CRISPR vs Base Editing vs Prime Editing: A Head-to-Head Comparison

Three Generations of Precision

Gene editing has evolved rapidly since CRISPR-Cas9 first demonstrated that targeted DNA cuts were possible. Today, scientists have three major editing strategies at their disposal, each with distinct mechanisms, capabilities, and trade-offs. Understanding the differences between CRISPR-Cas9, base editing, and prime editing is essential for anyone following the field — whether in the clinic, the lab, or the market.

CRISPR-Cas9: The Original Molecular Scissors

How It Works

CRISPR-Cas9 uses a guide RNA to direct the Cas9 nuclease to a specific genomic location, where it creates a double-strand break (DSB). The cell then repairs the break through one of two pathways: non-homologous end joining (NHEJ), which typically introduces insertions or deletions that disrupt the gene, or homology-directed repair (HDR), which uses a supplied DNA template to make a precise edit.

Strengths

• Gene knockout efficiency: NHEJ-mediated disruption is highly efficient, making Cas9 the tool of choice for disabling genes.
• Large modifications: Cas9 can facilitate large insertions, deletions, and even chromosomal rearrangements.
• Well-characterized: With over a decade of use, Cas9 has the deepest body of literature, validated protocols, and commercial support.

Limitations

• Double-strand breaks are risky: DSBs can trigger unintended large deletions, translocations, and chromothripsis (catastrophic chromosome shattering) at low but measurable frequencies.
• HDR is inefficient: Precise insertions via HDR work well only in dividing cells, limiting applications in post-mitotic tissues like the brain, heart, and muscle.
• Indel variability: NHEJ repair outcomes are stochastic, making it difficult to control the exact edit produced.

Base Editing: Chemistry Without Cutting

The Invention

In 2016, David Liu's laboratory at the Broad Institute introduced base editors — a fundamentally different approach that modifies individual DNA bases without creating double-strand breaks. The insight was to fuse a catalytically impaired Cas9 (called a nickase, which cuts only one DNA strand) to a deaminase enzyme that chemically converts one base into another.

Cytosine Base Editors (CBEs)

CBEs convert cytosine (C) to uracil (U), which the cell's replication machinery then reads as thymine (T). The net result is a C-to-T (or equivalently, G-to-A on the opposite strand) conversion. A uracil glycosylase inhibitor (UGI) is included to prevent the cell from repairing the uracil back to cytosine before replication occurs.

Adenine Base Editors (ABEs)

ABEs convert adenine (A) to inosine (I), which is read as guanine (G). This yields an A-to-G (or T-to-C on the opposite strand) conversion. Remarkably, the engineered adenosine deaminase used in ABEs (TadA*) does not exist in nature — Liu's team evolved it in the laboratory through directed evolution.

Strengths

• No double-strand breaks: Base editors nick only one strand, dramatically reducing the risk of large deletions, translocations, and chromothripsis.
• High efficiency in non-dividing cells: Because base editing does not rely on HDR, it works efficiently in post-mitotic tissues.
• Clean outcomes: The product is predominantly the intended single-nucleotide change, with far less indel formation than Cas9.

Limitations

• Limited edit types: CBEs can only make C-to-T changes; ABEs can only make A-to-G changes. Together they cover four of the twelve possible single-nucleotide transitions and none of the transversions.
• Bystander editing: If multiple target bases fall within the editing window (typically positions 4-8 of the protospacer), all of them may be edited, not just the intended one.
• Off-target RNA editing: Some base editor variants can cause transcriptome-wide off-target deamination, though engineered versions have largely mitigated this.

Prime Editing: Search-and-Replace for DNA

The Mechanism

In 2019, David Liu's lab unveiled prime editing, which they described as a "search-and-replace" tool for the genome. Prime editors use a Cas9 nickase fused to a reverse transcriptase enzyme. The guide RNA — called a prime editing guide RNA (pegRNA) — contains both a targeting sequence and a template encoding the desired edit.

Here is how it works step by step:

1. The pegRNA directs the prime editor to the target site.
2. The Cas9 nickase cuts only one strand of the DNA.
3. The reverse transcriptase uses the template portion of the pegRNA to synthesize new DNA directly at the nick site.
4. The cell incorporates the newly synthesized DNA, replacing the original sequence.

Strengths

• All twelve point mutations: Prime editing can install any single-nucleotide change — all four transitions and all eight transversions.
• Small insertions and deletions: It can also make precise small insertions (up to ~40 base pairs) and deletions (up to ~80 base pairs) without double-strand breaks.
• Minimal byproducts: Because it does not rely on DSB repair pathways, prime editing produces very few indels at the target site.
• No donor DNA template required: The edit information is encoded in the pegRNA itself.

Limitations

• Lower efficiency: Prime editing is generally less efficient than either Cas9-mediated knockout or base editing, though optimization is ongoing. PE4 and PE5 strategies, which transiently inhibit mismatch repair, have substantially improved rates.
• Larger cargo size: The prime editor protein plus pegRNA is larger than standard Cas9, creating challenges for delivery via adeno-associated virus (AAV) vectors.
• pegRNA design complexity: Designing effective pegRNAs requires careful optimization of scaffold, template length, and primer binding site.

Head-to-Head Comparison

Feature	CRISPR-Cas9	Base Editing	Prime Editing
Mechanism	Double-strand break	Chemical base conversion	Reverse transcription at nick
DSB required	Yes	No	No
Edit types	Knockouts, large indels, insertions via HDR	C-to-T, A-to-G only	All 12 point mutations, small indels
Efficiency	High (knockout); low (HDR)	High	Moderate (improving)
Indel byproducts	Common	Low	Very low
Works in non-dividing cells	Knockout yes; HDR no	Yes	Yes
Off-target DSBs	Possible	Minimal	Minimal
Delivery challenge	Moderate	Moderate	High (large cargo)
Maturity	Most validated	Clinical trials underway	Preclinical; advancing rapidly

When to Use Each Tool

Choose CRISPR-Cas9 when:

• The goal is to knock out a gene entirely.
• Large insertions or deletions are needed.
• Maximum editing efficiency is critical and DSB risks are acceptable.
• Working with well-established protocols in model organisms.

Choose base editing when:

• A single C-to-T or A-to-G change will correct the disease-causing mutation.
• The target tissue is non-dividing (e.g., liver, neurons).
• Minimizing DSB-associated risks is a priority.
• High on-target efficiency is required.

Choose prime editing when:

• The needed edit is a transversion, small insertion, or small deletion that base editors cannot make.
• Precision with minimal byproducts is the top priority.
• The application can tolerate somewhat lower efficiency.
• The goal is to model or correct mutations that require exact sequence specification.

Clinical Landscape

Base editing is furthest along in the clinic among the newer tools. Verve Therapeutics has advanced a base editing therapy targeting PCSK9 for cardiovascular disease, and Beam Therapeutics is running trials for sickle cell disease and acute leukemia using base-edited cells.

Prime editing remains predominantly in the preclinical stage, though Prime Medicine is advancing candidates toward clinical trials for chronic granulomatous disease and other indications.

CRISPR-Cas9 retains the deepest clinical footprint, with Casgevy already approved and dozens of additional trials underway across oncology, hematology, and rare diseases.

The Bigger Picture

These three tools are not competitors so much as complementary layers in an expanding toolkit. The optimal choice depends on the specific biological question, the target tissue, the nature of the mutation, and the acceptable risk profile. As delivery methods improve and editing efficiencies increase, the boundaries between what each tool can accomplish will continue to blur — but understanding their fundamental differences remains essential for anyone working at the frontier of genetic medicine.

Sources & Further Reading

• Komor, A.C. et al. "Programmable editing of a target base in genomic DNA without double-stranded DNA cleavage." Nature 533, 420–424 (2016). — First base editor paper.
• Anzalone, A.V. et al. "Search-and-replace genome editing without double-strand breaks or donor DNA." Nature 576, 149–157 (2019). — First prime editor paper.
• Liu, D.R. "2025 Breakthrough Prize Lecture: Base Editing and Prime Editing." Broad Institute (2025).
• As of 2026, at least 23 clinical trials use base editing or prime editing (source: Broad Institute).

Last updated: March 2026.