Investigating epithelial-mesenchymal heterogeneity of tumors and circulating tumor cells with transcriptomic analysis and biophysical modeling

Abstract

Introduction: Cellular heterogeneity along the Epithelial-Mesenchymal Plasticity (EMP) spectrum is a paramount feature observed in tumors and circulating tumor cells (CTCs). High-throughput techniques now offer unprecedented details on this variability at a single- cell resolution. Yet, there is no current consensus about how EMP in tumors propagates to that in CTCs. To investigate a relationship between EMP associated heterogeneity of tumors and that of CTCs, we integrated transcriptomic analysis and biophysical modeling. Methods: We apply three EMT (Epithelial-Mesenchymal Transition) scoring metrics to multiple tumor samples and CTC datasets from several cancer types. Moreover, we develop a biophysical model that couples EMT associated phenotypic switching in a primary tumor with cell migration. Finally, we integrate EMT transcriptomic analysis and in silico modeling to evaluate the predictive power of several measurements of tumor aggressiveness, including tumor EMT score, CTC EMT score, fraction of CTC clusters found in circulation, and CTC cluster size distribution. Results: Analysis of high-throughput datasets reveals a pronounced heterogeneity without a well-defined relation between EMT traits in tumors and CTCs. Moreover, mathematical modeling predicts different phases where CTCs can be less, equally, or more mesenchymal than primary tumor depending on the dynamics of phenotypic transition and cell migration. Consistently, various datasets of CTC cluster size distribution from different cancer types are fitted onto different regimes of the model. By further constraining the model with experimental measurements of tumor EMT score, CTC EMT score, and fraction of CTC cluster in bloodstream, we show that none of these assays alone can provide sufficient information to predict the other variables. Conclusions: By integrating analysis of single cell gene expression and in silico modeling, we propose that the relationship between EMT progression in tumors and CTCs can be variable, and in general, predicting one from the other may not be as straightforward as tacitly assumed.

Systems-level network modeling deciphers the master regulators of phenotypic plasticity and heterogeneity in melanoma

“…Other EMT regulators such as TWIST and AP-1 ( Feldker et al., 2020 ) are also reported to drive phenotypic plasticity in melanoma ( Caramel et al., 2013 ; Verfaillie et al., 2015 ). Second, heterogeneity along the EMT axis seen in circulating tumor cells (CTCs) ( Yu et al., 2013 ; Bocci et al., 2021 ) is recapitulated in melanoma CTCs too ( Aya-Bonilla et al., 2020 ). Third, cooperation among phenotypes with varying EMT status has been witnessed during metastasis and tumor formation ( Jolly and Celia-Terrassa, 2019 ), reminiscent of recent observations in melanoma ( Rowling et al., 2020 ).…”

Cluster Size Distribution of Cells Disseminating from a Primary Tumor

“…Due to this cluster breakup in the medium, smaller clusters are mainly composed of cells with lower motility, hence are relatively more epithelial-like and the larger clusters are mainly composed of cells with higher motility, and are relatively more mesenchymal-like. This can be a reason for CTCs to be relatively more mesenchymal compared to the primary tumor, as found in the case of Stage II-III breast cancer (GSE111842) [47, 48]. In general, it would be of interest to compare EMT-scores from clusters of different sizes taken from patient blood.…”

Systems-level network modeling deciphers the master regulators of phenotypic plasticity and heterogeneity in melanoma

“…Other EMT regulators such as TWIST and AP-1 (Feldker et al, 2020) are also reported to drive phenotypic plasticity in melanoma (Caramel et al, 2013; Verfaillie et al, 2015). Second, heterogeneity along the EMT axis seen in circulating tumor cells (CTCs) (Bocci et al, 2021; Yu et al, 2013) is recapitulated in melanoma CTCs too (Aya-Bonilla et al, 2020). Third, cooperation among phenotypes with varying EMT status has been witnessed during metastasis and tumor formation (Jolly and Celia-Terrassa, 2019), reminiscent of recent observations in melanoma (Rowling et al, 2020).…”