Abstract
Programmed cell death 1/programmed cell death ligand (PD-1/PD-Ls) axis is crucial for the modulation of immune responses and self-tolerance. Also, aberrant PD-L1 expression on the tumor cells or tumor-associated inflammatory cells accelerates immune evasion of tumor cells. In the past decade, PD-1/PD-Ls immune checkpoint inhibitors were introduced to cancer treatment trials and, in some cases, showed significant anti-cancer effects. PD-L1 immunohistochemical staining is considered a potential predictor of clinical response to PD-1/PD-Ls immune checkpoint inhibitors treatment. However, immunohistochemical data on PD-L1 expression in different types of cancer especially rare entities remains incomplete. In this study, PD-L1 expression was immunohistochemically analyzed in 5536 tumors including germ cell, epithelial, mesenchymal, melanocytic/neuroectodermal, and lymphohematopoietic tumors as well as in a set of human normal tissues including a fetus. Immunohistochemistry was performed with E1L3N rabbit monoclonal antibody and Leica Bond Max automation using multitumor blocks containing up to 70 tumor samples. PD-L1 was constitutively and strongly expressed in placental trophoblasts as well as choriocarcinomas and trophoblastic components of germ cell tumors. Also, the neoplastic cells of classical Hodgkin’s lymphoma, anaplastic large cell lymphoma, schwannoma, thymoma, and squamous cell carcinoma of various sites frequently expressed PD-L1. In gastrointestinal adenocarcinomas, PD-L1-expression was associated with EBER-positivity and mismatch repair-deficiency. In addition, PD-L1 was variably expressed in non-neoplastic macrophages and dendritic cells. PD-L1 immunohistochemistry may have some role in immunophenotypic differential diagnosis of tumors and pinpointing potential candidates for anti-PD-1/PD-Ls immune checkpoint therapy.
Keywords: programmed cell death ligand 1 (PD-L1), immunohistochemistry, choriocarcinoma, classical Hodgkin’s lymphoma, anaplastic large cell lymphoma (ALCL), schwannoma, mismatch repair deficiency, Epstein-Barr virus
INTRODUCTION
Programmed cell death 1/programmed cell death ligand (PD-1/PD-Ls) axis is crucial for the modulation of the immune system to reduce collateral tissue damage from the inflammatory response to infectious microorganisms in peripheral tissues. In addition, direct evidence that PD-1-deficient mice develop spontaneous autoimmune diseases suggests an inhibitory and regulatory role for PD-1/PD-Ls interaction in T-cell responses and the maintenance of self-tolerance. (1, 2)
PD-1 (CD279) is a cell surface receptor that belongs to the immunoglobulin superfamily and also a member of the extended CD28/CTLA-4 family. The extracellular region of PD-1 is 28% identical to CTLA-4, another immune checkpoint molecule. (1, 3) PD-1 is mainly expressed on activated T cells whereas other non-T lymphocytes such as B cells and natural killer (NK) cells express PD-1 only upon induction. (4, 5) PD-1 is a co-stimulatory molecule and, once engaged by its ligands PD-Ls, PD-1 (probably along with the other cooperative pathways) inhibits kinases that are involved in T cell activation through the phosphatase SHP2. (6) When T cells were exposed to chronic antigen stimulation such as chronic viral infection or cancer, high levels of persistent PD-1 expression is induced and leads to T cell exhaustion or anergy.
PD-L1 (B7-H1, CD274) and PD-L2 (B7-DC, CD273), two physiological ligands for PD-1, have been identified as cell-surface glycoproteins belonging to the B7 family. (6–9) These B7 family members share 37% sequence homology and are located within 100kb of the genomic locus at 9p24, probably evolving through gene duplication. (9) In peripheral tissues, PD-L1 is primarily induced by interferon-γ (IFNγ), from T helper 1 (TH1) cells under inflammatory conditions.
In early studies, melanoma, ovarian, colon and lung cancer tissues were reported to express high levels of PD-L1. (10) Subsequently, other human cancers such as esophageal, gastric, and hepatocellular carcinoma were reported to express PD-L1. Moreover, PD-L1 expression was linked to tumor aggressiveness and a poor clinical prognosis. (11–13) It was reported that aberrantly activated oncogenic signals induce PD-L1 expression on tumor cells and allow them to evade the host immune system by dampening antitumor immunity through anergy or apoptosis of antigen-specific T-cells. (14, 15) Furthermore, PD-L1-expressing tumor-associated inflammatory cells (TAIs, those includes macrophages and lymphocytes infiltrating and/or surrounding neoplastic tissue) are also reported to induce tumor immune evasion. (16–18)
In the past decade, several immune checkpoint inhibitors (nivolumab and pembrolizumab for PD-1/PD-Ls inhibition; tremelimumab and ipilimumab for CTLA-4/B7 inhibition) were introduced for the treatment of several cancer types to conquer tumor immune evasion. Recently, significant anti-cancer effects of nivolumab were demonstrated in phase 3 clinical trials for the treatment of non-small cell lung cancer (NSCLC) as well as other malignancies. (19–22) PD-L1 expression, as detected by immunohistochemistry, is a potential biomarker to predict clinical response to PD-1/PD-Ls checkpoint inhibitors in the clinical setting. (19–22) Several types of cancer have been analyzed for PD-L1 expression, but a comprehensive analyses, including rare germ cell and mesenchymal tumors has not been performed. (10–13, 23–25)
The aim of this study was to evaluate potential utility of PD-L1 immunohistochemistry in diagnostic pathology and identify additional tumor types for PD-1/PD-Ls axis inhibitor treatment.
MATERIAL and METHODS
5536 anonymized tumors, including germ cell, epithelial, mesenchymal, melanocytic/neuroectodermal, and lymphohematopoietic tumors, and normal tissues derived from surgical specimens were assembled to multitumor blocks containing 30 to 70 rectangular tissue samples as previously described. (26) The size of tumor tissue samples was estimated to exceed the size of a single 0.6mm2 core by a factor of 10–15. All tumors, selected for this study, were extensively characterized histologically and immunohistochemically. In addition, a 10-week-old fetus was immunohistochemically analyzed.
The rabbit monoclonal antibodies clone E1L3N (#13684) and 28-8 (ab205921) against PD-L1 were obtained from Cell Signaling Technology, Inc. (Danvers, MA) and Abcam Inc. (Cambridge, MA) , respectively. The antibodies were applied at a dilution of 1:200 (E1L3N) and 1:500 (28–8) to selected normal and tumor tissue arrays. Both of the antibodies showed almost similar staining patterns in placental trophoblast, choriocarcinoma, anaplastic large cell lymphoma (ALCL), and squamous cell carcinoma of lung. However, we selected clone E1L3N because of its lower background signal.
Immunostaining was performed with the Leica Bond-Max automation and Leica Refine detection kit. (Leica Biosystems, Bannockburn, IL) The protocol included in situ deparaffinization and high-pH epitope retrieval for 25 minutes, incubation with primary antibody for 30 minutes, polymer for 15 minutes, postpolymer for 15 minutes, and DAB as the chromogen for 10 minutes, followed by 5-minute hematoxylin counterstaining. MLH1, MSH2, MSH6, and PMS3 immunohistochemistry was performed to analyze mismatch repair (MMR) system status as previously reported. (27) For the detection of Epstein-Barr virus (EBV) infection, Bond™ Ready-to-Use ISH EBER Probe was used in Leica Bond-Max automation system according to the manufacturer instructions. (Leica Biosystems, Bannockburn, IL)
The stained sections were independently evaluated by two pathologists (SI and MM). PD-L1 immunoreactivity in placental trophoblasts and peripheral nerves were used as external and internal positive controls, respectively. PD-L1 has been reported to be expressed on not only tumor cells but also dendritic cells and TAIs, therefore, we evaluated PD-L1 expression in both neoplastic cells and TAIs with a detection cut-off of 5%.
Chi-square test or Fisher’s exact test were performed by SPSS software (IBM, Armonk, NY) to analyze the statistical correlation between PD-L1-expression and other tumor status such as MMR-deficiency, EBER-positivity, p16-, and ALK-expression.
RESULTS
Normal tissues and fetus
Among the normal adult tissue components, highest PD-L1-expression was detected in placental trophoblasts. (Fig. 1A) Macrophages such as alveolar ones, dendritic cells of lymphoid tissue, peripheral nerves probably schwann cells, smooth, and skeletal muscle cells showed variable PD-L1 positivity. (Fig. 1B) Organs such as adrenal gland, brain, genitourinary-tract, kidney, liver, lung, mammary gland, pancreas, salivary gland, and thyroid gland showed no PD-L1 expression. Fetal tissues with the exception of placental trophoblasts were also negative for PD-L1 expression.
FIGURE 1. CD171 expression in normal tissue.
A, Placental trophoblast cells strongly expressed PD-L1 on the cell membrane. B, Peripheral nerve bundles and smooth muscle cells of the stomach showed weak PD-L1 expression.
Germ cell tumors
The results of PD-L1 immunostaining in germ cell tumors have been summarized in Table 1. Choriocarcinomas were nearly uniformly positive for PD-L1 with a membrane pattern (Fig. 2A). Similarly, 1 placental site trophoblastic tumor and 6 mixed germ cell tumors of testis showed PD-L1 positivity in the trophoblastic cellular elements. Other germ cell tumors, such as seminoma, embryonal carcinoma, and yolk sac tumor showed PD-L1 expression (57–81%), only in TAIs, especially macrophages. (Figure 2B, Supplementary Table S1) Teratoma cells showed no PD-L1 expression.
Table 1.
PD-L1 expression in germ cell tumors
| Cases | / | All | (%) | Range (Median) | |
|---|---|---|---|---|---|
| Uterus, choriocarcinoma | 8 | / | 8 | 100.0 | 90–100 (100) |
| Germ cell tumor with trophoblastic elements* | 8 | / | 8 | 100.0 | 10–100 (50) |
| Testis, embryonal carcinoma | 1 | / | 34 | 2.9 | 5 |
| Testis, seminoma | 0 | / | 91 | 0.0 | - |
| Testis, teratoma | 0 | / | 17 | 0.0 | - |
| Testis, Yolk sac tumor | 0 | / | 7 | 0.0 | - |
These include 1 placental site trophoblastic tumor and 6 testicular germ cell tumors with trophoblastic differentiation.
FIGURE 2. PD-L1 expression in germ cell tumors.
A, Choriocarcinomas showed strong PD-L1 expression on the cell membrane. B, In seminomas, neoplastic cells expressed no PD-L1 whereas tumor-associated inflammatory cells showed strong PD-L1 expression.
Other epithelial neoplasms
Expression status of PD-L1 in epithelial non germ-cell tumors is shown in Table 2. Thymic tumors showed frequent PD-L1 expression. In thymomas, 79% of cases showed PD-L1 positivity in neoplastic epithelial cells, whereas, tumor-associated thymocytes were negative for PD-L1. (Fig. 3A, Table 2) Thymic carcinoma cells showed lower PD-L1 positivity than thymomas (38% of cases). (Table 2)
Table 2.
PD-L1 expression in epithelial tumors
| Cases | / | All | (%) | Range (Median) | |
|---|---|---|---|---|---|
| Tumors of squamous cell differentiation | |||||
| Thymoma | 26 | / | 33 | 78.8 | 5–100 (100) |
| Tongue, squamous cell carcinoma | 26 | / | 39 | 66.7 | 5–100 (70) |
| Tonsil, squamous cell carcinoma | 25 | / | 41 | 61.0 | 5–100 (50) |
| Skin, squamous cell carcinoma | 20 | / | 33 | 60.6 | 5–100 (50) |
| Lung, squamous cell carcinoma | 33 | / | 56 | 58.9 | 5–100 (80) |
| Gingiva, squamous cell carcinoma | 25 | / | 43 | 58.1 | 5–100 (80) |
| Esophagus, squamous cell carcinoma | 38 | / | 98 | 38.8 | 5–100 (20) |
| Thymic carcinoma | 6 | / | 16 | 37.5 | 60–100 (100) |
| Tumors of adenocarcinoma-like differentiation | |||||
| Lung, adenocarcinoma | 30 | / | 137 | 21.9 | 5–100 (75) |
| Pancreas, invasive ductal carcinoma | 35 | / | 224 | 15.6 | 5–100 (60) |
| Cholangiocellular carcinoma | 6 | / | 41 | 14.6 | 20–100 (90) |
| Stomach, adenocarcinoma | 27 | / | 190 | 14.2 | 5–100 (40) |
| Colorectum, adenocarcinoma | 61 | / | 478 | 12.8 | 5–100 (40) |
| Uterus, endometrioid adenocarcinoma | 25 | / | 192 | 13.0 | 5–100 (10) |
| Breast, invasive ductal carcinoma | 20 | / | 366 | 5.5 | 5–100 (60) |
| Ovary, endometrioid adenocarcinoma | 2 | / | 45 | 4.4 | 10–40 (25) |
| Ovary, serous carcinoma | 2 | / | 50 | 4.0 | 10–90 (50) |
| Prostate, adenocarcinoma | 1 | / | 28 | 3.6 | 100 |
| Hepatocellular carcinoma | 4 | / | 128 | 3.1 | 10–90 (40) |
| Breast, invasive lobular carcinoma | 1 | / | 56 | 1.8 | 5 |
| Other epithelial tumors | |||||
| Thyroid gland, papillary carcinoma | 16 | / | 44 | 36.4 | 10–100 (30) |
| Malignant mesothelioma | 56 | / | 186 | 30.1 | 5–100 (80) |
| Uterus, carcinosarcoma | 14 | / | 52 | 26.9 | 5–100 (35) |
| Urothelial Carcinoma | 47 | / | 185 | 25.4 | 5–100 (80) |
| Kidney, chromophobe renal cell carcinoma | 15 | / | 86 | 17.4 | 10–100 (100) |
| Kidney, papillary renal cell carcinoma | 10 | / | 96 | 10.4 | 10–100 (60) |
| Lung, small cell carcinoma | 4 | / | 48 | 8.3 | 40–100 (90) |
| Skin, Merkel cell carcinoma | 2 | / | 27 | 7.4 | 100 (100) |
| Kidney, clear cell renal cell carcinoma | 15 | / | 287 | 5.2 | 5–100 (85) |
FIGURE 3. PD-L1 expression in epithelial tumors.
A, Neoplastic epithelial cells of B2 thymoma showed strong PD-L1 expression on the cell membrane. B, Membranous PD-L1 expression in squamous cell carcinoma of tonsil. C and D, MMR-deficient colorectal adenocarcinoma showed strong PD-L1 expression. (C) PD-L1 staining. (D) MLH1 staining. E and F, EBER-positive gastric adenocarcinoma showed PD-L1 expression. (E) PD-L1 staining. (F) EBER in situ hybridization.
Squamous cell carcinoma cells of the tongue and tonsils (Fig. 3B) showed higher PD-L1 labeling (61–67% of cases), while those of esophagus showed lowest PD-L1 expression (39% of cases). No statistical correlation was detected between p16- and PD-L1-expression in any of the oral squamous cell carcinomas. (Supplementary Table S2) The former was most often detected in tonsillar carcinoma.
PD-L1 was only rarely expressed in colorectal adenocarcinoma cells (13% of cases, with 5–100% of positive cells, median 40%). (Figure 3C and Table 2) 17% of colorectal adenocarcinomas showed MMR-deficiency. (Figure 3D) Statistically positive correlation was detected between the MMR-deficiency and PD-L1-expression in colorectal adenocarcinoma. (Table 3)
Table 3.
PD-L1 expression in colorectal adenocarcinomas
| Colorectal adenocarcinoma subtypes (n=506) |
PD-L1 (tumor cells; %) |
|---|---|
| MMR-preserved (n=424) | 6.8 |
| MMR-deficient (n=82) | 44.7** |
Chi-square test,
p<0.01
Gastric adenocarcinoma cells also showed rare PD-L1 expression (14% of cases, 5–100% of positive cells, median 40%, Fig. 3E and Table 2). In this study, 8% (15 of 180 cases) and 6% (11 of 190 cases) of gastric adenocarcinomas were determined as MMR-deficienct and EBER-positive tumors, respectively. (Fig. 3F) MMR-deficiency and EBER-positivity were detected in a mutually exclusive manner. Statistically positive correlation was detected between the PD-L1-expression and MMR-deficienct or EBER-positive status. (Table 4)
Table 4.
PD-L1 expression in gastric adenocarcinomas
| Gastric adenocarcinoma subtypes (n=180 in MMR analysis; n=190 in EBV analysis) |
PD-L1 (tumor cells; %) |
|---|---|
| MMR-preserved (n=165) | 12.1 |
| MMR-deficient (n=15) | 46.7** |
| EBER-negative (n=179) | 12.3 |
| EBER-positive (n=11) | 45.5* |
Fisher’s exact test,
p<0.01,
p<0.05
Neoplastic cells of endometrioido adenocarcinoma and serous carcinoma of the ovary, invasive ductal and lobular carcinomas of the breast, hepatocellular carcinoma, and prostate adenocarcinoma only rarely expressed PD-L1 (2–6% of cases). (Table 2)
In other epithelial tumors, PD-L1 expression was detected in neoplastic cells (5–36% of cases) or TAIs (1–59%). TAIs of Merkel cell carcinoma showed a high frequency of PD-L1 expression (59% of cases). (Supplementary Table S1)
Hematopoietic and lymphoid tumors
PD-L1 expression in lymphohematopoietic tumors has been summarized in Table 5. 96% of classical Hodgkin’s lymphoma cases showed PD-L1 expression in neoplastic Hodgkin’s and Reed-Sternberg cells (10–100% of positive cells, median 80%, Fig. 4A). In this study, 30% (14 of 47 cases) of classical Hodgkin’s lymphoma showed EBER-positivity. Although all of the EBER-positive cases showed PD-L1 expression, no statistical correlation was detected between EBER- and PD-L1-expression. (Supplementary Table S3)
Table 5.
PD-L1 expression in tumors of hematopoietic system
| Cases | / | All | (%) | Range (Median) | |
|---|---|---|---|---|---|
| Classical Hodgkin’s lymphoma | 45 | / | 47 | 95.7 | 10–100 (80) |
| Anaplastic large cell lymphoma | 13 | / | 15 | 86.7 | 10–100 (100) |
| Myeloid sarcoma | 3 | / | 13 | 23.1 | 10–30 (20) |
| Diffuse large B-cell lymphoma | 15 | / | 84 | 17.9 | 5–100 (60) |
| T, NK-cell lymphoma | 5 | / | 31 | 16.1 | 40–100 (100) |
| Follicular lymphoma | 0 | / | 51 | 0.0 | |
| Lymphoblastic lymphoma | 0 | / | 8 | 0.0 | |
| Mantle cell lymphoma | 0 | / | 32 | 0.0 | |
| Nodal marginal zone lymphoma | 0 | / | 6 | 0.0 | |
| Small cell lymphoma | 0 | / | 19 | 0.0 | |
| Spleen, chronic myeloid leukemia | 0 | / | 6 | 0.0 |
FIGURE 4. PD-L1 expression in lymphomas.
A, PD-L1 expression in classical Hodgkin’s lymphoma. B, Anaplastic large cell lymphoma showed strong and diffuse PD-L1 expression.
ALCL cells also showed frequent and strong PD-L1 expression on the cell membrane (87% of cases, 10–100% of positive cells, median 100%, Fig. 4B). No statistical correlation was detected between ALK- and PD-L1-expression. (Supplementary Table S4)
A minor populations of diffuse large B-cell lymphoma (DLBCL) and NK- and T-cell lymphoma cells showed positivity for PD-L1 (18% and 16% of cases, respectively). Although only 2 cases of DLBCL showed EBER-positivity, statistical correlation was detected between EBER- and PD-L1-expression. (Table 6) Neoplastic cells of mantle cell lymphoma, nodal marginal zone lymphoma, T- and B-lymphoblastic lymphoma, follicular lymphoma, and small cell lymphoma showed no PD-L1 expression. Conversely, lymphoma-associated inflammatory cells showed frequent PD-L1 expression (29–75% of cases), except for small cell lymphoma. (Supplementary Table S1)
Table 6.
PD-L1 expression in diffuse large B-cell lymphoma (DLBCL)
| DLBCL subtypes (n=84) |
PD-L1 (tumor cells; %) |
|---|---|
| EBV-negative (n=82) | 15.9 |
| EBV-infected (n=2) |
100.0* |
Fisher’s exact test,
p<0.05
Mesenchymal and neuroectodermal tumors
PD-L1 expression in mesenchymal tumors has been summarized in Table 7. Among mesenchymal tumors, the highest PD-L1 positivity was detected in peripheral schwannomas (89% of cases, 5–100% of positive cells, median 90%, Fig. 5A). Gastric shwannomas showed less consistent and weaker PD-L1 expression than peripheral schwannomas. (67% of cases, 5–100% of positive cells, median 90%) TAIs in schwannoma were negative for PD-L1. (Supplementary Table S1) Neurofibroma showed less frequent positivity (44% of cases, 5–100% of positive cells, median 70%) while granular cell tumors were negative for PD-L1. Potential schwannoma mimics such as solitary fibrous tumor, fibrous histiocytoma, and perineurioma were all negative for PD-L1. In both malignant peripheral nerve sheath tumor (MPNST, Fig. 5B) and neuroblastoma, 21% of cases showed PD-L1 expression. In neuroblastomas, PD-L1 expression was limited to the schwannian stroma.
Table 7.
PD-L1 expression in mesenchymal tumors
| Cases | / | All | (%) | Range (Median) | |
|---|---|---|---|---|---|
| Peripheral, schwannoma | 58 | / | 65 | 89.2 | 5–100 (90) |
| Stomach, schwannoma | 4 | / | 6 | 66.7 | 5–100 (90) |
| Neurofibroma | 17 | / | 39 | 43.6 | 5–100 (70) |
| MPNST | 12 | / | 57 | 21.1 | 5–100 (50) |
| Neuroblastoma | 5 | / | 29 | 17.2 | 10–95 (95) |
| Embryonal rhabdomyosarcoma | 9 | / | 54 | 16.7 | 10–80 (40) |
| Uterus, leiomyoma | 18 | / | 125 | 14.4 | 5–80(20) |
| Meningioma | 18 | / | 192 | 9.4 | 5–100 (50) |
| Stomach, SDH-difficient GIST | 5 | / | 55 | 9.1 | 30–100 (90) |
| Alveolar soft part sarcoma | 1 | / | 11 | 9.1 | 100 |
| Desmoplastic melanoma | 3 | / | 33 | 9.1 | 20–80 (40) |
| Nephroblastoma | 2 | / | 24 | 8.3 | 30–50 (40) |
| Kaposi’s sarcoma | 3 | / | 37 | 8.1 | 30–90 (40) |
| Angiosarcoma | 7 | / | 100 | 7.0 | 10–100 (40) |
| Malignant melanoma | 5 | / | 78 | 6.4 | 20–100 (40) |
| Small intestine, GIST | 7 | / | 111 | 6.3 | 5–100 (50) |
| Leiomyosarcoma | 4 | / | 69 | 5.8 | 40–100 (80) |
| Stomach, conventional GIST | 6 | / | 218 | 2.8 | 40–100 (100) |
| PEComa | 2 | / | 75 | 2.7 | 10–30 (20) |
| Ewing’s sarcoma/PNET | 1 | / | 38 | 2.6 | 20 |
| Alveolar rhabdomyosarcoma | 1 | / | 42 | 2.4 | 20 |
| Dermatofibrosarcoma protuberans | 1 | / | 44 | 2.3 | 100 |
| Benign fibrous histiocytoma | 0 | / | 59 | 0.0 | - |
| Desmoplastic small round cell tumor | 0 | / | 14 | 0.0 | - |
| Glomus tumor | 0 | / | 58 | 0.0 | - |
| Granular cell tumor | 0 | / | 28 | 0.0 | - |
| Ovary, fibroma and thecoma | 0 | / | 31 | 0.0 | - |
| Perineurioma | 0 | / | 6 | 0.0 | - |
| Solitary fibrous tumor | 0 | / | 27 | 0.0 | - |
| Synovial sarcoma | 0 | / | 34 | 0.0 | - |
MPNST: malignant peripheral nerve sheath tumor, GIST: gastrointestinal stromal tumor, PNET: primitive neuroectodermal tumor.
FIGURE 5. PD-L1 expression in mesenchymal tumors.
A, Schwannoma showed diffuse cytoplasmic PD-L1 expression. B, Occasional cases of malignant peripheral nerve sheath tumors were weakly positive for PD-L1. C, A subpopulation of embryonal rhabdomyosarcoma cells showed faint PD-L1 expression. D, A case of alveolar soft part sarcoma showed weak but diffuse membranous PD-L1 expression.
Minor populations of embryonal rhabdomyosarcoma cells showed weak PD-L1 expression (17% of cases, 10–80% of positive cells, median 40%, Fig. 5C). An isolated case of alveolar soft part sarcoma (9% of cases) showed membranous PD-L1 expression. (Fig. 5D) Rare malignant melanoma 17% of cases) and SDH-deficient GIST (13% of cases) expressed PD-L1. Other soft tissue tumors such as glomus tumor, and synovial sarcoma were negative for PD-L1. In mesenchymal tumors, TAIs also showed much lower PD-L1 expression compared to carcinomas and lymphomas. (Supplementary Table S1)
DISCUSSION
In physiological conditions, the PD-1/PD-Ls axis is crucial for the regulation of immune responses to minimize collateral tissue damage as well as maintaining self-tolerance. However, in neoplastic disease, once PD-Ls are aberrantly expressed on tumor cells and/or TAIs, they help tumor cells evade the host immune system and promote tumor growth. In the past decade, PD-1/PD-Ls immune checkpoint inhibitors have been introduced in anti-cancer therapy. (19–22) To predict a clinical response to these drugs, PD-L1 immunohistochemical staining has been used as a potential biomarker. (19–22) In this study, PD-L1 expression was immunohistochemically analyzed in normal and neoplastic tissue to assess potential diagnostic utility of PD-L1 immunohistochemistry and to identify specific neoplasms that may be amenable to anti-PD-1/PD-Ls immune checkpoint therapy.
In normal tissues, the most striking PD-L1-expression was detected in placental trophoblast (Fig. 1A). Interestingly, the fetus showed no PD-L1 expression. These results reflect the nature of the trophoblast as a maternal-fetal barrier protecting the fetus from the maternal immune system. (28, 29) Indeed, critical roles for the PD-L1 in fetomaternal tolerance have been demonstrated in a mouse model. PD-L1-deficient females exhibited decreased allogeneic fetal survival rates compared to littermates and heterozygote controls. (30) It is not surprising that choriocarcinoma, a malignant trophoblastic tumor, showed strong PD-L1 expression among germ cell tumors. These observations indicate that PD-L1 could be a supplementary marker to identify choriocarcinoma or trophoblastic elements in other germ cell tumors.
PD-L1 was consistently expressed in normal peripheral nerves. The physiological roles of PD-L1 in nervous tissue have not fully explored. However, PD-L1 was indicated as a suppressor for the inflammatory response and neuropathic pain occurring after peripheral nerve injury in PD-L1 knock out mice. (31) Among neoplasms, schwannomas showed consistent PD-L1 expression, whereas neurofibroma and MPNST showed rarer and weaker expression. Schwannoma mimics such as gastrointestinal stromal tumor (GIST), fibrous histiocytoma, perineurioma, solitary fibrous tumor, and meningioma were almost always negative. These results indicate that PD-L1 could have a role in the differential diagnosis of schwannoma. However, it should be considered that minority of melanotic and desmoplastic melanomas are also positive for PD-L1.
In analyzed thymic tumors, 79% and 38% of thymoma and thymic carcinoma cells showed PD-L1 expression, respectively. (Table 2) Recently, two groups reported frequent PD-L1 expression in thymic tumors. (24, 25) One study showed that 68% of thymoma and 75% of thymic carcinoma neoplastic cells were positive for PD-L1 using rabbit monoclonal antibody 15 from Sino Biological (Beijing, China). This study also showed that PD-L1 high-expressing thymic tumors had a significantly worse overall survival in an adjusted (age/sex) analysis. (25) However, other study using E1L3N rabbit monoclonal antibody showed that that 23% of thymomas and 70% of thymic carcinomas were positive for PD-L1 and PD-L1-positivity was not a significantly negative factor for overall survival in a multivariate analysis. (24). Thus, additional studies are needed to explain these divergent results.
Recently, gastric cancers were classified into 4 categories such as EBV-infected, microsatellite unstable, genomically stable, and chromosomally unstable tumors, by the comprehensive molecular analysis by the Cancer Genome Atlas (TCGA) project. (33) It was also found that EBV-infected tumors carry PIK3CA mutations, extreme DNA hypermethylation, and amplification of 9p24 region including JAK2, PD-L1 and PD-L2. This study confirmed the positive correlation of EBER- and PD-L1-expression by typical methods such as in situ hybridization and immunohistochemistry. (Table 4) Our study also showed a positive correlation between MMR-deficiency and PD-L1-expression (Table 4) and that only 11% of other two types (genomically stable and chromosomally unstable tumors) were positive for PD-L1.
Aberrantly activated oncogenic signals due to PTEN-loss, EGFR-mutation, or ALK-translocation were reported to induce PD-L1 overexpression in neoplastic cells. (14, 15, 32) It was also reported that ALCLs, carrying nucleophosmin (NPM)/anaplastic lymphoma kinase (ALK) translocation, were induced to PD-L1 overexpression via the NPM/ALK-STAT3 axis activation. (14) However, no correlation between PD-L1- and ALK-expression statuses was demonstrated in this study. (Supplementary Table S4) Moreover, 9 of 10 ALK-negative ALCLs also showed strong PD-L1 expression. These results strongly indicated that there could be alternative pathway(s) regulating PD-L1-expression in ALCLs.
EBV is significantly associated with classical Hodgkin’s lymphoma. (34) It was reported that the induction of the EBV latent membrane proteins, latent membrane protein 1 (LMP1) or LMP2a, in normal germinal center B cells is sufficient to mimic a Hodgkin’s Reed-Sternberg cell-like phenotype. (35, 36) Furthermore, LMP1 was reported to increase PD-L1 expression by up-regulating its promoter activity via a JAK3-dependent manner. (37) Thus, EBER-positive classical Hodgkin’s lymphomas would be induced into PD-L1 expression via this pathway. (Supplementary Table S3) On the other hand, EBER-negative classical Hodgkin’s lymphoma cases also showed frequent PD-L1 expression (94%, 31 of 33 cases, Supplementary Table S3). It was reported that genomic amplification of 9p24 region including PD-L1 leads to PD-L1 expression in Hodgkin’s lymphoma cells. (38) These EBER-negative classical Hodgkin’s lymphoma cases might carry genomic amplification of 9p24 region.
In other viral infections, HPV-infection was reported to correlate with PD-L1-expression in squamous cell carcinomas of tonsil. (39) In this study, 90% and 93% of tonsil squamous cell carcinoma showed PD-L1 and p-16-expression, respectively. However, no statistical correlation was detected between PD-L1- and p16-expression. (Table 2 and Supplementary Table S2)
It has been reported that PD-L1-expressing dendritic cells or TAIs are able to induce tumor immune evasion. (16–18) In current study, seminoma and various carcinomas often showed such PD-L1-positive cells whereas mesenchymal tumors were less frequently associated with PD-L1-expressing inflammatory cells. (Supplementary Table S1) Clinical or experimental investigation is needed to determine whether tumors with PD-L1-positve TAIs could be targets for immune check point inhibition therapy.
In clinical trials, PD-1/PD-Ls inhibitors were introduced to the treatment of the patients with PD-L1 expressing tumors, such as melanoma, NSCLC, renal cell cancer, and Hodgkin’s Lymphoma. (19–22) Recently, advanced squamous-cell and other non-squamous-cell NSCLC patients were treated with nivolumab or docetaxel to compare their antitumor activity. (20, 21) Both squamous-cell and non-squamous-cell NSCLC patients treated by nivolumab showed significantly better overall survival, response rate, and progression-free survival than docetaxel treated patients. However, the hazard ratio for death was lower in squamous-cell carcinoma patients [0.59 (95% CI, 0.44–0.79), P<0.001] than non-squamous cell NSCLC patient [0.73 (96% CI, 0.59–0.89), P=0.002] indicating better treatment success for squamous cell NSCLC patients. In addition, refractory Hodgkin’s lymphoma patients showed a response to nivolumab treatment. (19) These results indicate that PD-L1-expressing tumors, such as germ cell tumors with trophoblastic differentiation and MPNSTs, might be successfully treated by PD-1/PD-Ls checkpoint inhibitors.
In conclusion, PD-L1 immunohistochemistry using rabbit monoclonal antibody clone E1L3N might have a role in diagnostic immunophenotyping of germ cell tumors with trophoblastic differentiation, schwannoma, thymoma, classical Hodgkin’s and ALCL because of their consistent PD-L1 positivity. Our data presented here may also be useful in identifying potential targets for anti-PD-1/PD-Ls immune checkpoint inhibition therapy.
Supplementary Material
Acknowledgments
Source of Fundings: Supported as a part of National Cancer Institute’s intramural research program.
Footnotes
Conflicts of Interest: The authors have disclosed that they have no relationships with, or financial interest in, any commercial companies pertaining to this article.
REFERENCES
- 1.Nishimura H, Honjo T. PD-1: an inhibitory immunoreceptor involved in peripheral tolerance. Trends Immunol. 2001;22:265–268. doi: 10.1016/s1471-4906(01)01888-9. [DOI] [PubMed] [Google Scholar]
- 2.Nishimura H, Nose M, Hiai H, et al. Development of lupus-like autoimmune diseases by disruption of the PD-1 gene encoding an ITIM motif-carrying immunoreceptor. Immunity. 1999;11:141–151. doi: 10.1016/s1074-7613(00)80089-8. [DOI] [PubMed] [Google Scholar]
- 3.Ishida Y, Agata Y, Shibahara K, et al. Induced expression of PD-1, a novel member of the immunoglobulin gene superfamily, upon programmed cell death. EMBO J. 1992;11:3887–3895. doi: 10.1002/j.1460-2075.1992.tb05481.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Fanoni D, Tavecchio S, Recalcati S, et al. New monoclonal antibodies against B-cell antigens: possible new strategies for diagnosis of primary cutaneous B-cell lymphomas. Immunol Lett. 2011;134:157–160. doi: 10.1016/j.imlet.2010.09.022. [DOI] [PubMed] [Google Scholar]
- 5.Terme M, Ullrich E, Aymeric L, et al. IL-18 induces PD-1-dependent immunosuppression in cancer. Cancer Res. 2011;71:5393–5399. doi: 10.1158/0008-5472.CAN-11-0993. [DOI] [PubMed] [Google Scholar]
- 6.Freeman GJ, Long AJ, Iwai Y, et al. Engagement of the PD-1 immunoinhibitory receptor by a novel B7 family member leads to negative regulation of lymphocyte activation. J Exp Med. 2000;192:1027–1034. doi: 10.1084/jem.192.7.1027. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Dong H, Zhu G, Tamada K, et al. B7-H1, a third member of the B7 family, co-stimulates T-cell proliferation and interleukin-10 secretion. Nat Med. 1999;5:1365–1369. doi: 10.1038/70932. [DOI] [PubMed] [Google Scholar]
- 8.Latchman Y, Wood CR, Chernova T, et al. PD-L2 is a second ligand for PD-1 and inhibits T cell activation. Nat Immunol. 2001;2:261–268. doi: 10.1038/85330. [DOI] [PubMed] [Google Scholar]
- 9.Tseng SY, Otsuji M, Gorski K, et al. B7-DC, a new dendritic cell molecule with potent costimulatory properties for T cells. J Exp Med. 2001;193:839–846. doi: 10.1084/jem.193.7.839. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Dong H, Strome SE, Salomao DR, et al. Tumor-associated B7-H1 promotes T-cell apoptosis: a potential mechanism of immune evasion. Nat Med. 2002;8:793–800. doi: 10.1038/nm730. [DOI] [PubMed] [Google Scholar]
- 11.Gao Q, Wang XY, Qiu SJ, et al. Overexpression of PD-L1 significantly associates with tumor aggressiveness and postoperative recurrence in human hepatocellular carcinoma. Clin Cancer Res. 2009;15:971–979. doi: 10.1158/1078-0432.CCR-08-1608. [DOI] [PubMed] [Google Scholar]
- 12.Ohigashi Y, Sho M, Yamada Y, et al. Clinical significance of programmed death-1 ligand-1 and programmed death-1 ligand-2 expression in human esophageal cancer. Clin Cancer Res. 2005;11:2947–2953. doi: 10.1158/1078-0432.CCR-04-1469. [DOI] [PubMed] [Google Scholar]
- 13.Wu C, Zhu Y, Jiang J, et al. Immunohistochemical localization of programmed death-1 ligand-1 (PD-L1) in gastric carcinoma and its clinical significance. Acta Histochem. 2006;108:19–24. doi: 10.1016/j.acthis.2006.01.003. [DOI] [PubMed] [Google Scholar]
- 14.Marzec M, Zhang Q, Goradia A, et al. Oncogenic kinase NPM/ALK induces through STAT3 expression of immunosuppressive protein CD274 (PD-L1, B7-H1) Proc Natl Acad Sci U S A. 2008;105:20852–20857. doi: 10.1073/pnas.0810958105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Parsa AT, Waldron JS, Panner A, et al. Loss of tumor suppressor PTEN function increases B7-H1 expression and immunoresistance in glioma. Nat Med. 2007;13:84–88. doi: 10.1038/nm1517. [DOI] [PubMed] [Google Scholar]
- 16.Kuang DM, Zhao Q, Peng C, et al. Activated monocytes in peritumoral stroma of hepatocellular carcinoma foster immune privilege and disease progression through PD-L1. J Exp Med. 2009;206:1327–1337. doi: 10.1084/jem.20082173. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Curiel TJ, Wei S, Dong H, et al. Blockade of B7-H1 improves myeloid dendritic cell-mediated antitumor immunity. Nat Med. 2003;9:562–567. doi: 10.1038/nm863. [DOI] [PubMed] [Google Scholar]
- 18.Bloch O, Crane CA, Kaur R, et al. Gliomas promote immunosuppression through induction of B7-H1 expression in tumor-associated macrophages. Clin Cancer Res. 2013;19:3165–3175. doi: 10.1158/1078-0432.CCR-12-3314. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ansell SM, Lesokhin AM, Borrello I, et al. PD-1 blockade with nivolumab in relapsed or refractory Hodgkin’s lymphoma. N Engl J Med. 2015;372:311–319. doi: 10.1056/NEJMoa1411087. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Borghaei H, Paz-Ares L, Horn L, et al. Nivolumab versus Docetaxel in Advanced Nonsquamous Non-Small-Cell Lung Cancer. N Engl J Med. 2015;373:1627–1639. doi: 10.1056/NEJMoa1507643. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus Docetaxel in Advanced Squamous-Cell Non-Small-Cell Lung Cancer. N Engl J Med. 2015;373:123–135. doi: 10.1056/NEJMoa1504627. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Motzer RJ, Escudier B, McDermott DF, et al. Nivolumab versus Everolimus in Advanced Renal-Cell Carcinoma. N Engl J Med. 2015;373:1803–1813. doi: 10.1056/NEJMoa1510665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Archibald DJ, Neff BA, Voss SG, et al. B7-H1 expression in vestibular schwannomas. Otol Neurotol. 2010;31:991–997. doi: 10.1097/MAO.0b013e3181e40e4f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Katsuya Y, Fujita Y, Horinouchi H, et al. Immunohistochemical status of PD-L1 in thymoma and thymic carcinoma. Lung Cancer. 2015;88:154–159. doi: 10.1016/j.lungcan.2015.03.003. [DOI] [PubMed] [Google Scholar]
- 25.Padda SK, Riess JW, Schwartz EJ, et al. Diffuse high intensity PD-L1 staining in thymic epithelial tumors. J Thorac Oncol. 2015;10:500–508. doi: 10.1097/JTO.0000000000000429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Miettinen M. A simple method for generating multitissue blocks without special equipment. Appl Immunohistochem Mol Morphol. 2012;20:410–412. doi: 10.1097/PAI.0b013e318245c82f. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lasota J, Kowalik A, Wasag B, et al. Detection of the BRAF V600E mutation in colon carcinoma: critical evaluation of the imunohistochemical approach. Am J Surg Pathol. 2014;38:1235–1241. doi: 10.1097/PAS.0000000000000229. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Kelly RW, Carr GG, Elliott CL, et al. Prostaglandin and cytokine release by trophoblastic villi. Hum Reprod. 1995;10:3289–3292. doi: 10.1093/oxfordjournals.humrep.a135904. [DOI] [PubMed] [Google Scholar]
- 29.King A, Hiby SE, Gardner L, et al. Recognition of trophoblast HLA class I molecules by decidual NK cell receptors--a review. Placenta. 2000;(21 Suppl A):S81–S85. doi: 10.1053/plac.1999.0520. [DOI] [PubMed] [Google Scholar]
- 30.Guleria I, Khosroshahi A, Ansari MJ, et al. A critical role for the programmed death ligand 1 in fetomaternal tolerance. J Exp Med. 2005;202:231–237. doi: 10.1084/jem.20050019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Uceyler N, Gobel K, Meuth SG, et al. Deficiency of the negative immune regulator B7-H1 enhances inflammation and neuropathic pain after chronic constriction injury of mouse sciatic nerve. Exp Neurol. 2010;222:153–160. doi: 10.1016/j.expneurol.2009.12.026. [DOI] [PubMed] [Google Scholar]
- 32.Azuma K, Ota K, Kawahara A, et al. Association of PD-L1 overexpression with activating EGFR mutations in surgically resected nonsmall-cell lung cancer. Ann Oncol. 2014;25:1935–1940. doi: 10.1093/annonc/mdu242. [DOI] [PubMed] [Google Scholar]
- 33.Comprehensive molecular characterization of gastric adenocarcinoma. Nature. 2014;513:202–209. doi: 10.1038/nature13480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Murray P, Bell A. Contribution of the Epstein-Barr Virus to the Pathogenesis of Hodgkin Lymphoma. Curr Top Microbiol Immunol. 2015;390:287–313. doi: 10.1007/978-3-319-22822-8_12. [DOI] [PubMed] [Google Scholar]
- 35.Portis T, Cooper L, Dennis P, et al. The LMP2A signalosome--a therapeutic target for Epstein-Barr virus latency and associated disease. Front Biosci. 2002;7:d414–d426. doi: 10.2741/portis. [DOI] [PubMed] [Google Scholar]
- 36.Vockerodt M, Morgan SL, Kuo M, et al. The Epstein-Barr virus oncoprotein, latent membrane protein-1, reprograms germinal centre B cells towards a Hodgkin’s Reed-Sternberg-like phenotype. J Pathol. 2008;216:83–92. doi: 10.1002/path.2384. [DOI] [PubMed] [Google Scholar]
- 37.Green MR, Rodig S, Juszczynski P, et al. Constitutive AP-1 activity and EBV infection induce PD-L1 in Hodgkin lymphomas and posttransplant lymphoproliferative disorders: implications for targeted therapy. Clin Cancer Res. 2012;18:1611–1618. doi: 10.1158/1078-0432.CCR-11-1942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Green MR, Monti S, Rodig SJ, et al. Integrative analysis reveals selective 9p24.1 amplification, increased PD-1 ligand expression, and further induction via JAK2 in nodular sclerosing Hodgkin lymphoma and primary mediastinal large B-cell lymphoma. Blood. 2010;116:3268–3277. doi: 10.1182/blood-2010-05-282780. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Lyford-Pike S, Peng S, Young GD, et al. Evidence for a role of the PD-1:PD-L1 pathway in immune resistance of HPV-associated head and neck squamous cell carcinoma. Cancer Res. 2013;73:1733–1741. doi: 10.1158/0008-5472.CAN-12-2384. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.