Back

P5 promoter-mediated incorporation explains REP/CAP manufacturing contaminants in patient liver after rAAV gene therapy

Brimble, M. A.; Tan, S.; Buddle, S.; Brown, L.-A. K.; Breuer, J.; Crawford, J. C.

2026-09-14 genetic and genomic medicine
10.64898/2026.09.10.26362787 medRxiv
Show abstract

Sequencing of liver tissue from a patient treated with the rAAV gene therapy Zolgensma for spinal muscular atrophy recently revealed contaminating plasmid sequences derived from rAAV manufacturing within the patient's hepatocytes. In particular, REP/CAP-derived sequences were remarkably abundant, corresponding to 0.5-1% of the therapeutic transgene. We hypothesized that these contaminants originated through Rep-mediated incorporation initiated at the AAV P5 promoter. Through reanalysis of the sequencing data, we inferred that an intact P5 promoter had been placed directly downstream of the rAAV capsid gene in the manufacturing plasmid. De novo assembly revealed a contiguous contaminant sequence spanning the rAAV REP and CAP genes and terminating within P5 at the Rep nicking site, immediately downstream of the Rep-binding element (RBE). This analysis also revealed a distinct vector-plasmid backbone contig consistent with reverse packaging. Among partially aligned REP/CAP reads, nearly 17% were linked to rAAV ITR-derived sequence at heterogeneous junctions. Long-read data independently identified the P5 promoter as the most frequent recombination breakpoint region. Together, these findings identify a defined, avoidable mechanism by which REP/CAP manufacturing contaminants arise. Positioning of the P5 promoter within the manufacturing plasmid is therefore a modifiable determinant of rAAV product purity and the transfer of rAAV DNA contaminants to recipient patients.

Matching journals

The top 6 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.