AUTHOR=Israeli Sapir , Krakow Elizabeth F. , Maiers Martin , Summers Corinne , Louzoun Yoram 

TITLE=Trans-population graph-based coverage optimization of allogeneic cellular therapy

JOURNAL=Frontiers in Immunology

VOLUME=Volume 14 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2023.1069749

DOI=10.3389/fimmu.2023.1069749

ISSN=1664-3224

ABSTRACT=Pre-clinical development and in-human trials of ‘off-the-shelf’ immune effector cell therapy (IECT) are burgeoning. Many IECTs are dependent on HLA compatibility, such as virus-specific T cells that exert antiviral activity through HLA-mediated recognition or NK cells that kill through KIR-HLA mismatch.  HLA compatibility is not commonly considered in the development of other allogeneic IECTs, such as CAR-modified cells. However, HLA matching, or knocking out problematic HLAs, could improve the persistence and duration of efficacy of such “HLA-independent” therapies. IECT offers many potential advantages over autologous products, and strategically creating an off-the-shelf IEC bank (as opposed to on-demand donor collection and manufacturing) can speed delivery of these products. However, even in situations where finding HLA compatible starting material is usually easy, there is no need to manufacture “HLA redundant” products. Optimally sizing the bank to maximize population coverage while minimizing manufacturing costs is an open problem. 

To address this aspect of IECT development, we developed an optimal coverage problem, combined with graph algorithms for donor selection, under different, clinically plausible scenarios where the cost of additional donor recruitment and/or the cost of preventing class I or class II HLA expression through gene editing were of interest.

We compared the efficiency of different optimization algorithms – a greedy solution, a linear programming (LP), and integer linear programming (ILP) vs.  random donor selection over millions of candidate donors. The  additional population coverage per donor decreases with the number of donors. All proposed algorithms consistently achieve the optimal coverage with far fewer donors than the random choice. The number of randomly selected donors required to achieve a desired coverage increases with increasing population. However, when optimal donors are selected, the number of donors required may counterintuitively decrease with increasing population size. 

editing was generally more expensive than recruiting additional donors. When choosing the donors and patients from different populations, the number of random donors required drastically increases, while the number of optimal donors does not change. Random donors fail to cover populations different from their original populations, while a small number of optimal donors from one population can cover a different population.