CRISPR-Cas9 has moved from a curiosity in bacterial immunity to an FDA-approved platform therapy in roughly a decade. The first drug, Casgevy, reached patients in December 2023. Base editing and prime editing — second- and third-generation variants that avoid the blunt instrument of double-strand breaks — are now entering human trials. AI tools are designing entirely novel gene-editing enzymes that have no counterpart in nature. And one company, Intellia Therapeutics, placed its lead Phase 3 programme on clinical hold in late 2025 after a serious liver adverse event, re-opening questions about safety that the He Jiankui scandal of 2018 never fully closed.
This review organises the available primary evidence under five headings: what CRISPR is, the current state of the industry, emerging technical breakthroughs, critical failures and controversies, and market sentiment. Every factual claim is linked to a primary or institutional source. No generic AI-generated summaries; every data point is traceable.
1. What Is CRISPR? Mechanism and Origins
CRISPR stands for Clustered Regularly Interspaced Short Palindromic Repeats — a defence system discovered in bacteria and archaea that stores fragments of invading viral DNA. In its therapeutic form, CRISPR has been repurposed into a precision gene-editing tool using the Cas9 (CRISPR-associated protein 9) endonuclease guided by a short synthetic RNA sequence.
How CRISPR-Cas9 edits DNA — four sequential steps
Guide RNA design. A synthetic 20-nucleotide guide RNA (sgRNA) is designed to match the genomic target sequence, complementary to one strand of the DNA double helix.
PAM recognition. The Cas9 protein scans the genome for a protospacer adjacent motif (PAM) — typically NGG in the most common SpCas9 system. No PAM, no cleavage.
Double-strand break (DSB). Once bound, Cas9 generates a blunt-ended DSB three base pairs upstream of the PAM sequence, severing both DNA strands.
DNA repair. The cell repairs the break via NHEJ (error-prone: insertions/deletions that disrupt gene function) or HDR (precise: template-guided correction, enabling specific sequence changes).
The 2020 Nobel Prize in Chemistry was awarded jointly to Jennifer Doudna (University of California, Berkeley) and Emmanuelle Charpentier (Max Planck Unit for the Science of Pathogens) for the development of CRISPR-Cas9 as a genome editing tool — recognition of the foundational 2012 Science paper demonstrating that Cas9 could be directed to any genomic sequence by redesigning the sgRNA.
The therapeutic logic is straightforward: if a disease is caused by a known genetic variant, CRISPR can in principle correct that variant in the patient's own cells. In practice, the challenge lies in delivery — getting the Cas9 and guide RNA to the right cells, in sufficient quantity, without triggering an immune response, editing off-target sites, or causing double-strand breaks that initiate malignant transformation.
Key terminology
- Ex vivo editing: Patient cells are removed, edited in the laboratory, and reinfused. Casgevy uses this approach — haematopoietic stem cells are extracted, CRISPR-edited to reactivate fetal haemoglobin, and transplanted back.
- In vivo editing: The CRISPR machinery is delivered directly into the patient's body — typically via lipid nanoparticles (LNPs) or adeno-associated virus (AAV) vectors. Intellia's NTLA-2001 uses LNP delivery to liver hepatocytes.
- HDR (Homology-Directed Repair): Precise sequence correction using a supplied DNA template. Requires cell division; efficient in haematopoietic cells, difficult in post-mitotic tissue (neurons, cardiomyocytes).
- NHEJ (Non-Homologous End Joining): Error-prone repair that creates small insertions or deletions (indels), useful when the goal is gene disruption rather than correction.
2. Current State of the Industry
The most important milestone in the history of therapeutic CRISPR arrived on December 8, 2023, when the US Food and Drug Administration approved Casgevy (exagamglogene autotemcel, exa-cel) for sickle cell disease (SCD) in patients aged 12 and older.[1] A second FDA approval for transfusion-dependent beta-thalassemia (TDT) followed in January 2024. Casgevy was developed jointly by CRISPR Therapeutics and Vertex Pharmaceuticals.
Mechanism of action. Casgevy edits patients' own haematopoietic stem cells to reactivate BCL11A, a gene whose natural function is to suppress fetal haemoglobin (HbF) after birth. By disrupting BCL11A in erythroid precursors, the therapy reinduces HbF expression. In SCD patients, HbF prevents haemoglobin S polymerisation — the proximate cause of painful vaso-occlusive crises. In TDT patients, HbF compensates for deficient adult haemoglobin.
Eligible population. Approximately 16,000 patients in the United States are estimated to be eligible for Casgevy across both indications.[2]
Commercial traction. As of Q1 2026, approximately 500 patients had been treated with Casgevy globally. CRISPR Therapeutics reported $170 million in total revenue from Casgevy since commercialisation commenced, with $43 million in Q1 2026 alone.[1] Revenue is shared with Vertex; Vertex books the product revenue and pays CRISPR Therapeutics tiered royalties.
List price. Casgevy's US list price is approximately $2.2 million per patient — reflecting the one-time, potentially curative nature of the treatment and the high manufacturing cost of personalised cell therapy.
Beyond Casgevy, the CRISPR pipeline is broad. A 2025 update from the Innovative Genomics Institute (IGI, UC Berkeley) documented more than 50 CRISPR-based clinical trials actively recruiting globally as of mid-2026, spanning haematological disease, oncology (CAR-T cell engineering, tumour microenvironment modulation), ophthalmology, and metabolic conditions including transthyretin (TTR) amyloidosis.[5]
Regulatory framework
CRISPR therapies that modify the human germline (heritable edits in eggs, sperm, or embryos) remain prohibited in clinical practice in the United States, the EU, and the UK, governed by a combination of FDA regulation, ICH guidelines, and the Declaration of Helsinki. Somatic cell editing — modification of non-reproductive cells — is subject to the same IND/BLA regulatory pathway as any biological therapy, with additional FDA guidance on long-term safety follow-up and insertional mutagenesis monitoring.
3. Emerging Technical Breakthroughs
Standard CRISPR-Cas9 creates a double-strand break (DSB) — a blunt cut through both DNA strands. While effective, DSBs carry inherent risk: they activate DNA damage response pathways, can trigger large chromosomal deletions or translocations at the cut site, and rely on error-prone NHEJ repair when precise HDR correction is not the goal. Three classes of next-generation technology are now moving into or through early clinical testing.
Base Editing — Chemical Rewriting Without a Break
Developed principally by David Liu (Broad Institute), base editing uses a catalytically disabled Cas9 (nickase or dead Cas9) fused to a deaminase enzyme. The system converts one DNA base to another — adenine to guanine (ABEs) or cytosine to thymine (CBEs) — without creating a DSB. This eliminates large indels at the on-target site and reduces DSB-associated mutagenesis risk. Liu was awarded the 2025 Breakthrough Prize in Life Sciences specifically for base editing and prime editing.[7]
Beam Therapeutics (BEAM) is the lead clinical-stage company using base editing. Its stock rose 53% between April 17 and early July 2026.[12]
Prime Editing — The Most Versatile Editor
Also from the Liu laboratory, prime editing uses a Cas9 nickase fused to a reverse transcriptase and a prime editing guide RNA (pegRNA) that carries both the targeting sequence and the desired edit as an RNA template. Prime editing can install all 12 types of point mutation, small insertions, and small deletions without a DSB and without requiring a separate HDR template.
In 2026, Prime Medicine reported first-in-human data for PM359, a prime editing therapy for chronic granulomatous disease (CGD), showing safety and early evidence of efficacy — the first time prime editing has demonstrated activity in a human patient.[7] Error rates have improved from approximately 1 in 7 edits containing an unintended change (early systems) to 1 in 101 in current optimised platforms (MIT data).[7] A further system, proPE, has demonstrated a 6.2× efficiency increase over prior prime editing variants.
Source: crisprmedicinenews.com; Innovative Genomics Institute
AI-Designed Nucleases — Beyond Nature's Toolkit
Two landmark papers in 2025–2026 demonstrated that artificial intelligence can design functional gene-editing enzymes with no natural counterpart. OpenCRISPR-1, described in Nature (2025), was the first AI-generated CRISPR-Cas protein to successfully edit human DNA — and achieved sharply reduced off-target effects compared with SpCas9.[9]
In July 2026, a team led by Jennifer Doudna at the Innovative Genomics Institute published in Science describing AI-designed synthetic TnpB-family nucleases (SynTnpBs). These synthetic enzymes, smaller than Cas9, matched or exceeded the activity of natural reference enzymes and may be especially valuable in delivery contexts where payload size is limiting — including certain plant gene editing applications and potentially in vivo somatic cell therapy.[8]
Sources: Phys.org, July 2026; Nature, 2025
In Vivo LNP Delivery — Editing Inside the Body
All first-generation CRISPR therapies required cells to be removed from the patient, edited in the laboratory, and reinfused (ex vivo). In vivo delivery — administering the CRISPR machinery directly into the bloodstream — is more scalable and avoids the toxic conditioning regimen required to prepare bone marrow for a cell transplant.
Intellia Therapeutics pioneered LNP-mediated in vivo CRISPR delivery with NTLA-2001 (nex-z), targeting transthyretin (TTR) in hereditary ATTR amyloidosis. Phase 1 data showed mean TTR reductions of 87% in the higher-dose cohort after a single infusion — a depth of suppression comparable to established RNAi therapies (patisiran, inclisiran).[6] Unlike RNAi, which requires repeated dosing, a single in vivo CRISPR edit is intended to be permanent. This Phase 1 result provided the first proof-of-concept that in vivo CRISPR gene editing is both feasible and clinically meaningful in humans.
4. Critical Failures and Controversies
CRISPR's rapid clinical translation has been accompanied by three categories of serious concern: a landmark ethical violation involving germline editing, evidence of off-target mutagenesis in the cells of patients from that violation, and a 2025 clinical hold on a Phase 3 programme following a serious adverse hepatic event.
The He Jiankui Affair — The First CRISPR Babies
🚨 Regulatory violation · Ethical breach · Criminal prosecution
In November 2018, Chinese biophysicist He Jiankui announced — at the International Summit on Human Genome Editing in Hong Kong — that he had used CRISPR-Cas9 to edit the CCR5 gene in human embryos, resulting in the birth of twin girls and a subsequent third child. The stated rationale was to confer resistance to HIV infection by disrupting CCR5, the co-receptor exploited by most HIV strains.
The announcement triggered immediate international condemnation from virtually every major scientific body, including the National Academies of Sciences, Engineering, and Medicine; the Wellcome Trust; the Royal Society; and the Chinese Academy of Sciences. He Jiankui had proceeded without adequate oversight, without genuine informed consent from the parents (who were recruited through an HIV patient advocacy group under misleading terms), and without the experimental or therapeutic justification required to expose healthy embryos — and future children — to heritable genome edits.[10]
He Jiankui was convicted in December 2019 by a Chinese court and sentenced to three years in prison and a 3 million yuan fine for illegal medical practice. Two collaborators received shorter sentences. He was released in 2022.
Off-Target Effects in the He Jiankui Cohort
Beyond the ethical violations, the scientific record now carries specific evidence that the CCR5 edits performed in 2018 produced off-target genomic damage. In 2024, Swiss researchers documented that the CCR5 disruption may have introduced off-target mutations at sites associated with increased cancer risk — a finding consistent with earlier modelling showing that CCR5-disrupted cells may have a proliferative advantage that could accelerate pre-existing malignant clones.[10] These mutations will persist in the affected children throughout their lives and may be transmitted to their offspring — the heritable consequence that made germline editing so categorically different from somatic gene therapy.
Intellia's Phase 3 Clinical Hold — October 2025
⚠️ FDA Clinical Hold · Grade 4 Liver Adverse Event · Phase 3 Paused
On October 27, 2025, Intellia Therapeutics disclosed that an 80-year-old participant in the NTLA-2001 (nex-z) trial had developed Grade 4 liver transaminase elevation and increased total bilirubin, requiring hospitalisation. The FDA subsequently placed a clinical hold on NTLA-2001 and on both Phase 3 trials — MAGNITUDE (650+ patients) and MAGNITUDE-2 (47 patients) — pending investigation.[11]
The event re-opened questions about hepatotoxicity from LNP delivery at scale. LNPs have known tropism for the liver (explaining why NTLA-2001 works for TTR amyloidosis — TTR is produced in liver hepatocytes), but LNP-associated hepatotoxicity has been a recognised risk since early pharmacology studies. The severity in this patient — Grade 4, meaning clinically serious — was unexpected given the Phase 1/2 safety profile.
Critically, some analysts and clinicians questioned whether the adverse event was attributable to the LNP vehicle, the Cas9 protein (which carries immunogenicity risk in patients with prior Cas9 exposure — a risk present in virtually all adults who have had natural bacterial infections), or the patient's pre-existing hepatic function.[4]
The Enduring Off-Target Problem
Even in rigorously conducted somatic CRISPR programmes, off-target editing — Cas9 cleaving at genomic sites other than the intended target — remains a quantifiable risk. Standard CRISPR-Cas9 generates off-target cuts at sites that share partial sequence homology with the guide RNA. These cuts are typically detected with GUIDE-seq, CIRCLE-seq, or integrated digital error suppression sequencing (iDES-seq) in cell-line and animal models, but translating that detection sensitivity to primary patient cells and to post-infusion monitoring remains technically challenging. Ongoing FDA guidance requires long-term safety follow-up (15 years in some trial designs) to detect delayed adverse events, including insertional oncogenesis.
5. Market Sentiment — 2026 Equity Landscape
The CRISPR equity sector peaked in early 2021, driven by optimism following the Casgevy regulatory filing, and has since undergone significant consolidation and repricing. As of July 2026, the sector is bifurcated: CRISPR Therapeutics, with commercial revenue from Casgevy, has stabilised with analyst support; Intellia, Editas, and Beam trade at varying discounts to their late-2020 highs reflecting trial timelines, dilution from equity issuances, and — for Intellia — the October 2025 clinical hold.
| Company / Ticker | Technology | Key Data Point (2025–2026) | Analyst View |
|---|---|---|---|
| CRISPR Therapeutics (CRSP) | CRISPR-Cas9 (ex vivo & in vivo) | ~$44 share price; $170M cumulative Casgevy revenue; $43M Q1 2026; cash $423M; total assets $2.7B | Median PT $77 (64% upside); 14 of 39 analysts = Buy; Morningstar fair value $106; projected net income 2029 |
| Beam Therapeutics (BEAM) | Base editing | Stock up 53% April 17 to early July 2026; pipeline in haematology and liver disease | Sector momentum driven by PM359 prime editing data and base editing validation |
| Intellia Therapeutics (NTLA) | In vivo LNP-CRISPR | Phase 3 on clinical hold since Oct 2025; $180M secondary offering April 2026; cash through 2028; BLA timeline deferred | Binary event risk; de-risking dependent on FDA hold resolution; strong Phase 1 data (87% TTR reduction) provides fundamental support |
| Editas Medicine (EDIT) | CRISPR-Cas9 & Cas12a (ex vivo) | FY2025 revenue $40.5M (+25.4% YoY); net loss $160.1M; lower P/S ratio vs CRSP | Analysts flag greater share price upside potential but ongoing operating losses; pipeline restructuring ongoing |
| Prime Medicine (PRME) | Prime editing | PM359 first human prime editing data (CGD) — safety + early efficacy, mid-2026 | Catalytic moment for prime editing as a modality; PM359 data seen as proof-of-concept milestone for entire class |
Sources consulted for market data include Motley Fool's gene-editing sector analysis (May 2026); Public.com CRSP analyst consensus; and GEN StockWatch, July 2026.[12][13][14]
What the Evidence Agrees On — Across All Five Sections
- CRISPR-Cas9 has produced an FDA-approved, commercially available therapy (Casgevy) that achieves durable functional haemoglobin improvement in SCD and TDT — the first gene editing therapy of any type to reach approved clinical use at scale.
- Next-generation editors (base editing, prime editing) systematically reduce the double-strand break risk that underlies the most serious safety concern about first-generation CRISPR; their arrival in Phase 1/2 trials is the most significant technical development since Cas9 itself.
- AI-designed nucleases are no longer theoretical — two independently published systems (OpenCRISPR-1, SynTnpBs) edit human DNA with efficacy matching or exceeding natural enzymes, with superior off-target profiles in published assays.
- In vivo delivery via LNP is clinically validated as a concept (Intellia's 87% TTR reduction) but carries a hepatotoxicity signal that the October 2025 clinical hold has made undeniable — the field must now solve the safety-versus-efficacy balance for direct systemic delivery before in vivo CRISPR can scale to mass-market indications.
- Market capitalisation in the sector reflects genuine uncertainty about timelines and safety, not scepticism about the underlying science — analysts' price targets imply 40–75% upside for the leading names from current levels, contingent on clinical hold resolution and commercial Casgevy ramp.
- The He Jiankui case remains the regulatory and ethical anchor point for all human germline CRISPR discussions; his prosecution and the documented off-target effects in the resulting children are the clearest empirical evidence that oversight infrastructure — not scientific capability — is the rate-limiting factor for responsible clinical translation.
Implications for Clinical Trial Teams
For oncology and haematology trial sites operating under ICH GCP E6(R3), the CRISPR landscape generates several near-term protocol questions.
CAR-T eligibility and prior CRISPR exposure. Allogeneic ("off-the-shelf") CAR-T products increasingly use CRISPR to knock out TCR and MHC genes, enabling donor T-cells to be used without causing graft-versus-host disease. Eligibility criteria for CRISPR-edited CAR-T trials must now specify prior CRISPR or Cas9 exposure — because prior exposure can generate anti-Cas9 antibodies that suppress editing efficiency or trigger immune-mediated adverse events in re-exposed patients.
Long-term safety follow-up obligations. FDA's 2023 guidance on long-term follow-up after administration of gene therapy products (which encompasses CRISPR) requires a minimum 15-year post-treatment observation period for products that introduce genetic changes that may persist or replicate. Trial sites must have data management infrastructure capable of supporting follow-up that outlasts most standard trial contracts.
Companion diagnostics for CRISPR eligibility. Like checkpoint inhibitor trials, emerging CRISPR trials — particularly those targeting specific mutant alleles in haematological malignancies — will increasingly require pre-treatment molecular characterisation to confirm the target variant is present. The same CDx infrastructure challenges documented for PD-L1 assays (see our CDx Debates article) will apply to CRISPR eligibility testing: assay choice, specimen type, and result turnaround time will all become protocol-critical variables.
References
- CRISPR Therapeutics / Vertex Pharmaceuticals. "Casgevy Q1 2026 commercial update." CRISPR Therapeutics Investor Relations, 2026. ir.crisprtx.com Commercial
- National Institutes of Health / PMC. "Casgevy: first CRISPR-based gene editing therapy approved for sickle cell disease." PMC11305803. pmc.ncbi.nlm.nih.gov/articles/PMC11305803 PMC
- US Food and Drug Administration. "FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease." FDA.gov, December 8, 2023. fda.gov press announcement FDA
- Inside Precision Medicine. "Is Intellia's Patient Death the Nail in the Coffin for Cas9?" November 2025. insideprecisionmedicine.com Clinical
- Innovative Genomics Institute. "CRISPR Clinical Trials: A 2025 Update." University of California, Berkeley, 2025. innovativegenomics.org IGI
- Intellia Therapeutics / Regeneron. "Landmark Clinical Data Showing Deep Reduction in Disease-Causing Protein After Single Infusion of NTLA-2001." Intellia IR press release. ir.intelliatx.com Clinical
- CRISPR Medicine News / The Scientist. "Prime Medicine PM359 first-in-human data; base and prime editing David Liu 2025 Breakthrough Prize." 2025–2026. crisprmedicinenews.com; the-scientist.com Review
- Phys.org / Science. "AI-designed gene-editing enzymes expand the CRISPR toolbox." July 2026. Doudna / IGI SynTnpBs publication. phys.org, July 2026 Science 2026
- Nature / GEN Engineering News. "AI expands the repertoire of CRISPR-associated proteins for genome editing." OpenCRISPR-1 publication, 2025. nature.com/articles/d41586-025-02135-3; genengnews.com Nature 2025
- Innovative Genomics Institute. "CRISPR Ethics — He Jiankui." IGI CRISPRpedia. innovativegenomics.org/crisprpedia/crispr-ethics; off-target documentation: 2024 Swiss researcher findings cited in science.org coverage. Ethics
- CGTlive. "Intellia Puts Phase 3 Trials for Transthyretin Amyloidosis Gene Editing Therapy Nex-Z on Hold Following Grade 4 Liver AE." October 2025. cgtlive.com Clinical Hold
- The Motley Fool. "Better Gene-Editing Stock: CRISPR Therapeutics or Beam Therapeutics?" May 25, 2026. fool.com, May 2026; "4 Best CRISPR Companies for 2026." fool.com/crispr-companies Market
- Public.com. "CRISPR Therapeutics (CRSP) Analyst Price Target 2026." 39 analysts; median $77. public.com/stocks/crsp/forecast-price-target Market
- GEN Engineering News. "StockWatch: After 1- and 6-Month Gains, Gene Editing Companies Await Q3 Results." July 2026. genengnews.com Market
- Nature. "CRISPR gets a power boost from AI-designed 'molecular scissors'." 2026. nature.com/articles/d41586-026-02217-w; C&EN. "AI-designed nucleases build on nature's design." July 2026. cen.acs.org Nature 2026
Disclaimer: This article is for informational and educational purposes only. It does not constitute medical, clinical, or investment advice. Market data is sourced from publicly available analyst consensus and investor reports. Clinical and regulatory developments cited reflect information available as of July 18, 2026. Readers should consult qualified professionals for medical decisions and licensed financial advisors for investment decisions.