CRISPR Gene Editing in 2026 — What's Been Cured, What's Coming, What's Real vs Hype

Photo: Google DeepMind / Pexels
DNA sequences in living cells with letter-level accuracy.">CRISPR-Cas9 is a molecular scissors system that can cut and rewrite DNA with precision no prior tool achieved. In December 2023, the FDA approved Casgevy — the first CRISPR-based cure — for sickle cell disease and beta-thalassemia. As of 2026, several more therapies are in late-stage trials. The technology is real, the first cures exist, but significant limits in cost, delivery, and germline safety remain.
Image: CRISPR-Cas9 mechanism — the Cas9 protein (blue-grey) is directed to a specific DNA sequence by a guide RNA (orange), where it makes a precise double-strand break (red).
What CRISPR Really Is
CRISPR — Clustered Regularly Interspaced Short Palindromic Repeats — is a natural immune memory system found in bacteria. Bacteria use it to store fragments of viral DNA so they can recognize and destroy that virus on re-infection. Scientists Jennifer Doudna (UC Berkeley) and Emmanuelle Charpentier (Max Planck Institute) repurposed this system into a programmable gene-editing tool, work for which they received the 2020 Nobel Prize in Chemistry.
Cas9 — CRISPR-associated protein 9 — is the molecular scissors. It is an endonuclease, meaning it cuts both strands of a DNA double helix at a precise location.
Guide RNA (gRNA) — a short synthetic RNA sequence, approximately 20 nucleotides long, that acts as the address label. The gRNA is designed to match the target DNA sequence in the patient's genome. The Cas9 protein follows the gRNA to the correct location and makes the cut.
The mechanism works in three steps. First, the researcher or clinician designs a guide RNA that matches the target gene. Second, the Cas9-gRNA complex is introduced into the patient's cells — either directly in the body (in vivo) or in cells removed from the patient and edited in the laboratory (ex vivo). Third, after the cut is made, the cell's natural repair machinery either disables the gene (via a messy repair called NHEJ) or replaces it with a corrected sequence (via homology-directed repair, HDR).
Somatic editing refers to changes made to non-reproductive cells — blood, liver, muscle. These edits affect only the treated individual and are not inherited. All currently approved CRISPR therapies are somatic.
Germline editing refers to edits made to embryos, eggs, or sperm, meaning the change would be inherited by all future generations of that person's descendants. Germline editing in humans remains banned or subject to severe moratorium in the United States, European Union, United Kingdom, and China following the 2018 He Jiankui incident (discussed in section five).
This distinction matters enormously. Somatic editing is medicine. Germline editing is the territory where ethics, evolution, and regulation intersect — and where no government currently permits clinical use.
Understanding the full genetic basis of this technology requires understanding DNA's foundational structure, which was decoded by James Watson and Francis Crick in 1953 building on X-ray crystallography data produced by Rosalind Franklin. CRISPR would be impossible without the double-helix model they revealed. The discovery that genes can jump and rewrite themselves — the conceptual precursor to intentional gene editing — was made by Barbara McClintock, whose Nobel-winning work on transposable elements in maize now reads as the earliest chapter of this story.
The 2023–2026 Cures: Casgevy and Beyond
The first CRISPR therapy approved by a regulatory body anywhere in the world was Casgevy (exa-cel, or exagamglogene autotemcel), developed jointly by Vertex Pharmaceuticals and CRISPR Therapeutics. The FDA approved it on December 8, 2023, for sickle cell disease (SCD) and transfusion-dependent beta-thalassemia (TDT). The UK's MHRA approved it on November 16, 2023 — beating the FDA by three weeks.
How Casgevy Works
Sickle cell disease and beta-thalassemia are both caused by defects in the genes encoding adult hemoglobin. Casgevy does not repair those defective genes directly. Instead, it exploits a developmental switch. Humans produce fetal hemoglobin (HbF) in the womb, then switch to adult hemoglobin after birth. A gene called BCL11A acts as the "off switch" for fetal hemoglobin. Casgevy uses CRISPR-Cas9 to disrupt BCL11A in the patient's own hematopoietic stem cells, reactivating fetal hemoglobin production at levels high enough to compensate for the defective adult hemoglobin.
The process is not a simple injection. It is a months-long ordeal. Stem cells are harvested from the patient's blood over several sessions. The cells are edited ex vivo in the laboratory. The patient then undergoes myeloablative chemotherapy — the same bone-marrow-destroying regimen used in leukemia treatment — to clear space in the marrow. The edited cells are then infused back. The patient typically spends four to six weeks in hospital during engraftment.
Clinical Results
In the pivotal trial published in the New England Journal of Medicine (Frangoul et al., 2021; updated 2023), 29 of 29 sickle cell disease patients who received Casgevy and had at least 12 months of follow-up were free of severe vaso-occlusive crises for at least 12 consecutive months — a 100% response rate in the trial cohort. For beta-thalassemia patients, 39 of 42 (93%) were transfusion-free at 12-month follow-up (Locatelli et al., 2023, NEJM). Fetal hemoglobin levels reached 40% or higher in treated patients — well above the approximately 20% threshold associated with clinical protection in sickle cell disease.
The word "cure" is used cautiously by regulators. Long-term durability beyond five years is not yet established. However, the trial data available as of 2026 continues to show sustained fetal hemoglobin elevation and maintained efficacy with no new safety signals.
Image: The Casgevy treatment timeline for a sickle cell patient — from stem cell collection through engraftment to the 12-month efficacy assessment. The full process takes approximately six to eight months from start to hospital discharge.
Approved and Late-Stage CRISPR Therapies as of 2026
| Therapy | Disease | Developer | Approval Year | List Price (USD) | Mechanism |
|---|---|---|---|---|---|
| Casgevy (exa-cel) | Sickle cell disease | Vertex / CRISPR Therapeutics | 2023 | $2,200,000 | BCL11A disruption → HbF reactivation |
| Casgevy (exa-cel) | Beta-thalassemia | Vertex / CRISPR Therapeutics | 2023 | $2,200,000 | BCL11A disruption → HbF reactivation |
| Lyfgenia (lovo-cel) | Sickle cell disease | bluebird bio | 2023 | $3,100,000 | Lentiviral vector (non-CRISPR; gene addition) |
| NTLA-2001 | Transthyretin amyloidosis | Intellia / Regeneron | Phase III (2025–26) | Not yet set | In vivo Cas9, liver-targeted LNP delivery |
| CTX001 successor | Acute myeloid leukemia | CRISPR Therapeutics | Phase II | Not yet set | CAR-T enhancement via CRISPR |
| Beam-101 | Sickle cell disease | Beam Therapeutics | Phase I/II | Not yet set | Adenine base editing (A-to-G) at HBG promoter |
Table 1: FDA-approved and late-stage CRISPR-based therapies as of May 2026. Lyfgenia is included for comparison as a competing gene therapy for the same indication. Sources: FDA, NEJM, company disclosures.
The most significant pipeline development as of 2026 is Intellia's NTLA-2001 for transthyretin amyloidosis — the first in vivo CRISPR therapy to advance into Phase III. Unlike Casgevy's ex vivo approach, NTLA-2001 delivers Cas9 directly into the patient's bloodstream via lipid nanoparticles (LNPs) targeted to the liver. Phase I data (Gillmore et al., NEJM, 2021) showed up to 87% reduction in serum transthyretin protein in a single dose. If Phase III confirms this, in vivo delivery will open CRISPR to diseases that cannot be treated by the ex vivo stem cell approach.
Beyond Cas9: Base Editing, Prime Editing, and the New Toolkit
The original CRISPR-Cas9 system introduced double-strand breaks. Those breaks are powerful but imprecise in repair — the cell's own machinery can introduce errors. A new generation of CRISPR variants operates without cutting both strands, dramatically improving precision.
Base editing — developed by David Liu's laboratory at the Broad Institute in 2016 — uses a modified, partially-inactivated Cas9 (called a nickase) fused to a chemical deaminase enzyme. Instead of cutting DNA and letting the cell repair it, base editors directly convert one DNA base letter into another. Adenine base editors (ABEs) convert A•T base pairs to G•C. Cytosine base editors (CBEs) convert C•G to T•A. Because approximately 30,000 human genetic disease variants are single-nucleotide changes (point mutations), base editing addresses a large class of genetic diseases with higher precision and lower off-target risk than Cas9 (Anzalone et al., Nature, 2020).
Prime editing — also from the Liu laboratory, introduced in 2019 — is described as a "search and replace" function. It uses a Cas9 nickase fused to a reverse transcriptase enzyme and a "prime editing guide RNA" (pegRNA) that contains both the address and the new sequence to be inserted. Prime editing can make all 12 possible point mutation conversions plus small insertions and deletions, without requiring double-strand breaks or donor DNA templates. It is more versatile than base editing but currently less efficient in vivo (Anzalone et al., Nature, 2019).
Cas12a (Cpf1) cuts DNA with staggered rather than blunt ends, uses a smaller guide RNA, and is more specific in some off-target analyses. It is particularly suited to multiplexed editing — cutting several genes simultaneously from a single guide RNA array.
Cas13 targets RNA rather than DNA, making it suited for transient knockdown of gene expression rather than permanent edits. SHERLOCK (Specific High Sensitivity Enzymatic Reporter UnLOCKing), developed by the Broad Institute and published in Science (Gootenberg et al., 2017), uses Cas13 for rapid, field-deployable diagnostic detection of viral RNA — including SARS-CoV-2.
CRISPR Variant Comparison
| System | Cut Type | Key Advantage | Primary Limitation | Stage (2026) | Best Applications |
|---|---|---|---|---|---|
| Cas9 (SpCas9) | Double-strand break | Well-characterized, versatile | Off-target risk, repair unpredictability | Approved (Casgevy) | Gene disruption, ex vivo editing |
| Cas12a (Cpf1) | Staggered DSB | Multiplexed guide RNA, higher specificity | Narrower PAM requirements | Phase I/II | CAR-T, multi-gene editing |
| Cas13 | RNA cleavage only | Reversible; diagnostic use | No permanent DNA edit | Diagnostic; preclinical therapeutic | Viral diagnostics, RNA knockdown |
| Adenine base editor (ABE) | Nick (single strand) | Precise A→G conversion, no DSB | Only two conversion types | Phase I/II (Beam-101) | Point mutation diseases (SCD, PKU) |
| Cytosine base editor (CBE) | Nick (single strand) | Precise C→T conversion, no DSB | Bystander editing risk | Phase I | Hypercholesterolemia, rare diseases |
| Prime editing (PE3) | Nick (single strand) | All 12 conversions + indels, no donor needed | Lower efficiency in vivo | Preclinical / early Phase I | Complex mutations, large corrections |
Table 2: Comparison of major CRISPR systems as of 2026. DSB = double-strand break. PAM = protospacer adjacent motif. Sources: Anzalone et al. Nature 2020; Liu laboratory publications; ClinicalTrials.gov.
The progression from Cas9 to base editing to prime editing represents a clear trajectory: more precision, less collateral DNA damage, broader mutation coverage. However, each new system also requires new delivery engineering, new regulatory data, and new manufacturing processes. Prime editing, despite its theoretical elegance, faces real efficiency gaps in tissue types outside the liver and blood — the two compartments where delivery solutions (LNPs and viral vectors, respectively) are most mature.
You can read more about the AI systems now being used to design guide RNAs and predict off-target effects in our article on AI and CRISPR genetic engineering with machine learning.
The He Jiankui Case and Where Germline Editing Stands
In November 2018, Chinese biophysicist He Jiankui announced at the Second International Summit on Human Genome Editing in Hong Kong that he had used CRISPR-Cas9 to edit human embryos that were subsequently implanted and carried to term. Two girls — referred to publicly as Lulu and Nana — had been born with edited CCR5 genes, a co-receptor that HIV uses to enter immune cells. A third edited child was born in 2019. He claimed the goal was to confer HIV resistance.
The announcement was met with near-universal condemnation from the scientific community. The problems were multiple and serious.
The CCR5 deletion He introduced is not a guaranteed protection against HIV — some HIV strains use a different co-receptor (CXCR4) and are unaffected by CCR5 disruption. The children were not at exceptional HIV risk. No institutional review board approved the experiment. The informed consent process for the parents was later determined to be inadequate. Off-target edits in the germline cannot be reversed or corrected in subsequent generations. He Jiankui was convicted by a Chinese court in December 2019 and sentenced to three years in prison and a three million yuan fine.
The scientific stakes are distinct from the legal ones. The children carry edits that, if they reproduce, will pass to their descendants permanently. The long-term health consequences — positive or negative — of the CCR5 deletion in a germline context are unknown. A 2019 analysis by Xinzhu Wei and Rasmus Nielsen published in Nature Medicine found that individuals naturally homozygous for the CCR5-delta32 deletion (the deletion He introduced) had a 21% higher mortality rate in a UK Biobank cohort — suggesting the mutation has off-target organismal effects beyond HIV susceptibility.
As of 2026, the international regulatory landscape remains one of effective prohibition for clinical germline editing. The WHO's Expert Advisory Committee on Developing Global Standards for Governance and Oversight of Human Genome Editing published its framework in 2021 calling for a global registry of any human genome editing research and recommending that no germline clinical applications proceed until safety and efficacy standards can be established. No country has licensed a germline editing clinical trial since He Jiankui's experiment.
The distinction is frequently misunderstood by the public and in media coverage: the CRISPR therapies in clinical use today — Casgevy, NTLA-2001, Beam-101 — are all somatic therapies. They do not alter the germline. They are not "designer babies." They are targeted medicines for specific diseases in individual patients.
What CRISPR Still Cannot Do
CRISPR's media coverage frequently outruns its actual capabilities. Understanding the real limitations is essential for calibrating both expectations and investment decisions.
Polygenic diseases — conditions driven by thousands of genetic variants each contributing a small effect — cannot currently be addressed by CRISPR. Height, intelligence, most psychiatric disorders, most forms of heart disease, and Type 2 diabetes are polygenic. Editing one or two genes does nothing useful for these conditions. Casgevy works because sickle cell disease is caused by a single-gene defect; most common diseases are not single-gene.
Delivery to most tissues remains unsolved at scale. Liver cells and blood stem cells are the two compartments where reliable delivery exists. Muscle (relevant for Duchenne muscular dystrophy), the brain (relevant for Huntington's, ALS, and neurological diseases), and the lung (relevant for cystic fibrosis) all require delivery methods — viral vectors, lipid nanoparticles, or novel carriers — that remain in active development. Adeno-associated viral (AAV) vectors can reach muscle and brain but carry immunogenicity risk, limited cargo capacity (approximately 4.7 kb, too small for some editing machinery), and concerns about insertional mutagenesis at high doses.
Off-target editing is reduced but not eliminated by any existing system. Base editing can cause bystander edits at neighboring bases. Cas9 can cut at sequences with partial homology to the guide RNA. Prime editing has lower off-target risk but is also less efficient. Every clinical program must characterize off-target profiles comprehensively, and this analytical burden adds years and cost to development timelines.
The mosaicism problem arises in embryo editing: because cells divide very rapidly after fertilization, edits made at the one-cell stage may not propagate uniformly, resulting in an organism where some cells carry the edit and others do not. This problem contributed to uncertainty in He Jiankui's experiment and remains a fundamental challenge for any germline application.
Immune responses to Cas9 protein are common in humans. A 2018 study in Nature Medicine (Charlesworth et al.) found pre-existing antibodies to SpCas9 (derived from Streptococcus pyogenes) and SaCas9 (from Staphylococcus aureus) in the majority of blood donors tested. This means in vivo delivery of Cas9 may trigger immune clearance of the editing machinery before it can act, or — more dangerously — an inflammatory response. Engineering humanized Cas9 variants and managing immunogenicity is an active area of research.
Cost, Access, and Equity
Casgevy costs $2.2 million per patient in the United States. Lyfgenia, the competing non-CRISPR gene therapy for sickle cell disease, costs $3.1 million. These figures make both therapies effectively inaccessible to the vast majority of patients with sickle cell disease worldwide.
This matters at scale because sickle cell disease disproportionately affects populations of sub-Saharan African, South Asian, Middle Eastern, and Mediterranean ancestry. According to the WHO, approximately 300,000 infants are born with sickle cell disease annually worldwide, with over 75% in sub-Saharan Africa. The geography of disease burden and the geography of ability to pay for $2 million therapies are almost perfectly non-overlapping.
In the United States, the Institute for Clinical and Economic Review (ICER) published an assessment in 2023 estimating that Casgevy would be cost-effective at approximately $1.85 million, slightly below its list price, given the long-term cost of chronic sickle cell disease management (hospitalizations, transfusions, pain management). However, the upfront nature of the cost, the lack of installment payment mechanisms, and Medicaid reimbursement complexities mean that even in the U.S., access has been limited in the first two years post-approval. As of early 2026, fewer than 100 patients had received Casgevy globally, despite the therapy having been available for over two years (Vertex earnings calls, Q1 2026).
Manufacturing complexity is part of the cost problem. Because Casgevy is an ex vivo autologous therapy — meaning each patient's own cells must be collected, shipped to a central manufacturing facility, edited, quality-tested, and shipped back — there is no economy of scale. Every dose is unique to its patient. In vivo therapies like NTLA-2001, if approved, would be manufactured as a standardized product and could in principle be distributed and administered at far lower per-patient cost.
Global access programs, tiered pricing agreements, and generic or biosimilar pathways for gene therapies remain largely theoretical as of 2026. Several academic groups and the Medicines Patent Pool (an organization backed by the UN) are in early discussions about licensing lower-cost CRISPR manufacturing to middle-income country manufacturers, but no formal agreement has been executed.
The equity gap is not a niche concern. It is arguably the central challenge of the gene therapy era. A technology that works but reaches only the wealthiest 0.1% of those who need it is not yet a public health intervention. For further context on how AI is intersecting with access to personalized medicine more broadly, see our article on AI and personalized medicine with genomics.
The Next Five Years: Cautious Predictions
Based on the peer-reviewed literature, regulatory pipeline, and clinical trial data available as of May 2026, several developments are likely — and several claims circulating in mainstream media are not.
What is likely by 2031:
In vivo CRISPR therapy approval is the most probable near-term milestone. Intellia's NTLA-2001 for transthyretin amyloidosis is in Phase III with strong Phase I/II data. If the trial completes on schedule (anticipated 2027 readout), FDA approval could follow in 2027 or 2028. This would open the door to in vivo editing of the liver for other targets, including PCSK9 (a gene whose disruption dramatically lowers LDL cholesterol) and HAO1 (relevant to primary hyperoxaluria).
Base editing therapies for sickle cell disease and other hemoglobinopathies will produce Phase II/III data. Beam-101's mechanism — base editing rather than gene disruption — may offer a cleaner profile than Casgevy for some patient populations.
CRISPR-enhanced CAR-T therapies for hematological cancers are likely to receive initial approvals. Several programs use CRISPR to make CAR-T cells from healthy donors (allogeneic) rather than from the patient, potentially reducing manufacturing time and cost substantially.
What is unlikely by 2031:
Germline editing for disease prevention will not receive regulatory approval in any major jurisdiction. The scientific, ethical, and regulatory consensus against clinical germline editing remains strong. Any researcher who claims otherwise is either misreading the regulatory environment or proposing illegal research.
"Designer babies" — editing embryos for non-disease traits like intelligence, height, or athletic ability — remain in the realm of science fiction in any responsible clinical context. The polygenic nature of virtually all complex traits makes meaningful editing toward such outcomes both scientifically impossible with current tools and categorically prohibited by law in every major jurisdiction.
CRISPR-based cures for Alzheimer's, Type 2 diabetes, or schizophrenia are not coming by 2031. These conditions have complex, partially understood, polygenic, and environmental etiologies that current CRISPR tools cannot address.
Cost trajectory: Manufacturing costs for autologous ex vivo therapies are unlikely to fall dramatically by 2031 given the fundamental constraint that each dose is patient-specific. In vivo therapies with standardized manufacturing are more likely to achieve meaningful price reduction. Tiered pricing in middle-income countries remains possible but is dependent on political will, patent agreements, and regulatory pathways that have not yet been established.
For the broader genomics landscape that CRISPR operates within, see our article on DNA replication and genetics explained and our overview of epigenetics and how environment changes your genes.
Frequently Asked Questions
Q: Is CRISPR gene editing safe? A: For the approved somatic therapies — Casgevy for sickle cell disease and beta-thalassemia — the safety profile through available follow-up (two-plus years post-approval) is acceptable, with no treatment-related deaths reported in pivotal trials. The major safety burden is the myeloablative chemotherapy required before cell infusion, not the CRISPR editing itself. Off-target editing has been characterized in each approved program and found at low levels. Longer-term data beyond five to ten years is not yet available for any CRISPR therapy. (Frangoul et al., NEJM, 2021; FDA prescribing information, 2023)
Q: Can I get CRISPR therapy right now? A: If you have sickle cell disease or transfusion-dependent beta-thalassemia and are in the United States, United Kingdom, or EU, Casgevy is an approved option. However, treatment is only available at a small number of authorized treatment centers, the process requires months of preparation and weeks of hospitalization, and the list price of $2.2 million means access is heavily dependent on insurance coverage and payer agreements. For conditions other than these two, no CRISPR therapy is currently approved. Patients with other diseases may be eligible for clinical trials; ClinicalTrials.gov lists active CRISPR studies.
Q: What about designer babies? Can CRISPR be used to enhance normal traits? A: Not responsibly, not legally, and not practically. Germline editing — changes to embryos that would be inherited — is prohibited in all major regulatory jurisdictions. Additionally, traits like intelligence, athleticism, and height are polygenic, influenced by thousands of genetic variants plus environment. CRISPR cannot currently edit thousands of variants simultaneously with the required precision, and the polygenic architecture means editing any subset of variants would have unpredictable effects. The He Jiankui case (2018) demonstrated the consequences for researchers who ignore this regulatory and scientific consensus.
Q: How is CRISPR different from older gene therapy? A: Earlier gene therapy approaches primarily used viral vectors to add a working copy of a defective gene alongside the existing broken one — a gene addition strategy. These approaches do not correct the underlying mutation; they compensate for it. CRISPR offers targeted correction of the mutation itself, disruption of a harmful gene, or activation of a compensatory gene. CRISPR is also programmable — the same molecular machinery can be redirected to a new target by simply changing the guide RNA sequence — whereas viral vector delivery systems must be redesigned for each new application. (Anzalone et al., Nature Biotechnology, 2020)
Q: What diseases will CRISPR cure next, after sickle cell? A: Based on current trial data, the most likely next approvals involve transthyretin amyloidosis (Intellia's NTLA-2001, Phase III), hemoglobinopathies via base editing (Beam-101), and CRISPR-enhanced CAR-T therapies for blood cancers. Duchenne muscular dystrophy (DMD) is a longer-term target with active preclinical and early clinical work, but delivery to muscle at therapeutic scale remains a significant unsolved problem. Liver-targeted in vivo therapies are the near-term frontier.
Q: Does CRISPR editing last forever? A: For somatic ex vivo therapies like Casgevy, the edited cells are hematopoietic stem cells — self-renewing cells that replenish the blood for life. If the edited stem cells engraft successfully, the effect should be lifelong. Durability data in the Casgevy trials shows maintained fetal hemoglobin levels through follow-up periods of up to four years in the earliest-treated patients, with no evidence of loss of effect. However, true lifelong durability data does not yet exist and will take decades to accumulate. (Locatelli et al., NEJM, 2023)
Q: What is the difference between base editing and prime editing? A: Base editing directly converts one DNA base letter to another (A-to-G or C-to-T) at a target site, without creating a double-strand break, using a chemically modified Cas9 fused to a deaminase enzyme. It is highly efficient and increasingly precise. Prime editing uses a Cas9 nickase fused to a reverse transcriptase and a specially designed guide RNA (pegRNA) that carries the new sequence to be written. Prime editing can make all possible single-base changes plus small insertions and deletions — effectively a "find and replace" for DNA — but is currently less efficient in many tissues than base editing. Base editing is further ahead in the clinic; prime editing is still largely in preclinical stages. (Liu laboratory, Nature, 2016–2023)
Q: How does CRISPR relate to the discovery of DNA structure? A: CRISPR is built entirely on the molecular biology that followed the discovery of the DNA double helix. Without Rosalind Franklin's X-ray diffraction data and the double-helix model of Watson and Crick, there would be no framework for understanding sequence-specific molecular recognition — the principle that allows a guide RNA to find its exact target in three billion base pairs. Barbara McClintock's discovery that genes can move and rearrange within a genome was the first evidence that the genome is not static — a conceptual foundation without which programmable gene editing would have been unimaginable.
Sources
Frangoul, H., et al. (2021). "CRISPR-Cas9 Gene Editing for Sickle Cell Disease and beta-Thalassemia." New England Journal of Medicine, 384(3), 252–260. doi:10.1056/NEJMoa2031054
Locatelli, F., et al. (2023). "Exagamglogene Autotemcel for Transfusion-Dependent beta-Thalassemia." New England Journal of Medicine, 390(18), 1663–1676. doi:10.1056/NEJMoa2218935
Gillmore, J.D., et al. (2021). "CRISPR-Cas9 In Vivo Gene Editing for Transthyretin Amyloidosis." New England Journal of Medicine, 385(6), 493–502. doi:10.1056/NEJMoa2107454
Anzalone, A.V., et al. (2020). "Genome editing with CRISPR–Cas nucleases, base editors, transposases and prime editors." Nature Biotechnology, 38(7), 824–844. doi:10.1038/s41587-020-0561-9
Anzalone, A.V., et al. (2019). "Search-and-replace genome editing without double-strand breaks or donor DNA." Nature, 576, 149–157. doi:10.1038/s41586-019-1711-4
Charlesworth, C.T., et al. (2018). "Identification of preexisting adaptive immunity to Cas9 proteins in humans." Nature Medicine, 25, 249–254. doi:10.1038/s41591-018-0326-x
Wei, X., & Nielsen, R. (2019). "CCR5-∆32 is deleterious in the homozygous state in humans." Nature Medicine, 25, 909–910. doi:10.1038/s41591-019-0459-6
Gootenberg, J.S., et al. (2017). "Nucleic acid detection with CRISPR-Cas13a/C2c2." Science, 356(6336), 438–442. doi:10.1126/science.aam9321
U.S. Food and Drug Administration. (2023). "FDA Approves First Gene Therapies to Treat Patients with Sickle Cell Disease." FDA press release, December 8, 2023. fda.gov
WHO Expert Advisory Committee on Developing Global Standards for Governance and Oversight of Human Genome Editing. (2021). Human Genome Editing: A Framework for Governance. World Health Organization. who.int
Institute for Clinical and Economic Review (ICER). (2023). "Sickle Cell Disease: Effectiveness and Value." Evidence Report. icer.org
Doudna, J.A., & Charpentier, E. (2012). "A programmable dual-RNA–guided DNA endonuclease in adaptive bacterial immunity." Science, 337(6096), 816–821. doi:10.1126/science.1225829
Follow new articles
Add SciHub101 to your feed reader to see new science and technology stories.
About the Author
SciHub101 is an independent science and technology publication founded and edited by Buğra Sözeri, a software engineer and app developer based in Türkiye. AI tools assist with drafting and research. Source documentation in the existing archive is uneven, and articles are being reviewed and corrected. The Editorial Team byline identifies editorial responsibility; it does not certify specialist or independent human review. Where we compare products, anything built by our own team is labelled as such — see the editorial guidelines and About page for how corrections and disclosures work.
Related Articles

Why Do We Forget Things — The Neuroscience of Forgetting Explained
The neuroscience of forgetting: encoding failure, decay, interference, motivated forgetting, sleep's role in memory consolidation, and why forgetting is adaptive.

Alzheimer's Disease in 2026 — What We Know, Can Treat, and Still Can't
Leqembi and Kisunla slow Alzheimer's progression by 27–35%. Here's what science finally knows about amyloid, tau, blood biomarkers, and the 14 modifiable risk factors.

The CRISPR Gene Editing Breakthrough
CRISPR turned a bacterial immune system into a gene-editing tool. This article explains the discovery, mechanism, clinical results, and ethical stakes.

CRISPR 3.0: The Gene-Editing Revolution That's Quietly Changing Medicine
In December 2023, the FDA approved Casgevy — the first CRISPR-based therapy for a human disease. It works in a single treatment session: a patient's stem cells are removed, their DNA edited, and the corrected cells returned. The results in sickle cell disease trials were striking. But Casgevy uses first-generation CRISPR, which cuts DNA like scissors. A newer class of tools — base editing and prime editing — can rewrite individual letters of the genetic code without cutting the double helix at all, dramatically reducing the risk of unintended edits. Clinical trials of these next-generation systems began in 2024. The field is not waiting for the conversation to catch up.

CRISPR Gene Editing: How It Works in 2026
From bacterial immune system to approved therapies: how CRISPR-Cas9, base editing, and prime editing are rewriting the genetic code of medicine.

AI Meets CRISPR: How Machine Learning Is Supercharging Genetic Engineering
Machine learning is making CRISPR more precise and predictable, designing better guide RNAs and anticipating off-target edits before they happen.