoreceptors, this is not a problem; for bone marrow, it is.
Lentiviral vectors integrate stably into the genome, ensuring that the therapeutic gene is copied with the cell every time it divides. This is essential for therapies targeting hematopoietic stem cells, which give rise to all blood cells and must be capable of long-term self-renewal. The Hemgenix hemophilia B therapy and the Casgevy sickle cell therapy both involve ex vivo lentiviral or CRISPR-based modification of stem cells.
CRISPR: A Revolution in Gene Editing
The development of CRISPR-Cas9 as a practical gene editing tool, described in a landmark 2012 paper by Jennifer Doudna and Emmanuelle Charpentier — for which they were awarded the Nobel Prize in Chemistry in 2020 — transformed the possibilities of gene therapy.
Earlier editing tools — zinc finger nucleases and TALENs — could also make targeted cuts in DNA, but designing them was laborious, expensive, and technically demanding. CRISPR-Cas9 is radically simpler: the targeting is done by a guide RNA (gRNA), a short RNA sequence that matches the genomic target. Designing a new gRNA is a matter of hours, not months. The same Cas9 protein can be directed to any target simply by changing the gRNA. The ease of CRISPR democratized gene editing research and dramatically accelerated the field's pace.
When CRISPR-Cas9 cuts DNA, the cell's repair machinery acts in one of two ways. Non-homologous end joining (NHEJ) is the faster, more common pathway: it re-joins the cut ends, often introducing small insertions or deletions (indels) at the cut site that disrupt the gene. NHEJ-based editing is useful when you want to knock out a gene. Homology-directed repair (HDR) uses a provided DNA template to repair the cut in a specific sequence, allowing precise correction of a mutation. HDR is much less efficient than NHEJ and occurs primarily in dividing cells, which limits its current therapeutic utility.
Newer editing modalities address some of CRISPR's limitations. Base editors, developed by David Liu's laboratory, can convert one DNA base to another — for example, converting an adenine to a guanine — without making a double-strand break. Because double-strand breaks are the primary source of off-target damage, base editing is significantly safer. Prime editing, also from Liu's laboratory and described in 2019, is essentially a "search and replace" for the genome: it can make any of the twelve possible base-to-base conversions and small insertions and deletions with high precision and without double-strand breaks.
The Gelsinger Case and Its Legacy
Jesse Gelsinger's death in 1999 deserves extended analysis because it shaped the regulatory and ethical framework within which gene therapy operates today.
The OTC deficiency trial at the University of Pennsylvania was not a trial for patients who desperately needed the therapy — it was a Phase I trial in mildly affected adults. Phase I trials assess safety, not efficacy, and are designed to identify dose-limiting toxicities. Jesse, who managed his condition adequately with diet and medication, was enrolled to help establish the safe dose range for infants who would be treated in future trials.
The trial used an adenoviral vector, which triggers an immune response more robustly than AAV. Jesse received a high dose. His immune response was catastrophic: within hours, he had systemic inflammatory response syndrome; within days, his kidneys, liver, and lungs had failed. He was eighteen years old and had gone to Philadelphia to help sick babies.
The subsequent investigation by the FDA found multiple failures. The research team had not fully disclosed serious adverse events in earlier participants to the FDA. The consent form had not adequately described the risks. James Wilson, the principal investigator, had a financial interest in a company that held patents on the vector. Institutional review and oversight had been inadequate.
The reforms that followed were extensive. The FDA significantly strengthened oversight of gene therapy trials. The Recombinant DNA Advisory Committee (RAC) enhanced its review processes. Conflict of interest disclosure requirements were tightened. The culture of the field shifted toward greater conservatism about dosing, more rigorous preclinical safety assessment, and clearer separation of financial interests from research oversight.
He Jiankui and the Ethics of Germline Editing
If Gelsinger's death was the field's darkest moment in somatic gene therapy, the announcement by He Jiankui in November 2018 was its most alarming ethical rupture in germline editing.
He, who had been trained in biophysics in the United States before returning to China, used CRISPR-Cas9 to edit human embryos at the CCR5 gene and had the edited embryos implanted, resulting in the birth of twin girls in 2018 and a third child in 2019. CCR5 encodes a co-receptor used by HIV to enter cells; loss-of-function variants of CCR5 (found naturally in a small percentage of Northern European populations) confer resistance to most HIV strains. He justified the experiment on the grounds that the children's father was HIV-positive and that he was giving them a lifelong protection.
The scientific community's condemnation was swift and comprehensive. He had violated the consensus of the international scientific community, which had repeatedly called for a moratorium on heritable human genome editing pending the development of appropriate safety, efficacy, and governance frameworks. The editing was technically flawed: the twins were mosaic, meaning different cells had different editing outcomes, with uncertain biological consequences. CCR5 deletion may increase susceptibility to West Nile virus and potentially influenza. The consent process appears to have been inadequate. The institutional review board approval He claimed to have obtained was disputed and appears to have been obtained through misrepresentation.
He was convicted of illegal medical practice by a Chinese court in December 2019 and sentenced to three years in prison. The three children born from his experiments are living with genetic modifications whose long-term consequences are unknown and that they had no ability to consent to.
The He Jiankui affair accelerated international work on governance frameworks for human germline editing. The International Commission on the Clinical Use of Human Germline Genome Editing, established by the US National Academies of Sciences, Engineering, and Medicine and the UK Royal Society, reported in 2020 that heritable human genome editing should not proceed until there are "stringent independent oversight, reasonable guarantees of safety and efficacy, and broad societal consensus about the appropriateness of the proposed application." No country currently authorizes reproductive germline editing for clinical use.
Approved Therapies: The Current Landscape
The current portfolio of approved gene therapies represents a dramatic change from the near-cessation of the field after 1999.
Luxturna (voretigene neparvovec), approved by the FDA in December 2017, treats a specific form of inherited retinal dystrophy caused by RPE65 mutations. The photoreceptor cells of affected patients gradually degenerate because they lack a functional RPE65 protein essential for the visual cycle. Luxturna delivers functional RPE65 via AAV2 injected into the subretinal space. Clinical trial participants showed meaningful improvements in light sensitivity. The therapy is not a cure for the underlying condition but substantially slows its progression. Pricing at $850,000 for both eyes reflected the rarity of the disease and the high cost of development.
Zolgensma (onasemnogene abeparvovec) for spinal muscular atrophy represents gene therapy's most dramatic early achievement. Infants with SMA Type 1, the most severe form, typically lose all motor function within the first two years of life and require ventilator support. Treated early — in the first weeks of life, before motor neurons are irreversibly lost — Zolgensma allows many children to achieve developmental milestones, including independent sitting and in some cases walking, that would have been impossible without treatment. The $2.1 million list price at approval generated intense debate about how society should price transformative one-time therapies, but cost-effectiveness analyses comparing it to the lifetime cost of supportive care have generally found it within acceptable health economic ranges for severe pediatric disease.
Casgevy (exagamglogene autotemcel), approved for sickle cell disease and transfusion-dependent beta-thalassemia in late 2023, is notable both as the first approved CRISPR-based therapy and as the first that approaches a functional cure for two common, severe, historically undertreated diseases. The mechanism is elegant: rather than correcting the hemoglobin mutation directly, Casgevy edits out a repressor of fetal hemoglobin production (BCL11A), allowing the patient's own cells to produce fetal hemoglobin that compensates for the defective adult hemoglobin. Early clinical trial data showed that the majority of treated patients became transfusion-free. The treatment process is demanding — stem cell mobilization and collection, myeloablative conditioning, the editing process, and reinfusion — and the price is approximately $2.2 million.
CAR-T Cell Therapy: Gene Editing Adjacent
CAR-T cell therapy deserves mention as a related approach that has achieved significant clinical success. Chimeric antigen receptor T-cell therapy involves genetically modifying a patient's own T cells to express a synthetic receptor that targets cancer cells expressing a specific antigen. The modified T cells are expanded in the laboratory and reinfused into the patient, where they seek and destroy cancer cells displaying the target antigen.
CAR-T therapies are approved for several blood cancers, including certain leukemias and lymphomas, and have produced remarkable remissions in patients with disease that had not responded to other treatments. They are not strictly gene therapy in the traditional sense — they do not correct a genetic defect — but they use the same ex vivo cell modification and reinfusion approach and depend on lentiviral or retroviral vectors to deliver the CAR gene into T cells.
The success of CAR-T therapy has expanded the commercial and institutional infrastructure for ex vivo cell and gene therapy and demonstrated that genetic modification of patient cells can produce durable therapeutic benefit at scale. It has also confronted the same access challenges as other gene therapies: approved CAR-T products are among the most expensive treatments in medicine.
Cost, Access, and the Future
The pricing of gene therapies represents one of the most difficult challenges in health policy. The scientific and economic arguments for high prices are real: development costs are enormous, patient populations are typically very small, and the one-time nature of the therapy means the manufacturer recovers costs from a single transaction rather than recurring purchases. The moral and social equity arguments against those prices are equally real: treatments that cost more than two million dollars are effectively unavailable to patients without extraordinary insurance coverage or wealth, and in most of the world, they are simply inaccessible.
New payment models — annuity payments contingent on sustained efficacy, subscription models, outcomes-based contracts — are being explored. But the fundamental tension between the economics of rare disease drug development and the principle of equitable access to transformative treatments is not resolved by payment model innovation alone. It requires broader thinking about how societies fund the development and distribution of high-cost, high-benefit medical technologies.
The future of gene therapy is moving in several directions simultaneously. In vivo CRISPR editing — delivering editing components directly to target tissues in the body, without cell removal and reinfusion — would dramatically simplify the treatment process and extend the range of conditions that can be addressed. Base editing and prime editing approaches, with their improved precision and safety profiles, are moving toward clinical trials. The growing understanding of the genetic architecture of common complex diseases — cardiovascular disease, diabetes, Alzheimer's — raises the longer-term question of whether polygenic common diseases might eventually be addressable through genetic approaches.
The distance from Jesse Gelsinger's death in a Philadelphia hospital to Victoria Gray's cure in a Mississippi clinic is a measure of what twenty-four years of science, regulation, failure, and persistence produced. The field that emerged from its near-death experience is more cautious, more rigorous, and vastly more capable than the one that began with so much premature confidence in the 1990s.
For related reading, see how CRISPR works, how epigenetics works, and how genetic engineering works.
References
- Bainbridge, J. W. B., et al. (2008). Effect of gene therapy on visual function in Leber's congenital amaurosis. New England Journal of Medicine, 358(21), 2231-2239. https://doi.org/10.1056/NEJMoa0802268
- Mendell, J. R., et al. (2017). Single-dose gene-replacement therapy for spinal muscular atrophy. New England Journal of Medicine, 377(18), 1713-1722. https://doi.org/10.1056/NEJMoa1706198
- Jinek, M., et al. (2012). A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science, 337(6096), 816-821. https://doi.org/10.1126/science.1225829
- Doudna, J. A., & Charpentier, E. (2012). The new frontier of genome engineering with CRISPR-Cas9. Science, 346(6213), 1258096. https://doi.org/10.1126/science.1258096
- Frangoul, H., et al. (2021). CRISPR-Cas9 gene editing for sickle cell disease and beta-thalassemia. New England Journal of Medicine, 384(3), 252-260. https://doi.org/10.1056/NEJMoa2031054
- Raper, S. E., et al. (2003). Fatal systemic inflammatory response syndrome in a ornithine transcarbamylase deficient patient following adenoviral gene transfer. Molecular Genetics and Metabolism, 80(1-2), 148-158. https://doi.org/10.1016/j.ymgme.2003.08.016
- International Commission on the Clinical Use of Human Germline Genome Editing. (2020). Heritable Human Genome Editing. National Academies Press. https://doi.org/10.17226/25665
- Anzalone, A. V., et al. (2019). Search-and-replace genome editing without double-strand breaks or donor DNA. Nature, 576(7785), 149-157. https://doi.org/10.1038/s41586-019-1711-4
- Gaudelli, N. M., et al. (2017). Programmable base editing of A to G in genomic DNA without DNA cleavage. Nature, 551(7681), 464-471. https://doi.org/10.1038/nature24644
Frequently Asked Questions
What is gene therapy?
Gene therapy is a medical approach that treats or prevents disease by delivering functional genetic material into cells. Rather than treating symptoms with drugs or surgery, gene therapy addresses the genetic root cause — replacing a defective gene, silencing an overactive one, or introducing a new gene whose product can fight disease.The basic logic is straightforward: many diseases are caused by mutations in single genes that result in a missing or non-functional protein. Cystic fibrosis, for example, is caused by mutations in the CFTR gene that disrupt chloride transport in epithelial cells. Hemophilia B is caused by mutations in the Factor IX gene that result in failure to produce a clotting factor. If functional copies of these genes can be delivered to the relevant cells, the protein is produced and the disease is treated or cured.Gene therapy operates through several broad strategies. Gene addition delivers a functional copy of a defective gene without removing the original — the cell now has the working copy alongside whatever mutated version it started with. Gene silencing reduces or eliminates the expression of a gene producing a harmful product. Gene editing — most prominently with CRISPR-Cas9 — makes precise targeted changes to the DNA sequence itself, potentially correcting a mutation at its exact location.The field also distinguishes between somatic gene therapy, which modifies cells in the patient's body (changes are not heritable), and germline gene therapy, which modifies embryonic cells and would be inherited by future generations. Germline modification for reproductive purposes is currently prohibited in most jurisdictions and is considered ethically off-limits by the scientific mainstream. The somatic/germline distinction is central to the ongoing debates about the ethics of gene therapy and gene editing.
How do viral vectors deliver genes into cells?
The fundamental challenge of gene therapy is delivery: how do you get a functional gene into the relevant cells of the body? The DNA molecule is large and fragile, is actively excluded by cellular defenses, and cannot simply be injected into the bloodstream and expected to find its way to the right cells. Viruses, which have evolved over millions of years to enter cells and deliver their genetic cargo, have been adapted as the primary delivery vehicles — viral vectors.A viral vector is a virus that has been genetically modified to remove the genes responsible for causing disease (the pathogenic genes) and replace them with the therapeutic gene of interest. The virus retains its ability to enter cells and deliver its payload, but cannot replicate or cause illness. Several viral platforms are used, each with different characteristics.Adeno-associated virus (AAV) is currently the most widely used vector in gene therapy clinical applications. AAV is a small, naturally occurring virus that does not cause any known human disease. Its small genome limits the size of the therapeutic payload it can carry — approximately 4.7 kilobases — but it can infect a wide range of cell types (different AAV serotypes have different tissue tropism), can persist in cells without integrating into the genome (reducing the risk of insertional mutagenesis), and has a relatively benign immune profile. Zolgensma (onasemnogene abeparvovec), the gene therapy for spinal muscular atrophy, uses an AAV9 vector delivered intravenously.Lentiviruses, derived from HIV and related retroviruses, can carry larger payloads and integrate stably into the host cell genome. Integration is useful for therapies targeting dividing cells, where a non-integrating vector would be diluted as cells divide. However, integration carries the risk of insertional mutagenesis — if the vector inserts near a cancer-promoting gene, it could activate it. This risk led to leukemia cases in early retroviral vector trials, pushing the field toward safer integration-site selection. Lentiviral vectors are used in many CAR-T cell therapies.
What is the difference between gene therapy and gene editing?
Gene therapy and gene editing are related but distinct approaches. Gene therapy is the broader category — delivering genetic material into cells to treat disease. Gene editing is a specific approach within that category that makes precise, targeted changes to the DNA sequence itself.In traditional gene therapy, a functional copy of a gene is delivered to cells alongside whatever existing genetic material is there. The original mutation is not corrected; the cell simply also has a working copy of the gene. This approach works well for conditions where the missing gene product needs to be supplied, but it has limitations. The therapeutic gene must be expressed persistently, which requires either stable integration into the genome or a vector that persists for a long time. And it cannot address conditions where the problem is a gain-of-function mutation — where the mutated gene is actively producing something harmful.Gene editing, most prominently using CRISPR-Cas9 technology, can correct the mutation at its precise chromosomal location. CRISPR-Cas9 works by using a guide RNA (gRNA) to direct the Cas9 protein — a molecular scissors — to a specific DNA sequence, where it makes a double-strand cut. The cell's DNA repair machinery then either joins the cut ends together (non-homologous end joining, NHEJ — which typically disrupts the gene) or uses a provided template to repair the cut with a specific sequence (homology-directed repair, HDR — which can correct a mutation to the normal sequence). NHEJ is efficient but imprecise; HDR allows precise correction but is less efficient in most cell types.Newer editing approaches — base editing and prime editing — offer more precision with fewer side effects than classical CRISPR-Cas9. Base editors can convert one DNA letter to another without making a double-strand break. Prime editing, sometimes described as a 'search and replace' for the genome, can make small insertions, deletions, and all twelve possible base-to-base conversions with high precision. These technologies are moving rapidly toward clinical application.
What gene therapies are currently approved?
As of 2024, several gene therapies have received regulatory approval in the United States, European Union, or both, representing a transformation from the field's near-death experience after the Gelsinger tragedy in 1999.Luxturna (voretigene neparvovec), approved by the FDA in December 2017, was the first directly administered gene therapy approved in the United States. It treats a form of inherited retinal dystrophy caused by mutations in the RPE65 gene, which is essential for the visual cycle. The therapy delivers a functional RPE65 gene via an AAV2 vector injected directly into the subretinal space. Clinical trials showed meaningful improvements in light sensitivity and visual navigation in treated patients. The price at approval was \(850,000 for both eyes.Zolgensma (onasemnogene abeparvovec), approved by the FDA in May 2019, treats spinal muscular atrophy (SMA), a devastating neurodegenerative disease caused by mutations in the SMN1 gene. SMA is the leading genetic cause of infant mortality. Zolgensma delivers a functional SMN1 gene via an AAV9 vector in a single intravenous infusion. Clinical trials showed remarkable outcomes in treated infants: children who would otherwise lose motor function and require ventilation were sitting, standing, and in some cases walking. At approval, Zolgensma was priced at \)2.1 million, making it at the time the most expensive drug ever approved. The price reflects the one-time nature of the therapy and the severity of the disease; its cost-effectiveness relative to lifetime supportive care is defensible in health economic terms, though the sticker price raises profound equity concerns.Casgevy (exagamglogene autotemcel), approved by the FDA in December 2023, is the first approved therapy to use CRISPR-Cas9 gene editing. It treats sickle cell disease and transfusion-dependent beta-thalassemia by editing patients' own hematopoietic stem cells to reactivate fetal hemoglobin production, which compensates for the defective adult hemoglobin. The therapy requires extraction of the patient's stem cells, editing them in the laboratory, and reinfusing them after the patient's existing bone marrow is ablated — a demanding and lengthy process.
What are the risks of gene therapy?
Gene therapy carries several categories of risk, some well-characterized and some still being defined through clinical experience. The history of the field includes fatalities and serious adverse events that have shaped current regulatory and safety standards.Immune reactions are among the most serious risks. Viral vectors are recognized by the immune system as foreign, and the immune response to a vector can be severe. Jesse Gelsinger died in 1999 from a massive inflammatory immune response to a high-dose adenoviral vector — the most serious gene therapy adverse event in the field's history and the one that most directly altered its trajectory. Contemporary gene therapy uses vectors with better immune profiles, lower doses where possible, and immunosuppression protocols, but immune reactions to AAV vectors remain a real concern, particularly at the high doses required for systemic delivery.Insertional mutagenesis is the risk that a vector integrating into the genome does so in a location that disrupts a tumor suppressor gene or activates an oncogene. This risk was dramatically illustrated in early clinical trials of retroviral gene therapy for X-linked severe combined immunodeficiency (X-SCID, or 'bubble boy' disease): several treated children developed leukemia due to vector insertions near the LMO2 proto-oncogene. These cases transformed the field's approach to vector design and integration-site analysis. Newer lentiviral vectors have significantly reduced but not eliminated integration-related risks.Off-target effects are a specific concern for gene editing: the possibility that CRISPR-Cas9 or other editing tools cut DNA at unintended locations in the genome, potentially causing mutations elsewhere. Significant effort in the field is devoted to improving the specificity of guide RNAs and Cas9 proteins to minimize off-target cutting.Germline transmission — the risk that a somatic gene therapy inadvertently modifies germline cells, making changes heritable — is a theoretical concern, particularly for therapies administered during pregnancy or involving cell types with some germline access. Current therapies are not believed to carry significant germline risk, but it remains a regulatory and ethical concern.
What is the He Jiankui scandal?
In November 2018, He Jiankui, a Chinese biophysicist who had trained in the United States, announced at the Second International Summit on Human Genome Editing in Hong Kong that he had used CRISPR-Cas9 to edit human embryos that were subsequently implanted and had resulted in live births — twin girls, Lulu and Nana, and later a third child. The edit targeted the CCR5 gene, which encodes a co-receptor that HIV uses to enter cells. He's stated rationale was to give the children resistance to HIV infection, since their father was HIV-positive.The scientific and bioethical response was swift and near-universal condemnation. The consensus was unambiguous: He's experiment was scientifically unjustified, ethically indefensible, and dangerous. The children were exposed to risks that could not be consented to and that remain unknown — CRISPR editing in embryos does not achieve perfect editing of all cells, raising concerns about mosaicism; CCR5 deletion, while providing HIV resistance, may increase susceptibility to some other infections including West Nile virus; and the long-term consequences of the edit throughout the affected individuals' lives are entirely unknown.Beyond the direct risks to the children, the experiment violated existing Chinese regulations on clinical research, did not receive proper institutional oversight, and appears to have involved misrepresentation to the review board that nominally approved it. Investigators found that He had failed to disclose the true nature of his research and that participants in the study may not have been adequately informed of what they were agreeing to.He Jiankui was convicted of 'illegal medical practice' by a Chinese court in December 2019 and sentenced to three years in prison. The case accelerated international discussions about governance frameworks for human germline editing and led to calls for a global moratorium on heritable human genome editing, which the international scientific community has repeatedly endorsed pending the development of appropriate safety, efficacy, and ethical standards.