Gene Replacement, Gene Editing, CRISPR and more … what’s the difference?
Key Takeaways:
• Baby KJ Muldoon received a custom base-editing cure for CPS1 deficiency at CHOP in 2025 — a single-letter DNA fix delivered via IV, avoiding a liver transplant.
• Base editing, CRISPR, gene replacement and other therapies are often conflated - this article provides a detailed account of their nuances, and when they might be appropriate.
Your child has a genetic disease. Someone has mentioned gene therapy. Maybe gene editing too. You have heard of CRISPR. You may have seen the story about Baby KJ. You want to understand what is real, what is ready, and what matters for your family.
This guide breaks down the two major categories of genetic medicine, gene replacement and gene editing, with real clinical data, honest risk analysis, and analogies that make the science accessible without dumbing it down.
The Core Difference: Replace vs. Repair
Think of your child's genome as a massive instruction manual, roughly 3 billion characters long. Somewhere in that manual, there is an error. Maybe a single typo. Maybe a missing paragraph. That error is causing disease.
Gene replacement and gene editing represent two fundamentally different strategies for fixing the problem.
Gene replacement delivers a brand new, working copy of the broken gene into your child's cells. It does not touch the original error. It adds a second set of instructions alongside the broken ones, like taping a corrected page into the manual next to the damaged one. The original typo is still there, but the cell now has a clean copy to read from.
Gene editing goes to the exact location of the error and modifies it directly. Depending on the tool, it might cut a problematic section to disable it, swap a single wrong letter for the right one, or rewrite a short passage. The original manual itself gets changed.
That distinction, adding a workaround vs. modifying the root cause, drives nearly every practical difference between the two approaches: how they are delivered, how long they last, what risks they carry, and which patients they can help.
Gene Replacement Therapy: The Established Approach
Gene replacement therapy (often called "gene therapy" or "gene addition") has been in clinical development for over three decades. It is the more mature of the two approaches, with multiple FDA-approved products on the market and thousands of patients treated worldwide.
How It Works
The most common version uses a modified virus called an adeno-associated virus (AAV) as a delivery vehicle. AAVs are small, naturally occurring viruses that have been engineered to be harmless. Scientists remove the virus's own genetic material and replace it with a working copy of the gene your child needs. When the AAV enters your child's cells, it deposits that working gene, which then starts producing the missing protein.
The analogy: imagine your child's cells are running a factory, but the original blueprint for a critical machine part is damaged. Gene replacement therapy does not fix the blueprint. Instead, it ships in a fresh copy of the blueprint on a USB drive (the AAV). The factory reads the new instructions and starts producing the part it was missing.
FDA-Approved Gene Replacement Therapies
Several gene replacement therapies have reached patients:
Zolgensma (onasemnogene abeparvovec) treats spinal muscular atrophy (SMA), a devastating neuromuscular disease and a leading genetic cause of infant death. It delivers a functional copy of the SMN1 gene via a one-time IV infusion. Long-term follow-up data out to 7.5 years show that children treated before symptom onset maintained or gained motor milestones, with 100% of presymptomatic patients achieving independent walking. Over 3,000 children have been treated worldwide.
Luxturna (voretigene neparvovec) treats inherited retinal dystrophy caused by RPE65 mutations. It is injected directly under the retina of each eye and delivers a working copy of the RPE65 gene. In clinical trials, treated patients showed meaningful improvements in the ability to navigate in low light. Luxturna was the first FDA-approved gene therapy for an inherited disease in the United States, approved in December 2017.
Elevidys (delandistrogene moxeparvovec) treats Duchenne muscular dystrophy. Because the full dystrophin gene is too large to fit inside an AAV, this therapy delivers a shortened "micro-dystrophin" gene that produces a smaller but partially functional version of the protein. It received accelerated approval from the FDA in June 2023.
Kebilidi (eladocagene exuparvovec) treats aromatic L-amino acid decarboxylase (AADC) deficiency, an ultra-rare neurotransmitter disorder. It is delivered directly into the brain and received FDA approval in November 2024.
Waskyra (etuvetidigene autotemcel) became the first cell-based gene therapy approved for Wiskott-Aldrich syndrome in December 2025, using a lentiviral vector to deliver the corrected gene to the patient's own stem cells.
Strengths of Gene Replacement
Clinical maturity. This approach has the longest track record. We have multi-year follow-up data showing sustained benefit in conditions like SMA. Regulatory pathways are well established. Physicians and treatment centers have experience administering these therapies.
Mutation-agnostic within a disease. Gene replacement does not need to know the specific mutation. If the gene is broken, regardless of where or how it is broken, delivering a new copy can work. This is a significant advantage for diseases with many different causative mutations (called allelic heterogeneity). You do not need a custom therapy for each patient's unique mutation.
Limitations and Risks of Gene Replacement
Durability questions. The delivered gene typically does not integrate into the patient's chromosomes. It exists as a separate piece of DNA (called an episome) inside the cell nucleus. In non-dividing cells like neurons, this can persist for years. But in dividing cells, like liver cells in a growing child, the episome can be diluted and eventually lost as cells replicate. This is a particular concern in pediatric patients, where organ growth means the therapeutic gene copies do not keep pace with new cells being generated.
Overexpression risk. Because gene replacement adds a new copy of the gene rather than fixing the original, there is no natural feedback loop controlling how much protein the new gene produces. The transgene operates from a synthetic promoter, and expression levels can be higher or lower than what the body normally makes. Overexpression of certain proteins can be toxic or disruptive to cell function. This is a fundamentally different risk profile than gene editing, where the corrected gene remains under its native regulatory control and produces protein at physiological levels.
Gene Editing: The Precision Approach
Gene editing directly modifies the patient's own DNA. Rather than adding a workaround, it changes the genome itself. The field has evolved rapidly since the CRISPR-Cas9 system was first characterized in 2012, and the toolbox now includes several approaches with meaningfully different capabilities.
How Gene Editing Works
CRISPR-Cas9, the original and most clinically advanced tool, works like molecular scissors. A guide RNA directs the Cas9 protein to a specific location in the genome, where it cuts both strands of the DNA. The cell then repairs the break, usually by joining the cut ends back together in a way that introduces small insertions or deletions. This effectively disrupts the targeted gene, a strategy called gene knockout.
The analogy: if the problem is a factory machine producing something harmful, CRISPR-Cas9 is like cutting the power cable to that machine. The machine stops working. You have not fixed it, but you have stopped it from doing harm.
This is how Casgevy (exagamglogene autotemcel) works. Casgevy is the first FDA-approved CRISPR-based therapy. It uses Cas9 to cut and disrupt the BCL11A gene in a patient's blood stem cells, which reactivates fetal hemoglobin production and compensates for the defective adult hemoglobin that causes sickle cell disease and beta-thalassemia. In clinical trials, 97% of sickle cell patients were free from vaso-occlusive crises for at least 12 consecutive months. Follow-up data now extends beyond five years, with stable fetal hemoglobin levels and durable editing confirmed in long-term hematopoietic stem cells.
Base editing is where true precision correction enters the picture. Instead of cutting the DNA, base editors chemically convert one DNA letter to another at a precise location. There are two main types: cytosine base editors (CBEs) convert C to T, and adenine base editors (ABEs) convert A to G. Together, these can address a large share of known disease-causing point mutations, the single-letter typos that account for a substantial portion of rare genetic diseases.
The analogy: if CRISPR-Cas9 cuts a power cable, base editing is more like using correction fluid on a single mistyped letter. No cutting, no structural damage to the page. Just a clean chemical swap.
Base editing is what treated Baby KJ, the infant with CPS1 deficiency who received the first personalized in vivo CRISPR therapy in February 2025. His treatment, published in the New England Journal of Medicine, used a custom adenine base editor delivered via lipid nanoparticles to his liver. After three IV infusions over three months, he went home from the hospital and has continued to develop normally.
Base editing is now in multiple clinical trials. CS-101, developed by CorrectSequence Therapeutics, has treated nearly 20 patients with beta-thalassemia in China, with all patients achieving transfusion independence and hemoglobin levels reaching near-normal within three months. Beam Therapeutics dosed its first patient in a glycogen storage disease type I trial in May 2025. At least 15 base editing and prime editing clinical trials are currently active across five countries.
Prime editing is the most versatile and newest tool. It can make any single-letter change (all 12 possible base-to-base conversions), plus small insertions and deletions, all without cutting both strands of the DNA. It uses a modified Cas9 fused to a reverse transcriptase enzyme, guided by a prime editing guide RNA (pegRNA) that contains both the target location and a template for the desired edit.
The analogy: prime editing is a find-and-replace function. It searches for the specific text, highlights it, and overwrites it with the corrected version.
In May 2025, Prime Medicine reported the first clinical data from a prime editing therapy. A patient with chronic granulomatous disease (CGD) received a single dose of PM359. By day 30, NADPH oxidase activity was restored in 66% of neutrophils, well above the 20% threshold considered clinically meaningful. No serious adverse events were reported. The FDA had cleared the prime editing IND in April 2024, making it the first prime editing therapy authorized for human testing.
Why Gene Editing Matters for Rare Disease
It fixes the root cause. Gene editing changes the actual DNA that is causing the disease. Unlike gene replacement, which adds a workaround alongside the broken gene, editing corrects or disables the problem at its source. For diseases caused by a single genetic error, this is the most direct possible intervention.
The fix is durable. Because gene editing modifies the patient's own genome, the correction is integrated into the DNA itself. In stem cells or other long-lived cell types, this means the fix is passed on to every daughter cell indefinitely. There is no episome to dilute as cells divide. For a growing child, this is a critical distinction.
Expression stays physiological. When gene editing corrects a mutation, the repaired gene remains under its native regulatory control. The cell produces the right amount of protein, in the right place, at the right time. There is no synthetic promoter driving expression at abnormal levels, which eliminates the overexpression risk that comes with gene replacement.
The tools are getting more precise. The progression from CRISPR-Cas9 (which cuts DNA) to base editing (which swaps a single letter) to prime editing (which can rewrite short sequences) represents a steady march toward greater precision with less collateral disruption. Each generation of tools introduces fewer unintended changes to the genome. David Liu, the Harvard scientist who invented base editing and prime editing, was awarded the 2025 Breakthrough Prize in Life Sciences for this work.
Limitations and Risks of Gene Editing
Mutation specificity. The biggest practical limitation of precision gene editing (base editing, prime editing) is that each therapy must be designed for a specific mutation or narrow class of mutations. A base editor built for one patient's point mutation will not work for a different patient with a different mutation in the same gene. For diseases with hundreds of known causative variants, designing a unique editor for each patient is a massive operational and regulatory challenge. CRISPR-Cas9 knockout strategies like Casgevy are more broadly applicable but can only be used when disabling a gene is therapeutically useful.
Off-target editing. Any time you modify DNA, there is a risk of making unintended changes elsewhere in the genome. CRISPR-Cas9 can cut at sites that resemble but are not identical to the target. Base editors can cause "bystander" edits at nearby bases. Prime editors appear to have the cleanest off-target profile of the three, but the data is still early. Because edits are permanent, any off-target change is also permanent. The permanence that makes gene editing powerful also makes its mistakes irreversible.
Tissue access. In vivo gene editing is currently most effective in the liver, because LNPs naturally accumulate there. Reaching the brain, muscles, heart, kidneys, and other organs remains a major unsolved delivery challenge. Ex vivo editing sidesteps this problem for blood diseases but is not applicable to diseases affecting solid organs.
Clinical immaturity. Gene editing has far less clinical history than gene replacement. Casgevy was approved in December 2023. Base editing has treated approximately 20 patients in a published clinical trial. Prime editing has treated one patient. We do not have 7-year follow-up data like we do for Zolgensma. Long-term safety, durability, and late-emerging risks remain open questions.
These Are Not Competing Technologies
The most important thing to understand: gene replacement and gene editing are not rivals. They are complementary tools suited to different situations. The right choice depends entirely on your child's specific mutation, which gene is involved, what organ is affected, and how urgently treatment is needed.
For a child with SMA, Zolgensma (gene replacement) has 7+ years of outcome data and has treated thousands of patients. It is the proven option today.
For a child with sickle cell disease, Casgevy (CRISPR gene disruption) has shown transformative results and is FDA-approved.
For a child with a disease caused by a very large gene, gene editing may be the only viable path, since AAV cannot carry genes above a certain size.
What This Means for Your Family
The field of genetic medicine is moving faster than at any point in its history. In the last 18 months alone, we have seen the first personalized in vivo gene editing treatment (Baby KJ), the first clinical data from prime editing, the first base editing therapy to achieve transfusion independence in nearly 20 patients, and continued long-term data confirming the durability of gene replacement therapies approved years ago.
But speed in science does not automatically translate to speed for your child. The gap between "this technology exists" and "this technology is available for my child's specific mutation" is where most families get stuck. Navigating which modality fits, which clinical trials are recruiting, which regulatory pathway applies, and which research teams have the right expertise requires a systematic evaluation that most families cannot do alone.
That is what Nome does. We evaluate your child's specific mutation against every available therapeutic modality, including gene replacement, gene editing, ASO therapy, drug repurposing, and active clinical trials. We find the fastest realistic path and, for families who want to move forward, manage the development process from variant analysis through FDA submission.
The technology is real. The question is whether someone is looking at your child's mutation with the full picture.
Submit your child's genetic diagnosis for a free evaluation at nome.bio.
References
Ahrens-Nicklas R, Musunuru K, et al. Patient-Specific In Vivo Gene Editing to Treat a Rare Genetic Disease. New England Journal of Medicine. 2025.
Frangoul H, et al. Exagamglogene Autotemcel for Severe Sickle Cell Disease. New England Journal of Medicine. 2024.
Vertex Pharmaceuticals. Long-term data for Casgevy presented at EHA Congress. 2024.
Novartis. Zolgensma long-term follow-up data: sustained durability up to 7.5 years. LT-001 and LT-002 studies. 2023.
Anzalone AV, et al. Search-and-replace genome editing without double-strand breaks or donor DNA. Nature. 2019;576:149-157.
Prime Medicine. First clinical data for PM359 in chronic granulomatous disease. May 2025.
Chen J, et al. Base editing therapy for beta-thalassemia (CS-101). Nature. 2026.
Ertl HCJ. Immunogenicity and toxicity of AAV gene therapy. Frontiers in Immunology. 2022;13:975803.
Innovative Genomics Institute. CRISPR Clinical Trials: A 2025 Update.
HHMI. David Liu Awarded 2025 Breakthrough Prize. 2025.
FDA. FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease. December 2023.
Genomics Education Programme. Casgevy: How It Works. 2023.
Nature Biotechnology. FDA Clears Prime Editors for Testing in Humans. 2024.
CRISPR Medicine News. Overview of CRISPR Clinical Trials 2026.
This article is for informational purposes only and does not constitute medical advice.