Vector Integration Site & Integrant Structure

Vector Integration Site & Integrant Structure

Integration-level characterization for integrating vectors and delivery systems — lentivirus/retrovirus, AAV, and transposons (Sleeping Beauty/piggyBac): Using target solution-phase capture combined with PacBio HiFi long reads to precisely localize integration sites, resolve abnormal integrant structures, and estimate vector copy number (VCN), providing molecular-level evidence for assessing insertional mutagenesis and genotoxicity risks, designing long-term follow-up strategies, and supporting IND/BLA submissions.

1. Background

Integration is the basis of stable expression and a potential source of genotoxicity. Lentivirus and retrovirus, in ex vivo cell products such as CAR-T, achieve long-term expression by integrating the transgene into the host genome. AAV is predominantly episomal but exhibits low-frequency integration. Transposons (Sleeping Beauty, piggyBac), as non-viral integrating systems, are increasingly used in new platforms such as non-viral CAR-T. Although the integration behavior of the three systems differs, the core questions are the same: where does the integration occur, is the integrant structure intact, and how many vector copies does each cell carry?

Integration-related risks have three principal dimensions. First, integration is not entirely random; lentivirus tends to insert near transcriptionally active genes, and activation of a proto-oncogene or disruption of a tumor-suppressor gene can drive insertional mutagenesis and clonal expansion — the LMO2-related leukemia events in the SCID-X1 clinical trials are a classic precedent. Second, in neonatal mouse models, AAV has been observed to integrate in clusters at the Rian locus (commonly referred to as the "AAV-HCC site") and to up-regulate adjacent miRNAs, associated with hepatocellular carcinoma; its human homologous cluster DLK1-DIO3 is correlated with HCC prognosis, making rare AAV integration an actively monitored safety topic among regulators. Third, vector copy number (VCN) is a common release attribute, with higher VCN generally implying greater insertional mutagenesis and genotoxicity risk.

Integration sites are often accompanied by structural rearrangements and repetitive sequences, and short-read sequencing struggles to span the vector–host junction and to distinguish complex integrants (tandem integration, truncation, inversion, chimerism). Capturing integration events at low abundance while reading the full junction structure requires targeted enrichment combined with single-molecule long-read sequencing.

From a regulatory standpoint, ICH S12 (Nonclinical Biodistribution Considerations, adopted by the FDA in 2023) requires evaluation of vector biodistribution, persistence and clearance in vivo, with integration risk assessed based on vector type, replication competence, integration potential, dose, and route of administration. The FDA guidance "Long Term Follow-Up After Administration of Human Gene Therapy Products" (2020) requires that integrating vectors or products causing genomic alterations be subject to long-term follow-up of up to 15 years. The FDA guidance "Considerations for the Development of CAR T Cell Products" (Final, 2024) sets requirements for CAR-T CMC (including RCR/RCL, VCN, purity, and potency) and for insertional-mutagenesis and copy-number control of integrating vectors. China's "Technical Guideline for Research and Evaluation of Cell Therapy Products (Trial)" (2017) and CDE's "Technical Guideline for Pharmaceutical Research and Evaluation of Immune Cell Therapy Products (Trial)" (2022) set clear requirements for tumorigenicity and safety evaluation. This service provides supporting evidence for these objectives through characterization of integration sites and integrant structure.

2. Technical Principle

This service is built on solution-phase target capture combined with PacBio HiFi long-read sequencing, using the customer-provided vector/transposon sequence and the human reference genome as custom alignment references. At the single-molecule level, individual reads span the vector–host junction and faithfully read out the full integrant structure. The technical characteristics align with the integration-analysis requirements of catching the events, reading through the junctions, and judging with accuracy. The core workflow is as follows:

(1) Sample receipt and QC

Receive cell pellets or extracted genomic DNA; assess concentration, purity, and integrity; and confirm compliance with library construction and capture requirements.

(2) Library preparation and custom capture

Following long-read library preparation, custom probes designed against the customer-provided vector are used to enrich the vector and its flanking host sequences from the bulk genome. Custom capture probes are an included step of this service.

(3) PacBio HiFi sequencing

High per-base accuracy long reads are obtained via circular consensus sequences (CCS), with single reads spanning the vector–host junction to fully resolve integrant structure; sequencing depth is set according to target sensitivity.

(4) Bioinformatic analysis and risk interpretation

Alignment is performed against GRCh38 (and T2T-CHM13 where appropriate) together with the customer's vector reference. Integration sites are localized, neighboring genes annotated, abnormal integrant structures identified (rearrangement, truncation, deletion, inversion, tandem), VCN is estimated, and risk is graded according to predefined rules. A complete report is then issued.

Figure 1. Workflow of integrant analysis using target capture combined with PacBio HiFi long-read sequencing.

3. Technical Features and Advantages

(1) Single-molecule junction-spanning reads

PacBio HiFi single long reads span the vector–host junction, eliminating short-read assembly ambiguity and faithfully reflecting abnormal integrant structure.

(2) Access to low-abundance clones

Custom solution-phase capture enriches integration events at low abundance, with detection of low-abundance integration clones achievable down to approximately 0.1% LOD (depending on sequencing depth and enrichment efficiency).

(3) Compatible with three integrating systems

With probe and reference adaptation, the same target-capture plus long-read platform covers lentivirus/retrovirus, AAV, and transposons (Sleeping Beauty/piggyBac).

(4) VCN and integration profile in one data set

Integration sites, integrant structure, and vector copy number are obtained from the same dataset, avoiding the unit-of-measure inconsistencies that arise when stitching data across multiple platforms.

(5) Customizable risk grading

The general version outputs high/medium/low grading; criteria can be narrowed by project (for example, focusing only on rearrangements and tandems near proto-oncogenes) to align with the submission data package.

4. Applications

CAR-T and ex vivo gene-modified cells: Characterization of integration profile, clonal dynamics, and VCN for lentiviral or transposon-based CAR-T.

In vivo AAV gene therapy: Surveillance of low-frequency AAV integration (particularly at sensitive sites such as the Rian/DLK1-DIO3 region) to support tumorigenicity assessment.

Non-viral integrating systems: Characterization of integration profile and copy number for transposon systems such as Sleeping Beauty and piggyBac.

Long-term follow-up support: Provide molecular baseline data for the up-to-15-year follow-up strategy required by FDA.

IND/BLA characterization: Provide integration-level evidence for the safety and CMC packages of integrating gene therapy and cell therapy products.

5. Report and Deliverables

The report provides quantitative, locus-resolved, comparable integration-level evidence rather than a single "pass/fail" verdict. Core contents include:

·Integration site localization and annotation: Precise coordinates, neighboring genes, and whether the site falls within a proto-oncogene or sensitive region.

·Integrant structure characterization: Abnormal events including rearrangement, truncation, deletion, inversion, and tandem integration.

·Vector copy number (VCN) estimation: Depth-based population-level VCN and locus-level abundance.

·Tumorigenicity risk grading: General version includes high/medium/low grading; criteria customizable per project.

·Clonal abundance distribution: Distribution of integration clonal abundances and indicators of heterogeneity.

·Actionable recommendations: Input for vector design optimization, process improvement, and long-term follow-up strategy.

·Data deliverables: Complete analysis report (PDF), integration-site list, structural-variant files, and raw sequencing data.

Figure 2. Schematic of integration site distribution, integrant structure, and VCN analysis results.

6. Service Workflow

* Standard turnaround: 45–55 business days (including custom probe preparation).

7. Sample Requirements

* The latest Sample Submission Guide takes precedence. This service is not applicable to severely degraded samples. Please schedule and confirm the study plan before sample submission.

8. Technical Specifications

9. References

[1] International Council for Harmonisation (ICH). S12: Nonclinical Biodistribution Considerations for Gene Therapy Products. Step 4 version, adopted 26 April 2023. (FDA implemented 2023)

[2] ICH. Q2(R2): Validation of Analytical Procedures. Step 4 version, adopted 1 November 2023.

[3] U.S. Food and Drug Administration (FDA). Long Term Follow-Up After Administration of Human Gene Therapy Products; Guidance for Industry. Final, January 2020.

[4] FDA. Considerations for the Development of Chimeric Antigen Receptor (CAR) T Cell Products; Guidance for Industry. Final, January 2024. (Docket No. FDA-2021-D-0404)

[5] National Medical Products Administration of China (NMPA). Technical Guideline for Research and Evaluation of Cell Therapy Products (Trial) [in Chinese]. 2017.

[6] Center for Drug Evaluation, National Medical Products Administration of China (NMPA-CDE). Technical Guideline for Pharmaceutical Research and Evaluation of Immune Cell Therapy Products (Trial) [in Chinese]. 2022.

[7] Hacein-Bey-Abina S, et al. LMO2-associated clonal T cell proliferation in two patients after gene therapy for SCID-X1. Science, 2003, 302(5644): 415-419.

[8] Chandler RJ, et al. Vector design influences hepatic genotoxicity after adeno-associated virus gene therapy. Journal of Clinical Investigation, 2015, 125(2): 870-880.

[9] Nguyen GN, et al. A long-term study of AAV gene therapy in dogs with hemophilia A identifies clonal expansions of transduced liver cells. Nature Biotechnology, 2021, 39(1): 47-55.

[10] Tipanee J, et al. Transposons: moving forward from preclinical studies to clinical trials. Human Gene Therapy, 2017, 28(11): 1087-1104.