From CRISPR Breakthrough to Clinical Reality
Casgevy: From "Capable of Editing" to "Capable of Becoming a Drug"
CRISPR/Cas9 gene editing has rapidly evolved from a fundamental
research tool to an approved clinical therapeutic modality. Exemplified by the FDA
approval therapy of Casgevy for sickle cell disease and transfusion-dependent beta-thalassemia,
this therapy achieves sustained clinical benefits
by specifically disrupting the
BCL11A erythroid enhancer to
reactivate fetal
hemoglobin within autologous hematopoietic stem cells
[1][2][3][4].
This ex vivo editing strategy fully leverages the self-renewal capacity of hematopoietic stem
cells, enabling durable hematopoietic reconstitution
through a single-dose cellular therapy. However, broader clinical adoption of this technology
will continue to depend on improving gene
editing efficiency, minimizing off-target risks, and streamlining large-scale manufacturing
processes[5][6].
This evolving view of "immune ecology" suggests that neuroinflammation is a highly coordinated,
multicellular, and temporally regulated process involving interactions among innate immunity,
adaptive immunity, glial cells, and neurons.
Figure 1. Casgevy (Vertex Pharmaceuticals and CRISPR Therapeutics): Mechanism of action
involving the reactivation of γ-
globin
expression and the induction of therapeutic HbF
[7].
The therapeutic mechanism of Casgevy lies in targeting the erythroid enhancer site of
BCL11A to
induce DNA double-strand breaks, which are
subsequently repaired through the
non-homologous end joining
(NHEJ) pathway, resulting in insertion or deletion mutations.
This disruption weakens the repressive effect of
BCL11A on γ-globin expression and thereby
reactivates fetal hemoglobin production.
Clinical trials have demonstrated that treated patients not only achieved high rates of
transfusion independence but also showed a
significant reduction in the frequency of vaso-occlusive crises (VOCs), with some patients
experiencing complete elimination of
VOC-related events
[2][7].
As one of the first approved gene-editing therapies, Casgevy also helped establish emerging
regulatory frameworks for long-term follow-up,
off-target risk assessment, and genomic safety evaluation. Three-year follow-up data reported in
2024 indicated that 94% of sickle cell disease
patients and 81% of beta-thalassemia patients achieved their respective clinical
endpoints—including becoming transfusion-independent or
eliminating vaso-occlusive crises. For eligible patients, this one-time treatment holds the
potential to significantly reduce the lifelong
burden of transfusions, chronic medication use, and repeated hospitalizations[1][3].
NTLA-2002: Clinical Validation of in vivo Gene Editing
Figure 2. KLKB1-targeting NTLA-2002 mechanism of action[9].
NTLA-2002 is a CRISPR-based
in vivo gene editing therapy designed to treat
hereditary angioedema (HAE). The therapy utilizes lipid
nanoparticles (LNPs) for systemically deliver CRISPR components directly to hepatocytes, where
editing of the
KLKB1 gene reduces
kallikrein production and lowers the risk of HAE attacks
[8][9].
Unlike
ex vivo editing approaches, NTLA-2002 does not require cell
harvesting,
ex vivo manipulation, or reinfusion. Instead, gene editing is performed directly
within the patient's body.
NTLA-2002 delivers Cas9 mRNA together with a guide RNA targeting
KLKB1 through
LNP-mediated hepatic delivery. Following uptake by
hepatocytes, CRISPR/Cas9 system mediates the knockout of KLKB1, reducing plasma
kallikrein
production and preventing
downstream inflammatory signaling
[8].
Phase I/II clinical data showed that patients receiving the highest dose experienced a 95%
reduction in HAE attack frequency relative
to baseline, while all patients remained attack-free 16 weeks after treatment. NTLA-2002 has now
advanced to Phase
III clinical trials, and a regulatory submission has been filed with the FDA[9].
As the world's first in vivo gene editing therapy to achieve Phase III success, NTLA-2002
validates the feasibility of LNP-mediated systemic delivery for hepatic gene editing and
highlights the potential of one-time curative therapies for liver-associated genetic diseases.
DNA Damage Response Pathways Remain a Central Bottleneck
Figure 3. Cell cycle-dependent DDR activation at DSBs[12].
Despite rapid progress in genome engineering, efficient and safe
in vivo editing technologies
remain heavily constrained by cellular
DNA damage response (DDR) pathways. Nuclease-induced
DNA double-strand breaks can activate NHEJ or homology-directed repair
(HDR) pathways. However, uncontrolled repair can lead to chromosomal rearrangements or trigger
p53-mediated apoptosis.
In theory, modulating DDR components—such as inhibiting NHEJ factors or enhancing HDR
function—can shift repair outcomes and increase
the frequency of precise genome editing. However, without stringent control over repair pathway
choice and the editing window, such
interventions may introduce substantial genotoxic risks. Therefore, DDR modulation should not be
pursued solely to improve editing
efficiency; it must also comprehensively evaluate its effects on chromosomal structure, cellular
viability, and the generation of
unintended mutations[10][11].
The use of small-molecule DDR modulators remains highly controversial. Early studies reported
that SCR7, a small molecule inhibitor
targeting
DNA ligase IV, could enhance HDR efficiency; however, numerous subsequent independent
studies failed to replicate
these findings. In addition, concerns regarding its limited inhibitory specificity and the
potential to exacerbate genomic
instability have prevented its progression into clinical trials. Furthermore, since HDR
primarily occurs during the S and G2 phases
of the cell cycle, whereas many target cells
in vivo remain in a quiescent state, HDR efficiency
is often below 0.1%. This severely
limits the
in vivo application of traditional precision editing technologies mediated
by double-strand breaks
[11].
In the field of drug discovery and development, genomic editing strategies are increasingly
shifting toward approaches that minimize
persistent activation of the DDR, including the use of transiently expressed editing systems,
optimizied delivery windows, and
self-inactivating vectors. In the context of in vivo editing therapies, the central objective of
DDR modulation is to maximize
targeted editing efficiency while simultaneously minimizing chromosomal rearrangements,
p53-mediated cytotoxicity, and the
generation of unintended repair products.
The Rise of Base Editing: Toward Precision Genome Repair
Figure 4. General overview of DNA base editing technologies[15].
Base editing technologies enable programmable nucleotide conversion without inducing DNA
double-strand breaks. By
avoiding nuclease-mediated DNA cleavage, base editors substantially reduce the formation of
insertion-deletion (indel)
mutations and broaden the therapeutic potential for pathogenic point
mutations associated with rare diseases[12][13][14].
Cytosine base editors (CBEs) convert C•G pairs to T•A pairs, while adenine base editors
(ABEs) convert A•T pairs to G•C pairs; furthermore, novel base editing systems are now being
utilized to address certain transversion mutations.
Most base editors consist of a catalytically impaired Cas protein—such as a Cas9 nickase or
an inactivated Cas9 (dCas9)—fused to a deaminase. Guided by programmable RNAs sequences,
these systems perform localized chemical modification of DNA bases within a defined editing
window[12][13].
This design facilitates precise nucleotide conversions while minimizing the rate of indel
formation. Current estimates suggest that base
editing could theoretically correct approximately 95% of pathogenic transition mutations
cataloged in the ClinVar database[12][15].
Preclinical studies using dual adeno-associated virus (AAV) vectors and LNP-based delivery
have demonstrated successful correction of
pathogenic mutations in disease models such as progeria and spinal muscular atrophy. At the
translation level, base editing offers
novel therapeutic avenues for monogenic diseases previously limited by delivery constraints,
repair inefficiency, or
nuclease-associated safety risks. Concurrently, advances in off-target detection tools—such
as CIRCLE-seq and
CREATE—along with the engineering of high-fidelity editing enzymes, are improving the
systematic safety evaluation
of base-editing platforms[15][16].
The World's First Case of Personalized in vivo Base Editing: N-of-1 Therapy for CPS1
Deficiency
CPS1-related urea cycle disorders are life-threatening
inherited metabolic diseases with limited treatment options. In a landmark
clinical case, researchers rapidly developed a personalized LNP-delivered adenine base
editing therapy for a neonate
with
Carbamoyl Phosphate Synthetase I (CPS1) deficiency.
Following treatment, the patient showed a marked reduction in
disease burden, with no serious adverse events observed during a 7-week follow-up period
[9][17].
This therapeutic approach employs a patient-specific guide RNA to target the pathogenic
mutation and utilizes LNPs to deliver mRNA encoding ABE8e into hepatocytes. This LNP-mRNA
delivery system provides a transient expression window, enabling in vivo base editing while
reducing the risks of sustained expression inherent to traditional viral vectors.
As a quintessential "N-of-1" (single-patient) therapeutic case, this study not only
validates the rapid programmability of base-editing platforms for precise correction of an
individual patient's pathogenic mutation, but also pioneers a new paradigm for personalized
treatment of ultra-rare genetic diseases driven by unique mutations, which are often
impractical to address through conventional drug development pipelines.
The Emerging Paradigm of N-of-1 Medicine
Figure 5. The state-of-the-art of N-of-1 therapies and the development roadmap[21].
The high programmability of base editing has fundamentally revolutionized traditional R&D
paradigms, shifting the focus from the development of universal drugs targeting specific
diseases or mutation types toward "N-of-1" personalized therapies centered on sequence
design, editing-tool selection, delivery-system optimization, and rapid synthesis.
Through a platform-based R&D model—where standardized manufacturing processes ensure
relative consistency in editing enzymes, delivery vectors,
and quality control frameworks—the timeline for preparing personalized treatment regimens
has been drastically shortened from years
to mere months[18].
In the field of rare disease therapeutics, this approach enables deep integration with
modular delivery vectors, such as LNPs and
adeno-associated viruses (AAVs), thereby substantially reducing the cost of developing
treatments for each new indication from
scratch. Its principal advantage does not lie in circumventing safety evaluation, but rather
in the high reusability of platform
components and the markedly accelerated cycles for sequence design and candidate therapeutic
production. Collectively, these
features provide a more practical and scalable framework for developing precision therapies
tailored to small patient populations
or even individual patients[18].
Technical Challenges and Research Tool Requirements for Base Editing
Site-Dependent Editing Efficiency
The editing efficiency of ABEs and CBEs is influenced by multiple factors, including
chromatin accessibility at the target locus, local DNA
secondary structure, cell cycle status, and DNA methylation patterns. Because chromatin
environments and deaminase activity windows differ
across genomic loci, editing efficiencies can exhibit significant disparities[21].
This site-dependent variability requires researchers to conduct empirical screening during
candidate target selection, rather than relying solely on guide RNA sequences to predict
editing outcomes. In practical applications, multiple candidate sites are typically
evaluated by considering factors such as target cell type, editing window, distribution of
bystander bases, and the position of the target base, in order to identify editing
strategies that achieve an optimal balance between both efficiency and specificity.
Off-Target Effect Profiles
Base editing avoids the DNA double-strand breaks induced by Cas9 nucleases; however, its
off-target risks still require systematic evaluation.
The primary risks include off-target DNA editing, transcriptome-wide off-target RNA editing,
and "bystander editing", in which
non-target cytosines or adenines within the editing window are inadvertently converted[22][23].
Therefore, prior to clinical translation, it is necessary to integrate unbiased sequencing
methods with targeted validation experiments to
analyze the DNA and RNA off-target profiles of candidate editors within target cells. For
therapeutic base editing, safety evaluation should
extend beyond simply reporting average on-target editing efficiency and instead
systematically assess bystander-editing frequency,
off-target site distribution, and the long-term stability of edited products.
Regulatory Requirements for Mismatch Repair Pathways
The heteroduplex DNA generated during base editing must undergo endogenous DNA repair
processes to achieve permanent fixation of the
edited sequence. The activity of mismatch repair (MMR) influences editing outcomes:
excessive MMR activity may reverse edited products,
whereas MMR deficiency can increase the overall mutational burden. Consequently, MMR
activity impacts both the efficiency and fidelity
of base editing, underscoring the need for studies to validate editing outcomes and evaluate
the risks of unintended editing under
physiologically relevant conditions[10].
When optimizing base editing strategies, researchers typically integrate analyses of target
site chromatin accessibility, DNA
repair pathway activity, and the specific characteristics of the cell type. Epigenetic
modulators, such as
HDAC inhibitors
that enhance chromatin accessibility, as well as DNA repair pathway regulators (e.g.,
PARP inhibitors or
DNA-PK
inhibitors)
can be employed to investigate the interplay between editing efficiency, repair bias, and
off-target risks. Importantly, these
interventions should be coupled with sequencing validation to definitively determine whether
they enhance on-target editing
efficiency, reduce bystander editing, or alter the spectrum of unintended editing events.
The Delivery Race: From Liver Targeting to Tissue Selectivity
Figure 6. Advantages and limitations of viral, non-viral, and hybrid AAV-LNP vectors for in
vivo delivery systems[25].
Structural Limitations of AAV Vectors
Due to their relatively favorable safety profile, inherent tissue tropism, and capacity for
long-term gene expression, AAV vectors
remain widely utilized for in vivo gene delivery. However, their effective packaging
capacity is limited to approximately 4.7–5 kb;
exceeding this threshold often results in reduced packaging efficiency, genomic truncation,
or diminished expression levels. Consequently,
it is challenging to simultaneously accommodate full-length base editors or prime
editors—along with their requisite gRNAs, promoters,
and regulatory elements—within a single AAV vector[16][24].
To address these packaging limitations, researchers have developed dual-AAV delivery
strategies based on split-intein-mediated reconstitution.
In this approach, base editors or prime editors are divided into fragments, packaged
separately into distinct AAV vectors, and subsequently
reassembled within target cells to generate functional editing complexes[16]. While this
strategy expands the capacity of AAV vectors to deliver
large-scale editing systems, it relies on dual-vector co-infection and intracellular
reconstitution. This increases system complexity and renders
editing efficiency susceptible to various factors, including co-delivery ratios,
reconstitution efficiency, and the physiological state of
the target cells[16].
In addition, the prolonged expression of Cas9 or other editing proteins mediated by AAV
vectors may exacerbate immune responses against Cas proteins
and increase the risk of off-target editing[24][25]. From an immunological perspective,
approximately 30–60% of the human population possesses
pre-existing antibodies against AAV, which can reduce delivery efficiency and complicate
patient enrollment. Furthermore, humoral and cellular
immune memory responses induced after the initial administration may prevent the effective
re-administration of vectors utilizing the same
capsid serotype. Regarding safety, high-dose systemic administration of AAV vectors has been
associated with adverse events such as severe
hepatotoxicity, thrombotic microangiopathy, renal injury, and multi-organ failure.
Collectively, packaging capacity limitations, restrictions
on re-administration, immune responses, and tissue toxicity remain the primary bottlenecks
in the application of AAV vectors for in vivo gene
editing delivery.
The LNP Platform: From Liver Preference to Tissue Selectivity
LNPs have emerged as widely used non-viral vectors in gene editing and RNA delivery;
however, their clinical applications remain markedly
biased toward the liver. Most successful examples of LNP delivery—including NTLA-2002 and
the individualized CPS1 therapy—primarily target
hepatic tissues. This "structural hepatotropism" is linked to the
in vivo uptake mechanisms
of LNPs: LNPs can bind to plasma
apolipoprotein
E (ApoE), which promotes their uptake and accumulation in hepatocytes through members of the
low-density lipoprotein receptor family,
particularly
LRP1 and
LDLR,
that are highly expressed on the surface of hepatocytes.
While this hepatic tropism renders LNPs particularly effective for targeting pathogenic
genes expressed in hepatocytes, it also limits their
utility in extrahepatic tissues. Through strategies such as ligand conjugation, lipid
composition modulation, or surface modification,
LNPs can be engineered to facilitate delivery to extrahepatic tissues—such as the lungs,
spleen, and endothelium—thereby broadening
their therapeutic applicability[8][25]. Furthermore, Selective Organ Targeting (SORT)
nanoparticles have demonstrated that the
rational modulation of LNP composition can reshape in vivo distribution patterns, thereby
enabling more
tissue-selective delivery[26].
Summary
Programmable medicine has entered clinical reality through CRISPR-based therapies such as
Casgevy and NTLA-2002, which have demonstrated durable
efficacy in rare genetic diseases via both ex vivo and in vivo gene-editing approaches[1][17]. Base editing technologies further expand
this paradigm by enabling precise single-nucleotide corrections without double-strand breaks,
offering improved safety and strong potential
for ultra-rare "N-of-1" therapies[17][12].
Meanwhile, continuous innovation in delivery systems—including dual-AAV platforms and
tissue-selective nanoparticles—are expanding the spectrum
of treatable diseases while addressing challenges related to payload capacity, immunogenicity,
and delivery specificity[16][26]. As platform-based
development models shorten R&D cycles from years to months, rigorous evaluation of off-target
effects
and long-term safety remains essential[18][12].
Together, these advances mark the emergence of an era in which rare diseases therapies can be
programmed at the molecular level, with
future progress dependent on the integration of molecular engineering, translational science,
regulatory innovation, and equitable access.
Recommended Products targeting Gene Editing
| Product Name |
Cat. No. |
Category |
Mechanism of action |
| VL-422 |
HY-159709 |
LNP system |
An ionizable cationic lipid, delivers CRISPR/Cas9 mRNA to the liver. |
| THP1 Lipid |
HY-171953 |
LNP system |
An ionizable cationic lipid, can deliver mRNA to muscle tissue. |
| BCP-NC2-C12 |
HY-171904 |
LNP system |
An ionizable cationic lipid, can deliver CRISPR mRNA in vivo. |
| OptiLNP Gene Editing Kit |
HY-K2027 |
Gene Editing Kit |
Suitable for the co-transfection of Cas9 mRNA and sgRNA for gene editing in standard
cell lines (adherent or suspension cells). |
| OptiLNP Gene Editing Kit (Immune Cells) |
HY-K2028 |
Gene Editing Kit |
Suitable for the co-transfection of Cas9 mRNA and sgRNA during the gene editing of
immune cells. |
| High-Efficiency Gene Editing Compound Library |
HY-L244 |
Gene Editing Kit |
Contains 702 small molecules capable of explicitly or potentially enhancing gene
editing efficiency. |
| Peptide A5K |
HY-P5307 |
CRISPR Engineering |
Non-covalently binds to CRISPR ribonucleoproteins to efficiently deliver CRISPR mRNA
into cells. |
| INF7TAT-P55 |
HY-P10999 |
CRISPR Engineering |
Achieves CRISPR-engineered ribonucleoprotein delivery. |
| Peptide A5K acetate |
HY-P5307A |
CRISPR Engineering |
Non-covalently binds to CRISPR ribonucleoproteins to efficiently deliver CRISPR mRNA
into cells. |
Note: MCE can provide products for research use only. We do not sell to patients.
References
[1]
Cetin B, et al. Expert Rev Mol Med. 2025 Mar 31;27:e16.
[2]
Greco F, et al. Front Med (Lausanne). 2024 Feb 15;11:1356578.
[3]
Jacinto FV, et al. J Cell Mol Med. 2020 Apr;24(7):3766-3778.
[4]
Laurent M, et al. Cells. 2024 May 8;13(10):800.
[5]
Ugalde L, et al. Cytotherapy, 2023.
[6]
Colella P, et al. Front Genome Ed. 2023 Jul 6;5:1248904.
[7]
Kang CY, et al. Neurospine. 2025 Jun;22(2):421-440.
[8]
Rahmanian M, et al. Hum Gene Ther. 2026 Mar;37(5-6):170-182.
[9]
Danny M, et al.
1 Jun 2023 NCT05120830.
[10]
Masarwy R, et al. Adv Drug Deliv Rev. 2024 Aug;211:115359.
[11]
Gil JS, et al. Gene Ther. 2026 Jan;33(1):97-106.
[12]
Groelly FJ, Nat Rev Cancer. 2023 Feb;23(2):78-94.
[13]
Nambiar TS, et al. Mol Cell. 2022 Jan 20;82(2):348-388.
[14]
Li H, et al. Signal Transduct Target Ther. 2020 Jan 3;5(1):1.
[15]
Porto EM, et al. Nat Rev Drug Discov. 2020 Dec;19(12):839-859.
[16]
Komor AC, et al. Nature. 2016 May 19;533(7603):420-4.
[17]
Kantor A, et al. Int J Mol Sci. 2020 Aug 28;21(17):6240.
[18]
Cabré-Romans JJ, et al. Front Genome Ed. 2025 Apr 2;7:1553590.
[19]
Wang D, et al. Cell. 2020 Apr 2;181(1):136-150.
[20]
Musunuru K, et al. N Engl J Med. 2025 Jun 12;392(22):2235-2243.
[21]
Jonker AH, et al. Nat Rev Drug Discov. 2025 Jan;24(1):40-56.
[22]
Urnov F, et al. Cytotherapy. 2025 Oct;27(10):1151-1163.
[23]
Huang TP, et al. Nat Protoc. 2021 Feb;16(2):1089-1128.
[24]
Guo C, et al. Front Bioeng Biotechnol. 2023 Mar 9;11:1143157.
[25]
Sandoval-Villegas N, et al. Mol Ther. 2024 Oct 2;32(10):3211-3214.
[26]
Rees HA, et al. Nat Rev Genet. 2018 Dec;19(12):770-788.
[27]
Kong X, et al. Sci China Life Sci. 2024 Dec;67(12):2540-2553.
[28]
Moyo B, et al. Biomaterials. 2025 Oct;321:123314.
LinkedIn
Researchgate
Facebook
X
Instagram
Mail
From CRISPR Breakthrough to Clinical Reality
The Rise of Base Editing: Toward Precision Genome
Repair
The Delivery Race: From Liver Targeting to Tissue
Selectivity
