New insights into the anti-PD-L1 and anti-PD-1 reagents in cancer therapy

Abstract

Immunotherapy that inhibits the interaction between programmed death ligand 1 (PD-L1), present on the surface of tumor or antigen-presenting cells, and programmed death 1 (PD-1), present on the surface of activated lymphocytes, is generating much excitement and enthusiasm. Although considerable knowledge has been accumulated on anti-PD-L1 and anti-PD-1 reagents, discovering immunotherapy-associated issues still remains a pressing task for the researchers and clinicians.

Introduction

Since anti-programmed death ligand 1 (PD-L1) and anti-programmed death 1 (PD-1) mAbs revolutionized the cancer immunotherapy in 2012, an impressive variety of clinical trials on checkpoint blockade, ranging from head-to-head to combinations, were either proposed or already underway.^1–4 We believe these intriguing reports contribute to our understanding of the immune checkpoint targeting in cancer therapy. Despite progress in these fronts, securing long-term strategies of immunotherapy to cancer requires addressing more urgent issues. Here, this review attempts to propose several new insights on the anti-PD-L1 and anti-PD-1 mAb immunotherapy against cancer, and to discuss how to improve the immunotherapy leading to more effective eradication of cancer.

PD-L1/PD-1 axis blockade enhances immune response by multiple mechanisms

Like the CTLA-4, the PD-1/PD-L1 pathway down-modulates T-cell response by regulating overlapping signaling proteins in the immune checkpoint pathway. However, their functions are slightly different. The CTLA-4 focuses on regulating the activation of T cells, while the PD-1/PD-L1 pathway regulates the effector T cell activity in peripheral tissues.^5–7 Current cancer immunotherapy strategies mostly aim at restoring T cell-mediated antitumor immunity. T cell-mediated immunity includes multiple sequential steps: clonal selection of antigen-specific T cells and T cell activation; proliferation; trafficking; and cytolytic effect (execution of direct effector function). In addition, PD-L1 is not only expressed on cancer cells, but also in cells of antigen-presenting cells (APCs) and myeloid-derived suppressor cells (MDSCs).^8–14 Thus, the PD-L1 expression of dendritic cell (DCs) and macrophages might also be responsible for the fine-turning of APCs-mediating T cell activation.^15–18 Taken together, the PD-1/PD-L1 axis is an inhibitory pathway physiologically counterbalancing the co-stimulatory pathway leading to fine-tuning the immune responses. Blockade of this interaction either by cancer cells or APCs could enhance the anti-tumor immunity, and decrease proliferation of the tumor infiltrating T-regulatory (Treg) cells, thus leading to a new era in cancer treatment through the multilevel immune interference.^19,20

Predictive biomarkers of anti-PD-1/PD-L1 therapies

PD-L1-positive cancers correlated with poor prognosis may indicate tumors sensitive to anti-PD-1 and/or PD-L1 therapies. As supported, recent evidence has shown that tumor PD-L1 expression is intimately correlated with tumor regression by the anti-PD-1 blockade, and to a greater extent than tumor PD-1 and PD-L2 expressions.^1,21 It was also found that PD-L1 expressed on tumor infiltrated immune cells is also a biomarker of therapy, suggesting that the functional restoration of tumor infiltrated immune cells can be derived from the PD-L1/PD-1 blockade.^22,23

However, there are still some controversies for this viewpoint. Recently, Julie Brahmer reported, among patients with advanced, but previously treated squamous-cell non-small cell lung cancer (NSCLC), overall survival, response rate, and progression-free survival were significantly better with PD-1 antibody treatment than with Docetaxel, regardless of tumor PD-L1 expression level. ²⁴ Further complicating matters is that while PD-L1 expression analysis was required before administration of anti-PD-L1 and anti-PD-1 reagents: (1) It was once believed that the lack of standardization of the PD-L1 testing makes it difficult to compare results derived from different trials; (2) the two leading hypotheses on PD-L1 expression were described as an innate versus adaptive immune response. In some settings, expression of PD-L1 in tumors was constitutive, and was (neither) not associated with the degree of T cell infiltration.^25,26 Conversely, in the adaptive model, tumors started to express PD-L1 only after exposure of tumors to tumor infiltrating T lymphocytes in an attempt to resist immune surveillance;^27,28 and (iii) In addition to tumor cells, this interaction has also shown to involve in proliferation of Treg cells, and antigen presentation of APCs. So, at this stage, it does not seem a necessity to say that PD-L1 expression analysis will be the key step to immune checkpoint therapies.

Clinical evaluation criteria of immunotherapy

Clinical experience using immunotherapy approaches indicates that the Response Evaluation Criteria in Solid Tumors (RECIST) and WHO criteria may not adequately or accurately characterize the therapeutic responses to checkpoint inhibitors. The current data suggest that disease stabilization can be durable for months or even years, and may even lead to regression of the disease long after cessation of therapy. Furthermore, complete or partial responses or disease stabilization can occur in patients who have had some brief disease progression by RECIST or WHO criteria.^29,30 It has also been noted in other malignancies that the anti-PD-1/ PD-L1 therapy may induce transiently worsening and disease progression, and that responses of tumors to the treatment may be delayed. Then, it is required to develop a set of criteria more suited for measuring clinical responses to immunotherapeutics, which is called an immune-related response criteria.

Side-effect of anti-PD-L1 and anti-PD-1 mAbs

Immunologically, the activation of PD-1/PD-L1 pathway weakens immune responses, limits inflammations and avoids development of immune injuries. The relationship between an overactive immune response and a clinical outcome of the checkpoint inhibitors has not been delineated. Overall, the immune therapy was relatively well tolerated. However, drug-related adverse events were frequently found. The mechanism of toxicity related to the PD-1/PD-L1 inhibition is autoimmune in nature and is distinct from that of other cancer therapies. These side effects consist of immune-related adverse events that are defined by inflammatory conditions, including dermatitis, colitis, hepatitis, pancreatitis, pneumonitis, and hypophysitis.^31–33 If a sustainable therapy with PD-1/PD-L1 antagonists is available, the immune associated damage will (was) not be evitable. Then, an optimal dose and a course of immunotherapy are yet to be definitively established.

Resistance to anti-PD-L1 and anti-PD-1 mAb immunotherapy

Acquired resistance to reagent is the most important limiting factor for treatment efficiency in cancer. But there are limited evidences for the existence of resistance induced by the continuation of anti-PD-L1 and anti-PD-1 mAb treatments. This new pharmacological reagent would have been developed for acquired resistance via bystander immune pathways, called cell-intrinsic pathways that regulate T cell activation. ³⁴ Other contributing factors for the failure of cancer vaccines are perhaps the antibody internalization and anti-idiotype antibodies.

Beyond PD-L1/PD-1 axis

It is well established that tumors use several mechanisms to avoid elimination by the immune system, one of which involves hijacking these checkpoint pathways. It is unlikely that a single immunologic target mediating immune escape can be identified. In addition to the PD-L1/PD-1, multiple other immune checkpoints play roles in immune regulation. For example, B7 homology 4 (B7-H4) and T cell immunoglobulin and mucin-domain containing protein 3 (Tim3) can also inhibit T-cell activation, proliferation, and cytokine production.^35–38 Therefore, inhibition of the PD-L1/PD-1 interaction alone could result in an incomplete rescue of immune cells fighting cancers. Given the complexity of regulation of T cell responses by multiple signaling pathways, both negative and positive, it will be necessary to determine other components within the tumor microenvironment for development of combinational strategies with greater clinical benefits.

Conclusion

An in-depth understanding of how tumors evade immune surveillance will help to develop more effective therapeutic strategies that can be used for the benefit of cancer patients. PD-L1 and PD-1 are crucial molecules fine-tuning the immune responses. It is well established that tumors use this pathway to avoid elimination by the immune system. Blockade therapy utilizes monoclonal antibodies to release brakes from suppressing T cells might allow them to be activated and recover their antitumor activity. However, several key issues regarding the anti-PD-L1 and anti-PD-1 immunotherapy need to be highlighted. For example, the development of pharmacodynamic, side effect, predictive, or prognostic biomarkers faces unique challenges. Especially, those reagents that block immune checkpoints unleash the dynamic and complexed immune responses.