Rewriting Genetic Destiny: Prime Editing Leads the Future in Fixing Genetic Disorders

Clustered regularly interspaced short palindromic repeats (CRISPR) associated protein 9 (Cas9)

In 2012, Jennifer Doudna and Emmanuelle Charpentier developed a revolutionary technology using CRISPR to enhance the precision and efficiency of gene editing, for which they were awarded the Nobel Prize in Chemistry in 2020. CRISPR are repeat DNA sequences that form a component of the bacterial adaptive immune system against viral pathogens. This natural phenomenon was harnessed by Doudna and Charpentier for human genome editing.

The CRISPR-Cas9 technique uses a protein molecule, Cas9 (CRISPR-associated protein 9), that forms a complex with a guide RNA (gRNA). This complex is introduced into the target cell, where the gRNA directs Cas9 to act as a pair of ‘molecular scissors’ to make double-stranded breaks (DSBs) or cuts in the DNA double helix for insertions, deletions, and/or substitutions of genes. The cell’s natural repair mechanisms then fix the DSBs, which are joined end-to-end (nonhomologous end joining, NHEJ) or repaired using a template (homology-directed repair [HDR]). NHEJ is prone to mutations resulting from the insertion and deletion (indel) of base pairs, while the relatively error-free HDR is limited by the availability of exogenous donor DNA repair templates and is shown to be inefficient in most therapeutically relevant cell types.^1,2

In 2023, CASGEVY^® (exagamglogene autotemcel, Vertex Pharmaceuticals Incorporated, MA, USA), a one-time therapy to treat sickle cell disease (SCD) and transfusion-dependent beta thalassaemia, became the first and only currently available gene-editing therapy on the market that is CRISPR-based and US Food and Drug Administration (FDA) approved.³

While CRISPR-Cas9 is versatile and can be used to introduce gene edits in a wide range of organisms (plants, animals, humans),⁴ the aforementioned unintended or ‘off-target’ mutations (indels) necessitate the sequencing of edited cells to eliminate erroneous end products. This restricts its use to ex vivo and in vitro applications, although recent years have seen an increase in vivo therapies advancing into clinical trials.⁵ There are additional concerns in the form of DNA-damage toxicity, where DSBs stimulate cell apoptosis through p53 activation. On the other hand, p53-suppressed populations of cells, which have a higher chance of successful DNA repair, would promote oncogenesis.⁶ Additionally, CRISPR-Cas9 cannot make precise modifications to individual nucleotides.

Base editing (BE) and PE

Recent advances from the laboratory of Dr David Liu, namely, BE (2016) and PE (2019), build upon the CRISPR-Cas9 system and circumvent these limitations by introducing the novel factor of single-stranded breaks (SSBs) instead of DSBs. BE employs a ‘base editor’, which is a modified Cas9, called nCas9 (Cas9 combined with an RNA-programmable nickase), fused with an enzyme (deaminase) capable of chemically converting one nucleotide base to another. This base editor is guided to the target site by gRNA, where it induces transition mutations primarily through two systems—the cytosine BE for C to T/T to C and the adenosine BE for A to G/G to A.^2,7

PE

PE takes it further by allowing all 12 base-to-base conversions, insertions, deletions, and more precise and efficient edits than BE and CRISPR-Cas9. It employs a ‘prime editor’, which is nCas9 combined with PE guide RNA (pegRNA) carrying both the desired edit as well as the reverse transcriptase (RT) enzyme. RT uses pegRNA as a template to synthesise a new DNA strand with the desired edit at the target site. This way, PE has ushered in a novel concept of ‘search-and-replace’ genome editing, replacing the previous versions of ‘error correction’.^2,7 Table 1 provides a comparison of CRISPR-Cas9, BE, and PE.^8,9

Pre-clinical and clinical applications of PE

The wide editing range of PE expands its scope to potentially correcting 89% of human genetic disorders.⁷ Several studies have demonstrated successful gene modification in various cell types, organoids, zebrafish, Drosophila, mice, and plants.² Studies in vivo, ex vivo, and the potential to correct monogenic diseases such as sickle cell anaemia (SCA), Duchenne muscular dystrophy (DMD), cystic fibrosis, and Huntington’s chorea have been examined,¹⁰ and 5 proof-of-principle studies in mammalian animal models, with a promise of future clinical trials are described by Zhao et al.²

A landmark study by Sousa et al.¹¹ demonstrated successful in vivo PE in mice to alleviate alternating hemiplegia of childhood (AHC). AHC is a rare neurological genetic disorder affecting approximately 1:100 000 to 1:1 000 000, and caused by mutations in ATP1A2 (AHC1) and ATP1A3 (AHC2), which encode different α subunits of the Na+/K+ ATPase. Symptoms appear before the age of 18 months, with paroxysmal, reversible episodes of hemiplegia and other abnormalities such as nystagmus, strabismus, seizures, dystonia, choreoathetosis, and developmental delays. The current pharmacological mainstay, flunarizine, offers only symptomatic relief.

In the study, PE was explored following the failure of conventional gene therapy and BE, which corrected only one mutation with important bystander edits. In the subsequent in vivo efficacy test, intracerebroventricular injections were used to deliver PE-AAV9 (adeno-associated virus 9, a neurotropic delivery vector) systems. Disease manifestations in mutant mice were characterized by reduced lifespan and body weight, paroxysmal events, and motor and cognitive defects. Following the in vivo treatment, efficient correction of pathogenic alleles was observed in the cortex and hippocampus, marked by an increase in hippocampal Atp1a3 ATPase. Clinically, both mouse models displayed long-term improvements in survival, improvements in motor deficits, and reduced disease severity and frequency of convulsive episodes.

PE witnessed its clinical trial debut in May 2025 with a US biotechnology company, Prime Medicine, running a phase 1/2, multinational trial to test the safety, biological activity, and preliminary efficacy of PM359. This autologous ex vivo patient-derived haematopoietic stem cell was prime-edited to correct the NCF1 gene defect causing chronic granulomatous disease, a rare condition that disables a variety of immune cells, especially neutrophils. The (teenager) patient was administered a single intravenous dose of PM359, which was well tolerated a month after the dose, with no serious side-effects, and has been reported to have restored the function of a crucial enzyme in two-thirds of the neutrophils. Establishing cure is expected to take 6 months to 1 year.^12,13

Thus, the advantages of PE are evident (Table 1).⁹

Limitations of PE

Nevertheless, PE is still in its infancy, requiring substantial optimization before it can be implemented in vivo or deliver success with clinical trials. Apart from low editing efficiency compared to CRISPR-Cas9 and BE, one of the major drawbacks of PE is the size and complex structure of prime editors, which allow large-scale base insertions/substitutions up to 44 nucleotides and deletions up to 80 nucleotides.¹⁴ This limits the delivery efficiency and thereby the therapeutic effect.

Despite attempts to reduce the size of the prime editor, it was still found unsuitable for AAV delivery. One AAV vector (packing capacity ~4.7–5 kb) does not have the capacity to fit the large size of the editor protein and pegRNA complex (~6.3 kb).¹⁵

Modifications such as a stem-loop aptamer (split prime editor, sPE) to the 3' extension of pegRNAs have been shown to improve PE efficiency in mammalian cells.¹⁶ A 2022 study demonstrated successful editing of the β-catenin gene in the mouse liver (leading to tumour formation) and correcting a mutation in a mouse model of type I tyrosinaemia using a dual AAV vector system.¹⁶

Protospacer adjacent motif (PAM) requirements also present a constraint to PE efficiency, which has been attempted to be resolved using alternative Cas9 orthologs, e.g. SaCas9, cjCas9, and Cas12a,¹⁶ and creating PAM variants such as the PE2-SpRY variant that allows the targeting of a higher number of potential sites.¹⁷

Delivery mechanisms for prime editors

Various delivery mechanisms have been attempted, including physical (e.g. electroporation, microinjection of plasmid DNA, mRNA, gRNA, ribonucleoprotein), non-viral (lipid nanoparticle [LNP], virus-like particle, extracellular vesicles), and viral vectors (AAV, adenoviral vectors, lentiviral vectors, vexosomes). An optimal delivery vehicle must successfully navigate biological parameters such as pulse duration, possible inflammation from the delivery vehicle, degradation by nucleases, and several intra- and extracellular barriers. A failure in these increases the risk of toxicity within the cell/tissue. Applications of the physical methods have been restricted to ex vivo and in vitro in a limited number of cells. CRISPR-engineered chimeric antigen receptor– T/natural killer (CAR-T/NK) cell therapy has been shown to overcome these limitations.¹⁴ A 2020 study¹⁸ employed a ‘selective organ targeting’ in vivo approach that successfully used LNPs for the intravenous co-delivery of single-gRNA and Cas9 mRNA to the lungs, spleen, and liver of mice and demonstrated precise tissue-specific delivery and editing. Viral vectors have strong applications in vivo and are actively being studied in ongoing clinical trials. They provide major advantages over physical and non-viral vehicles: negligible/no pathogenicity toward the human host, low cytotoxicity and immunogenicity, compatibility with a wide range of cell types (muscles, heart, lungs, neurons), and the ability to modify non-dividing cells, thereby eliminating the need for a DNA donor template and DSBs. To prevent the AAV from inserting its genetic material into the host genome, recombinant AAVs are being engineered and studied. Nevertheless, several AAV-associated gene therapies for retinal diseases and Pompe disease have received FDA approval, which has enabled the expansion of CRISPR/Cas9–based research scope into blood disorders (SCD, beta-thalassaemia), liver ailments, muscular dystrophy, and neurodegenerative diseases. Triple AAV systems are currently being designed for genes such as those associated with DMD (11.1 kb) and Usher syndrome (10 kb) that exceed the capacity (9 kb) of dual AAV vectors.¹⁴

Non-medical constraints of PE

The current exorbitant costs of PE and gene editing in general, sometimes running into crores (millions) of rupees, prove to be prohibitive for the most vulnerable groups, for example, tribal populations in India. Of the estimated 300 million individuals living with rare diseases globally, almost 70 million are in India, primarily due to consanguinity and founder effects.¹⁹ The long time to diagnosis; dearth of awareness among medical professionals, carriers of genetic defects among the general population, and patients; and lack of viable treatment options and supportive care contribute to the burden on patients and their families.²⁰

The past few years in India have witnessed the establishment of National Guidelines for Gene Therapy and Product Development (2019), National Policy on Rare Diseases (2021), National Registry for Rare and other Inherited Disorders, indigenous generics for Gaucher disease and Wilson disease, and Centres of Excellence for rare diseases. The implementation of these policies, however, is irregular and suffers from multiple administrative loopholes. Expensive imported treatment modalities cannot be covered by the US $57 000 ( around `50 lakh) cap on financial assistance.^21,22 For example, CASGEVY^® costs more than US $2 million per dose.^23,24 The aforementioned Prime Medicine, a company with high economic means, has decided not to develop further the PM359 therapy on its own, but has called for financial investors.³

However, this also presents us with an opportunity to develop indigenous treatments through collaboration and government support. For example, the locally developed NexCAR19 (actalycabtagene autoleucel), a CAR-T cell therapy for specific cases of relapsed or refractory B cell malignancies, has reduced treatment costs by around 87% to US $45 000 (around `40 lakhs), with hopes to further reduce costs as the treatment is scaled.²⁵ Similarly, an indigenous gene therapy vector to correct the mutation causing SCD is expected to enter phase I clinical trial in 2025, developed under the Council of Scientific and Industrial Research (CSIR) with the support of the Ministry of Tribal Affairs (MoTA), Government of India.^26,27

It is worth noting that all genome editing treatments currently under development are in somatic cells. Human germline gene editing presents a myriad of regulatory, ethical, social, and scientific challenges, distinct from somatic cell editing.¹⁰

Outlook and future applications

PE is being further refined in terms of cargo size, optimal editing efficiency, pegRNA modifications to enhance precision, DNA-RNA substrate domain, delivery vehicles, and strategies to inhibit the mismatch repair pathway to prevent the cell from undoing the edit. Seven advancements in prime editors (Prime Editors 1-7) are currently being tested in vitro and in vivo.^2,7,16

While PE seems to be the future for precise, efficient, and safe gene editing, it will also be important to administer the most optimal gene editing approach based on considerations such as the underlying genetic defect, type of editing required, delivery mechanism for the cell/tissue, and the extent of genetic correction for maximized therapeutic effect. Ongoing research will pave the way for more advanced editing tools, such as DNA polymerase editors (DPEs) and click editors (CEs), both of which induce SSBs, thereby evading the risks associated with traditional CRISPR-based tools. DPEs use pegRNA-guided DNA-dependent polymerases to introduce edits, and CEs involve additional chemical modifications further to enhance the precision and efficiency of genome editing.¹⁶ The next big thing is the introduction of artificial intelligence and machine learning in the form of computational tools (such as DeepPE, PrimeDesign, and pegFinder) for CRISPR data analysis, refining target recognition, biomarker discovery and detection, and therapeutic optimization,¹⁴ which are poised to reduce the need for extensive empirical testing and simultaneously offer major advancements in personalized medicine.

As we navigate the complex and exciting world of gene editing, there is a growing need for the medical fraternity and regulatory bodies to stay abreast of recent developments; bridge the gap between scientists and communities with effective communication strategies including digital communication; advocate for patients, informed patient consent, fair accessibility, safety assurance, strong data privacy and cybersecurity policies; and facilitate the introduction of indigenous treatments at par with foreign counterparts.