Prognostic value and immunological role of PD-L1 gene in pan-cancer

Despite being one of the most feared diseases of the 20th century, cancer is still widespread in the 21st century [1]. Every fourth person has a lifelong risk of developing cancer, which is a shocking state of affairs [2]. As a powerful anticancer strategy, cancer immunotherapy using immunotherapeutics has aroused people's wide concern [3]. In the above process, immune checkpoint control plays a pivotal role, which has become a research hotspot [4]. One of the primary causes of cancer is the differential expression and action of immune checkpoint molecules. Repairing this immune checkpoint malfunction is therefore a crucial treatment approach for tumors [5]. For instance, some studies demonstrated that PD-1 and PD-L1 inhibitors effectively inhibit lung cancer and melanoma by causing tumor cells to undergo apoptosis by disrupting the PD-1/PD-L1 signaling pathway [6, 7].

Immunotherapy is a method of treating disease by regulating the immune function by targeting the body's immune status [8]. At present, almost all patients with mNSCLC are treated with PD-1 or PD-L1 in the first-line setting, except for mNSCLC carrying targeted oncogenes [9]. Furthermore, the PD-L1’s correlation with immunotherapeutic response provides patients with the most promising choices of anti-gastric cancer drugs [10]. An immunotherapeutic drug stimulates the immune system and eliminates malignancies in cancer immunotherapy. Using immunosuppressive drugs known as immunological checkpoints, the body's immune activation can be controlled [11‐13]. At present, the most commonly prescribed medications are immunotherapy inhibitors (PD-L1, CDLA4, PD-1, and so on) [14]. One of the biological characteristics of malignancy is that cells are able to evade immune response through different pathways like the PD-1/PD-L1 pathway [15]. Among them, some kinases play important roles in the activation of PD-L1 [16]. For instance, the overexpression of CDC28 protein kinase regulatory subunit 1B (CKS1B) could promote cell viability and invasion of papillary thyroid carcinoma cells through activation of STAT3/PD-L1 signaling and Akt phosphorylation [17]. When programmed death protein 1 binds to programmed death-ligand-1, it triggers a downstream signaling cascade that prevents T-cell activation and suppresses the immune system's capacity to mount any inflammatory response [18]. TNBC can develop in the context of PD-1 overexpression in malignant cells because these cells are capable of dodging the immune system’s reaction and multiplying out of control [19‐21]. Furthermore, relevant molecular dynamics simulations revealed that the docking time between PD-L1 and the ligand was at least 300 nanoseconds [22]. Anti-PD-1/PD-L1 immune checkpoint inhibitors (ICIs) can be used to stop the binding relationship between PD-1/PD-L1 to halt this rapid growth and spread [23]. This will trigger a powerful immune response that will kill malignant cells [24].

PD-1 and PD-L1’s combination can protect healthy body tissue by reducing the overreaction of autoimmunity and reducing the damage caused by an overactive immune response [25]. However, for tumor patients, PD-1 and PD-L1’s combination can decrease the vitality and proliferation ability of T cells in the tumor microenvironment and lose the ability to normally identify or kill tumor cells, which will comparatively improve the propagation speed, proliferation ability, and invasion of tumor cells, as well as promote tumor cell metastasis [26]. PD-L1's function in tumor immunity and its underlying mechanisms, however, are unknown.

A comprehensive analysis of 33 different cancer types was performed to further study the association between PD-L1 (also known as CD274) and prognosis. Additionally, we investigated PD-L1's potential roles in a variety of malignancies and found evidence that it may be a predictive biomarker and is strongly connected with immune infiltration for several tumors.

Pan-cancer analysis, a study of molecular abnormalities data from several cancer kinds [35], can reveal commonalities and differences between tumors, giving insight into the design of diagnostic targets [1, 36]. In previous studies, researchers developed various prognostic prediction models. For instance, AC010973.2, one of six stemness-related genes, can promote cell proliferation and predict overall survival in renal clear cell carcinoma [37]. Besides, the role of autophagy-related genes in COAD was revealed to facilitate the design of new targets for improving cancer therapy [1]. Apart from these, pan-cancer analysis can spot patterns in critical biological functions that are dysregulated in cancer cells of various ancestries [38]. The B7 family of immune-regulatory molecules includes PD-L1, also referred to as B7-H1 or PD-L1, which is an immunological co-signaling molecule [39]. However, pan-cancer analysis of CD274 regulation in human pan-cancer has not yet been clarified [40].

In the current study, we discovered that PD-L1 is abnormally expressed in 25 different cancer types and its levels and DNA methylation are highly linked with MMR gene mutation levels. Additionally, patients’ prognoses, particularly those with LGG, SKCM, THYM, PAAD, OV, TGCT, BRCA, KIRC, LIHC, SARC, and UCEC, were linked to PD-L1 expression. The expression of immune checkpoint markers and immune infiltration levels were also found to be favorably linked with PD-L1 expression, particularly in BRCA, COAD, HNSC, SKCM, TGCT, THCA, and UCEC. These findings clearly suggest that PD-L1 may be a predictive biomarker for those malignancies.

The DNA damage repair mechanism known as MMRs is made up of several heterodimers. The accumulation of DNA replication mistakes caused by the functional loss of important genes in this pathway increases somatic mutation rates, MSI, and cancer [41, 42]. The previous study showed poor prognosis in prostate cancer patients and their sensitivity to Olaparib was closely related to mutations in the DNA damage response pathway [43]. While in our correlation analysis, we discovered in this study that PD-L1 expression was strongly correlated with the mutation levels of 5 MMR genes in human pan-cancer. Changes in DNA methylation status also play a role in the growth of cancer. According to research, cancer frequently exhibits hypermethylation of the gene promoter [44, 45]. Additionally, we found a significant link between PD-L1 expression and four DNMTs, particularly in the cases of BRCA, OV, UVM, and BLCA. The results above support the hypothesis that aberrant PD-L1 expression might significantly influence carcinogenesis by modulating DNA methylation and MMR gene mutation levels.

TME is made up primarily of the extracellular matrix (ECM), the vasculature, and other benign cells that surround the tumor. It also contains elements that either encourage or prevent the growth of tumors, such as immune cells that are present in and around the tumor but are not carcinogenic (B cells, TILs, and T cells) [32, 46]. These non-cancerous elements have been demonstrated to play a crucial function in tumors as a double-edged sword to promote or inhibit tumor progression. They also significantly affect tumor sample genomic analyses and may alter how the data are biologically interpreted. PD-1 is a crucial immunosuppressive transmembrane protein expressed on the surface of T cells. In the tumor microenvironment, tumor cells can express the ligand of PD-1, namely PD-L1 or PD-L2. A well-known method for cancer cells to avoid T cell surveillance is that PD-L1 binds to the PD-1 of T cells [47, 48]. An essential mechanism for preserving immunological tolerance and preventing autoimmune disorders is the PD-1/PD-L1 axis [49]. The balance between tumor immune surveillance and immunological resistance is also influenced by the PD-1/PD-L1 axis. T cells become exhausted as a result of increased PD-L1 expression on a tumor cell or TIL, reducing tumor-specific immunity and accelerating tumor growth [50]. Additionally, tumor-infiltrating T cells and tumor cells compete with one another. High quantities of aerobic glycolysis are seen in the first. Additionally, studies have demonstrated that highly glycolytic tumor cells are probable to deplete the microenvironment of glucose and other nutrients, which is essential for tumor-infiltrating T cells. So infiltrating T cells’ ability to respond to the tumor cells is dampened [50, 51]. Additionally, PD-L1 has the ability to prevent activation of the RAS-ERK1/2 signaling pathway, which in turn prevents the proliferation of T lymphocytes, inhibits the activation of PKCδ, and lowers the level of IL-2 secreted by T cells [48, 52]. In the tumor microenvironment, some studies have confirmed that PD-L1 is highly expressed in tumor cells as well as immune cells (Tregs, DCs, macrophages, and so on) [53]. Also, some studies have found PD-L1 is known to be upregulated in response to interferon-γ produced by infiltrating T cells [54]. In our study, for most tumors, it was found that PD-L1 expression had a strong positive correlation with T cell CD8 + and T cell CD4 + , and macrophages, which is consistent with the results of previous studies. It was found in previous studies that late GC B cells upregulate PD-L1 [55]. Also, with high levels of PD-L1 expression, humoral reactions can be significantly inhibited by regulatory B (Breg) cells [56]. For instance, B cells in terminal differentiation toward antibody secretors, in the presence of certain antigens, can be transcriptionally reprogrammed to produce high levels of inhibitory molecules, such as PD-L1, which suppress pro-inflammatory populations in the bone marrow and lymph [57]. In our study, for most tumors, it was found that PD-L1 expression was strongly positively correlated with B cells, which is consistent with the results of the above studies.

Apart from wide expressions on the surface of macrophages, B lymphocytes, DCs, and T lymphocytes, the surface of many tumor cells also presents high PD-L1 expression, which leads to T cell exhaustion and immune tolerance, causing immune escape [21]. To sum up, the expression of PD-L1 can be induced when a variety of cytokines and exosomes exist in the TME, which contributes to strengthening the PD-L1/PD-1 signal to inhibit CTL activation in the TME and thereby boost tumor escape [44, 58].

However, there is not enough research on PD-L1's functions in the immunological microenvironment. In this investigation, we discovered a significant association between PD-L1 expression and immune cells that were infiltrating OV, BRCA, UCEC, COAD, LGG, KIRC, LIHC, PAAD, SKCM, TGCT, THCA, PRAD, and HNSC, which indicates that PD-L1 may lead to the inhibition of tumor or tumorigenesis by altering the TIL status. The innovative researches represent a significant advancement in understanding PD-L1’s critical function in immune infiltration.

We use immunological scoring to gauge the quantity of invading CD3 + /CD45RO + , CD3 + /CD8 + , or CD8 + /CD45RO + lymphocytes at the tumor's center and borders. A higher ImmuneScore or StromalScore represents that the TME has more immune or matrix components [59]. Our findings showed a substantial positive connection between PD-L1 expression and stromal and immunological scores in these malignancies, showing that the quantity of stromal or immune cells rises concurrently with an increase in PD-L1 expression levels. Additionally, we can see that immunological scores in the DLBC, KIRC, MESO, SARC, TGCT, and USC have a favorable connection with PD-L1 expression. Additionally, the relationship between immunological check-point markers and PD-L1 expression suggests that PD-L1 has a function in controlling tumor immunology in malignancies, particularly in BRCA, PRAD, LUAD, BLCA, and OV. These findings further support the crucial part PD-L1 plays in tumor immunity.

By combining bioinformatic analysis and laboratory experiments, the characteristics of PD-L1 were presented in a systematic manner by the pan-cancer study in a number of areas, such as expression pattern, genetic mutation, survival prognosis, MSI, TMB, MMR, tumor immune micro milieu, medication sensitivity, and signaling pathway. Since PD-L1 showed abnormal expression in a number of cancers and predicted prognosis in patients suffering from these cancers, particularly for those with tumors like LGG, SKCM, THYM, PAAD, OV, TGCT, BRCA, KIRC, LIHC, SARC, and UCEC, it provides guidance for better strategies for the clinical treatment of immune checkpoint inhibitors. Furthermore, across a variety of cancer types, the aberrant PD-L1 expression was connected to the MSI, MMR, TMB, and TIME. This work shows the several functions of PD-L1 in pan-cancer and offers fresh information on PD-L1's potential role in controlling chemoresistance. However, we do not have enough experiments to support this. Integrated meta-analysis in combination with existing studies and larger samples are needed to validate the role and mechanism of PD-L1 in pan-cancer.