A 65-year-old (62 kg) female patient was admitted to our department on 19 September 2021, with the chief complaint of chest tightness, dyspnoea, and lower abdominal distension for 5 days. Her previous history was unremarkable, but she had hypertension for 10 years. Two masses in the right atrium were observed on echocardiography; of concern, one of the masses was lodged in the tricuspid valve orifice. To prevent pulmonary embolism caused by tumour shedding, an emergency operation was carried out.
Complete surgical resection of the tumour was performed through median sternotomy under standard cardiopulmonary bypass, superior vena cava, inferior vena cava, and aortic cannulation with aortic cross-clamping. After incision of the right atrium, the base of the larger tumour was localized to the right atrial wall, filling the right atrium (Fig. 1, A). The base of the smaller tumour was located above the tricuspid valve annulus, and the tumour crossed the tricuspid valve and penetrated deep into the right ventricle (Fig. 1, B). The tumour sizes were 3.0 × 7.0 × 7.0 cm3 and 2.0 × 4.0 × 6.0 cm3 (Fig. 2, A). The tumours were removed carefully. The tricuspid valve was seriously evaluated, and intact valve function was noted. The cardiopulmonary bypass time was 66 min, and the patient was easily weaned from cardiopulmonary bypass. Postoperative echocardiogram showed good right ventricular function without tricuspid regurgitation. The patient recovered uneventfully and was discharged on postoperative day 4. Echocardiography at postoperative month 6 showed an ejection farction of 55%, without recurrent tumours.
The tumour was rich in cellular myxoma with chondroid metaplasia and sarcomatoid differentiation (Fig. 2, B). Immunohistochemistry showed the following: Vimentin (+), Filami-1 (FIL-1) (+), Cluster of differentiation 99 (CD99) (+), Cluster of differentiation 34 (CD34) (+), Neuron specific enolase (NSE) (+), Casein kinase (CK) (-), S100 (-), Epithelial membrane antigen (EMA) (-), Cluster of differentiation 117 (CD117) (-), Specific AT sequence binding protein-2 (SATB-2) (-), and KI-67 proliferation index 40%.
To identify the features of the tumour, we determined its cell composition and growth pattern via scRNA-seq. The tumour samples were used for scRNA-seq after informed consent from patient and approval from the ethics committee of Bengbu medical college (No.2021170). All methods were performed in accordance with the Declaration of Helsinki. The method of scRNA-seq is summarized below.
The tumour sample was minced on ice to less than 1 mm cubic pieces, followed by enzymatic digestion using an enzymatic cocktail [collagenase I/dispase II (1 µg/ml) tissue] with manual shaking every 5 min. The sample was then centrifuged at 300 rcf for 30 s at room temperature, and the supernatant was removed without disturbing the cell pellet. Next, 1×PBS (calcium and magnesium free) containing 0.04% weight/volume BSA (400 µg/ml) was added and then centrifuged at 300 rcf for 5 min. The cell pellet was resuspended in 1 ml lysis buffer and incubated for 15 min at 4°C. Then, the sample was resuspended in 1 ml PBS containing 0.04% BSA. Next, the sample was filtered over Scienceware Flowmi 40-µm cell strainers. Nuclei were gently collected and suspended in ice-cold PBS containing 0.4% BSA.
Chromium System (10x Genomics, Pleasanton, California) was used and processed following the manufacturer’s instructions. Library construction was performed using Chromium Single Cell 3’ Library & Single Cell 3ʹ v3 Gel Beads. Libraries were pooled based on their molar concentrations. To demultiplex samples, process barcodes, align and filter reads, and generate feature barcode matrices, we used the 10x Genomics Cell Ranger (v3.1.0) pipeline according to the manufacturer’s instructions in raw data quality assessment. We processed the unique molecular identifier (UMI) count matrix using the R package Seurat (version 3.0). To remove low-quality cells and likely multiplet captures, we applied criteria to filter out cells with unique molecular identifiers (UMIs)/gene numbers out of the limit of the mean value +/- 2-fold the standard deviation. We further discarded low-quality cells when a certain percentage of counts belonged to mitochondrial genes. Library size normalization was performed in Seurat on the filtered matrix to obtain the normalized count. Principal component analysis (PCA) was performed to reduce the dimensionality of the log-transformed gene-barcode matrices of the top variable genes. Cells were clustered based on a graph-based clustering approach and visualized in 2 dimensions using tSNE. A likelihood ratio test was used to identify significantly differentially expressed genes between clusters. Differentially expressed genes (DEGs) were identified using the R package. A P value < 0.05 and |log2foldchange| > 1 were set as the thresholds for significant differential expression. Gene Ontology (GO) enrichment and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses of DEGs were performed using the hypergeometric distribution. The sequencing and bioinformatics analysis were performed by OE Biotech Co., Ltd. (Shanghai, China).
After quality control by eliminating multicellular and apoptotic cells, the transcriptomes of 2095–11815 cells were analysed. A total of 3564 DEGs were detected. Compared with the public data in published studies , we found that the major cell components of the tumour to be chondrocytes, mesenchymal stromal cells, and smooth muscle cells. After filtering, GO analysis of 36 hub genes showed that they are involved in multicellular organismal processes, anatomical structure development, cellular response to stimulus, and cell differentiation. KEGG pathway enrichment showed that the cyclic guanosine monophosphate-cyclic guanosine monophosphate (cGMP-PKG) signalling pathway, transcriptional misregulation in cancer, and the p53 signalling pathway are related to the growth of these tumours. Protein–protein interactions of these genes were also analysed (Fig. 3).