Found 1 Presentation For Request "1694P"

Translational research (agnostic)

1694P - Single cell characterization of longitudinal biopsies from breast cancer patients treated with the aromatase inhibitor letrozole and the CDK4/6 inhibitor ribociclib

Presentation Number
1694P
Speakers
  • Xavier Tekpli (Oslo, Norway)
Date
Sat, 10.09.2022

Abstract

Background

The recent introduction of CDK4/6 inhibitors has been one of the most pivotal breakthroughs in breast cancer therapy in the last decades. A growing body of evidence proposes that CDK4/6 inhibitors potentially influence the recruitment of immune cells in the tumor microenvironment, but also induce a senescent-like phenotype in the tumor cells. The NEOLETRIB-study is a multicenter, single-arm, open-label, neoadjuvant, phase II trial. Locally advanced breast cancer patients belonging to the luminal-A and luminal-B subtypes, received neoadjuvant therapy for 6 months with ribociclib (600 mg daily, 21 days on / 7 days off) and letrozole (2.5 mg daily).

Methods

Pre-treatment, on-treatment (end of the first 21 days of therapy) and end-of-treatment biopsies were subjected to single-cell transcriptome using the Chromium Single-Cell v2 5′ Chemistry. Raw gene expression matrices were analyzed with the Seurat package (v4.0.2). First, empty droplets and droplets containing two cells were filtered out using the EmptyDrops and scDblFinder algorithms respectively. In addition, cell barcodes with < 500 UMIs, < 300 expressed genes, > 5000 expressed genes or > 15% of reads mapping to mitochondrial RNA were filtered out. The gene expression of the remaining good quality cells was normalized.

Results

Variably expressed genes were used to construct principal components (PCs). Significant PCs were further used to cluster cells. We identified 8 main cell types: T cells, B cells, epithelial cells, fibroblasts, endothelial cells, mast cells and dendritic cells. To discriminate between cancer and normal epithelial cells, we estimated for each epithelial cell the copy number variations using the InferCNV algorithm. Furthermore, to identify specific and specialized cell subtype, we clustered each of the eight cell type independently and annotated the clusters obtained using validated marker genes. We also compare our annotations with publish methods identifying cell states in single cell data (Luca et al., Cell 2021).

Conclusions

We highlight how scRNA-seq and our analytic pipelines are allowing us to study the tumor microenvironment of breast cancer biopsies at a very high resolution.

Legal entity responsible for the study

The authors.

Funding

Novartis.

Disclosure

All authors have declared no conflicts of interest.

Collapse