Overview of Cancer Immunotherapy
Cancer immunotherapy has developed rapidly in these two years. Numerous new targets and drug forms continue to emerge, including innate immunity and adaptive immunity, immune checkpoints & costimulatory molecules, T cells & macrophages, monoclonal antibodies, cell treatments, vaccines, small molecules. According to the statistics of the American Pharmaceutical Association, there were 248 cancer immunological drugs in clinical studies in 2017, including 45 immune checkpoint inhibitors, 96 therapeutic vaccines, 21 CART cell therapy, 14 oncolytic viruses, and bispecific antibodies, and 72 other drugs such as bispecific monoclonal antibody. And there are more than 1,000 clinical studies on these drugs and their combination. It is reported in The New York Times, the clinical research on American immunotherapy has been so advanced, however, the patients are not enough.
This overheating phenomenon has caused concern in the industry. On the one hand, there are too many repeated studies, on the other hand, people cannot keep pace with the development of clinical trials, and the clinical design that involves the combination of various drugs is not clear enough. In the long term, the availability of drugs with similar efficacy, especially the combination of expensive drugs, will increase the patient’s medical burden. What’s more, immunotherapy occupies too much R&D resources, indirectly affecting the development of other areas.
Increased Negative News on Cancer Immunotherapy
Several years ago, Most people were very optimistic about the prospects of immunotherapy and thought that PD-1 was just the beginning. However, with the results of various clinical trials are updated, people gradually realized that: (1) The efficacy and range of application of these new targets and new drug forms after PD-1 may be difficult to compare with PD-1 drugs; (2) PD-1 is not a panacea, its mechanism of action is far more complicated than the immune brakes, and drug combination based on PD-1 is not as simple as expected.
Taking several popular immune targets as examples, in addition to the more encouraging clinical data reported by Lag-3, IDO, and ASCO in 2017, TIM-3, TIGIT, 4-1BB, VISTA, XO40, and CSF-1R targets such as CD47, CD47 and A2A are not very clear or preliminary data are not satisfactory. Several OX40 projects have been abandoned by large companies. Even if the Lag-3, which function is also more complicated, it can not only inhibit the function of T cells, but also activate DC. Finally, there is still uncertainty about whether it can succeed.
Currently, PD-1 drugs are no longer fully progressing. On August 15, 2017, BMS announced the Opdivo/Yervoy combination as a first-line phase III clinical outcome for advanced kidney cancer, missed the primary endpoint of mPFS and required waiting for OS results. In general, the benefits of OS are more difficult, of course, there are some exceptions to immunotherapy. On July 27, AstraZeneca announced the phase III clinical results of a disappointing PD-1 drug. The PD-L1 monoclonal antibody Imfinzi (durvalumab) was used in combination with the CTLA-4 monoclonal antibody tremelimumab. The first-line PFS for non-small cell lung cancer with PD-L1 >25% did not reach the end. On July 24, MSD’s PD-1 monoclonal antibody Keytruda failed to reach the end of its overall survival OS in phase III clinical trial of failed head and neck cancer (HNSCC). This result was very surprising, not only because the drug was previously approved by the FDA for this indication because of its excellent ORR, and BMS’s Opdivo was successful in phase III of a similar trial.
Obviously, these setbacks of PD-1 drugs have awakened the latecomers. Immunotherapy is more complex than other drugs. Even if making mature targets, they may also fail due to molecular nuances, clinical development experience and strategies, selection of biomarkers, and differences in detection methods. However, these failures will not change the fundamentals of immunotherapy, and immune drugs such as small molecules, CARTs, and personalized vaccines will remain the most promising direction for development in the tumor field. The current failure, to some extent, is due to theoretical research and clinical experience cannot keep up with the explosive growth in this field. In order to make immunotherapy become a subversive treatment, the academic community and the industrial community need to work together to better solve the problems in both theory and practice through basic research and drug development:
(1) Development and improvement of immunotherapy theory
(2) The action mechanism of PD-1 drugs
(3) Development of biomarkers (how to accurately identify effective patients)
(4) A reasonable combination of drugs to increase efficiency (how to turn invalid or resistant patients into effective)
Development and improvement of immunotherapy theory
The concept of treating tumors through the immune system has been a hundred years. At the earliest time, Paul Ehrlich put forward a hypothesis, ‘the self-protection of host self can prevent the formation of tumors.’ In the fifties, Lewis Thomas and Frank Dr. Burnet further proposed the theory of immune surveillance (Immune Surveillance): the immune system is similar to a monitoring and defense system. Once a normal cell becomes a tumor cell, its mutated and expressed new antigen will trigger the human immune response, and the tumor cells will be destroyed by immune cells. This theory caused great controversy at that time, positive and negative experimental results have been reported. However, in 1974, it was MSKCC scientist Osias Stutman who published an article in Science that was considered irrefutable at the time. This article demonstrates through large-scale nude mice experiments that there is no difference in the incidence of tumors between nude mice and normal mice lacking the immune system, completely negating the possibility of tumor immunity.
A key figure in the resurgence of cancer immunity is Professor Robert D. Schreiber of the University of Washington. In the 1990s, people gradually realized that the nude mice used by Stutman only lacked acquired immunity, and the innate immune system was intact, that is, these mice still had partial immune functions. Schreiber skillfully obtained full-immune mice by removing the genes Rag2 and stat1 needed for immunity, proving that the immune system plays an important role in tumor formation. In 2002, the famous Immune Editing concept was proposed, which later became the most important theoretical basis for modern cancer immunotherapy.
The immunoediting theory summarizes the causal relationship between immunity and tumorigenesis as three E, namely Elimination, Equilibrium, and Escape. This is actually a “sustainable battle” between the tumor cells and the immune system. In the initial stage, the tumor cells are in a defensive position. After the appearance of tumor cells, NK and other innate immune cells initiate the first round of attack, secrete inflammatory factors such as INF-gamma, inhibit angiogenesis, and recruit other immune cells such as DC and macrophage. Part of the dead tumor cells further activate the acquired immune system, their released antigens are presented by DC cells to T cells in lymph nodes. The priming CD4+ and CD8+ T cells, with the help of various chemokines, enter the tumor tissue to specifically kill more tumor cells. After several rounds, the tumors are eliminated, and some T cells are transformed into memory cells, which can be used to quickly and repeatedly extinguish tumor cells. However, there is resistance where there is oppression, tumor cells have the characteristics of genome instability. Under the pressure of survival exerted by the immune system, they are continuously mutated and evolved. Some cells gradually have the ability to fight the immune system. For example, they do not express new antigens and are insensitive to INF-gamma signaling. At this time, the struggle between the tumor cells and the immune system enters a dynamic strategic stalemate phase. Sensitive tumor cells are constantly being extinguished, gradually becoming insensitive, and they have gradually acquired the ability to suppress and control the immune system, such as the induction of inhibitory inflammatory cytokines and immune cells, and changes in costimulatory signaling pathways. At this stage, the immune system cannot completely eliminate the tumor, but it can control its development. This process may take years or even decades. Due to changes in the environment, changes in living habits, illness, age, and other reasons, the body’s immune capacity has declined, and tumor cells have taken the opportunity to fully escape and turn into a strategic offensive stage. At this time, the immune system itself is corroded and controlled, suppressive immune cells are increased, and aggressive immune cells are negatively constructed, creating an immunosuppressive microenvironment, and the tumor begins to grow rapidly. Later, some tumor cells either have the ability to evade the monitoring of immune cells, or are insensitive to immune attack, begin to metastasize, rapidly increase and proliferate, and finally lead to the death of the host. Immune editing theory is different from immunodetection, which dictates the occurrence of tumors in the interaction between the immune system and tumor cells. In other words, the immune system constantly edits tumor cells and is fully engaged in the formation and development of tumors.
Tumor immunotherapy requires the use of drugs or other means to reverse these three phases. However, where do you start, when, what stage of intervention, and how to effectively reverse it? These problems are throughout the development of immunotherapy. In 2013, Daniel Chen and Ira Mellman, immunogen leaders of Genentech published a review of the Cancer Immune Cycle, which simplifying tumor immunity into 7 end-to-end links, including release of tumor antigens, presentation of antigens, priming and activation of T cells, migration of T cells, penetration of T cells into tumor tissue, recognition of tumor cells by T cells, and killing Tumor cells, also detailed the regulation of signaling pathways and intervention targets and factors in this seven steps.
In 2016, Prof. Schumacher of the Dutch Cancer Research Institute and others published a paper in Science that proposed the concept of Cancer Immunogram. They hypothesized that the effects of immunotherapy, whether direct or indirect, would eventually fall on the function of potent T cells. There are seven factors that affect the function of T cells, and the biomarkers of these factors are discussed. This theory can be used to guide the patient’s screening and develop a combination of drugs.
In 2017, Chen and Mellman of Genentech published a review article in Nature, and proposed the concept of Cancer Immune Setpoint, trying to use a framework to describe individual differences in tumor immunotherapy. The theory assumes that if the immunotherapy is to produce a therapeutic effect, the patient’s immune status must cross a gap. And this gap can be simply expressed by the immune critical point. It can also be understood as a balance between all immune stimuli and inhibitory factors. The purpose of the drug is to increase immune stimulation or reduce immunosuppression, push the immune system beyond this equilibrium, and reactivating T cells to enter the attack state. Because the tumor itself and the patient’s genome are different, microbiological and viral infections, treatment and even the degree of sunshine are not the same as the internal and external factors, and the immune threshold is different. The effect of drug treatment will also be different. Although the immune status should be dynamic and variable, the available evidence suggests that this change is generally temporary and it is easy to return to the original state. It seems that the patient’s genes and the primary state of the tumor largely determine this critical point of immunity in advance. Therefore, turning an ineffective patient into an effective state is not easy. It is worth mentioning that this article incorporates traditional Chinese medicine into one of the factors affecting the immune balance. Immunotherapy may be the starting point for combining Chinese and Western medicine and is worthy of clinical exploration.
The Mechanism of PD-1
PD-1/PD-L1 was discovered in the late 1990s. However, although PD-1 drugs have become the core of the tumor immunotherapy drug combination today, our understanding of the action mechanism of PD-1 remains murky. In order to understand the issue of immunization, we must grasp the three keywords ‘integrity, dynamics, and contradiction.’ The contradictions are particularly important. The thymus that produces T cells is the only organ that begins to age from birth. All immune cells have both stimuli and inhibition, and all signaling pathways have positive and negative effects.
The physiological function of PD-1 is simply an immune brake. The immune system discovers invaders and initiates immune attacks while not injuring itself, but starting the immune brake. Therefore, T cells are activated and simultaneously express PD-1, while the aggressive inflammatory factor INF-r stimulates cells to express its receptors PD-L1 and PD-L2. When PD-1 binds to its receptor, it suppresses the immune response. Some tumor cells gradually learned this trick in the fight against the immune system and achieved immune escape by expressing PD-L1 in large quantities. PD-1 drugs, including anti-PD-1 or anti-PD-L1 monoclonal antibodies, reactivate the immune system by blocking the binding of PD-1 to its receptors to attack tumor cells.
The mechanisms by which PD-1 suppresses immunity or PD-1 drugs activate immunity are extremely complex. This complexity is reflected in three levels: the molecular, cellular and organizational systems. At the molecular level, the molecular signaling pathway downstream of PD-1 is still not clear. The most studied downstream of PD-1 is a dephosphorylation enzyme called SHP-2. PD-1 activates the recruitment of SHP-2, and dephosphorylates and negatively regulates the common kinase pathways such as LCK and PI3K to achieve molecular signal transmission. Researchers have believed that dephosphorylation of the T cell receptor (TCR) is the main regulatory mechanism of PD-1 immune brakes. However, two articles published on Science in the same period of March completely subverting previous opinions. Scientists at UCSF and Genentech found that the co-stimulatory molecule CD28 is the primary target of PD-1/SHP-2. Scientists at Emory University confirmed this finding, they found that the function of PD-1 drugs to restore T cells depends on CD28.
This finding has important implications for current immunotherapy. We know that T cell activation requires both TCR and antigen binding, as well as CD28 costimulatory molecules. It is generally believed that CD28 is mainly involved in the activation, proliferation and migration of naïve T cells. The first immune checkpoint drug Yorvey was to block the inhibitory effect of CTLA4 on CD28. Apparently, CD28 is also involved in the proliferation of CD8+ potent T cells and the upstream is regulated by PD-1. That is, the role of PD-1 drugs is not only to restore the function of “depleted” CD8+ potent T cells, but also to stimulate their proliferation through CD28, and several studies have shown that T cell proliferation contributes more. This partly explains some clinical problems, such as, the expression of PD-L1 in tumor cells does not fully reflect the efficacy of PD-1 drugs.
With the development of research, the molecular regulation mechanism of PD-1 continues to unfold. In particular, the application of CRISPR-Cas9 gene editing technology has greatly promoted this process. The latest two articles published on Nature introduce the PD-L1 stability-regulatory molecule CMTM6/4 was discovered by whole-genome CRISPR-Cas9 screening. Previously, Nature also reported new targets for immunization PTPN2 and APLNR identified by the same techniques.
Immunotherapy differs from previous drugs in that it targets interactions between cells. The complexity of PD-1 is also reflected at the cellular level. PD-1 drugs not only interfere with the interaction between T cells and tumor cells. Since PD-1 and PD-L1 are widely expressed in various cells other than T cells and tumor cells, they must be functional and their presence is reasonable. For example, the above mentioned PD-1 regulates CD28, as tumor cells do not express CD28 receptor B7.1, it is mainly the communication between T cells and APC cells expressing B7 and PD-L1. Nature reported that tumor-associated macrophage TAM highly expressed PD-1 and inhibited macrophage phagocytosis in May. The expression level of PD-L1 in tumor cells and host immune cells all influence the effect of immunotherapy. Moreover, T cells themselves have a variety of subtypes. James Allison reported in a new issue of Cell that they used mass cytometry techniques to find that PD-1 and CTLA4 drugs modulate completely different T cell subsets. Also, some tumor cells themselves may rely on PD-1 signal growth. In 2015, an article on Cell reported that PD-1 promotes the growth of melanoma cells through mTOR signaling. it was thought that this may be one of the reasons that PD-1 drugs have a better curative effect than CTLA-4 drugs. As for PD-1, indirect regulation of other cell functions through inflammatory factors such as INF-r is even more common. For example, INF-r-induced weakened Treg function is a necessary condition for PD-1 drugs to be effective.
The study of immunotherapy focuses on the tumor microenvironment, but immunity is the result of the coordination of multi-tissue and multi-cellular systems. In January 2018, scientists at Stanford University reported in Cell that the efficacy of tumor immunity requires peripheral immune activation, while the upregulation of PD-L1 inhibits systemic immunity, but the specific mechanism of action is still unclear.
The immune system is an extremely complex dynamic network containing countless switches, thresholds, feedforward and feedback loops. However, this chaotic system, under normal circumstances, is extremely accurate and predictable for killing viruses and tumor cells. On the other hand, in rare cases, a subtle change, amplified by various loops, can lead to unpredictable changes. For example, CART can lead to a lethal inflammatory cytokine storm and septic shock in a short time. In some patients with advanced cancer, with the use of a PD-1 drug, the tumor can completely disappear within a few weeks. Therefore, the immune system is similar to social, economic, and biological complex systems, with steady states, transition states, and equilibrium points. PD-1 seems to be on a key node of the complex network of the immune system. PD-1 has PD-L1 and PD-L2 receptors, and PD-L1 and PD-L2 can also bind to other molecules such as CD80 and RGMb. The expression of these molecules has cell and tissue specificity and is influenced by internal and external factors. Including metabolic and epigenetic regulation. The PD-1 drug regulates these molecular pathways across cells, organs, and tissues. It loosens brakes and presses on the accelerator. It promotes the immune system to cross a certain equilibrium point in some lucky cancer patients and enters an activated state, showing long-term effectiveness. At present, there are not other targets have been found, the functions are complex and diverse, or they are on such a powerful node that can affect countless other pathways. Or we can say that PD-1 is on a strong node that can affect countless other pathways.