On December 8, 2023, the FDA approved Casgevy — exagamglogene autotemcel, developed by Vertex Pharmaceuticals and CRISPR Therapeutics — for the treatment of sickle cell disease in patients aged 12 and older with recurrent vaso-occlusive crises. Six weeks later, the FDA approved it again, this time for transfusion-dependent beta-thalassemia. The approvals marked the first time a CRISPR-based therapy had been approved anywhere in the world by a major regulatory agency.

One month after the sickle cell approval, in January 2024, the FDA published its final guidance on Human Gene Therapy Products Incorporating Human Genome Editing. The timing was not coincidental. The guidance was written in the context of a field that had just crossed from experimental to approved — and it reflects what the FDA learned from reviewing Casgevy and the broader pipeline of genome editing programs that preceded it through IND and into clinical development.

Understanding what the guidance actually says, why it was written when it was, and what it signals about the regulatory trajectory of genome editing is more valuable than a section-by-section summary of its contents.

What Casgevy Actually Did

Casgevy’s mechanism is elegant and illustrative of how CRISPR-based editing can address inherited disease without correcting the underlying mutation directly. Sickle cell disease and beta-thalassemia are both caused by mutations in the beta-globin gene, which normally produces adult hemoglobin. Fetal hemoglobin — produced by a different gene and naturally suppressed after birth — is functional and can compensate for defective adult hemoglobin if its expression is restored.

The BCL11A gene encodes a transcriptional repressor that suppresses fetal hemoglobin production in adult cells. Casgevy uses CRISPR-Cas9 to edit a specific enhancer region of BCL11A in autologous hematopoietic stem cells, disrupting the repressor’s activity in erythroid cells and allowing fetal hemoglobin to be re-expressed. The edit is delivered ex vivo: a patient’s stem cells are collected by leukapheresis, electroporated with Cas9 protein and guide RNA as a ribonucleoprotein complex, and reinfused after myeloablative conditioning.

The clinical results from the CLIMB trials that supported approval were substantial. Final results published in the New England Journal of Medicine showed that 96.7 percent of sickle cell disease patients in the pivotal trial were free of vaso-occlusive crises for at least twelve months after treatment, and 100 percent were hospitalization-free for the same period. Longer-term follow-up data presented at ASH in December 2024 extended the observation window to more than five years, with median follow-up of 33.2 months for SCD patients and 38.1 months for TDT patients, continuing to demonstrate durable clinical benefit.

These results established that CRISPR-based ex vivo editing of hematopoietic stem cells could achieve durable, clinically meaningful outcomes in a human disease — and that the regulatory pathway, while demanding, was navigable.

What the FDA Learned and Codified

The January 2024 guidance reflects several specific lessons from the Casgevy development program and the broader early-phase genome editing clinical experience that the FDA reviewed in the years leading up to it.

The guidance’s emphasis on genome-wide off-target analysis reflects the FDA’s recognition that computational prediction of off-target editing sites is insufficient for regulatory purposes. During the review of early genome editing programs, the agency observed that candidate off-target sites identified by bioinformatics approaches did not always correlate with the sites that showed actual editing activity in empirical assays. The guidance’s recommendation for genome-wide methods — approaches that detect editing events across the entire genome without relying on predictions about where editing will occur — codifies this lesson as a standard expectation.

The guidance’s detailed treatment of on-target editing efficiency, therapeutic editing thresholds, and the distinction between confirming the edit and demonstrating the downstream biological effect reflects the FDA’s experience with potency assay development in gene editing programs. Early submissions for genome editing INDs frequently proposed potency assays based on editing efficiency alone — the percentage of cells in which the target sequence was modified. The agency’s position, now explicit in the guidance, is that editing efficiency is a necessary but not sufficient measure of potency. The downstream biological consequence — fetal hemoglobin expression in Casgevy’s case, corrected protein expression in a gene correction program, or tumor cell killing in an oncology program — is what must ultimately be measured to demonstrate that the product does what it is intended to do.

The guidance’s specific provisions for ex vivo versus in vivo programs reflect the FDA’s growing experience with both modalities in parallel. As Casgevy moved through development as an ex vivo program, other developers were pursuing in vivo CRISPR delivery via LNP and AAV. The regulatory challenges of these two approaches — ex vivo manufacturing control and cell characterization on one side, delivery vehicle biodistribution and in vivo editing characterization on the other — are sufficiently different that the guidance addresses them with distinct frameworks. The convergence of these two development tracks in the FDA’s review experience informed the guidance’s structure.

The guidance’s 15-year long-term follow-up recommendation reflects a straightforward regulatory logic about permanence: CRISPR editing introduces heritable changes to the genome of treated cells. Unlike a small molecule drug that is cleared from the body, or an mRNA therapy whose effects are transient, a genome edit in a hematopoietic stem cell persists in all the stem cell’s progeny for the life of the patient. The long-term safety profile of an edit — including the possibility that edited cells undergo clonal expansion driven by the genomic location of the edit — cannot be assessed on the timelines of typical Phase 3 clinical trials. The 15-year follow-up requirement creates a monitoring infrastructure that is proportionate to the permanence of the intervention.

What the Guidance Signals About the Field’s Trajectory

The 2024 genome editing guidance is not a restrictive document. It is a clarifying one. The FDA is not signaling that genome editing is uniquely dangerous or that the regulatory pathway is unusually difficult. It is signaling that the field has matured to the point where clear expectations are appropriate — and that programs which build their development strategies around those expectations are better positioned than those that treat guidance documents as aspirational rather than operational.

Several specific signals in the guidance are worth noting for their strategic implications.

The guidance’s acknowledgment that surrogate genome editing products — products in which human-specific elements are replaced with species-specific counterparts for use in animal models — are acceptable for nonclinical studies, with appropriate justification, reflects a pragmatic FDA position on the limitations of animal models for human-specific editing machinery. This acknowledgment reduces one of the practical barriers to nonclinical development for programs whose human guide RNAs do not cross-react with the animal model target sequence.

The guidance’s specific recommendation that first-in-human studies should enroll patients who have no other treatment options available — or that enrollment of patients with moderate disease should be justified — reflects the agency’s benefit-risk calculus for a technology whose long-term safety profile is still being established. This is not a permanent restriction on genome editing development. It is an appropriate early-stage position that will evolve as the long-term safety data from Casgevy and other approved programs accumulate.

The guidance’s treatment of accelerated approval for serious or life-threatening diseases — noting that surrogate or intermediate clinical endpoints that predict clinical benefit may be appropriate for accelerated approval — opens a pathway that is particularly relevant for rare diseases where conventional Phase 3 enrollment timelines are impractical and where the unmet need is severe. Programs targeting rare monogenic diseases with no existing treatment options are well-positioned to build accelerated approval strategies around editing efficiency, on-target correction rates, or biomarker endpoints that predict clinical response.

Where the Field Goes Next

Casgevy established that CRISPR-based therapy can achieve regulatory approval and durable clinical benefit in humans. The 2024 guidance established what the FDA expects from programs that follow. The pipeline of genome editing programs in clinical development — base editors, prime editors, in vivo LNP-delivered editing systems, multiplex editing for allogeneic cell therapy — will now navigate that guidance as a defined regulatory framework rather than a set of evolving expectations.

The most significant near-term regulatory questions in genome editing are not about whether CRISPR works. They are about the long-term durability of editing in different cell types, the safety signal that will or will not emerge from 15 years of follow-up in Casgevy patients, the sufficiency of current off-target characterization methods for detecting rare but consequential editing events, and whether the ex vivo manufacturing model that worked for hematopoietic stem cells can be extended to other cell types and indications at commercially sustainable cost.

These questions will be answered by the clinical programs now in development, and the data they generate will shape the next iteration of genome editing guidance. The FDA’s 2024 document is not the final word on genome editing regulation. It is the first comprehensive word, written in the context of the field’s first approved product. For developers building programs today, it is the starting point.

MKA Insights works with gene therapy and genome editing developers to build regulatory strategy, CMC frameworks, and development plans that reflect the current regulatory landscape and anticipate where it is heading. If your program is approaching IND or navigating the clinical development questions that genome editing raises, we bring the analytical depth and regulatory experience that these programs require.