How Are Viruses Used for Gene Therapy?

Viruses are used for gene therapy by stripping them of the genes that cause disease and replacing those genes with therapeutic DNA that a patient’s cells need. The resulting particles, called viral vectors, still retain the virus’s natural talent for getting inside human cells, but instead of causing infection, they deliver a corrective gene that can treat or even cure a genetic disorder. The concept is elegantly parasitic in reverse: the same molecular machinery that evolved to hijack your cells now works on your behalf. How researchers actually pull this off, which viruses they choose, and what can go wrong are questions worth unpacking, because the details matter more than the elevator pitch suggests.

Turning a Virus Into a Delivery Vehicle

Every virus is, at its core, a package of genetic material wrapped in a protein shell. That shell is remarkably good at latching onto specific cell types and slipping its contents inside. Gene therapy exploits this by keeping the shell but swapping the cargo. In practice, scientists separate the virus’s own genes from the structural and regulatory sequences the shell needs to form. The disease-causing coding genes are removed and replaced with a therapeutic gene, often called a transgene. The key engineering trick is making sure the pieces that build the shell and the pieces that carry the therapeutic payload are kept on different DNA molecules, so the virus can never reassemble into something that could replicate and spread.

1Premier Science. Viral Vectors: “Friends or Foes” in Gene Therapy – Section: How Viruses are Manipulated to Work as Carriers or Vectors

The result is a particle that looks and behaves like a virus on the outside but is a one-way courier on the inside. It can enter a target cell, release its therapeutic DNA, and prompt the cell to start producing the protein it was missing. But it cannot make copies of itself, so the “infection” stops right there. This is what researchers mean when they say a vector is replication-deficient.

The Major Types of Viral Vectors

Not all viruses are equally suited for gene therapy. Each type brings a different combination of strengths and trade-offs, and the choice depends on the disease, the target tissue, and how long the therapeutic effect needs to last.

Adeno-Associated Virus (AAV)

AAV has become the workhorse of gene therapy, and for good reason. It is a tiny, non-enveloped virus that does not cause any known human disease, which gives it a major safety head start. Engineered AAV particles can be made completely free of viral genes while still efficiently delivering therapeutic DNA to target cells. That combination of safety and effectiveness has made AAV one of the most extensively tested platforms in the field.

2PubMed Central. Adeno-Associated Virus (AAV) as a Vector for Gene Therapy

AAV also comes in multiple “serotypes,” which are essentially different natural flavors of the virus, each with a slightly different protein shell. Those shell differences change which cell-surface receptors the virus recognizes and how it moves through cells once inside, which means different serotypes naturally home in on different tissues.

3PubMed. Adeno-associated virus serotypes: vector toolkit for human gene therapy

The main limitation of AAV is its small cargo capacity. The virus can only carry a relatively short stretch of DNA, which means it cannot deliver very large genes. For many single-gene disorders, though, the gene fits comfortably.

Adenovirus

Adenoviruses are the viruses behind many common colds, and their vectors are engineered by deleting key early genes. The combined deletions give enough room for a therapeutic gene cassette of up to about 7.7 kilobases, which is larger than what AAV can handle. Adenoviral vectors can transfer genes to most non-dividing cells, making them versatile delivery tools.

4Molecular Therapy. Adenovirus vectors: History, biology, and current applications in gene therapy

The trade-off is that adenoviral vectors tend to provoke stronger immune reactions than AAV. The body recognizes the adenovirus coat readily, which can limit how long the therapeutic effect lasts and can cause inflammatory side effects. This has made adenoviral vectors less popular for long-term gene replacement, though they remain valuable in settings where a short, potent burst of gene expression is the goal, such as cancer immunotherapy.

Lentivirus

Lentiviruses, a subgroup of retroviruses, have one standout ability: they can integrate their genetic cargo directly into the host cell’s chromosomes. That means the therapeutic gene gets copied every time the cell divides, potentially providing a permanent fix. This property makes lentiviral vectors especially useful for treating blood disorders, because you can modify a patient’s blood-forming stem cells outside the body, then transplant them back.

Lentiviral vectors are under clinical development for conditions like beta-thalassemia, where the goal is to give stem cells a working copy of the gene for beta-globin so they can produce healthy red blood cells for the rest of the patient’s life.

5PubMed. The development of an advanced lentiviral gene therapy for beta-thalassemia

Lentiviral platforms are also being adapted for cancer treatment. In one approach, lentiviral vectors deliver chimeric antigen receptor (CAR) genes to a patient’s T cells, reprogramming them to attack leukemia. A newer platform called PACK-IT can generate CAR-T cells in a rapid four-day manufacturing process without needing to first purify or activate the T cells, and preclinical testing against leukemia in mice showed these cells outperformed conventionally engineered CAR-T cells.

6Cancer Research. T cell targeted lentiviral gene delivery using the PACK-IT Platform generates CAR-T cells with superior potency compared to conventional lentivirus and enables in vivo generation of CD19-CAR T cells capable of controlling leukemia in preclinical models

Herpes Simplex Virus (HSV)

HSV-based vectors carve out a niche in neurological applications. The virus has a natural tendency to infect neurons and establish a quiet, long-lasting presence in nerve cells without killing them. That makes it a logical choice for delivering genes to the brain and nervous system, where you want the therapeutic gene to persist in non-dividing cells for years.

7PubMed Central. Herpes simplex virus-based vectors 8PubMed. Potential use of herpes simplex virus (HSV) vectors for gene therapy of neurological disorders

HSV also has a large genome, which means it can accommodate bigger therapeutic genes than AAV or adenovirus. The downside is that taming a virus this complex for clinical use has proved difficult, and HSV vectors have not yet reached the same level of commercial success as AAV.

In Vivo Versus Ex Vivo Delivery

There are two broad strategies for getting a viral vector to the right cells. In the “in vivo” approach, the vector is injected directly into the patient, either into the bloodstream or into a specific tissue like the eye or muscle. The vector then finds its target cells on its own. In the “ex vivo” approach, cells are removed from the patient, modified with the vector in a lab dish, and then returned to the body.

The choice between these strategies depends on the disease. Eye diseases are well suited to direct injection because the eye is a small, contained space. Blood disorders lend themselves to ex vivo treatment because stem cells can be harvested, modified, and transplanted back. Cancer immunotherapy often blends both: one approach involves modifying a patient’s immune cells (dendritic cells) outside the body with an adenovirus vector carrying a tumor antigen gene, then injecting those cells back to rally the immune system against micrometastatic cancer.

9PubMed. In vivo and ex vivo gene therapy strategies to treat tumors using adenovirus gene transfer vectors

Engineering Smarter Vectors

Simply gutting a virus and loading it with a therapeutic gene is only the starting point. A major focus of current research is making vectors more precise, both in where they go and in which cells they switch on once they arrive.

One important lever is the promoter, a stretch of DNA that acts as an on-switch for the therapeutic gene. Using a tissue-specific promoter means the gene will only activate in the intended cell type, even if the vector happens to enter other cells along the way. Choosing the right promoter is one of the most critical steps in designing a gene therapy, because it determines whether the therapeutic protein is made only where it is needed or everywhere indiscriminately.

10PubMed Central. Evaluation of promoters for use in tissue-specific gene delivery

Beyond promoters, researchers also use elements like microRNA target sequences embedded in the expression cassette to further restrict where a gene is active. The idea is that certain microRNAs are abundant in “off-target” tissues but absent from the target tissue, so including their binding sites ensures the transgene is silenced in the wrong cells.

11PubMed Central. Viral expression cassette elements to enhance transgene target specificity and expression in gene therapy

Getting Past the Blood-Brain Barrier

Delivering genes to the brain is one of the hardest problems in gene therapy. The blood-brain barrier, a tightly sealed layer of cells lining the brain’s blood vessels, keeps most large molecules and viruses out. Historically, treating brain disorders with viral vectors required invasive direct injection into the brain, which is risky and only reaches a limited area.

Recent breakthroughs have changed the outlook. Researchers engineered an AAV capsid called BI-hTFR1 that binds to the human transferrin receptor, a protein on blood-brain barrier cells. In mice engineered to carry the human version of that receptor, this capsid delivered roughly 40 to 50 times more gene expression in the central nervous system compared with AAV9, a standard serotype, after a simple intravenous injection.

12PubMed Central. An AAV capsid reprogrammed to bind human transferrin receptor mediates brain-wide gene delivery

Other groups are using directed-evolution platforms to rapidly screen huge libraries of capsid variants for the ones that cross the barrier most efficiently. One such platform, called TRACER, uses RNA-guided selection to identify capsids with strong brain penetration in animals that have not been genetically modified, making the results more likely to translate to humans.

13Molecular Therapy Methods & Clinical Development. RNA-Driven Directed Evolution of AAV Capsids Enables Efficient Blood-Brain Barrier Penetration and Widespread CNS Transduction

A variant called VCAP-102, generated using the same TRACER platform, showed 20- to 400-fold increased gene transfer across multiple brain regions compared with AAV9, and it works in both rodents and primates. Researchers identified alkaline phosphatase as the receptor it uses to cross the barrier, and they demonstrated that binding to this receptor triggers active transport across barrier cells.

14Molecular Therapy. VCAP-102 is a cross-species, blood-brain barrier-penetrant AAV capsid that targets alkaline phosphatase for gene therapy

These advances matter because dozens of devastating neurological conditions, from lysosomal storage disorders to spinal muscular atrophy, could benefit from brain-wide gene delivery via a simple IV infusion rather than brain surgery.

Approved Gene Therapies and Landmark Trials

The clearest proof that viral vectors work is the growing list of approved treatments. The first FDA-approved gene therapy for a genetic disease, voretigene neparvovec (brand name Luxturna), uses an AAV2 vector to deliver a working copy of the RPE65 gene to retinal cells in patients with an inherited form of blindness. In the pivotal trial, patients who received the treatment showed dramatic improvement in navigating a mobility course under dim light, and most of the treated group achieved the maximum possible improvement. No serious product-related adverse events or harmful immune responses were reported.

15The Lancet. Efficacy and safety of voretigene neparvovec (AAV2-hRPE65 vector) in patients with inherited retinal disease: a phase 3 randomised controlled trial 16PubMed Central. Clinical Perspective: Treating RPE65-Associated Retinal Dystrophy

Since Luxturna, additional AAV-based therapies have gained approval for conditions including spinal muscular atrophy and hemophilia. Lentiviral vector therapies have been approved for sickle cell disease and beta-thalassemia. The field has expanded from rare eye diseases to blood disorders, neuromuscular conditions, and cancer. There are currently more than 250 clinical trials involving AAV vectors alone worldwide.

17PubMed. Evading and overcoming AAV neutralization in gene therapy

The Immune System Problem

The human immune system does not care that a viral vector is trying to help. It sees a foreign protein shell and reacts. Innate and adaptive immune responses to viral vectors and the proteins they produce remain one of the biggest obstacles to wider clinical use.

18PubMed Central. Immune Responses to Viral Gene Therapy Vectors

One particularly frustrating aspect is pre-existing immunity. Many people have been exposed to natural AAV during childhood and carry antibodies against it. Those antibodies, called neutralizing antibodies, can intercept and disable the vector before it ever reaches the target tissue. This means a significant fraction of the population may be ineligible for certain AAV therapies without some kind of immune management.

Re-dosing is another challenge. Even patients who did not have prior antibodies will develop them after the first treatment. If the therapeutic effect fades over time, giving a second dose of the same vector may not work because the immune system is now primed to destroy it. Researchers are exploring several strategies to get around this, including using different AAV serotypes for subsequent doses, engineering capsids the immune system does not recognize, and administering immunosuppressive drugs around the time of treatment.

19Trends in Biotechnology. Evading and overcoming AAV neutralization in gene therapy

A systematic review of immunosuppressive protocols across dozens of clinical trials and real-world studies found that a wide range of regimens have been used, but no clear consensus has emerged on the best approach. Better understanding of which immunosuppressive strategies work best for which vector types and delivery routes remains a priority.

20PubMed. A systematic review of immunosuppressive protocols used in AAV gene therapy for monogenic disorders

Safety Risks That Shaped the Field

Gene therapy’s history includes setbacks that forced the field to rethink its approach. The most sobering involved early retroviral vectors used to treat severe combined immunodeficiency (SCID-X1), often called “bubble boy disease.” Because retroviruses integrate their cargo into the host genome, they can accidentally insert near a gene that controls cell growth, switching it on and triggering cancer. Five children in clinical trials in France and the United Kingdom developed leukemia as a result of this insertional mutagenesis, and one of them died in 2004.

21PubMed Central. Development of safer gene delivery systems to minimize the risk of insertional mutagenesis-related malignancies: a critical issue for the field of gene therapy

Those cases drove the development of safer vectors. Modern lentiviral vectors use self-inactivating designs that reduce the risk of activating nearby genes after integration. AAV vectors, which typically do not integrate into chromosomes, have become preferred partly because of this safety advantage, though AAV-mediated gene transfer has induced tumors in some animal models. No vector system is entirely free of risk, but the safety engineering has come a long way since the early trials.

21PubMed Central. Development of safer gene delivery systems to minimize the risk of insertional mutagenesis-related malignancies: a critical issue for the field of gene therapy

Manufacturing Challenges

Making viral vectors at clinical scale is harder than it sounds. AAV production, for instance, generates a mixture of capsids, and a large proportion of those capsids are empty shells that do not contain the therapeutic gene. Separating the “full” capsids (the ones carrying the gene) from the empty ones is a significant manufacturing bottleneck.

22Molecular Therapy: Methods & Clinical Development. Assessing the on-column stability of adeno-associated virus capsids during anion exchange chromatography

Anion exchange chromatography is one of the most scalable methods for this separation, but when the starting proportion of full capsids is low, achieving a highly enriched product without throwing away too much material is a real challenge.

23Molecular Therapy. A Two-Pass Anion-Exchange Chromatography Method for Maximum Enrichment of Full Capsids in Adeno-Associated Viral Vectors

These manufacturing difficulties are one reason gene therapies carry staggering price tags. Luxturna costs around $850,000 for a course of treatment. Some newer gene therapies for conditions like hemophilia or beta-thalassemia have been priced above $2 million. The combination of complex production, small patient populations, and uncertainty about how long the treatment’s benefits will last has fueled widespread concern about whether health systems can afford these therapies.

24PubMed Central. Managing the challenges of paying for gene therapy: strategies for market action and policy reform in the United States

How Non-Viral Alternatives Compare

Viral vectors are not the only option. Non-viral delivery systems, including lipid nanoparticles, polymers, and other synthetic carriers, avoid many of the problems viruses bring. They do not risk replicating, do not integrate into the genome, provoke less immune reaction, and can carry much larger genetic payloads. They are also generally easier and cheaper to manufacture.

25eBioMedicine. How Are Viruses Used for Gene Therapy? – Section: Non-viral vectors in gene therapy

The catch is efficiency. Non-viral vectors are substantially worse at getting their cargo into cells and producing meaningful levels of the therapeutic protein. Lipid nanoparticles proved spectacularly successful for delivering mRNA in COVID-19 vaccines, where a short burst of protein production is all you need. For gene therapy, which often requires sustained, high-level expression in specific tissues, viral vectors still have a clear performance advantage.

26PubMed Central. Lipid nanoparticles for gene delivery

The two approaches are increasingly seen as complementary rather than competing. Non-viral methods may eventually dominate applications like transient gene editing, where you only need the editing machinery active briefly. Viral vectors are likely to remain the platform of choice for conditions that require long-lasting gene expression from a single treatment, at least until non-viral technology catches up on delivery efficiency.

Where the Technology Is Heading

The current generation of engineered capsids that can cross the blood-brain barrier, combined with tissue-specific promoters and microRNA-based safeguards, points toward therapies that are far more targeted than what was possible even five years ago. If a vector can be injected into the bloodstream and reliably reach only the intended cell type in the brain, liver, or muscle, the therapeutic dose drops, side effects shrink, and manufacturing becomes more feasible.

The re-dosing problem is arguably the single biggest scientific hurdle left. A gene therapy that works once but can never be repeated places enormous pressure on getting the dose and timing exactly right the first time. Several groups are engineering capsids with surfaces the immune system has never seen before, essentially stealth viruses that slip past existing antibodies. Others are testing whether short courses of immunosuppression can create a window for re-administration. Neither approach has been proven in large-scale human trials yet, but both have shown promise in animal models. If re-dosing becomes routine, gene therapy shifts from a one-shot gamble to something closer to a renewable treatment, and the economics change with it.