FDA draft guidance ushers in a new approach: scalable long-read sequencing for cell and gene therapy QC

Why is scalable long-read sequencing biopharma’s new priority?

Cell and gene therapies hold extraordinary promise for tackling previously untreatable diseases. However, the same tools that enable these breakthroughs, such as adeno-associated virus (AAV) vectors, CRISPR-Cas9, and other genome-editing methods, also carry risk.

For example, CRISPR-Cas9 genome editing can unintentionally generate large structural variants (SVs), including insertions and deletions ≥50 bp, at both on-target¹ and off-target sites and such mutations can segregate across generations². Unintended changes to a patient’s genome could have significant clinical consequences, which is why regulators are sharpening their focus on genomic integrity.

Recently, the U.S. Food and Drug Administration (FDA) published draft guidance on the safety assessment of genome editing in human gene therapy products using next-generation sequencing (NGS)³. For the first time, the agency explicitly discusses the use of long-read sequencing for insertion site analysis (ISA) to investigate complex on- and off-target editing events.

This draft guidance signals an evolution in regulatory expectations. Where short-read sequencing was once considered adequate, the FDA now calls out scenarios where short reads are not enough: detecting SVs such as large insertions, deletions, or chromosomal rearrangements. In simple terms, the era of partial insight is over. Regulators want sponsors (organisations that initiate and oversee the development of a new medicine) to show the full picture, and long reads are the way to deliver it.

What does the FDA draft guidance actually say?

The FDA guidance spells out the expectation for a ‘fit-for-purpose’ sequencing strategy for genome-edited therapies. Small changes, like single-base edits, can still be assessed using short-read methods. But for edits that introduce larger genomic changes, the FDA notes that long sequencing reads may be needed.

When the editing method causes double-strand breaks in DNA or involves any mechanism that could produce large or complex changes, the guidance recommends applying a long-read sequencing strategy to the on-target edit site. The rationale is straightforward: long reads, which can span thousands to even millions of bases in a single read, allow scientists to capture entire regions of interest along with their surrounding genomic context without the reconstruction that is needed with short reads. That’s what’s needed to detect large insertions, deletions, tandem duplications, or translocations — events that short fragments of a few hundred bases struggle to reveal.

This recommendation has strong implications for regulatory submissions such as investigational new drug (IND) applications and Biologics License Applications (BLAs)⁴. Sponsors will need to demonstrate that their sequencing strategy can support a robust safety narrative. For many, this means upgrading from fragmented approaches to something more comprehensive, future-ready, and scalable.

Why aren’t short reads enough for cell and gene therapy safety checks?

Legacy short-read sequencing methods face inherent limitations. When you’re dealing with reads of just 50–300 bp, reconstructing long stretches of DNA becomes an assembly puzzle prone to ambiguity, especially in repetitive regions.

This gap matters because unintended consequences of genome editing rarely limit themselves to neat boundaries. For example, research has shown that CRISPR-induced off-target mutations can arise in difficult-to-map ‘dark’ genomic regions, where short reads often fail to provide a coherent picture⁵. Amplification-free long-read sequencing uncovered previously unforeseen CRISPR-Cas9 off-target activity.

These findings and the FDA’s emphasis on long reads reflect a simple truth: piecemeal data can conceal risks. Without long sequencing reads that traverse entire loci and neighbouring regions, sponsors run the risk of overlooking integrity issues that could derail regulatory progress, or, worse, compromise patient safety.

How can Oxford Nanopore sequencing streamline ISA and quality control of cell and gene therapies?

ISA is a critical step in ensuring therapeutic integrity. The fundamental question for sponsors remains: has your genome edit happened exactly where you intended, without introducing harmful surprises? As discussed above, legacy short-read sequencing can struggle to answer this, whereas long reads may confirm the success and safety of your edit.

Oxford Nanopore technology delivers long reads that span the entire integration site and its genomic context in one go. This direct view reduces assembly gaps and reveals the full architecture of an integration event. Complex outcomes — whether it’s concatemerised transgene insertions, unwanted plasmid backbone integrations, or truncated or rearranged transgene insertions — can be resolved, alongside any large SVs flanking the site. This level of clarity is particularly critical in applications like chimeric antigen receptor T (CAR-T) cell therapy, where confirming that a lentiviral vector has not landed in a disruptive region of the genome is a key part of the safety assessment.

Beyond structural integrity, Oxford Nanopore sequencing adds another dimension with epigenetic insights. Our platform directly measures native DNA methylation without bisulfite conversion, providing a comprehensive view of the chromatin context around an edit site. Researchers have used targeted PCR-free nanopore sequencing to examine how CRISPR-Cas9-induced double-strand breaks influence local methylation patterns⁶. The team captured modifications before and after editing and observed localised changes near the Cas9 cleavage sites, suggesting that the repair process itself can reshape epigenetic landscapes in ways legacy methods would miss.

This multiomic approach means that Oxford Nanopore doesn’t just offer long reads, but a consolidated and streamlined single workflow that replaces multiple fragmented assays. Rather than running separate tests for structural integrity, insertion mapping, and methylation analysis, sponsors can generate many of the required datasets in a single run, reducing turnaround time and overall assay complexity.

How can biopharma labs implement scalable sequencing?

The draft guidance is clear: long reads should be part of your 'fit-for-purpose' strategy for genome-edited therapies where complex structural outcomes may occur. But how do you incorporate them at scale? Oxford Nanopore devices allow you to run the same workflows at the scale that suits your programme — from early discovery through to good manufacturing practice (GMP) environments. You can choose from MinION for targeted assessments, GridION for medium-throughput projects, or PromethION for large-scale, continuous use.

What’s next? Preparing for compliance and confidence

The FDA draft guidance may not yet be finalised, but it’s clear that future standards will demand comprehensive ISA as a baseline expectation for therapies involving genome editing. Acting now doesn’t just mean positioning your programme for regulatory success but also means accelerating the delivery of genome-edited therapies to patients. Oxford Nanopore sequencing can help you move therapies forward with confidence.

Want to learn more?

Explore our brochure on Oxford Nanopore sequencing solutions for cell and gene therapies.

Oxford Nanopore Technologies products are not intended for use for health assessment or to diagnose, treat, mitigate, cure, or prevent any disease or condition.