Department of Computer Science, University of Maryland, College Park, MD, 20742, USA.
Department of Computer Science, Northwestern University, Evanston, IL, 60208, USA.
Nat Commun. 2022 Mar 25;13(1):1613. doi: 10.1038/s41467-022-29154-2.
Mining a large cohort of single-cell transcriptomics data, here we employ combinatorial optimization techniques to chart the landscape of optimal combination therapies in cancer. We assume that each individual therapy can target any one of 1269 genes encoding cell surface receptors, which may be targets of CAR-T, conjugated antibodies or coated nanoparticle therapies. We find that in most cancer types, personalized combinations composed of at most four targets are then sufficient for killing at least 80% of tumor cells while sparing at least 90% of nontumor cells in the tumor microenvironment. However, as more stringent and selective killing is required, the number of targets needed rises rapidly. Emerging individual targets include PTPRZ1 for brain and head and neck cancers and EGFR in multiple tumor types. In sum, this study provides a computational estimate of the identity and number of targets needed in combination to target cancers selectively and precisely.
通过挖掘大量的单细胞转录组学数据，我们运用组合优化技术来描绘癌症的最佳联合治疗方案。我们假设，每一种治疗方法都可以针对 1269 个编码细胞表面受体的基因中的任意一个，这些基因可能是 CAR-T、缀合抗体或涂层纳米颗粒治疗的靶点。我们发现，在大多数癌症类型中，由最多四个靶点组成的个性化组合足以杀死至少 80%的肿瘤细胞，同时保留肿瘤微环境中至少 90%的非肿瘤细胞。然而，随着需要更严格和选择性的杀伤，所需靶点的数量迅速增加。新兴的个体靶点包括脑和头颈部癌症中的 PTPRZ1 和多种肿瘤类型中的 EGFR。总之，这项研究提供了一个计算估计，即需要联合使用多少个靶点来有针对性和精确地靶向癌症。
