Abstract 54P
Background
Lymph nodes (LN) are the most frequent metastasis sites in oesophageal squamous cell carcinoma (ESCC), but the associated mechanisms are poorly understood. Tumor cells seed and survive at LN during the early adaptation step, which is the bottleneck for successful metastasis. However, the phenotypic and molecular characteristics of heterogeneous metastatic cells that contribute to such a challenging process remain elusive.
Methods
We established a human xenograft spontaneous LN metastasis model to capture tumor cell dissemination behaviors. We conducted single-cell RNA sequencing on tumor cells collected from primary tumors and metastatic LN to profile the dynamic cellular and molecular changes in tumor cells associated with LN metastasis. We further identified and characterized the early metastasis tumor cells in LN through in vitro and in vivo functional assays.
Results
We revealed that LN metastasis is initiated by rare ESCC subpopulations with stemness and mesenchymal regulatory gene expression signatures. These tumor cell subpopulations are critical to driving metastasis in distant organs through the blood circulation in mice, supporting the metastasis-initiating cells (MICs) properties. Importantly, these key driver cells were observed to convert the metabolic strategy to mitochondrial oxidative phosphorylation (OXPHOS) during cell seeding in the LN. Enhanced OXPHOS in LN metastatic cells is induced by fatty acids and regulated by NRF2 signaling. Pharmacological inhibition of OXPHOS with IACS-010759 markedly attenuated metastasis in mice.
Conclusions
Our results demonstrate the important role of mitochondrial OXPHOS in promoting ESCC metastasis and suggest the clinical therapeutic potential of OXPHOS inhibitors for preventing metastatic spread in ESCC patients.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
National Natural Science Foundation of China (82103643, 82072738, 82072731).
Disclosure
All authors have declared no conflicts of interest.
Resources from the same session
514P - Immunophenotypic profile of glioblastoma microenvironment: A cohort study
Presenter: Lidia Gatto
Session: Poster session 09
515P - A MRI-based radiomics model for predicting the response to anlotinb combined with temozolomide in recurrent malignant glioma patients
Presenter: Shu Zhou
Session: Poster session 09
516P - Building a new prognostic score for patients with central nervous system (CNS) tumors enrolled in early phase clinical trials
Presenter: Kristi Beshiri
Session: Poster session 09
517P - Differentiating IDH-wildtype and IDH-mutant high grade gliomas with deep learning
Presenter: Katherine Hewitt
Session: Poster session 09