Increased tumor vascularization is associated with the amount of immune competent PD-1 positive cells in testicular germ cell tumors

Abstract

Testicular germ cell cancer in a metastatic state is curable with a cisplatin-based first line chemotherapy. However, 10-15% of these patients are resistant to first line chemotherapy and are thus left with only palliative options. Immunotherapies and inhibition of angiogenesis used in multiple types of cancer; however, the molecular context of angiogenesis and immune checkpoints in the development and progression of testicular cancers is still unknown. Therefore, the present study performed tissue micro array based analysis of 84 patients with immunohistochemistry of programmed cell death protein 1 (PD-1), programmed cell death ligand 1 (PD-L1) and vascular endothelial growth factor receptor 2 (VEGFR2) of testicular cancer and corresponding normal appearing testis tissue, matching the results with clinical data. The results demonstrated that PD-L1 was significantly upregulated in testicular tumors and that PD-1 positive cells significantly infiltrated the testicular tumor when compared with normal testicular tissue. VEGFR2 was significantly upregulated in testicular cancer. It was indicated that PD-1 expressing cytotoxic cells may require pathologic tumor vessels to pass the blood-testis-barrier in order to migrate into the tumor. Notably, when matching the clinical data for PD-1, PD-L1 and VEGFR2 there were no differences in expression in the different International Germ Cell Cancer Collaborative Group stages of non-seminoma. These data suggested that the anti-PD-1/PD-L1 immunotherapy and the anti-angiogenic therapy, sequentially or in combination, may be a promising option in the treatment of testicular cancer.

Keywords: immune; programmed cell death ligand 1; programmed cell death protein 1; testicular cancer; vascular endothelial growth factor receptor 2.

Figure 1.

Assurance of staining quality and specificity. (A) PD1-immunohistochemistry of normal appearing testicular tissue without primary antibody as a negative control. (B) PD1-immunohistochemistry of a tonsil as a positive control. Tonsil material was on every single tissue micro array as a positive control. (C) PD-L1-immunohistochemistry of normal appearing testicular tissue without primary antibody as a negative control. (D) PD-L1-immunohistochemistry of a tonsil as a positive control. Tonsil material was on every single tissue micro array as a positive control. (E) VEGFR2-immunohistochemistry of normal appearing testicular tissue without primary antibody as a negative control. (F) VEGFR2-immunohistochemistry of a tonsil as a positive control showing positive endothelial staining. Tonsil material was on every single tissue micro array as a positive control. Scale bars=100 µm. PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; VEGFR2, vascular endothelial growth factor receptor 2.

Figure 2.

Immunohistochemistry of PD-1, PD-L1 and VEGFR2 in human testicular cancer. (A) PD-1-immunohistochemistry of normal appearing testicular tissue. (B) PD-1-immunohistochemistry of a classical seminoma. (C) PD-1-immunohistochemistry of an embryonal testicular carcinoma. (D) PD-L1-immunohistochemistry of normal appearing testicular tissue. (E) PD-L1-immunohistochemistry of a classical seminoma. (F) PD-L1-immunohistochemistry of an embryonal testicular carcinoma. (G) VEGFR2-immunohistochemistry of normal appearing testicular tissue. (H) VEGFR2-immunohistochemistry of a classical seminoma. (I) VEGFR2-immunohistochemistry of an embryonal testicular carcinoma. Scale bars=50 µm. PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; VEGFR2, vascular endothelial growth factor receptor 2.

Figure 3.

Statistical analysis of immunohistochemical staining of tumor vs. normal tissues. (A) Contingency table of the presence of PD-1-immunopositive infiltrating cells (0=no, 1=yes) of normal appearing testicular tissue (n=15; no, 14/93.3%; yes, 1/6.67%) and testicular cancer (n=45; no, 27/60%; yes, 18/40%) P=0.0162. (B) Histogram of PD-L1 expression score of normal appearing testicular tissue (n=17; min: 0; max: 3; mean: 0) and testicular cancer (n=46; min, 0; max, 6; mean, 2.5), P<0.001. (C) Histogram of VEGFR2 expression score of normal appearing testicular tissue (n=14; min, 0; max, 3; mean, 1) and testicular cancer (n=67; min, 0; max, 3; mean, 2), P=0.002. PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; VEGFR2, vascular endothelial growth factor receptor 2; N, normal testicular tissues; T, testicular cancer tissues.

Figure 4.

Statistical analysis of immunohistochemical staining of tumor vs. normal tissues grouped in Seminoma and Non-Seminoma. (A) Contingency table of the presence of PD-1-immunopositive infiltrating cells (0=no, 1=yes) in seminoma of normal appearing testicular tissue (n=4; no, 4/100%; yes,/0%) and testicular cancer (n=23; no, 15/65.2%; yes, 8/34.8%) P=0.160. (B) Contingency table of the presence of PD-1-immunopositive infiltrating cells (0, no; 1, yes) in non-seminoma of normal appearing testicular tissue (n=11; no, 10/90.9%; yes, 1/9.1%) and testicular cancer (n=22; no, 12/54.5%; yes, 10/45.5%) P=0.037. (C) Histogram of the PD-L1 expression score of normal appearing testicular tissue of patients with seminomas (n=9; min: 0; max: 1; mean: 0) and seminoma tissue (n=25; min, 0; max, 6; mean, 3), P=0.002. (D) Histogram of the PD-L1 expression score of normal appearing testicular tissue of patients with non-seminomas (n=8; min, 0; max, 3; mean, 0) and non-seminoma tissue (n=21, min, 0; max, 6; mean, 2), P=0.008. (E) Histogram of the PD-1 expression score of normal appearing testicular tissue of patients with seminomas (n=8; min, 1; max, 2; mean, 1) and seminoma tissue (n=35; min, 0; max, 3; mean, 2), P=0.051. (F) Histogram of the PD-1 expression score of normal appearing testicular tissue of patients with non-seminomas (n=6; min, 0; max, 3; mean, 1.5) and non-seminoma tissue (n=32; min, 1; max, 3; mean, 3), P=0.002. PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; VEGFR2, vascular endothelial growth factor receptor 2; N, normal testicular tissues; T, testicular cancer tissues.

Figure 5.

Statistical analysis of immunohistochemical staining of Seminoma vs. Non-Seminoma. (A) Contingency table of the presence of PD-1-immunopositive infiltrating cells (0, no; 1, yes) of seminoma (n=23; no, 15/65.2%; yes, 8/34.8%) and non-seminoma (n=22; no, 12/54.5%; yes, 10/45.5%) P=0.465. (B) Histogram of the PD-L1 expression score of seminomas (n=25; min, 0; max, 6; mean, 3) and non-seminomas (n=21; min, 0; max, 6; mean, 2), P=0.507. (C) Histogram of the VEGFR2 expression score of seminomas (n=35; min, 0; max, 3; mean, 2) and non-seminomas (n=32; min: 1; max: 3; mean: 3), P=0.041. PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; VEGFR2, vascular endothelial growth factor receptor 2.

Figure 6.

Statistical analysis of immunohistochemical staining of the biomarkers in dependence of IGCCCG stage. (A) Contingency table of the presence of PD-1-immunopositive infiltrating cells (0=no, 1=yes) in the non-seminoma of different IGCCCG stages (stage 1: n=16; no, 10/62.5%; yes, 6/37.5%; stage 2: n=2; no, 0/0%; yes, 2/100%; stage 3: n=2; no, 0/0%; yes, 2/100%) P=0.82. (B) Histogram of the PD-L1 expression score of IGCCCG good prognosis (stage 1: n=13; min, 0; max, 6; mean, 3), IGCCCG intermediate prognosis (stage 2: n=3; min, 2; max, 4; mean, 2) and IGCCCG poor prognosis (stage 3: n=2; min, 2; max, 3; mean, 2.5) P=0.968. (C) Histogram of VEGFR2 expression score of IGCCCG good prognosis (1: n=22, min: 1; max: 3; mean: 3), IGCCCG intermediate prognosis (stage 2: n=4; min, 1; max, 3; mean, 2) and IGCCCG poor prognosis (stage 3: n=2; min, 2; max, 3; mean, 2.5) P=0.248. IGCCCG, International Germ Cell Cancer Collaborative Group; PD-1, programmed cell death protein 1; PD-L1, programmed cell death ligand 1; VEGFR2, vascular endothelial growth factor receptor 2.