Seeing life and death, people will become numb, probably people will always have such misunderstandings about doctors. But no one will be more excited than them when revolutionary therapies emerge.
Sixty-three children with B-cell acute lymphoblastic leukemia who have no other treatment options have a very low chance of survival. However, about 83% of patients had complete remission within three months after receiving a new drug treatment, and doctors were unable to find any disease in the blood of these patients.
This medicine is exactly Kymriah. In August last year, after nearly three decades of development, the first CAR-T drug was finally approved by the FDA.
These are patients who have no medicine to treat, and they will recover quickly after using CAR-T medication, even doctors will find it unbelievable. In July last year, when the FDA’s external review vote unanimously recommended that Kymriah be listed, a jury member said that this was the most exciting thing he had ever seen in his life.
It’s not just doctors and patients who are excited. After Kymriah’s approval, pharmaceutical companies and investors have become mad at the CAR-T sector.
That is, in the month when Kymriah was approved for listing, Gilead acquired Kite Pharma for $11.9 billion, and soon Kite’s CAR-T product, Yescarta, was also approved by the FDA.
After that, New Base acquired Juno therapeutics for $9 billion. At the same time, countless investment institutions have begun to spill a large amount of money on biotech companies engaged in the development of a new generation of CAR-T therapy.
Up to now, only Kymriah and Yescarta have been approved for marketing. The current patient population for the approved indications is not large, and because CAR-T therapy is a highly individualized therapy, the price is very expensive. Kymriah’s single treatment price is $475,000, so it was accepted after approval. Not many patients have been treated with these two drugs.
At the same time, patients receiving treatment may also have serious toxic reactions, such as cytokine release storms and cerebral edema. In clinical trials, there have been many cases of death due to toxic reactions, so the therapy is usually only applied to the patient’s last-line treatment.
At present, the research progress of CAR-T therapy in the field of solid tumor treatment is not great, and if the CAR-T therapy is extended to the treatment of solid tumors, the control of toxicity will face greater challenges.
Therefore, whether it is a doctor, a pharmaceutical company or an investor, the expectations for the future of CAR-T therapy have begun to become more cautious.
The success of a hematoma is inevitably or accidental?
I remember one time when listening to the report of the meeting, a child said that CD19 is a gift from God. Indeed, CD19 is indeed an excellent target for CAR-T treatment.
It can be clearly seen from the following figure that CD19 is widely present on the surface of B cells, and traces of CD19 can be found on the surface of the corresponding B cells from the entire stage of Pre B to mature B cells.
CAR-T cells targeting CD19 are able to clear B cells expressing this antigen: whether it is healthy B cells or B cells that cause leukemia or lymphoma, CAR-T cells are able to remove them very efficiently.
So it is clear that these CAR-T cell attacks are not tumor specific, as these CAR-T cells also cause damage to normal B cells.
But the key to the success of the therapy is that even if the B cells in the patient are removed, it usually does not have a serious impact on the health of the patient. Because patients can compensate for the effects of normal B cell damage by entering immunoglobulins, the risk of clearing normal B cells is controllable.
CAR-T therapy has extremely strict target requirements, and the requirements for CAR-T targets are:
- Expressed on the cell surface.
- This target cannot appear on important organs or cell types other than tumors (such as hematopoietic stem cells), and even low levels of expression may cause serious side effects.
- In order to avoid antigen escaping, all tumor cells must express the antigen. Even if the antigen cannot be expressed on the surface of all tumor cells, it must be crucial for the maintenance of tumorigenic phenotype.
The first one is easy to understand, which is determined by the MHC-independent nature of CAR-T cells. The second is critical for controlling the toxicity of CAR-T cells and will be explained in detail later in the section on solid tumor progression. Because the sensitivity of CAR-T cells is very high, even low levels of antigen expression can trigger T cells to attack the corresponding cells. Article 3 is essential to ensure the effectiveness of CAR-T therapy.
It is obvious that CD19 can meet the above three requirements. Moreover, in addition to CD19, other antigens on the surface of B cells can also satisfy the above characteristics, such as BCMA.
BCMA is also a very good target for CAR-T treatment. Although no related products have been approved for marketing, in clinical trials, CAR-T therapy targeting BCMA has been used successfully for the treatment of multiple myeloma.
Although CAR-T therapy targeting CD19 and BCMA has achieved great success, toxicity remains a major obstacle.
Although Kymriah and Yescarta were successfully approved, not all pharmaceutical companies are as lucky as Novartis and Kite.
In March 2016, Juno Therapeutics stopped clinical trials in five patients who died of cerebral edema in Phase II clinical trials. This severe brain edema is one of the main haze of CAR-T therapy research. Until now, researchers have not been very clear about the specific formation mechanism of cerebral edema.
Cerebral edema is not the only toxic reaction of CAR-T therapy, and the return of highly active CAR-T cells may lead to other serious consequences. In a clinical trial conducted in Novartis in 2015, approximately half of patients developed a severe inflammatory response: cytokine release syndrome (or cytokine storm cytokinestorm). We now know that this is a common side effect of CAR-T therapy.
CAR-T cells are very different from traditional drugs because they are living drugs. CAR-T cells can be exponentially amplified after being activated by antigens, and CAR-T cells also release cytokines when they attack tumor cells, thereby promoting inflammation and recruiting more immune cells. If these CAR-T cells attack too quickly, it can trigger a deadly cytokine release storm.
At present, cytokine release storms are generally controllable, and doctors can control the reaction by using drugs such as tocilizumab. However, in extreme cases, timely and rapid removal of CAR-T cells that cause toxic reactions will likely control severe toxicities.
The use of a suicide switch to control CAR-T cells is a good choice. In the event of serious toxic reactions, it can rapidly induce CAR-T cell apoptosis by controlling suicide switches on CAR-T cells. The toxic side effects are further aggravated.
There are many suicide switches designed by academic institutions and researchers in biotechnology companies. There are already some CAR-T therapies using suicide switches to enter preclinical/clinical research.
The safety switch used in CAR-T therapy developed by Bellicum Pharmaceuticals has two engineered proteins that bind to the small molecule drug rimiducid to dimerize and activate caspase-9 and ultimately induce CAR-T cell apoptosis.
Bellicum has recently launched Phase I clinical trials of the GoCAR-T product BPX-601. Another biotechnology company, Poseida Therapeutics, is also using the riniducid-activated CAR-T suicide switch.
In addition to controlling the switch by small molecules, suicide switches can also be controlled using antibodies. The CAR-T cells used by Cellectis are capable of binding to rituximab, and CAR-T cells may be recognized by normal immune cells and cleared by corresponding CAR-T cells after being bound by rituximab.
It is possible to clear CAR-T cells within a few hours by a small molecule-controlled suicide switch, and this rituximab-labeled CAR-T cell takes several days to be cleared. A big drawback of class switches.
It is not known whether these suicide switches can effectively control serious toxic reactions, and many people now do not think suicide switches are necessary.
In addition to these biological problems, CAR-T therapy faces the problem of time and money. Chemotherapy can be widely used in cancer treatment, not only because it can slow or prevent the progression of tumors, but also because these drugs are relatively high in accessibility, relatively inexpensive, and can be prepared, transported, and stored in large quantities.
In contrast, CAR-T therapy, in particular, CAR-T cells require patients to separate immune cells from the blood in the hospital during the preparation process, freezing (some products do not require freezing during the preparation process) and transporting these cells. To the CAR-T factory, after thawing, the virus vector is used to introduce the gene, frozen, and then transported back to the hospital, thawed, and finally returned to the patient.
Taking Novartis’s medicine as an example, this process takes about 22 days. Taking into account the complexity of this process and the urgency of time, Novartis priced its drug as a single treatment of $475,000.
In addition, many advanced patients have difficulty collecting T cells, and it is difficult for infants and younger children to extract sufficient amounts of T cells, which makes them unable to complete CAR-T therapy. first step.
If CAR-T therapy really wants to fight against chemotherapeutic drugs, as a first-line treatment, we must find a way that is easy to implement, easy to manufacture, and acceptable to patients at the same price as chemotherapy drugs.
A more popular way to solve this problem is to use a generic CAR-T, that is, to make CAR-T cells using T cells derived from others rather than the patient itself.
There are also many studies in this area. For example, UCL immunologist Waseem Qasim tried to use donor T cells and genetically engineered them to knock out TCR and CD52 on T cell surface to avoid immune rejection.
However, the general-purpose CAR-T also faces the same risk of side effects as common CAR-T. For example, in August last year, a 78-year-old patient died of cytokine release storm after receiving Cellectis’ universal CAR-T product UCART123. .
Despite the potential for cerebral edema and serious side effects such as CRS, CAR-T therapy is truly revolutionary for patients with B-ALL, multiple myeloma and other medically ill patients. change.
However, the recent high investment intensity in the CAR-T field is actually because many people have placed the future of CAR-T therapy on the field of solid tumors.
However, there are numerous obstacles to the application of CAR-T therapy in the field of solid tumors.
Marching solid tumors
Compared to the success of B cell-associated hematological cancer, CAR-T therapy is much more difficult to achieve breakthroughs in the field of solid tumors.
In fact, CAR-T therapy has been tried in the field of solid tumors, but so far no substantial progress has been made.
The three requirements for CAR-T therapy for antigens are mentioned above: 1 Expression on the cell surface. 2 This target cannot appear on important organs or cell types other than tumors (such as hematopoietic stem cells), and even low levels of expression may cause serious side effects. 3 In order to avoid antigen escaping, all tumor cells must be able to express the antigen, even if the antigen is not expressed on the surface of all tumor cells, it must be crucial to maintain tumorigenic phenotype.
We can indeed find some B-cell-related, very good targets for CAR-T therapy, and CD19 and BCMA are good examples. But in the field of solid tumors, it is extremely difficult to find such an ideal target.
This is a very difficult problem to balance. If the antigen is expressed too broadly, the toxicity may be difficult to control. If the antigen is expressed less, it will affect the effectiveness of CAR-T therapy.
Taking the target of HER2 as an example, HER2 can be overexpressed on the surface of many types of tumor cells. Whether it is a small molecule drug, a monoclonal antibody drug, or an ADC, the controllability of the drug and the side effects of this target have been verified. .
However, the CAR-T therapy targeting HER2 caused a fatal toxicity in the treatment of the first patient, leading to death.
This CAR-T therapy is based on the third generation CAR design (CD28/4-1BB/CD3(R)) and uses a high affinity scFv based on Herceptin. Later, it was found that this toxicity was caused by CAR-T cells recognizing HER2 on the surface of lung epithelial cells, and CAR-T cells were able to attack these cells extremely violently, causing severe lung failure.
Although the first patient in the clinical application died, there were still some CAR-T therapies targeting HER2. Subsequent CAR-T therapy products use lower doses and reduce the affinity of scFv for improved safety, but unfortunately even if these drugs can guarantee the safety of the therapy to some extent, in clinical trials No such CAR-T therapy has been shown to be sufficiently effective.
However, many researchers have not been discouraged. There are still a large number of clinical studies related to solid tumors, and some of them may give us a glimmer of hope.
A 50-year-old man with glioblastoma has already metastasized to the spine. At that time he had not responded to any other type of treatment. However, the condition was relieved after multiple injections of CAR-T cells directly injected into the central nervous system.
This CAR-T therapy targets tumor cells that express IL-13 receptor proteins, and most of these tumors are capable of overexpressing such proteins.
But this is really just a case. How convincing this example is, I believe that everyone’s views will be different. In addition to IL-13R and HER2, there are many solid tumor targets in clinical research, such as PSMA, mesothelin, FAP, EGFRvIII, EGFR, CEA, GD2 and so on.
I believe that many people will be like me, and the attitude of CAR-T therapy in the field of solid tumors will be very pessimistic. However, many people also think that TCR-T therapy will be a more potential way to solve solid tumor disorders, because TCR-T has some advantages over CAR-T in target selection in solid tumors.
The scFv of CAR-T is consistent with the mode of action of antibodies, and only recognizes antigens expressed on the cell surface, while cell surface proteins usually account for only 10% of cellular proteins.
However, TCR can recognize the cell surface presented by MHC as well as intracellular antigens, so for TCR-T therapy, it has a much wider range of choice than CAR-T therapy targets.
But it does not mean that TCR-T has no defects. A more important limiting factor is the need to perform HLA matching on patients before receiving treatment. Of course, there are also some new technologies to overcome this obstacle.
There are also many TCR-T targets in clinical stage, such as NY-ESO-1, MAGE-A4/A4/A6/A10, WT1, mesothelin, HPV-16E7, AFP and so on.
Although TCR-T therapy can expand the range of antigens to be selected, it is difficult to meet the three target requirements mentioned above, whether overexpressing tumor antigens or tumor-specific antigens. New technologies are still difficult to overcome obstacles to target selection, so it is equally difficult to achieve sufficient effectiveness while ensuring safety.
From the perspective of antigen selection, perhaps new antigens will be a more promising development direction. The new antigen refers to a new peptide antigen produced by tumor mutation and is therefore inherently tumor specific. However, because each patient carries a different new antigen, the therapy based on the new antigen is highly individualized.
I still believe in the potential of the new antigen. Steven Rosenberg of NCI has been conducting research in this field for many years. If there is a chance, it will focus on the progress of the new antigen field in a new article.
But even if the new antigen can prepare new antigens quickly and at a cost-controllable condition, the new antigen will not play much role for most patients with metastatic tumors, because the choice of antigen is only CAR-T, etc. One of the obstacles to cell therapy, even if a relatively good target is found, CAR-T cells are unlikely to achieve the shocking effects of hematoma.
Difficulties in solid tumors
To successfully clear a tumor, these CAR-T cells must accomplish four things: reaching the tumor, entering the tumor, fighting the tumor’s immunosuppressive microenvironment, and killing the tumor cells.
The first and last ones are the easiest to accomplish. It is the second and third that hinder the progress in the field of solid tumors.
Researchers at the University of Pennsylvania have conducted a clinical study of glioblastoma. They found that CAR-T therapy targeting EGF receptor III protein can indeed reach and enter brain tumors after local injection of drugs, killing tumors. cell. But after two weeks of injection, the number of T cells began to decline.
People always have the impression that these engineered T cells can spontaneously expand in the body and spontaneously attack tumor cells. But this is not the case. After returning to the human body, these cells will face the same dilemma as normal immune cells in the human body. It is still necessary to overcome the heavy immunosuppression of the tumor microenvironment.
In order to solve the problem of immunosuppression in solid tumors, many people are trying to use CAR-T therapy and other immunotherapy, such as the use of checkpoint inhibitors, to break through immunosuppression. Moreover, in addition to directly using checkpoint inhibitors, it is also possible to knock out checkpoint proteins on the surface of T cells by genetic engineering techniques, which theoretically makes it possible to achieve similar effects to the combined checkpoint inhibitors.
In addition to these methods, armored CAR-T is also a research area, such as by transforming CAR-T cells to secrete IL-12, IL-15, CD40L and 4-1BBL.
The first two days of the Rentier Brentjens lab also published a study that allowed the engineered CAR-T cells to secrete PD-L1 scFv.
Regardless of the extent to which armored CAR-T can enhance the potential of CAR-T therapy for solid tumors, these studies have many overlaps with other tumor immunotherapy studies. How to overcome tumor micro-environmental disorders and how to infiltrate tumor cells with immune cells are the main problems currently facing all tumor immunotherapy including PD1/L1 inhibitors.
In fact, if you are interested in CAR-T technology and often pay attention to the progress in this field, you will find that the new technologies in the CAR-T field are diverse, and many of these various technologies are the darling of the primary market and the weapon of financing.
Due to space limitations, this article will not cover all types of new CAR-T technologies. Because of the big difference from the traditional CAR-T technology, we will refer to these new technologies as CAR-T 2.0.
The first half of the article mentions that many researchers have been studying how to use suicide switches to regulate CAR-T cell activity to ensure safety. But in 2015, UCSF scientist Wendell Lim published a study in Science that they designed another unique switch.
This new technology is called Throttle. In this technique, small molecules are capable of dimerizing the CAR protein to activate T cells. That is, in the year the article was published, Lim formed a new company called Cell Design Labs.
In addition to Throttle, Cell Design Labs has another black technology.
In order to improve the ability of CAR-T cells to recognize antigens, Lim Labs designed an extremely flexible system called synNotch. A typical synNotch contains extracellular regions as well as transmembrane regions, as well as intracellular transcribed regions. When the Notch receptor is activated, the intracellular transcribed region can be released by splicing and transported into the nucleus to regulate the expression of the target gene.
Through the combination of synNotch and CAR design, synNotch can be used to regulate the expression of CAR to achieve the conditional expression of CAR.
We can take ovarian cancer treatment as an example. Mesothelin is a common target for solid tumor CAR-T treatment. It can be overexpressed in many types of tumors, but it can also be expressed in normal organs, such as the lungs.
As we mentioned earlier, common CAR-T therapy is difficult to achieve both safety and effectiveness in dealing with this target.
To solve this problem, we can look for a protein that is only expressed in the ovary, which binds to and activates synNotch, and binds the activation of synNotch to the expression of CAR that targets mesothelin: when synNotch is activated It is able to induce related CAR expression in T cells, which enables CAR-T cells to express CAR only in the ovary, avoiding damage to other tissues. But this is only theoretically the case.
Two years after Cell Design Labs was founded, Gilead announced that it had acquired the company for $567 million without any product entering clinical research.
In addition to the synNotch system, there is a more interesting research direction: all CAR-T cells express this CAR by designing a universal CAR that does not have antigen recognition and can bind to antibodies. The targeting specificity of CAR-T cells is then regulated by antibodies.
Thus, the targeting of CAR-T cells can be regulated by changing the type of antibody, and the activity intensity of CAR-T cells can be regulated by adjusting the dose of the antibody drug.
Unum Therapeutics has been conducting research in this area and has pushed related products to clinical trials. Unum’s technology platform is called Antibody-Coupled T cell Receptor (ACTR).
Although this technique seems very interesting, it has been found in clinical trials that this technology is equally likely to be extremely risky.
In the IPO Form S-1 submitted by Unum this year, investors found that its core product, ACTR087 (based on rituximab), treats relapsed and refractory non-Hodgkin’s lymphoma. In clinical trials, 10 patients died accidentally in 3 patients. Two patients died of cytokine release syndrome and one died of neurotoxicity. I believe that no one will want to see such a result. While these new technologies look fancy, they also have a lot of unknown risks.
In April of this year, Wilson Wong, who had worked in Lim Labs, made the concept of this universal CAR the ultimate.
In an article by Cell, Wong reports on a new technology that uses a universal receptor called SUPRA CAR. This CAR receptor consists of zipCAR and zipFv. The system is detachable and can be implemented. Very fine adjustment. The work in the paper is very beautiful and the article is very readable. Based on this technology, Senti Biosciences, a biotechnology company co-founded by Wong this year.
There is probably no more complex technology in the CAR-T field than this technology. But technical fancy doesn’t necessarily mean that it has a promising future. The more complex the technology, the less likely it is to develop a new treatment that can really be used for disease treatment, because each additional operation adds a layer of risk, which in turn increases Increased the chance of failure.
In addition to these more eye-catching new technologies, there are many new ideas in the CAR-T field. For example, by light control, ultrasound to control the activity of CAR-T cells.
It is difficult to predict whether these very fancy new technologies will fall to the ground, when they will land, and how many years later these technologies will really help patients and benefit them. But it is clear that these expectations are impossible to achieve in the short term.
Perhaps influenced by the approval of Kymriah and Yescarta, many investors have unrealistic fantasies about CAR-T technology, imagining that this technology can revolutionize the treatment of multiple tumors.
But the reality is always cruel, and the maturity of technology takes a very long time. B cell-associated antigens such as CD19 and DCMA are so-called low hanging fruits, and the discovery of these targets has contributed to the success of CAR-T therapy. But the success of CD19 and DCMA does not mean that we can extend the victory to other areas of cancer treatment.
Can CAR-T therapy capture the next city and win the next city?