Tag: Gene therapy

Gene editing therapies are among the most technically complex and scientifically promising treatments in modern medicine. The ability to make precise changes to the DNA of living cells has already produced approved therapies for conditions that were previously untreatable, including sickle cell disease, beta-thalassemia, and most recently the first gene therapy for genetic deafness. The pipeline is substantial and growing. But the path from a gene editing candidate to an approved therapy requires rigorous safety assessment, and one of the most important questions that must be answered before any gene editing therapy enters human clinical trials is: what happens when the editing tool goes somewhere it was not supposed to go?

On April 14, 2026, the FDA issued draft guidance specifically addressing how to answer that question. The document provides recommendations for using next-generation sequencing (NGS) methods in nonclinical studies to evaluate safety risks associated with off-target gene editing and loss of genomic integrity. It is a technical document aimed primarily at drug developers and researchers, but the questions it addresses are directly relevant to any patient or family considering a gene therapy clinical trial, and to anyone following the gene therapy field’s trajectory.

This post covers what off-target editing is and why it is a safety concern, how NGS is used to detect it, what the guidance specifically recommends, and why this particular regulatory step matters for the field.

The Problem the Guidance Is Solving: Off-Target Editing

To understand why this guidance exists, it helps to understand the specific risk it is designed to evaluate.

How gene editing works

Gene editing technologies, the most widely discussed being CRISPR-Cas9, work by directing a molecular complex to a specific sequence in the genome, where it makes a targeted cut or modification. In most therapeutic applications, the goal is to correct a harmful mutation, disrupt a disease-causing gene, or insert a therapeutic gene into a specific location.

The molecular machinery that performs this editing uses a guide sequence to find its target in the genome. The human genome contains roughly 3 billion base pairs. The guide sequence is designed to match a unique target, but no biological system is perfect. In some cases, the editing complex finds and modifies sites in the genome that are similar in sequence to the intended target but are not the target. These unintended modifications are called off-target edits.

Why off-target edits are a safety concern

The consequences of off-target edits depend entirely on where they occur in the genome. Many genomic locations are non-functional or contain genes with no role in cell survival or proliferation. An off-target edit in one of these locations may have no detectable consequence. But the genome also contains tumor suppressor genes, proto-oncogenes, and genes that regulate cell cycle progression. An off-target edit that disrupts a tumor suppressor or activates an oncogene could, in theory, initiate a process leading to cancer. This is not a theoretical concern invented by regulators: insertional mutagenesis, a related phenomenon in early viral gene therapy, caused leukemia in several patients in early trials in the 2000s, which fundamentally shaped how the field approaches vector safety.

A separate but related concern is chromosomal integrity. Gene editing tools make cuts in DNA. When the cellular repair machinery processes these cuts, it can sometimes cause larger structural changes: translocations (pieces of one chromosome joining to another), deletions spanning larger regions, or chromosomal rearrangements. These structural changes are assessed separately from single-site off-target edits and require different detection methods.

The FDA guidance addresses both categories of risk.

What Next-Generation Sequencing Is and Why It Is the Right Tool

Next-generation sequencing (NGS), also called high-throughput sequencing, refers to a family of technologies that can read millions or billions of short DNA sequences simultaneously. Unlike the original Sanger sequencing approach, which reads one sequence at a time, NGS generates massive parallel data that can characterize the entire genome of a sample at very high depth, meaning each region is read many times to detect even rare variants.

This depth of coverage is what makes NGS the right tool for detecting off-target editing. Off-target edits may occur in only a small fraction of cells in a treated sample, perhaps 0.1% or less of the total. Detecting these rare events requires reading each genomic region thousands of times to achieve sufficient statistical confidence that a signal is real rather than a sequencing error. The guidance specifically addresses the sequencing depth required for adequate detection of low-frequency off-target events.

NGS is also used for assessing chromosomal integrity. Whole genome sequencing and structural variant analysis can detect larger chromosomal rearrangements that would be missed by targeted approaches.

Short-read versus long-read sequencing

One of the more technically nuanced aspects of the guidance is its discussion of sequencing strategy. Not all off-target events are the same size:

Short stretches of DNA change at off-target sites (insertions, deletions, or base substitutions spanning a few to tens of base pairs) are well-characterized by short-read sequencing, where each read covers approximately 150 to 300 base pairs. This is the most widely used NGS approach.

Longer structural changes (larger deletions, translocations, inversions) may require long-read sequencing approaches, where individual reads span thousands to tens of thousands of base pairs, allowing the detection of events that short-read approaches might miss or mischaracterize.

The guidance advises sponsors to match their sequencing strategy to the type of event being evaluated, rather than applying a single approach to all safety questions. This is a scientifically rigorous position that acknowledges the genuine methodological trade-offs in the field.

What the Guidance Specifically Recommends

The draft guidance covers four main areas: sequencing strategies, sample selection, analysis parameters, and reporting.

Sequencing strategies

Sponsors should use sequencing approaches appropriate to the type of off-target event being assessed. For detection of small insertions and deletions (indels) at off-target sites, short-read approaches are generally appropriate. For detection of larger structural variants and chromosomal integrity assessment, long-read approaches or complementary methods such as optical genome mapping should be considered.

The guidance also addresses sequencing depth, recommending that sequencing be performed at a depth sufficient to detect off-target editing events that may occur at frequencies substantially lower than the on-target edit rate. Because off-target events are typically rare relative to on-target edits, inadequate sequencing depth can produce false-negative results that miss biologically relevant events.

Sample selection

The cells selected for safety assessment should reflect the actual therapeutic product. For ex vivo therapies (where cells are edited outside the body and then infused), the edited cell product itself is the appropriate test material. For in vivo therapies (where the editing tool is delivered directly into the patient), selecting appropriate tissue types for safety assessment is more complex and requires consideration of the delivery route and target tissues.

The guidance acknowledges that for individualized therapies, including personalized therapies being developed for patients with ultra-rare diseases where the specific mutation is unique to one individual, sample availability may be limited. It provides recommendations for how to approach safety assessment in these constrained scenarios.

Analysis parameters and bioinformatics

The guidance addresses how sponsors should approach the computational side of NGS analysis. Raw sequencing data must be processed through bioinformatics pipelines to identify candidate off-target sites, filter sequencing artifacts, and determine which signals represent genuine editing events. The document recommends that sponsors provide sufficient detail about their bioinformatics workflows to allow the FDA to evaluate the rigor of the analysis.

It also addresses how to identify candidate off-target sites to examine in the first place. Computational tools can predict likely off-target sites based on sequence similarity to the guide RNA target, and experimental methods such as GUIDE-seq and CIRCLE-seq can empirically identify editing sites in cell-based systems before sequencing. The guidance recommends using both approaches in combination.

Reporting

The guidance specifies what sponsors should include in their IND and BLA submissions regarding off-target safety assessment. This includes the complete list of candidate off-target sites evaluated, the sequencing methodology and depth, the bioinformatics pipeline used, the results at each evaluated site, and a risk assessment framework for interpreting any off-target events detected.

The Regulatory Context: Where This Guidance Fits

This is not the FDA’s first guidance document on gene editing safety. It builds directly on January 2024 guidance on human gene therapy products incorporating genome editing, which addressed broader nonclinical, clinical, and CMC considerations. The April 2026 draft guidance goes deeper specifically on the NGS methodology question, providing the technical detail that was implicit but not fully specified in the 2024 document.

It also relates to FDA’s February 2026 draft guidance supporting approval of ultra-rare disease therapies, which specifically addresses genome editing and RNA-based therapies including antisense oligonucleotides for conditions affecting so few patients that conventional randomized trial designs are not feasible. The NGS safety guidance applies in those individualized therapy contexts as well, and the February guidance specifically cited it.

The broader policy context is the current administration’s stated priority of accelerating gene therapy development. FDA Commissioner Marty Makary stated at the April 14 release that the guidance provides sponsors with clear, scientifically grounded recommendations for evaluating off-target editing risks using state-of-the-art sequencing technologies and that the agency is serious about moving this ball forward. CBER Director Vinay Prasad described the document as giving sponsors a roadmap for comprehensive safety assessment while supporting the efficient development of these promising therapies.

The practical significance is reduced regulatory uncertainty. Before standardized guidance existed, different sponsors might approach NGS-based off-target assessment very differently, leading to unpredictable FDA feedback and development delays. A clear framework means sponsors can design their safety assessment programs with confidence that the approach will be acceptable to regulators, potentially saving months of back-and-forth early in development.

Why This Matters for Patients and the Gene Therapy Field

Gene editing safety assessment is not a topic that patients following the field need to understand in technical detail. But the existence and quality of this guidance matters for several reasons that are directly relevant to anyone with a personal stake in gene therapy development.

Faster paths to clinical trials. The guidance is specifically designed to help sponsors design adequate nonclinical studies so that IND applications can move forward without extended regulatory delays. For a patient with a genetic disease watching a promising therapy move through development, regulatory efficiency at the nonclinical stage is a meaningful factor in how quickly human trials begin.

Individualized therapies for ultra-rare diseases. The guidance explicitly addresses scenarios where standard approaches cannot be fully applied because the patient population is too small to generate conventional safety datasets. This is directly relevant to the growing number of individualized gene therapy programs, some designed for single patients, where regulatory flexibility and clear scientific standards are both necessary.

The off-target safety question is real. For anyone following the first-in-class gene therapy approvals, including Casgevy (exagamglogene autotemcel) for sickle cell disease and Otarmeni for genetic deafness (covered in our post on the first gene therapy for deafness), understanding that rigorous off-target safety assessment underlies every approved gene editing therapy is reassuring context for both patients and families. This guidance represents the standardization of that rigor across the field.

Transparency through public comment. As a draft guidance, this document is open for public comment through July 14, 2026. Comments can be submitted via Regulations.gov using docket number FDA-2026-D-1255. Academic researchers, patient advocacy organizations, and industry sponsors are all invited to provide feedback that will inform the final guidance. Organizations like the Alliance for Regenerative Medicine and the American Society of Gene and Cell Therapy (ASGCT) will likely submit formal comments representing the field’s collective perspective.

What This Guidance Does Not Do

Clarity on scope matters. This guidance does not:

• Approve any gene editing therapy or change the status of any existing approved therapy
• Replace the 2024 genome editing guidance, which it supplements rather than supersedes
• Address clinical study design, patient safety monitoring during trials, or post-approval safety requirements
• Apply to non-genome editing gene therapies (such as AAV gene replacement without editing) except where editing tools are used
• Establish a lower bar for approval; it specifies what evidence is needed, not a reduced standard

The guidance is specifically about the nonclinical safety assessment phase: the studies done before human trials begin. Clinical trial safety monitoring, informed consent, adverse event reporting, and post-approval pharmacovigilance are governed by separate frameworks.

Are you a researcher, sponsor, or patient advocate who wants to comment on the draft guidance?

The comment period closes July 14, 2026. Comments can be submitted electronically at Regulations.gov, docket FDA-2026-D-1255. The full draft guidance document is available at FDA.gov. The FDA also encourages sponsors to engage early through INTERACT meetings and pre-IND meetings to discuss specific development strategies before formal submission.

For patients and families following gene therapy development, the National Human Genome Research Institute, the American Society of Gene and Cell Therapy, and the Alliance for Regenerative Medicine maintain current information on approved and investigational gene editing therapies.

Sources

FDA press announcement: FDA Issues Draft Guidance on Genome Editing Safety Standards to Advance Gene Therapy Development. FDA.gov. April 14, 2026.

Draft guidance document: Safety Assessment of Genome Editing in Human Gene Therapy Products Using Next-Generation Sequencing; Draft Guidance for Industry. FDA.gov.

Federal Register docket: FDA-2026-D-1255. Safety Assessment of Genome Editing in Human Gene Therapy Products Using Next-Generation Sequencing. Federal Register. April 15, 2026.

RAPS coverage: FDA drafts guidance on using next-generation sequencing to assess gene therapy safety. raps.org. April 2026.

BioSpace coverage: FDA bolsters bespoke therapy framework with new draft safety guidelines. biospace.com. April 2026.

Clinical Trials Arena: FDA shares guide on genome editing best practices. clinicaltrialsarena.com. April 2026.

European Pharmaceutical Review: New FDA draft guidance to enhance safety of genome editing therapies. europeanpharmaceuticalreview.com. April 2026.

January 2024 predecessor guidance: Human Gene Therapy Products Incorporating Human Genome Editing. FDA.gov. January 2024.

February 2026 ultra-rare disease guidance: Considerations for the Development of Individualized Antisense Oligonucleotide and Genome Editing Therapies. FDA.gov. February 2026.

Comment submission: Regulations.gov docket FDA-2026-D-1255.

Patient and researcher resources: National Human Genome Research Institute: Gene Therapy | American Society of Gene and Cell Therapy | Alliance for Regenerative Medicine | FDA INTERACT meetings