

CRISPR can fix a typo in your DNA. But what about replacing an entire missing gene? A new tool called INSTALL uses a clever immune-evasion trick to insert gene-sized DNA sequences, and it could unlock treatments for thousands of rare diseases that current editors can't touch.
Your body has about 20,000 genes. When one of them breaks, the consequences can be devastating: metabolic disorders, blood diseases, liver failure. For years, scientists have had tools to fix tiny typos in your DNA. Think of CRISPR as a precise pencil eraser that can rub out a single wrong letter. Base editing is like whiteout for one character. Prime editing can swap out a word or short phrase.
But what if the problem isn't a typo? What if an entire paragraph is missing?
That's the challenge a new tool called INSTALL is built to solve. And it might just crack open the next era of genetic medicine.
Published on March 11, 2026, research on INSTALL (which stands for "integration through nucleus-synthesized template addition of large lengths") tackles a problem that's haunted gene therapy for years: how do you insert something big into the genome without triggering an immune meltdown?
The existing CRISPR toolkit is brilliant at small fixes. Base editors can swap a single DNA letter (think C to T) without even cutting the double helix. Prime editors can handle slightly bigger jobs, but for diseases caused by large missing or mangled gene segments, these tools are like trying to renovate a kitchen with nothing but a screwdriver.
You need the equivalent of a full construction crew. INSTALL is designed to be that crew.
To understand why INSTALL matters, you need to know about your body's paranoia. Your cells are constantly scanning for threats, and one of the biggest red flags is double-stranded DNA floating in the cytoplasm (the gooey interior of a cell, outside the nucleus). When cells spot it, they assume a virus has invaded and slam the panic button. That triggers inflammation and cell death, which is, to put it mildly, not great for a therapy.
This is why delivering large chunks of replacement DNA has been so difficult. Previous methods relied on viral vectors (essentially hijacked viruses) or double-stranded DNA templates, both of which set off immune alarms or come with serious size limitations. AAV vectors, the most popular delivery vehicle in gene therapy, can only carry about 4,700 base pairs of cargo. Many genes are bigger than that.

An Iran-linked hacking group just hit Stryker, one of the world's largest medical device companies, wiping 200,000 systems and claiming to steal 50 terabytes of data. The attack didn't compromise surgical robots or defibrillators, but it exposes a terrifying vulnerability in the healthcare supply chain.


Join thousands of biotech professionals who start their day with our free, daily briefing.
INSTALL's workaround is elegant. Instead of double-stranded DNA, it uses circular single-stranded DNA, inspired by how certain bacteria and viruses store their genetic material. Here's the key insight: cells don't flag single-stranded DNA as a threat. It flies under the immune radar like a delivery driver in an unmarked van.
The INSTALL system, developed by Benjamin Kleinstiver and Connor Tou at Massachusetts General Hospital, uses cssDNA molecules formed into tiny spheres, with technology contributed by Full Circles Therapeutics' C4DNA™ platform. Small double-stranded snippets (too short for immune detection) are attached to provide docking sites for gene editing enzymes. The whole package gets delivered via lipid nanoparticles, the same fatty bubble technology that powered COVID mRNA vaccines. Once inside the nucleus, the editing enzymes integrate the large DNA payload into the precise target location.
No viruses required. No immune freakout. Gene-sized insertions, done cleanly.
Let's zoom out and talk about why inserting large DNA sequences is such a big deal.
More than 7,000 rare diseases remain underserved by current therapies. Many of them involve mutations too complex for a single-letter fix. A child with a large deletion in a critical gene doesn't need a spell-checker; they need an entire passage restored. Current tools simply can't do that with any reliability.
Traditional CRISPR (using the Cas9 enzyme) can theoretically insert bigger pieces through a repair process called homology-directed repair, or HDR. But HDR is wildly inefficient. Cells much prefer a sloppier repair pathway called NHEJ, which basically glues the cut ends back together and hopes for the best. The result: error-prone edits, potential large deletions, and a repair success rate that makes a .200 batting average look impressive.
There's also the safety issue. Standard CRISPR cuts both strands of DNA to make its edits. Those double-strand breaks can activate p53, a tumor-suppressor gene that acts like a cellular fire alarm. Repeatedly triggering p53 can bias the surviving cell population toward cancer-prone cells. Not exactly the therapeutic outcome anyone wants.
INSTALL sidesteps these problems by providing the template for large insertions without relying on error-prone repair or dangerous double-strand breaks.
The early reaction from the gene editing community has been positive. Kiran Musunuru, a leading figure in the CRISPR field, called INSTALL "beautiful work" that elegantly solves the challenge of controlled DNA insertion. Benjamin Kleinstiver, the researcher behind the tool, highlighted its potential for both broad applications (treating common mutations shared by tens of thousands of patients) and bespoke therapies (custom fixes for ultra-rare cases affecting only hundreds).
So far, INSTALL has been demonstrated in lab-grown human cells and mouse genomes, with early targets including liver diseases, blood disorders, and metabolic conditions. It's pre-clinical, meaning years of work remain before it could reach patients. But the proof of concept is there.
INSTALL doesn't exist in a vacuum. The gene editing landscape in 2026 is more competitive than ever, and several companies are pushing the boundaries of what's possible.
Beam Therapeutics is leading the charge in base editing, with its BEAM-302 program for alpha-1 antitrypsin deficiency showing promising early clinical data, including a 2.8-fold increase in functional AAT levels and mean total AAT of 12.4μM at Day 28, exceeding the 11μM protective threshold. Pfizer liked the tech enough to exercise global rights on an unnamed Beam base-editing candidate in February 2026.
Intellia Therapeutics is advancing in vivo CRISPR editing for transthyretin amyloidosis (a progressive disease where misfolded proteins damage organs), though its Phase 3 MAGNITUDE and MAGNITUDE-2 trials were paused in October 2025 due to a grade 4 liver adverse event. CRISPR Therapeutics is expanding beyond its landmark sickle cell therapy into cardiovascular disease, with clinical-stage programs targeting cholesterol and blood clotting.
Then there are the wildcards. Tessera Therapeutics is building "Gene Writers" that use mobile genetic elements to insert, replace, and rewrite entire genes. Chroma Medicine is developing epigenetic editors that silence genes without touching the DNA sequence at all. Each approach has different strengths, and it's too early to declare a winner.
What INSTALL brings to this mix is a distinct advantage in payload size. While base and prime editors excel at precision fixes, and epigenetic editors work by turning genes on or off, INSTALL is specifically designed for the heavy lifting: replacing whole genes or inserting large missing segments.
The timing couldn't be better. In late February 2026, the FDA released draft guidance on a "plausible mechanism" approval pathway for custom genome editing therapies targeting rare diseases. The idea is straightforward: if a therapy can demonstrate it targets the root cause of a disease, it can potentially win approval using data from just a handful of patients combined with natural history controls.
This pathway was inspired in part by the story of baby KJ, an infant with CPS1 deficiency (a deadly urea cycle disorder) who received a custom CRISPR therapy. Musunuru described efforts to convert that one-off treatment into a platform for urea cycle disorders more broadly, aiming for a single trial that covers multiple liver-centered diseases.
For a technology like INSTALL, which could theoretically serve as a platform for many different large-insertion therapies, this regulatory framework is a potential accelerator. One well-characterized tool, applied across multiple diseases, with a streamlined path to approval.
Gene editing has spent the last decade proving it can fix small problems precisely. INSTALL represents a bet that the field is ready for bigger ambitions: not just correcting typos, but restoring entire missing chapters of the genetic code.
It's early. The technology has only been validated in cells and mice. Manufacturing, delivery optimization, and safety studies will take years. The path from elegant lab result to approved therapy is long, expensive, and littered with fallen heroes.
But the unmet need is enormous. Approximately 95% of the 7,000+ rare diseases lack any approved treatment, and many involve the kind of large genetic disruptions that current tools can't address. INSTALL won't solve all of them. But if it works as advertised, it fills a gap that the gene editing field has been trying to close since CRISPR first captured the world's imagination.
Sometimes the most important breakthroughs aren't about doing something new. They're about doing something bigger.
A South Korean biotech with no drugs on the market just drew 839-to-1 institutional demand for its IPO, pricing at the top of its range. Behind the frenzy: a near-billion-dollar licensing deal with a US partner backed by one of biotech's biggest venture firms.