Is there already proof that in vivo CRISPR works in humans?
Yes, and the evidence is striking. The most direct proof comes from a 2021 phase 1 clinical trial of NTLA-2001, a CRISPR-Cas9 therapy delivered intravenously via lipid nanoparticles to patients with transthyretin (ATTR) amyloidosis, a fatal disease caused by misfolded protein buildup. In just 28 days after a single infusion, the therapy reduced the disease-causing protein in the blood by an average of 87% in the higher-dose group (0.3 mg/kg), with only mild side effects [6]. This was the first-ever human data showing that CRISPR can edit genes inside the body to treat a disease, and the effect was durable—preclinical studies had already shown lasting knockout of the target gene after a single dose.
Even more recent is a 2025 case report of a newborn with a severe, often fatal urea-cycle disorder (carbamoyl-phosphate synthetase 1 deficiency). Within weeks of receiving a customized lipid nanoparticle-delivered base editor (a more precise form of CRISPR that changes single DNA letters), the infant was able to tolerate more dietary protein and required half the dose of a nitrogen-scavenger medication, with no serious adverse events [3]. This demonstrates that in vivo CRISPR can be tailored to a patient's specific mutation and deployed rapidly in a life-threatening situation.
These two human studies, though small in number, are the strongest evidence here that in vivo CRISPR is not just a future promise—it is already delivering measurable clinical benefit. They also highlight a key advantage: a single treatment can produce a lasting effect, potentially replacing lifelong daily medications.
How is in vivo CRISPR accelerating the discovery of new treatments?
Beyond direct therapy, in vivo CRISPR is being used as a discovery engine to find new drug targets and understand cancer biology. The key innovation is the 'in vivo CRISPR screen'—a technique where thousands of genes are systematically disrupted inside a living animal to see which ones drive or suppress disease. This is far more realistic than studying cells in a dish, because it captures the full complexity of the tumor microenvironment and immune system.
For example, a 2025 study used a genome-wide in vivo CRISPR screen in a mouse model of glioblastoma (an aggressive brain cancer) to find genes that, when knocked out, made the tumor vulnerable to an immunotherapy called L19-TNF. The screen converged on a process called ferroptosis (a form of cell death), and when the researchers combined a ferroptosis-triggering drug with the immunotherapy, survival in mice was significantly prolonged [2]. The same study validated the targets in patient tumor samples, showing clinical relevance.
Similarly, a 2024 study screened 125 genes recurrently mutated in pancreatic cancer by mutating them directly in the pancreas of living mice. This identified two previously unknown tumor suppressor genes, USP15 and SCAF1, which are altered in 31% of pancreatic cancer patients. Tumors lacking these genes were more sensitive to existing chemotherapy (gemcitabine) and a PARP inhibitor, suggesting new treatment strategies for a notoriously hard-to-treat cancer [5]. Another 2025 screen in a mouse model of soft-tissue sarcoma found that knocking out the BAP1 gene cooperated with a common p53 mutation to drive tumor formation, and that FAT1 loss made tumors sensitive to a BRD4 inhibitor [1].
These screens are powerful because they don't just find genes—they find vulnerabilities that can be exploited with existing or new drugs, directly linking discovery to treatment. The glioblastoma study [2] and the pancreatic cancer study [5] are particularly strong because they validated their findings in patient samples or with orthogonal drug screens, bridging the gap between mouse models and human disease.
What are the biggest hurdles, and are they being solved?
The central challenge for in vivo CRISPR is delivery: getting the gene-editing machinery (Cas9 protein and guide RNA) into the right cells in the body, and only those cells, without being degraded or causing off-target edits. The papers here show that this hurdle is being tackled from multiple angles, with lipid nanoparticles (LNPs) emerging as a leading solution.
The human trials for ATTR amyloidosis [6] and the urea-cycle disorder [3] both used LNPs to deliver CRISPR components to the liver, and they worked. LNPs are already used in mRNA vaccines, so they have a proven safety track record. But the liver is an easy target because it naturally absorbs lipid particles. For other organs, researchers are developing viral vectors (like adeno-associated viruses, or AAVs) and virus-like particles, as well as non-viral nanocarriers that can be engineered to target specific tissues [9][4]. A 2023 review highlights that biodegradable polymers and microneedle patches are being designed for skin disorders, and optical or chemical 'switches' are being added to control when Cas9 is active, reducing off-target editing [4].
Another precision challenge is avoiding unintended edits at similar DNA sequences. Here, newer CRISPR tools offer an advantage. Base editors, which change one DNA letter to another without cutting the DNA double helix, are inherently more precise. The 2021 primate study using base editors to knock down the cholesterol gene PCSK9 achieved a ~90% reduction in the PCSK9 protein and ~60% reduction in 'bad' LDL cholesterol, with effects lasting at least 8 months after a single dose [8]. This level of durable, precise editing in a large animal model is a critical step before human trials. Additionally, a 2023 paper introduced the CRISPR-Csm complex, which targets RNA instead of DNA, achieving 90-99% knockdown of target RNA in human cells with minimal off-target effects—a promising alternative for diseases where you want to temporarily silence a gene rather than permanently edit DNA [7].
The evidence shows that delivery and precision are solvable problems, not fundamental barriers. The field is moving from 'can we edit?' to 'can we edit safely and specifically in the right tissue?'—and the answers are increasingly positive, with multiple complementary strategies under development.
About These Sources
This answer is built on 9 peer-reviewed studies — published from 2021 to 2025, 5 from 2024 or later, 9 in Q1 journals, collectively cited 2,541 times — selected as the most relevant from 13 studies that passed quality screening, drawn from 50 papers retrieved from a database of over 500 million.
Sources used in this answer
Abstract A025: Direct <i>in vivo</i> CRISPR screen identifies <i>BAP1</i> and <i>FAT1</i> as potent tumor suppressors in sarcomagenesis
An in vivo CRISPR screen in a mouse model of soft-tissue sarcoma identified BAP1 and FAT1 as tumor suppressors; FAT1 loss made sarcomas sensitive to a BRD4 inhibitor, suggesting a new targeted therapy.
KS01.7.A IN VIVO CRISPR SCREENING UNCOVERS VULNERABILITIES FOR CANCER IMMUNOTHERAPY AGAINST GLIOBLASTOMA
A genome-wide in vivo CRISPR screen in a mouse glioblastoma model revealed that vulnerabilities converge on ferroptosis; combining a ferroptosis-inducing drug with immunotherapy significantly prolonged survival, validated in patient samples.
Patient-Specific In Vivo Gene Editing to Treat a Rare Genetic Disease
A 2025 case report of a newborn with a severe urea-cycle disorder treated with a customized lipid nanoparticle-delivered base editor; the infant tolerated more protein and required half the dose of a nitrogen-scavenger drug within weeks, with no serious adverse events.
Development of CRISPR/Cas Delivery Systems for In Vivo Precision Genome Editing
A review of delivery systems for in vivo CRISPR, highlighting biodegradable polymers, microneedle patches, and optical/chemical switches to control Cas9 activity and reduce off-target editing.
In vivo CRISPR screens reveal SCAF1 and USP15 as drivers of pancreatic cancer
An in vivo CRISPR screen in a mouse model of pancreatic cancer identified USP15 and SCAF1 as tumor suppressors; their loss (seen in 31% of patients) increased sensitivity to gemcitabine and PARP inhibition.
CRISPR-Cas9 In Vivo Gene Editing for Transthyretin Amyloidosis
A phase 1 clinical trial of NTLA-2001 (CRISPR-Cas9 in lipid nanoparticles) in 6 patients with ATTR amyloidosis; a single dose reduced disease-causing serum TTR protein by up to 87% at day 28 with only mild adverse events.
Precise transcript targeting by CRISPR-Csm complexes
A study showing that the CRISPR-Csm complex achieves 90-99% knockdown of target RNA in human cells with minimal off-target effects, outperforming shRNA and Cas13, and can be used for live-cell RNA imaging.
In vivo CRISPR base editing of PCSK9 durably lowers cholesterol in primates
In cynomolgus monkeys, a single infusion of lipid nanoparticle-delivered CRISPR base editors targeting PCSK9 achieved ~90% reduction in PCSK9 protein and ~60% reduction in LDL cholesterol, lasting at least 8 months.
Viral and nonviral nanocarriers for in vivo CRISPR-based gene editing
A review of viral (AAV, virus-like particles) and nonviral (lipid nanoparticles) nanocarriers for in vivo CRISPR delivery, discussing their promise and remaining challenges.