Clinical-Scale Expansion of Human Bone Marrow-Derived Mesenchymal Stromal Cells to Treat Patients After Ischemic Stroke.

Abstract 3021

Ischemic stroke (IS) is the second leading cause of death worldwide and the leading cause of adult disability. IS patients have few options to limit neurological damage or augment the rehabilitation process. Tissue plasminogen activator, a fibrinolytic agent, is the only FDA approved therapy after IS but it must be administered within 3 hours of onset and is of limited benefit. Based on promising animal studies, another therapeutic option may be the application of mesenchymal stromal cells (MSCs). In one study, autologous MSCs were given to a subset of IS patients. This group showed better survival and neurological improvement, as judged by modified Rankin score and the Barthel index.

A significant limitation to the use of autologous MSCs for acute stroke is that cells cannnot be transfused until at least 5 weeks after IS due to the time required for expansion in vitro. We hypothesize that treating patients with MSCs within 48 hours of a stroke may improve outcomes. We plan to expedite the time-to-infusion in two ways. First, banked allogeneic MSCs will be used instead of autologous MSCs to eliminate the production time and convert this to an off-the-shelf therapy. Second, we have compared the use of an automated cell culture device (the Quantum by Terumo BCT), which uses 2.1 m² of hollow fibers in a bioreactor (equivalent to ∼120 T-175 cm² flasks), to our current flask-based expansion method.

In flasks, human bone marrow (BM) mononuclear cells (BMMC) were separated using the SEPAX device and were seeded at 5,000 BMMC/cm² in T-175 cm² flasks and split 1:4 when 70% confluent. After ∼1 month and 3–5 passages, ∼130 T-175 cm² flasks were harvested and cryopreserved. In the Quantum, ∼25 mL of whole unprocessed BM was added to the bioreactor through a 200 μm filter. After 10 days, the cells were harvested and 2.0–3.5×10⁷ cells were re-seeded in a new bioreactor at ∼1000 MSC/cm² for a total culture time of ∼17 days.

After 3–4 passages in flasks, an average of 2.69×10⁸ MSCs were expanded from three independent BM donors. In two BM donors, MSCs harvested after 2 passages in the Quantum yielded an average of 8.75×10⁸; these cells were frozen and can be used for subsequent expansions. MSCs expanded in both flasks and the Quantum were analyzed by flow cytometry and met the International Society of Cell Therapy (ISCT) minimum criteria for MSC (expression of CD73, CD105, CD90 etc). Expanded MSCs were sterile, were free from chromosomal abnormalities, and differentiated into adipocytes, chondrocytes, and osteoblasts. The production of 5×10⁸ MSCs in flasks required 101 hours of labor compared to 21 hours with the Quantum. We estimated that during the manufacture of MSCs in flasks, there were 600 open (defined as exposing the system to the environment) events compared to 9 in the Quantum.

To test the in vivo effects of manufactured MSCs, 24 aged male Long-Evans rats were randomized to receive flask-based human MSCs (P4) or saline vehicle at 7 days after stroke. At 28 days after stroke, animals treated with MSCs showed a significant reduction in neurological deficits compared with saline treated controls.

In conclusion, in addition to the advantage of manufacture in a functionally- closed system, large numbers of MSCs expanded in the Quantum were available at a lower passage number with a higher average-cell yield at a slightly higher cost per 5×10⁸ cells. When administered to rats with ischemic stroke, flask-based MSCs improved outcomes compared to placebo control treated animals. Whether MSCs expanded in the Quantum will function with equivalent efficacy is currently being tested and these data will be available for presentation in December. This project is supported by NHLBI-PACT, contract # HHSN268201000007C.