Quasi-perfusion studies for intensified lentiviral vector production using a continuous stable producer cell line

• Published in: Molecular therapy. Methods & clinical development 2024-06, Vol.32 (2), p.101264-101264, Article 101264
• Main Authors: Stibbs, Dale J., Silva Couto, Pedro, Takeuchi, Yasuhiro, Rafiq, Qasim A., Jackson, Nigel B., Rayat, Andrea C.M.E.
• Format: Article
• Language: English
• Subjects: fixed-bed bioreactor; lentiviral vector; process intensification; quasi-perfusion processing; scale-up; stable producer cell line
• ISSN: 2329-0501; 2329-0501
• DOI: 10.1016/j.omtm.2024.101264

Quasi-perfusion culture was employed to intensify lentiviral vector (LV) manufacturing using a continuous stable producer cell line in an 8-day process. Initial studies aimed to identify a scalable seeding density, with 3, 4, and 5 × 104 cells cm−2 providing similar specific productivities of infectious LV. Seeding at 3 × 104 cells cm−2 was selected, and the quasi-perfusion was modulated to minimize inhibitory metabolite accumulation and vector exposure at 37°C. Similar specific productivities of infectious LV and physical LV were achieved at 1, 2, and 3 vessel volumes per day (VVD), with 1 VVD selected to minimize downstream processing volumes. The optimized process was scaled 50-fold to 1,264 cm2 flasks, achieving similar LV titers. However, scaling up beyond this to a 6,320 cm2 multilayer flask reduced titers, possibly from suboptimal gas exchange. Across three independent processes in 25 cm2 to 6,320 cm2 flasks, reproducibility was high with a coefficient of variation of 7.7% ± 2.9% and 11.9% ± 3.0% for infectious and physical LV titers, respectively. The optimized flask process was successfully transferred to the iCELLis Nano (Cytiva) fixed-bed bioreactor, with quasi-perfusion at 1 VVD yielding 1.62 × 108 TU. [Display omitted] Rayat and colleagues intensified LV production by combining quasi-perfusion culture with a stable producer cell line. The impact of seeding density and quasi-perfusion rate on vector titers was evaluated. The optimized process was scaled in static vessels, demonstrating high reproducibility, before being transferred to a FBR.