How inclusive are cell lines in preclinical engineered cancer models?

Abstract

Diverse factors contribute to significant and dire disparities in cancer risk and treatment outcomes. To address this, there was a call for inclusion of sex as a biological variable, which resulted in more instances of careful inclusion of sex in preclinical studies of cancer. Another variable in cancer treatment is genetic ancestry. Although this is considered explicitly in clinical research, it is considerably neglected in preclinical studies. Preclinical research can use several 3D in vitro model systems, such as spheroids/organoids, xenografts, or other bioengineered systems that combine biomaterials and cellular material. Ultimately, the cellular base for all of these in vitro model systems is derived from human cell lines or patient samples, to investigate mechanisms of cancer and screen novel therapeutics, all of which aim to maximize successful outcomes in clinical trials. This in itself offers an opportunity to potentiate effective treatments for many groups of people, when diverse variables like genetic ancestry are consciously included into study design. This Perspective highlights the need for conscious inclusion of genetic ancestry in preclinical cancer tissue engineering, especially when it pertains to determining therapeutic outcomes.

Fig. 1.

The promise of preclinical research that consciously includes genetic ancestry as a variable is that precision medicine becomes a reality for all populations of people, and not just a majority. By expanding the genetic diversity of in vitro models in preclinical testing, we can improve clinical trial outcomes and treatment efficacy for all.

Fig. 2.

Reported diversity of patient samples and cell lines available through common databases and repositories. (A) Of 1018 cancer cell lines in the Catalogue of Somatic Mutations in Cancer (COSMIC), 317 had reported ancestry and 701 had unreported ancestry; 697 of these cell lines were of European ancestry (245 reported and 453 unreported), 253 were of East Asian ancestry (37 reported and 215 unreported), 56 were of African ancestry (26 unreported and 30 reported), and one cell line of unreported ancestry was found to be South Asian (Kessler et al., 2019). (B) The Affymetrix SNP6.0 arrays genotyping dataset contains 994 cancer cell lines that are deposited in the European Genome-Phenome Archive. Analysis of this dataset found that 633 of the cell lines were of European ancestry, 248 were of East Asian ancestry, 56 were of African ancestry, 47 were of American ancestry and ten were of South Asian ancestry (Nguyen et al., 2021). (C) Of 853 samples in the NCI patient-derived models repository, 135 had unreported ancestry, 645 were of European ancestry, 39 were of African ancestry, 25 were of mixed ancestry, eight were of East Asian ancestry, and one was of American ancestry (NCI-Frederick). (D) A literature review of 30 peer-reviewed manuscripts used in colorectal cancer preclinical research (Table S1); 22 of these studies used established cell lines, all of which were of European ancestry (https://web.expasy.org/cellosaurus), and eight of these studies used patient samples, all of which had unreported ancestry. *Numbers represent studies instead of individual cell lines or samples.