PD-1, PD-L1 (B7-H1) and Tumor-Site Immune Modulation Therapy: The Historical Perspective

The success of anti-PD drug development benefited from the advancement of our fundamental understanding of both cancer biology and the immune system. Cancer originates from the mutated self of the host in the setting of genotoxic insults with molecular hallmarks of genetic instability and heterogeneity [13, 14]. Most conventional therapies, including radiation, chemotherapy and even molecular-targeted therapies, directly target cancers themselves with improving, but still less desirable efficacy due to significant side-effects, drug resistance and recurrence of more aggressive malignant clones [8, 13, 15]. The limitations of conventional therapies are especially evident for late stage and advanced cancers. In contrast, the immune system, especially the adaptive arm of immunity such as T cells, is capable of surveilling against mutated antigens in tumor cells and controlling tumor growth through specific immune effector mechanisms. Importantly, these immune cells evolve in parallel with tumors, and are capable of sustainable tumor recognition and killing [16, 17]. However, surveillance efficacy often fails due to multiple tumor immune evasive mechanisms, resulting in the outgrowth and metastasis of cancer cells [18, 19]. Efforts in understanding the basic mechanisms of immune tolerance in general and the primary immune evasive mechanisms by cancer cells in particular have been the key to the successful development of drugs targeting the PD-1/PD-L1 pathway.

Many individuals contributed to the success of PD-directed cancer therapy (Table 1 and Fig. 1) [2]. Among them, Lieping Chen’s work is the focus of this review. In the early span of the 1990s, Chen was convinced that in the tumor microenvironment, there exist specific molecular pathways that are primarily responsible for immune evasion, and that this concept could be harnessed for more effective cancer therapy. His passion has been devoted towards the search for these elusive molecules ever since. Trained as a physician-scientist in China, Chen quickly became convinced that basic medical research holds the key to cancer cure. He earned his M.S. degree at the Cancer Institute of Peking Union Medical College in 1986 before obtaining his Ph.D. from Drexel University in 1989. He then undertook his postdoctoral training at the University of Washington, Seattle in 1990, all along focusing on immunology. During that period, important discoveries were made regarding the major immune cell populations with different immune functions, distinguished by cell surface molecules, including CD3, CD4, CD8, T cell receptor (TCR) and others [20–23]. More importantly, the guiding principle of T cell activation and tolerance began to emerge [21]. Among those important events, the discovery of CD28, CTLA-4 and their ligands B7 (B7.1 and B7.2) was the key development in the field [24, 25] which vindicated the two-signal hypothesis for lymphocyte activation proposed by Bretscher and Cohn [26], and extended by Lafferty and his colleagues [27]. The first signal is via TCR triggered by an MHC-antigenic peptide complex, whereas the second signal, also known as a co-stimulation signal, is provided by co-stimulatory molecules expressed on the surface of antigen presenting cells (APC) and T cells (like B7-CD28 pathway). Based on this principle, Chen and his colleagues, for the first time, demonstrated that stably-enforced expression of B7 molecules in cancer cells elicited strong anti-tumor immune responses, leading to eradication of distal tumors [28]. This effect is mainly CD28 dependent: upon ligation of the same B7 ligand, CD28 delivers an essential signal for naïve T cell activation [29], which is in contrast to the CTLA-4 molecule behaving as a T cell “checkpoint” receptor with inhibitory activities [30–33]. Blocking CTLA-4 for immunotherapy of cancer was later steered by James Allison who has played crucial roles in the renaissance of cancer immunology [34–37], whereas Chen’s earlier work laid the groundwork for the therapeutic potential of manipulating co-stimulatory molecules against cancer. However, the ectopic expression of costimulatory molecules like B7 on tumor cells worked effectively against multiple murine tumor models, its application is currently limited in cancer patients [38]. Additionally, since both B7 ligands and CD28/CTLA-4 receptors are expressed broadly without tumor specificity and that they play essential roles in the control of general immune homeostasis, targeting this pathway could be met with severe autoimmune toxicities, which was shown to be the case in the clinical trials of TGN1412, an anti-CD28 super agonist antibody [39], and the anti-CTLA-4 antibodies, ipilimumab and tremelimumab [40–42]. Chen was determined to uncover tumor-specific immune evasion mechanisms and find ways to block them.

During his seven-year tenure at Bristol-Myers Squibb (1990–1997), Lieping Chen evaluated the immune functions and potential anti-tumor effects of many cell surface molecules on T cells, especially 4-1BB (CD137), a molecule specifically expressed by activated T cells and serving as another co-stimulatory receptor. The Chen group were the first to reveal the potent anti-tumor effect of the anti-4-1BB agonistic antibody in both immunogenic and non-immunogenic tumor models [43], making 4-1BB an attractive target for immuno-oncology [44]. Some promising results from the anti-4-1BB clinical trials were recently reported [45]. Although the expression of 4-1BB ligand is relatively broad, 4-1BB expression could accurately identify tumor-reactive T cells [46]. Even so, Chen believed that immune-regulatory molecules with higher tumor specificity remained to be identified. He returned to academia and joined the Mayo Clinic to continue searching for these pathways. Inspired by the progress of the Human Genome Project, Chen turned to informatics to discover candidate molecules from the human expressed sequence tag (EST) libraries based on their predicted homology to B7 family molecules. This effort was incredibly successful; he made several seminal discoveries of a series of novel B7 family members, including B7-H1 (PD-L1) [47], B7-H2 [48], B7-H3 [49], B7-H4 [50], B7-H5 [51], PD-1H (VISTA) [52] and so on [53–55]. In a series of papers, Chen and his colleagues completed the fundamental characterization of the B7-H1’s biological function and provided the very first proof of anti-tumor effects via blockade of B7-H1 [47, 56], which mainly serve as a ligand to PD-1 receptor on T cells [57], a molecule discovered earlier by Tasuku Honjo in Japan [58]. In 2004, Chen joined the Johns Hopkins University School of Medicine and collaborated with Suzanne Topalian, Julie Brahmer and other clinical investigators to initiate the first clinical trial of anti-PD-1/PD-L1 pathway therapies in patients with cancer. It was this clinical study that opened the floodgate of PD-1/PD-L1 pathway-directed cancer immunotherapy.

The key events that led to the successful development of anti-CTLA-4 and anti-PD-1/PD-L1 drugs are chronicled below (Fig. 1 and Table 1):

The timeline in the history of anti-PD drug development (Fig. 1) spans from the understanding of the B7 pathway in regulating T cell responses to the discovery of the PD pathway with tumor-site immune modulation properties, reflecting our increased understanding of both T cell biology and tumor immunity. This advancement resulted from the progress of research in the field of both immunology and oncology, eventually leading to the birth of immuno-oncology. Ironically, anti-PD antibodies did not show clear anti-tumor effects in many animal models in the early days of research, raising skepticism among many over their therapeutic application. However, Lieping Chen and his colleagues proposed that the potency of anti-PD therapy depends on both the existence of immune cells especially T cells in the tumor site, as well as PD-L1 expression by the tumor cells [67, 70, 84]. He steadfastly and passionately championed the clinical development of agents targeting this pathway. Thus, Chen played key roles in advancing the anti-PD drug program in the areas of basic research as well as in clinical translation (Table 1), for which he was recognized with the 2014 William Coley Award in Basic and Tumor Immunology [85] (shared with Tasuku Honjo, Arlene Sharpe and Gordon Freeman), the 2015 Lifetime Achievement Award in Hematology and Oncology by CAHON (www.​cahon.​org, and this paper), and the 2016 American Association of Immunologists (AAI)-Steinman Award for Human Immunology Research (http://​www.​aai.​org/​Awards/​Archive/​index.​html).

Targeting the PD pathway for treatment of cancer is unique in several aspects, including the following (Figs. 2 and 3):

First, the action of anti-PD therapy is primarily localized to the tumor site. The ligand for PD-1, PD-L1 (B7-H1), has high expression on tumor cells. It is absent in the majority of normal tissues, but could be induced in the tumor microenvironment by ongoing immune responses, mostly by the cytokine IFN-γ [3, 56]. This tumor-localized effect of PD-L1 dictates that the antibodies targeting either ligand or receptor will work in the tumor microenvironment with ongoing immune responses. In contrast, PD-L2, another ligand of PD-1, has weak expression in tumors and is constitutively expressed by dendritic cells [53, 54]. The expression pattern of PD-L2 likely explains why targeting PD-L2 has only minor anti-tumor effects, although PD-L2 and PD-L1 have similar binding affinity to PD-1, and their inhibitory effects on T cells in vitro are comparable.

Second, targeting the PD pathway repairs or resets tumor-associated immune defects. The PD-L1 molecule is a key mechanism for tumor-mediated immune evasion [3]. The ligation of PD-L1 to PD-1 causes functional defects in T cells through several different mechanisms (Fig. 2), including anergy to antigen stimulation [86–88], functional exhaustion [72], apoptosis [56], induction of immune suppressor cells [89, 90], and secretion of inhibitory cytokines, such as IL-10 [47]. PD-1 on myeloid cells also impairs dendritic cell functions [91]. In addition, PD-L1 reverse-signaling on tumor cells was found to serve as a “Molecular Shield,” protecting the tumor from killing by CTLs [70, 92]. Importantly, this type of immune defect is not permanent, and can be restored by termination of this pathway, especially by the antibody blockade of either PD-L1 or PD-1 [70]. Moreover, anti-PD therapy may reset the global anti-tumor immune defects possibly by yet unknown “cascade reaction” mechanisms, as profound and sustainable therapeutic effects have been observed in many patients receiving this therapy [3].

Third, the PD blockade aims to normalize the anti-tumor immune response but not over-exuberate immune responses in general. Blockade of either PD-L1 or PD-1 is not to simply amplify immunity, but to re-adjust the anti-tumor immune responses to a desirable level (Fig. 3). Given the mild and rare autoimmune symptoms observed in PD-L1 and PD-1 gene-deficient mice [60, 68] and weak expression of these molecules by normal tissues [56, 93], targeting the PD pathway in the tumor settings will normalize anti-tumor immunity but spare the normal peripheral tolerance mechanism against self-antigens [3]. Thus, anti-PD treatment has an understandably great safety window [94].

These abovementioned features of the PD pathway are very unique among the many pathways currently tested for cancer immunotherapy. Strictly speaking, PD-1/PD-L1 is the only known pathway responsible for key tumor-specific immune evasion mechanisms so far. Yet, both PD-1/PD-L1 and CTLA-4 are often lumped together as “immune checkpoints” [6, 8, 31, 95], which is an immunological term defining a plethora of inhibitory pathways in the control of physiological immune responses [6]. This concept does not distinguish the immune modulating role of anti-PD therapy from the actions of anti-CTLA-4 therapy [3, 7] (Fig. 3). An alternative term of “Tumor-Site Immune Modulation Therapy” might better describe the attributes of anti-PD therapy [5], which is characterized by targeting molecules responsible for tumor-site specific immune evasion mechanisms, resetting and restoring the anti-tumor immunity back to a desirable level. Such action is clearly different from CTLA-4 blockade, which overdrives systemic T cell immunity including self-reactive T cell responses.

Lieping Chen continues his journey to discover more PD-like molecules, aiming to convert human cancers to chronic and manageable diseases via exploiting the power of the hardwired immune system. His contributions to this field have cemented his place in immuno-oncology.