How Gene Therapy Treats Disease

FIGURE 1 — Major milestones in gene therapy from the first human trial in 1990 to the first CRISPR-based approval in 2023. Sources: FDA, EMA, NIH historical records.

01 The Broken Blueprint

Every human cell carries a copy of the genome — roughly three billion letters of DNA that encode the instructions for building and operating a body. Most of the time, the text reads correctly. But a single misplaced letter, a duplicated passage, or a missing paragraph can be enough to cause disease. Sickle cell anemia traces to a single nucleotide change in the HBB gene. Cystic fibrosis most often stems from a three-letter deletion in CFTR. Huntington's disease is caused by an expanded repeat that grows beyond a critical length. These are not infections or environmental injuries — they are typos in the source code, and conventional pharmacology can only manage the downstream symptoms.

Gene therapy proposes a different logic: instead of masking the effects of a genetic error, correct the error itself. That might mean replacing a missing gene, silencing a harmful one, or editing a mutation back to its healthy sequence. The concept is elegant in principle and fiendishly difficult in practice. Delivering new genetic material into the right cells, getting it to the right place in the genome, and avoiding the immune system's alarm bells have each required decades of ingenuity. But after thirty years of setbacks, the field has arrived. As of 2026, dozens of gene therapies have received regulatory approval, and hundreds more are in clinical trials.

02 Viral Vectors and Delivery

The central engineering challenge in gene therapy is not designing the therapeutic gene — it is delivering it. DNA cannot simply be injected into the bloodstream and expected to find its way into the nucleus of the right cells. The molecule is too large, too charged, and too fragile to cross cell membranes on its own. The solution that the field converged on was to hijack the most efficient DNA delivery system evolution ever produced: the virus.

Viruses have spent millions of years optimizing the ability to insert their genetic material into host cells. Gene therapists strip out the viral genes that cause disease and replace them with the therapeutic payload. The most widely used platform is the adeno-associated virus (AAV), a small, non-pathogenic virus that can infect dividing and non-dividing cells alike. Different AAV serotypes have natural affinities for different tissues — AAV9 tends toward muscle and the central nervous system, AAV8 toward the liver, AAV5 toward the retina. This tropism lets clinicians target specific organs by choosing the right viral shell.

Retroviruses and lentiviruses are used when permanent integration into the genome is required. These vectors insert their payload directly into the host chromosome, ensuring the therapeutic gene is copied along with the cell's DNA every time the cell divides. This is essential for diseases affecting rapidly dividing tissues, but it carries a risk: integration can occur near a cancer-related gene and disrupt its regulation. Early gene therapy trials for X-linked severe combined immunodeficiency (SCID-X1) successfully cured the immune disease but caused leukemia in several patients when the retroviral vector integrated near oncogenes. This painful lesson reshaped the field's approach to vector design and safety monitoring.

03 The CRISPR Revolution

The gene therapy toolkit was transformed in 2012 when researchers demonstrated that CRISPR-Cas9, a bacterial immune system component, could be repurposed as a programmable genome editor. Unlike earlier approaches that could only add genes, CRISPR can make precise cuts at virtually any location in the genome specified by a short guide RNA molecule. The cell's own repair machinery then knits the cut back together — imperfectly in a way that can disable a gene, or precisely using a supplied template that corrects a mutation.

The practical implications are enormous. A disease like sickle cell anemia, caused by a single nucleotide change, can now be addressed by editing the offending letter directly rather than adding a supplementary gene. In December 2023, the FDA approved Casgevy (exagamglogene autotemcel), the first CRISPR-based therapy for sickle cell disease. Patients' own blood stem cells are extracted, edited ex vivo to reactivate fetal hemoglobin production, and reinfused. The edited cells produce healthy hemoglobin, preventing the sickling and vascular blockage that define the disease. Clinical trial results showed that the vast majority of treated patients remained free of vaso-occlusive crises for at least a year — an outcome that conventional treatments could never achieve.

CRISPR's versatility extends beyond correction. Base editors can change individual DNA letters without cutting the double helix, reducing the risk of large rearrangements. Prime editors can write short new sequences into the genome. Each generation of editors expands the range of mutations that gene therapy can address, moving the field from gene addition toward true gene correction.

04 From Rare Diseases to Common Ones

Gene therapy's earliest successes were almost entirely in rare monogenic diseases — conditions caused by a single defective gene. This is logical: if one gene is broken, fixing that one gene should be sufficient. Spinal muscular atrophy (SMA), a devastating neuromuscular disease that typically killed infants before age two, is treated by Zolgensma, an AAV-delivered gene that produces the missing SMN protein. Luxturna restores vision in patients with an inherited form of blindness caused by mutations in the RPE65 gene. These are dramatic, life-changing interventions in patient populations that had few or no treatment options.

The frontier now pushes toward polygenic and acquired diseases. Cardiovascular disease, certain cancers, and neurodegenerative conditions involve many genes and environmental factors. Gene therapy for these conditions does not aim to cure a single mutation but to modify a relevant pathway — for instance, lowering cholesterol by editing the PCSK9 gene in the liver, or enhancing immune responses to tumors by engineering T cells with chimeric antigen receptors (CAR-T therapy). CAR-T cell therapy, technically a form of ex vivo gene therapy, has produced remission rates in certain blood cancers that were previously unimaginable, and represents the most commercially successful gene therapy platform to date.

FIGURE 2 — Cumulative FDA-approved gene and cell therapy products, showing accelerating regulatory adoption. Data approximated from FDA CBER records through 2025.

05 Clinical Trials and Approved Therapies

The regulatory landscape has shifted dramatically. In 2012, Glybera became the first gene therapy approved in the Western world, treating a rare inherited lipoprotein disorder. It was commercially unsuccessful and later withdrawn, but it established a regulatory pathway. By 2017, the FDA approved Kymriah (tisagenlecleucel), a CAR-T cell therapy for acute lymphoblastic leukemia, and Luxturna (voretigene neparvovec), an AAV-based gene therapy for inherited blindness. Zolgensma (onasemnogene abeparvovec) followed in 2019 for spinal muscular atrophy, priced at approximately $2.1 million per dose — the most expensive drug in the world at the time.

As of 2026, the FDA has approved more than ten gene therapy products, and the pipeline is deep. The Alliance for Regenerative Medicine tracks over 2,000 active clinical trials worldwide spanning inherited diseases, oncology, ophthalmology, and neurodegeneration. The approvals span in vivo therapies (AAV delivered directly to patients) and ex vivo therapies (cells edited in the lab and returned to the patient). The distinction matters: ex vivo approaches allow quality control of the edited cells before reinfusion, while in vivo approaches are simpler logistically but harder to monitor and control.

06 Risks, Limits, and Ethics

Gene therapy is not without serious risks. The death of Jesse Gelsinger in 1999, from a massive immune response to an adenoviral vector, set the field back years and established the principle that delivery safety would be the rate-limiting challenge. Immune responses to viral vectors remain a concern: some patients carry pre-existing antibodies to AV or AAV from natural infections, excluding them from treatment. High doses of AAV, needed to reach large tissues like muscle or the central nervous system, have caused severe adverse events including thrombotic microangiopathy and hepatotoxicity in recent trials.

Off-target effects are the other major worry. CRISPR-Cas9 can occasionally cut at genomic sites that resemble the intended target, creating unintended mutations. While base editors and prime editors reduce this risk, no editing system is perfectly specific. For therapies that modify cells ex vivo, the edited cells can be sequenced and screened before reinfusion. For in vivo therapies, off-target editing in the body is harder to detect and monitor.

Access and cost are equally pressing. Gene therapies are among the most expensive drugs ever sold, with prices ranging from hundreds of thousands to millions of dollars per patient. The economics are driven by small patient populations for rare disease therapies, the complexity of manufacturing biological products, and the one-time nature of many treatments. Novel payment models — outcomes-based contracts, annuity payments tied to durability — are being explored, but the question of who pays for cures remains unresolved.

07 The Road Ahead

The next decade of gene therapy will be shaped by several converging trends. Delivery is expanding beyond AAV: lipid nanoparticles, the delivery system used for mRNA vaccines, are being adapted for gene editing cargo and offer the advantage of repeat dosing, which AAV's immune memory largely precludes. In vivo CRISPR editing — delivering editors directly into the body rather than editing cells in a lab — has entered clinical trials for liver, eye, and blood disorders, and early results are promising.

The scope of treatable diseases is broadening. CRISPR therapies for high cholesterol, HIV, and certain forms of blindness are in development. RNA-based therapies, including antisense oligonucleotides and siRNA, offer a complementary approach that modulates gene expression without altering the genome permanently. The distinction between gene therapy and conventional pharmacology is blurring as more genetic medicines reach the clinic.

Perhaps most importantly, manufacturing is maturing. The early gene therapy era was marked by batch-to-batch variability and supply shortages that delayed treatment for patients who needed it. Standardized vector production, closed-system bioreactors, and regulatory frameworks for potency assays are making gene therapy an industrial process rather than a bespoke one. The field that began with a single trial in 1990 has, after decades of struggle, reached the point where genetic medicine is no longer experimental — it is becoming ordinary.