Unveiling the Genomic Landscape of Intraductal Carcinoma of the Prostate Using Spatial Gene Expression Analysis

Abstract

Intraductal carcinoma of the prostate (IDCP) has recently attracted increasing interest owing to its unfavorable prognoses. To effectively identify the IDCP-specific gene expression profile, we took a novel approach of characterizing a typical IDCP case using spatial gene expression analysis. A formalin-fixed, paraffin-embedded sample was subjected to Visium CytAssist Spatial Gene Expression analysis. IDCP within invasive prostate cancer sites was recognized as a distinct cluster separate from other invasive cancer clusters. Highly expressed genes defining the IDCP cluster, such as MUC6, MYO16, NPY, and KLK12, reflected the aggressive nature of high-grade prostate cancer. IDCP sites also showed increased hypoxia markers HIF1A, BNIP3L, PDK1, and POGLUT1; decreased fibroblast markers COL1A2, DCN, and LUM; and decreased immune cell markers CCR5 and FCGR3A. Overall, these findings indicate that the hypoxic tumor microenvironment and reduced recruitment of fibroblasts and immune cells, which reflect morphological features of IDCP, may influence the aggressiveness of high-grade prostate cancer.

Keywords: hypoxic markers; immune cells; intraductal carcinoma of the prostate (IDCP); spatial gene expression analysis; tumor microenvironment.

Figure 1

Diagnosis of IDCP and clustering through spatial gene expression analysis. (A) A 62-year-old man, diagnosed with prostate adenocarcinoma (cT3aN0M0, high risk, PSA 31.5 ng/mL), underwent a robot-assisted laparoscopic prostatectomy. HE staining revealed tumor invasion within normal glandular ducts with surrounding invasive carcinoma (blue square: invasive cancer area; yellow square: IDCP area; yellow arrow: IDCP). (B) Basal cell staining was performed on a formalin-fixed paraffin-embedded slide of the total prostatectomy specimen with a prostate cancer background showing Gleason Score 4 + 5. A cribriform morphological growth of the tumor was found in the normal glandular ducts with preserved basal cells, and IDCP was diagnosed (blue square: invasive cancer area; yellow square: IDCP area; yellow arrow: IDCP). (C,D) Spatial gene expression analysis (CytAssist Visium) classified the cells of the prostate tissue into 10 clusters. Of these 10 clusters, cluster 10 matched the IDCP region. (E,F) The clusters of invasive cancer lesions outside normal glandular ducts were close to those of IDCP on the pathology slides and were also close to IDCP clusters on the t-SNE plot, suggesting a similar gene expression pattern. (G) Trajectory analysis showed that IDCPs were similar in lineage to the neighboring invasive carcinomas. (H) The 20 most highly expressed genes in the IDCP (cluster 10) and non-IDCP regions (clusters 1–9). (I) The volcano plot shows the highly expressed genes in the IDCP and non-IDCP regions (H) Heat maps showed differences in gene expression between the IDCP and non-IDCP regions. (J) The heat map illustrates a clear distinction between gene expression in IDCP regions (cluster 10) and non-IDCP regions (clusters 1–9).

Figure 1

Diagnosis of IDCP and clustering through spatial gene expression analysis. (A) A 62-year-old man, diagnosed with prostate adenocarcinoma (cT3aN0M0, high risk, PSA 31.5 ng/mL), underwent a robot-assisted laparoscopic prostatectomy. HE staining revealed tumor invasion within normal glandular ducts with surrounding invasive carcinoma (blue square: invasive cancer area; yellow square: IDCP area; yellow arrow: IDCP). (B) Basal cell staining was performed on a formalin-fixed paraffin-embedded slide of the total prostatectomy specimen with a prostate cancer background showing Gleason Score 4 + 5. A cribriform morphological growth of the tumor was found in the normal glandular ducts with preserved basal cells, and IDCP was diagnosed (blue square: invasive cancer area; yellow square: IDCP area; yellow arrow: IDCP). (C,D) Spatial gene expression analysis (CytAssist Visium) classified the cells of the prostate tissue into 10 clusters. Of these 10 clusters, cluster 10 matched the IDCP region. (E,F) The clusters of invasive cancer lesions outside normal glandular ducts were close to those of IDCP on the pathology slides and were also close to IDCP clusters on the t-SNE plot, suggesting a similar gene expression pattern. (G) Trajectory analysis showed that IDCPs were similar in lineage to the neighboring invasive carcinomas. (H) The 20 most highly expressed genes in the IDCP (cluster 10) and non-IDCP regions (clusters 1–9). (I) The volcano plot shows the highly expressed genes in the IDCP and non-IDCP regions (H) Heat maps showed differences in gene expression between the IDCP and non-IDCP regions. (J) The heat map illustrates a clear distinction between gene expression in IDCP regions (cluster 10) and non-IDCP regions (clusters 1–9).

Figure 2

Visualization of the expression of epithelial marker genes, AR signature genes, and other upregulated genes in the IDCP region. (A,B) Spatial gene expression analysis showing that the epithelial markers were upregulated in all clusters (A), with similar findings in the violin plot (B). (C,D) Spatial gene expression analysis (C) and violin plot (D) showing the expression of a group of AR signature genes. (E,F) Spatial gene expression analysis (E) and violin plots (F) demonstrating the expression of MUC6, MYO16, NPY, and KLK12.

Figure 3

Visualization of the expression of HRR genes. (A,B) Spatial gene expression analysis (A) and violin plot (B) showing homologous recombination repair (HRR) gene expression. (C,D) Spatial gene expression analysis (C) and violin plot (D) showing TMPRSS2 and PTEN expression.

Figure 4

Visualization of the expression of fibroblast, immune cell, endothelial cell, and hypoxia markers. (A,B) Spatial gene expression analysis (A) and the violin plot (B) showing fibroblast marker gene expression. (C,D) Spatial gene expression analysis (C) and the violin plot (D) showing immune cell marker gene expression. (E,F) Spatial gene expression analysis (E) and the violin plot (F) showing endothelial cell marker gene expression. (G,H) Spatial gene expression analysis (G) and the violin plot (H) showing hypoxia marker gene expression. (I) Intraductal carcinoma of the prostate may enhance malignancy and increase resistance to therapy by causing hypoxia and a reduced recruitment of immune cells caused by its morphological features.

Figure 4

Visualization of the expression of fibroblast, immune cell, endothelial cell, and hypoxia markers. (A,B) Spatial gene expression analysis (A) and the violin plot (B) showing fibroblast marker gene expression. (C,D) Spatial gene expression analysis (C) and the violin plot (D) showing immune cell marker gene expression. (E,F) Spatial gene expression analysis (E) and the violin plot (F) showing endothelial cell marker gene expression. (G,H) Spatial gene expression analysis (G) and the violin plot (H) showing hypoxia marker gene expression. (I) Intraductal carcinoma of the prostate may enhance malignancy and increase resistance to therapy by causing hypoxia and a reduced recruitment of immune cells caused by its morphological features.