**4. Programmed Death-1 (PD-1)**

#### **4.1 General description**

PD-1 gene was first identified in two different types of lymphoid cell lines be that 2B4.11 (a murine T cell hybridoma) and LyD9 (a murine hematopoietic progenitor cell line) following manipulations using ionomycin/phorbol 12-myristate 13-acetate (PMA) and IL-3-deprivation, respectively. Given the two cell lines shared the same feature in common that was programmed cell death, it was plausible that PD-1 was a player in the death-inducing process in the two manipulated cell lines. In addition, in mRNA level PD-1 was deemed to be one of the molecules whose *de novo* synthesis induced death in the two cell lines. Owing to its patterns in death-increasing manipulation-induced augmentation as well as in thymus-restricted expression, PD-1 is enforced as a cell-death-associated gene [40].

PD-1 as a checkpoint molecule is a 55 kDa transmembrane protein consisting of 288 amino acids with a membrane-permeating domain, an extracellular N-terminal domain (IgV-like) and a cytoplasmic tail at N and C ends [41] containing two tyrosine-based signaling motifs, tyrosine-based inhibitory motif (ITIM), and an immunoreceptor tyrosine-based switch motif (ITSM) [42]. This is expressed on B cells, T cells, natural killer T (NKT) cells, activated dendritic cells (DCs), and monocytes. In resting T cells, PD-1 is not expressed, but this can be induced. In normal human lymphoid tissue, PD-1 is expressed on germinal center-associated T cells [43]. PD-1 can be both beneficial and harmful. In the context of physiological condition, PD-1 reduces ineffective immune responses and maintains immune tolerance, to prevent autoimmune reactions [44]. As oppose, the expression of this checkpoint molecule in a tumor microenvironment (TME) mediates dilation of malignant cells and silence of the immune surveillance. Therefore, blocking interaction of PD-1 and its ligands either PD-L1or PD-L2 was deemed to potentially augment endogenous antitumor responses [45].

### **4.2 Signaling pathway of PD-1 in cancers**

The primary ligand of PD-1 is PD-L1 that is also known as CD279 or B7-H. This ligand is a 290 amino acid-containing 33 kDa type I transmembrane glycoprotein with Ig and IgC domains in its extracellular region [46]. PD-L1 is not only a ligand, but it also carries receptor functions. PD-1 could act as a ligand to mediate transmission of antiapoptotic signal to tumor cells via PD-L1. The second known counter receptor of PD-1 is PD-L2 or B7-DC which binds with RGMb (repulsive guidance molecule b). This interaction leads to induction of pulmonary tolerance [47].

PD-L1 is expressed by tumor cells as an immune escape strategy [48]. It is associated with production of Th1 cytokines and interferons. It has been demonstrated that IFN-γ causes PD-L1 upregulation in ovarian cancer cells, whereas the inhibition of this interferon leads to reduction of PD-L1 expression in acute myeloid leukemia in mouse models. Both have been reported to take mitogen-activated protein kinase kinase (MEK/extracellular signal-regulated kinase (ERK)) and MYD88/TRAF6 pathways [49].

The PD-L1 secreted-IFN-γ subsequently induces protein kinase D isoform 2 (PKD2); thus, inhibition of this PKD2 activity inhibits the expression PD-L1. NK and T cells produce IFN-γ via Janus Kinase (JAK)1, JAK2, and signal transducer and activator of transcription (STAT)1 pathways, to ultimately upregulate PD-L1 expression on the tumor cells' membranes [50]. PD-L1, therefore, acts as a protumorigenic factor in cancer cells via binding to its receptors and activating proliferative and survival signaling pathways. This finding further indicates that PD-L1 is implicated in subsequent tumor progression. In addition, PD-L1 has been shown to exert nonimmune proliferative effects on a variety of tumor cell types [41]. These are the basis of PD-1/PD-L1 blocking antibody development, which is intended for downregulating these expressions, to allow functional immune cells to perform robust tumor surveillance [51].

#### **4.3 PD-1/PD-L1 in cancer immunotherapy**

When a T cell recognizes antigen:MHC complex, expressed on the surface of the target cell, inflammatory process begins which is marked by the secretion of inflammatory cytokines that subsequently induces the expression of PD Ligand-1 (PD-L1) in the affected tissue. This PD-L1 activates secretion of PD-1 protein on the T cells which causes immune tolerance, an event where the immune system is no longer capable of mounting an inflammatory response even in the presence of antigen [52].

In the tumor microenvironment (TME), PD-1 and its ligand PD-L1 play a fundamental role in tumor progression and survival by escaping immune surveillance. As aforementioned, while PD-1 is expressed on various subsets of immune cells, PD-L1 is expressed on tumor cells and APCs. Upon their engagement, T cells become dysfunctional and exhausted. Moreover, interleukin-10 (IL-10) is produced largely in the tumor [53]. This is known as the cytokine synthesis inhibitory factor which inhibits the productions of diverse pro-inflammatory cytokines including IL-1α, IL-1β, IL-6, IL-8, IL-12, IL-18, tumor necrosis factor-α (TNF-α), and granulocyte macrophagecolony-stimulating factor (GMSF) in T cells as well as in macrophages. Moreover, IL-10 diminishes the expression of interferon-γ (IFN- γ) in T helper (Th) cells and peripheral blood mononuclear cells (PBMCs). On the other hand, cytokines stimulate proliferation of mast cells [54], a particular subset of tissue-resident myeloid cells that contain coarse granules of inflammatory mediators like histamine that contribute to

shaping of tumor cells and tumor microenvironment (TME) [55]. Foxp3+ CD4+ Tregs have recently been observed to maintain PD-1 expression on their surfaces in order to create a highly immunosuppressive TME [44]. The FDA-approved PD-1/PD-L1-based immune checkpoint inhibitor therapy apparently outnumbers CTLA-4. Like CTLA-4, PD-1/PD-L1 checkpoint can be administered as monotherapy or combination with CTLA-4 [56].

Nivolumab and pembrolizumab are antibodies designed to block binding of PD-1 to its ligands. First approval for nivolumab with brand name Opdivo was in December 2014 when nivolumab was evidently great for treating unresectable metastatic melanoma. Few months later, again nivolumab was approved to be harnessed to treat NSCLC which resulted in 23.7% of objective response rate (ORR) and 91 days of progression-free survival (PFS) [57]. It was also used for treating Hodgkin's lymphoma patients for which overall survival (OS) was observed at 3 years for 80% patients with median PFS between 12 and 18 months [58]. As nivolumab, pembrolizumab with brand name Keytruda was initially approved in 2014 for metastatic melanoma. The ORR obtained in this treatment was at 18%. In May 2017, it received its second approval for its use in locally advanced or metastatic urothelial carcinoma. In non-Hodgkin's lymphoma and head and neck squamous cell carcinoma (HNSCC), treatment using pembrolizumab resulted in ORR of 53% and 19%, respectively [51]. The other FDA-approved anti-PD-1 antibody is Cemiplimab, also known as Libtayo, which is the first checkpoint designed for advanced cutaneous squamous cell carcinoma (CSCC). It was shown that the effects of Cemiplimab was durable, and no recurrence was observed even after more than 16 months [59].

As anti-PD-L1 antibodies, Atezolimumab (Tecentriq), Avelumab (Bavencio), and Durvulumab (Imfinzi) were also approved for cancer type targets. Atezolimumab is widely used for urothelial carcinoma [60, 61] and differs to the other two as it is a phage-derived human IgG1, whereas Avelumab and Durvulumab are fully human anti-PD-L1-IgG1 [51]. Avelumab is used to treat patients with metastatic Merkel cell carcinoma and NSCLC with ORR values of 62.1% and 12%, respectively [62, 63]. Durvulumab is designed to directly target PD-L1, to prevent tumor immune escape and enhance immune responses. In head and neck squamous cell carcinoma (HNSCC) ORR of 9.2% was observed. Moreover, 6-month progression-free in HNSCC patients rose to 25% for those patients who were PD-L1+ [64]. For NSCLC patient cohort, the ORR was observed at 66.3% [65].
