In Vivo Gene Editing: The Next Frontier After CRISPR-Cas9

The ex vivo and in vivo distinction

The distinction between ex vivo and in vivo gene editing matters because the two approaches have fundamentally different technical requirements, clinical workflows, and addressable patient populations.

Ex vivo gene editing involves removing the target cells from the patient, editing them in a controlled laboratory environment, and then returning the edited cells. The approach has been most successful for blood cell-based diseases, where the target cells (hematopoietic stem cells) can be harvested from the patient’s bone marrow or peripheral blood, edited, and reinfused. The patient typically receives chemotherapy or other conditioning to make room for the edited cells to engraft.

In vivo gene editing delivers the editing machinery — the editor protein, the guide RNA, and any associated components — directly into the patient. The editor finds its way to the target cells inside intact tissues, performs the edit, and the resulting modified cells continue to function in their native environment.

The trade-offs between the two approaches are substantial. Ex vivo editing offers more control: the editing can be verified before reinfusion, the dose of edited cells can be calibrated, and the off-target effects can be assessed in detail. In vivo editing offers broader applicability: many tissues and cell types cannot be effectively removed and replaced, and for those, in vivo delivery is the only path to gene editing therapy.

Why in vivo gene editing is so much harder

The technical challenges of in vivo gene editing are substantial.

Delivery is the dominant problem. The editing machinery has to reach the target tissue at therapeutic concentrations, enter the target cells, and avoid causing problems in non-target tissues. Different delivery technologies — lipid nanoparticles, adeno-associated viral vectors, lentiviral vectors, and others — have different strengths in terms of tissue targeting, payload capacity, and immunogenicity.

Specificity is critical. An editing event in the wrong location in the genome can have significant consequences. The off-target editing profile of any in vivo gene editing therapy is a central focus of development and regulatory review.

Editing efficiency in the target tissue determines whether the therapeutic effect is achieved. The fraction of target cells that are successfully edited affects the clinical outcome, and the relationship between editing efficiency and therapeutic effect varies substantially across disease indications.

Immunogenicity is a meaningful concern. The editing machinery — particularly the editor protein — is recognized by the immune system as foreign in many cases. The immune response can limit repeat dosing, cause inflammation in the target tissue, and complicate the safety profile.

Durability of the editing effect varies. Editing a long-lived cell type (such as a hepatocyte or a stem cell) produces an effect that persists. Editing a short-lived cell type produces an effect that dilutes as the cells turn over.

The different editing technologies

The category of “gene editing” includes several distinct technical approaches, each with its own characteristics.

CRISPR-Cas9 is the most widely known and most extensively studied. It uses a bacterial protein, Cas9, guided to a specific genomic location by an RNA molecule. CRISPR-Cas9 has been the foundation of much of the gene editing field’s development.

Base editing and prime editing are more recent technologies derived from CRISPR but designed to produce different kinds of edits with greater precision. Base editors can change specific single nucleotides without creating double-strand breaks in the DNA, which reduces some classes of off-target effects.

ARCUS gene editing uses a different family of editor proteins, derived from meganucleases. The proteins recognize specific DNA sequences and create cuts that can be used to disrupt genes or to enable targeted insertions. The smaller size of these proteins compared to Cas9 has potential delivery advantages, and the precise recognition specificity has potential implications for off-target profiles.

Zinc finger nucleases and transcription activator-like effector nucleases (TALENs) are older technologies that preceded CRISPR but continue to have specific applications.

Each of these editing approaches has its own strengths and weaknesses, and the choice of editing technology for a specific therapeutic application depends on the requirements of that application — the type of edit needed, the target tissue, the delivery constraints, and the desired off-target profile.

Disease areas in active development

Several disease areas are at the frontier of in vivo gene editing development.

Liver-targeted editing has been the most active in vivo area to date. The liver is comparatively accessible to systemic delivery via lipid nanoparticles and viral vectors, and several genetic diseases affecting liver function are attractive targets. Familial hypercholesterolemia, ATTR amyloidosis, alpha-1 antitrypsin deficiency, and various inborn errors of metabolism are among the indications where in vivo gene editing programs have entered clinical development.

Eye-targeted editing is another active area. The eye is partially immune-privileged and can be accessed by local injection, which simplifies some of the systemic delivery challenges. Inherited retinal diseases are the most active disease category.

Hematopoietic cell editing, while historically pursued ex vivo, is increasingly being explored through in vivo approaches that target the cells directly in the bone marrow.

Central nervous system editing remains difficult because of the blood-brain barrier, but several approaches are in early development.

What investors should think about

For investors evaluating in vivo gene editing companies, several principles tend to apply.

Platform technology and individual program economics are distinct. A robust in vivo gene editing platform produces multiple drug candidates over time. The platform’s value is the option set it creates across indications and targets.

The delivery technology is often as important as the editing technology. Companies with proprietary delivery capabilities that pair well with their editing machinery have a more durable competitive position than those that depend on external delivery partnerships.

Early clinical readouts in this category have specific elements to watch — the editing efficiency in the target tissue, the durability of the editing effect, the off-target profile, the immunogenicity, and the clinical efficacy of the editing on the disease phenotype.

Regulatory engagement is meaningful. The FDA and EMA have been actively developing the regulatory framework for in vivo gene editing therapies, and the precedents established by current programs affect how future programs will be reviewed.

Partnership and licensing economics in this category have been substantial. Large pharmaceutical companies with interest in genetic medicines have shown willingness to engage in significant transactions for access to in vivo gene editing capability.

The longer-term picture

In vivo gene editing is at an inflection point. The category has moved from preclinical proof of concept through early clinical demonstration of feasibility into the early stages of clinical efficacy data. The next several years will be formative.

For companies positioned in this category, the operating environment combines substantial scientific opportunity, significant clinical and regulatory challenges, and a competitive landscape that is intensifying as more entrants pursue similar problems. The companies that succeed will be those that combine strong editing technology with effective delivery, deep development experience, and the financial and operational capability to navigate a long and expensive development pathway.