Stem cell-like circulating tumor cells identified by Pep@MNP and their clinical significance in pancreatic cancer metastasis.

Objective: The circulating tumor cells (CTCs) could be captured by the peptide functionalized magnetic nanoparticles (Pep@MNP) detection system in pancreatic ductal adenocarcinoma (PDAC). CTCs and the CXCR4 expression were detected to explore their clinical significance. The CXCR4+ CTCs, this is highly metastatic-prone stem cell-like subsets of CTCs (HM-CTCs), were found to be associated with the early recurrence and metastasis of PDAC. Methods: CTCs were captured by Pep@MNP. CTCs were identified via immunofluorescence with CD45, cytokeratin antibodies, and the CXCR4 positive CTCs were assigned to be HM-CTCs. Results: The over-expression of CXCR4 could promote the migration of pancreatic cancer cell in vitro and in vivo. In peripheral blood (PB), CTCs were detected positive in 79.0% of all patients (49/62, 9 (0-71)/2mL), among which 63.3% patients (31/49, 3 (0-23)/2mL) were HM-CTCs positive. In portal vein blood (PVB), CTCs were positive in 77.5% of patients (31/40, 10 (0-40)/2mL), and 67.7% of which (21/31, 4 (0-15)/2mL) were HM-CTCs positive CTCs enumeration could be used as diagnostic biomarker of pancreatic cancer (AUC = 0.862), and the combination of CTCs positive and CA19-9 increase shows improved diagnostic accuracy (AUC = 0.963). in addition, PVB HM-CTCs were more accurate to predict the early recurrence and liver metastasis than PB HM-CTCs (AUC 0.825 vs. 0.787 and 0.827 vs. 0.809, respectively). Conclusions: The CTCs identified by Pep@MNP detection system could be used as diagnostic and prognostic biomarkers of PDAC patients. We identified and defined the CXCR4 over-expressed CTC subpopulation as highly metastatic-prone CTCs, which was proved to identify patients who were prone to suffering from early recurrence and metastasis.

Single-cell transcriptome profiling of primary tumors and paired organoids of pancreatobiliary cancer.

Single-cell RNA-seq (scRNA-seq) and cancer organoid model have shown promise in investigating tumor microenvironment heterogeneity and facilitating chemotherapeutic drug testing to inform treatment selection. It is still unknown whether the scRNA-seq results based on organoid can faithfully reflect the heterogeneity of primary pancreatobiliary cancer. To reveal the similarities and differences between primary tumors and their matched organoids at transcriptome level, we conducted scRNA-seq for paired primary tumors and organoids from one cholangiocarcinoma (CCA) and two pancreatic ductal adenocarcinoma (PDAC) patients. We identified inter-patient and intra-tumor heterogeneity and found that the organoids retained copy number variation (CNV) patterns of primary tumors. There was no significant difference in cancer stem cell (CSC) properties between the primary tumors and the organoids, whereas organoid from one PDAC case had increased mesenchymal-score and decreased epithelial-score compared with the primary tumors. All organoids showed a transition tendency from the classical subtype to the basal-like subtype in the transcriptional level. Organoids and primary tumors differed in metabolic and unfolded protein response (UPR) signatures. In addition, we revealed the heterogeneity of cancer associated fibroblasts (CAFs) and T cells, and explored the developmental trajectory of T cells. Our findings facilitate further understanding of organoid model and confirm its application prospects in pancreatobiliary cancer.

Single-cell RNA-seq reveals characteristics in tumor microenvironment of PDAC with MSI-H following neoadjuvant chemotherapy with anti-PD-1 therapy.

Accumulating evidence suggests the minority of patients with advanced pancreatic ductal adenocarcinoma (PDAC) that have microsatellite instability high (MSI-H) can benefit from immune checkpoint inhibitors (ICIs). However, the effects of ICIs on the tumor microenvironment (TME) of PDAC remain elusive. We conducted single-cell RNA-seq (scRNA-seq) analysis on a residual lesion from a MSI-H PDAC patient who received a radical operation after eight cycles of neoadjuvant treatment (nab-paclitaxel/gemcitabine plus pembrolizumab). Multiple tumor subclusters were identified in residual lesion after neoadjuvant treatment, one of which was mainly composed of cells in the S and G2M phases. This subcluster also had enriched expression of MKI67 and PCNA and cell cycle-related signatures and was thus defined as a proliferating tumor subcluster. This subcluster had higher S_score, Fatty acid_score, UPR_score, and Glycolysis_score than others. We also identified characteristics of the TME after neoadjuvant treatment by comparing the excised primary tumors form nontreated PDAC and the residual lesion. The residual lesion was characterized with activated pancreatic stellate cells (PSCs) and exhausted T cells (Tex). We compared the receptor-ligand interactions between the two groups, and found that no checkpoint receptor-ligand pairs between T cells and tumor cells were identified in the residual lesion, while there were many checkpoint receptor-ligand pairs in the nontreated primary PDAC. In conclusion, our findings revealed the characteristics of residual lesion of advanced PDAC with MSI-H upon combination treatment of chemotherapy and immunotherapy, which might provide some valuable clues for solving the puzzle of ICI in PDAC.