A Non-Consensus, More Balanced Look at the CAR-T Development Efforts of Kite and Juno (Kite, $48.53: Juno, $30.42)
Definition of Key Terms Used in This Report
Engineered Autologous T-cells (eACTS): These are T-cells that are removed from a patient and genetically engineered to increase their activity against cancer cells. They are then expanded into billions of clones that are reinfused into the patient where they continue to differentiate and attack cancer targets.
Chimeric antigen receptor T-Cells (CAR-T): This is the first generation of eACTs. Genetic engineering leads to the expression on the surface of the T-cell of an artificial receptor that has a fragment of an antibody, which can target a cancer antigen. These cells combine the specificity of an antibody with the killing power of a T-cell.
Relapsed/ refractory (r/r): This describes a cancer that has begun to grow again (relapsed) after all standard of care treatment options have been exhausted (refractory).
Acute lymphoblastic leukemia, also known as acute lymphocytic leukemia (ALL): This is a cancer that affects progenitor B-cells. Stem cells in the bone marrow divide into a cell that then goes through a number of additional differentiations into other cell types before it matures into a plasma cell that produces an antibody. These progenitor B-cells can become cancerous and as they grow uncontrollably can crowd out the production of normal cells in the bone marrow. They also invade the blood and spread cancer widely throughout the body. The term acute is used to describe an aggressive form of the disease that can progress quickly and can be rapidly fatal if not treated.
Pediatric relapsed/ refractory acute lymphoblastic leukemia (pediatric r/r ALL): About 80% of pediatric ALL can be successfully treated with chemotherapy and possibly stem cells transplants. The remaining 20% makes up the r/r patients that may be treated with CAR-T cells.
Adult relapsed/ refractory acute lymphoblastic leukemia (adult r/r ALL): Treatment of the adult patients with ALL is much less successful than with pediatric ALL.
Diffuse large B-cell lymphoma (DLBCL): This is one of the most aggressive forms of non-Hodgkin’s lymphoma which is also a cancer of B-cells. It usually occurs in older adults. About 60% of patients can be cured with chemotherapy. The remainder relapse and usually die within three years. The remaining 40% makes up the r/r DLBCL patient population that may be treated with CAR-T cells
I have felt for some time that investors were overly focused on the impressive treatment effects of CAR-T while ignoring very troublesome side effect issues which have been brought into sharp focus with the deaths in Juno’s ROCKET clinical trial. Equally troubling from my perspective has been how sanguine analysts and investors have been about the early stage of development, limited amount of data from clinical trials, significant competitive uncertainties and the limited commercial opportunity in the initial markets addressed.
It seems to me that the current market valuations of Juno and do not reflect these issues. This is not a sudden awakening on my part as I first wrote of my concern about the stock prices of Kite and Juno in an April 29, 2015 report called CAR-T Companies-Kite (KITE, $55.09) and Juno (JUNO, $47.30): Is the Bloom Coming off the Rose? Are Investor Expectations Unrealistic? They have seemingly been priced for perfection as the majority of Wall Street analysts have fawned over their potential. (We cannot ignore the possibility that huge investment banking fees stemming from equity offerings of Kite, Juno and several other CAR-T companies could have led to positive biases.)
Stock psychology can have great pendulum like swings from exuberance to disillusionment and we could see this happening with Kite and Juno. I think that if CAR-T fundamentals were to be viewed with more balance, we could see a more cautious or negative consensus view emerge that could result in more weakness of these stocks in coming months. I must warn you that this opinion goes against Wall Street consensus.
Key Conclusions of this Report
This report is divided into two sections. The first section goes through some key clinical and commercial issues that investors should consider in regard to CAR-T products in general and Kite and Juno in particular. I think as you go through this first section, you may come to agree with me that while the therapeutic results of CAR-T have been extremely impressive (breakthrough may not be an exaggeration), but the side effects are as alarming as efficacy is impressive. I also think that investors may not realize that the initial indications in r/r ALL and r/r DLBCL have small patient populations and perhaps limited commercial potential. If CAR-T can be effectively and safely used in earlier stages of these cancers, the commercial potential is of blockbuster size. However, for this to happen, the side effect issues must be manageable and I am not sure that this can be readily achieved.
Very importantly, the competitive situation has been very much under-emphasized on the Street. There are three companies vying to be the first to gain approval of a CAR-T product; somewhat ominously the giant pharma company Novartis is the third player. It is difficult to be precise, but it seems likely that Novartis will gain the first ever CAR-T approval for the indication of pediatric r/r ALL in 2017. Novartis and/or Kite could possibly be the first to gain approval for the r/r DLBCL indication in 2017. Juno’s management has guided that its product would be approved for r/r adult ALL in 2017, but the recent clinical hold issue with the ROCKET trial could mean that it might be delayed until 2018.
The products of all three companies are based on the same mechanism of action which is CAR-T targeting of CD-19 on B-cell progenitor cells. There could be differences in the engineering of the chimeric receptor and/or manufacturing of the living cells that might make one product more effective than another in a given B-cell malignancy. However, early clinical studies indicate they are comparable in effectiveness, whether it be r/r ALL or r/r DLBCL.
If it is the case that the three CAR-T products are viewed as not being meaningfully differentiated in any indication, then the Company that gains the first regulatory approval and with that first mover advantage might have an enormous commercial advantage. Here is my reasoning. The marketing of CAR-T treatment is more likely to take the form of medical device marketing in which representatives sits in on the surgery and provides technical advice; they are an integral part of the procedure. This is indeed the sales model that Novartis is contemplating. See the article Novartis considers new sales model for experimental cancer therapy.
This sales model requires in a very close working relationship and trust between the Company and institution. If an institution views the CAR-T products as being largely undifferentiated, I think that it would want to deal with just one company, regardless of whether the treatment is for r/r pediatric ALL, r/r adult ALL or r/r DLBCL.
A counter argument is that physicians will only use the products in accordance with its approved indication. If this is the case, the Novartis product likely would be the first to be used for r/r pediatric ALL and the Juno product would likely to be the first for r/r adult ALL. In the case of r/r DLBCL it could be either Kite or Novartis. In my opinion, this is unlikely because of the very difficult logistics required to deal with two or even three companies.
If I had to put my money on a company that gains first mover advantage, it would be Novartis. Although, given the vagaries of clinical trials this is not a given. In addition to the possible first mover advantage, Novartis probably has much greater financial resources, cancer drug marketing skills and reimbursement understanding than Kite and Juno. If I were a key manager at Kite or Juno, I would be very worried about largely being shut out of the r/r ALL and r/r DLBCL markets. However, both managements have billed these markets as having enormous commercial potential for them.
The second section of this report is a layman’s (my own) view of immune-oncology. I give a brief overview of the whole field as I see it and how eACTS may fit in. I believe that immune-oncology has already caused a paradigm shift in treating cancer, but the best is yet to come. The clinical development work on eACTS is very early and it is difficult to discern when or if they will be widely used. They are almost certainly a breakthrough for relapsed/ refractory B-cell cancers, but their role beyond that that remains to be determined. I can imagine scenarios in which they are hugely important and also scenarios in which they are disappointing. My ambivalence reflects the sparsity of clinical data.
This second section is a bit heavy and generally elaborates on points made in the first. If you decide to skip it, you would still largely understand my points of view, but I would urge you to take the time to read it.
The key points I make in this section are:
- CAR-T cells have produced dramatic responses in r/r ALL and r/r DLBCL for patients who have failed most or all treatment options and who often have life expectancies of less than a year. In some cases, something like a cure has been achieved.
- Unfortunately, side effects are also dramatic and neurological toxicity has led to patient deaths.
- The mechanism of action for CAR-T is extremely applicable to hematological cancers (about 20% of cancer incidence) but perhaps to only a very few (if any) solid tumors (80% of cancer incidence).
- The side effect profile is sufficiently troubling that use may be limited to relapsed/ refractory patients who have failed all other therapeutic options.
- If so, this means that the initial markets addressed by the first generation CAR-T products may be small commercial opportunities.
- There are three companies that are leading the development of first generation CAR-T products with the giant pharmaceutical company Novartis being one along with Kite and Juno.
- The products of these three companies have essentially the same mechanism of action and may not be meaningfully differentiated.
- Novartis has enormous financial resources and marketing skills in comparison to Kite and Juno and could be the first company to gain regulatory approval which could give it first mover advantage.
- Market valuations of Kite and Juno seem based on the expectation of blockbuster commercial returns for their first generation CAR-T products.
- CAR-T cells are only the first generation of engineered autologous T-cells. There are many creative ideas and approaches that will address the shortcomings of first generation CAR-T cells. However, these are several years away and companies other than Kite, Juno and Novartis could be the ultimate winners.
Mechanism of Action of First Generation CAR-T Products
Led by work done at the NCI over the last 20 years, scientists have developed a technology that can engineer a receptor that contains an antibody fragment (referred to as chimeric) on the surface of T-cells that can seek out and bind to a single, specific antigen target; these are called chimeric antigen receptor T-cells or CAR-T cells. They combine the specificity of an antibody with the killing power of a T-cell.
This is a different type of drug. It is not an organic compound; it is a living cell. The manufacturing process starts with the collection of T-cells from a patient. They are then engineered ex vivo to express a receptor that recognizes and binds to the targeted antigen. The resultant CAR-T cells are induced to proliferate in the laboratory to produce billions of identical cells (clones). They are then re-infused into the body where they continue to proliferate and also attack any cell in the body that expresses the specific antigen on its surface. The CAR-T cells will attack normal as well as cancerous cells.
The mechanism of action with the first generation of CAR-T cell products targets the antigen CD-19. CD-19 is a very attractive target because it is expressed in significant amounts on the surface of B-cell lineage cells. The CAR-T cells attack and eliminate both cancerous and normal B-cells expressing CD19. You can think of them as indiscriminate killers that wipe out the bad guys (cancer cells) but in the process also wipe out the good guys (normal cells). Without B-cells, this means that for some time, the body can’t produce an adequate antibody response to fight infectious disease and the patient has to be given immunoglobulins. To a layman like me the inability to produce antibodies would seem to be a show stopper. However, the blockbuster monoclonal antibody Rituxan (introduced in 1997) has the same mechanism of action and has demonstrated an acceptable risk to benefit profile. You can think of CAR-T cells as a “super Rituxan”; metaphorically CAR-T cells are heavyweight fighters while Rituxan is a lightweight or middleweight.
Early Efficacy Results in B-cell Malignancies Have Been Impressive
The CD-19 targeted CAR-T cells have produced amazing results in r/r ALL. In patients who had exhausted all therapeutic options; these patients were facing a life expectancy of a year or less. In this patient universe they appear to generate complete responses of 80% to 90%. Investigators suggest that current treatments lead to CRs of 10% although this has not been documented in a controlled study. In some patients, the CAR-T s have resulted in something close to a cure or treatment allows the patient to go on to a stem cell transplant that can be curative. The next most widely studied cancer is relapsed/ refractory diffuse large B-cell lymphoma or r/r DLBCL. Here the results are still impressive with very small trials suggesting CRs on the order of 50% to 60%. r/r ALL and r/r DLBCL are both attractive targets because they broadly express CD-19 on their surfaces.
I think that the lay press has sometimes suggested that the results for CAR-T therapy in r/r ALL and r/r DLBCL are suggestive of what can be expected in all cancers. This is most definitely not the case. CAR-T is limited because it is a rifle shot approach that can only attack cancers that express a single antigen on their surface. In other cancer therapeutic modalities this single shot approach has proven to be a drawback. While cancer cells expressing this antigen target may be killed, other cancer cells that don’t express it continue to grow unimpeded. Also solid tumors often do not express key antigens on their cells surface and do not produce viable targets. From a side effect standpoint, the indiscriminate killing of normal cells is a major drawback.
Hematological cancers like r/r ALL and r/r DLBCL are perfect targets for CAR-T because they express high quantities in CD-19 on their surface. However, many solid tumors do not express antigens on their cell surface. The key takeaway is that the first generation CAR-T cells can address hematological cancers that account for 20% of the incidence of all cancers, but not solid tumors that account for the other 80%.
Side Effect Profile is Troubling
The second key issue for CAR-T cells is a very troubling side effect profile. T-cells and of course CAR-T cells produce cytokines (proteins) that attack targeted cells and also signal other cells in the immune system to join the attack. CAR-T cells expand rapidly in the body and this can lead to the over-production of cytokines which results in cytokine release syndrome, severe cases of which are called cytokine storms which can lead to fever, etc. etc. etc. Let me first explain how side effects are classified: grade 5 is causing death, grade 4 is life threatening and grade 3 is serious. There is very limited clinical data, but the data suggests that about 20% to 30% of patients treated with CAR-T experience grade 3 or 4 side effects which may require hospitalization.
Perhaps, more troubling is that neurological toxicity in the form of cerebral edema (swelling in the brain) has caused deaths. In the ROCKET trial, 3 of 20 patients treated died from this and in a DLBCL trial conducted by Novartis 1 of 15 patients died. In the ZUMA-1 DLBCL phase 1 portion, one person in the 7 patients treated died from the bursting of a cerebral artery, but investigators and Kite say this was not related to the drug.
The key takeaway point is that outstanding efficacy is offset by harsh life-threatening side effects. This is acceptable in relapsed/ refractory patients who have no treatment options. It should also be noted that death is also a risk with many chemotherapy regimens and this yet they are still widely used. However, this is a major deterrent to using the therapy in less severe cases.
Is There A Role in Solid Tumors?
The indiscriminate killing of normal cells is acceptable for hematological cancers as I explained earlier. Almost total destruction of normal cells can be managed and is an acceptable risk of I would also note that many cancer therapies kill normal cells. This drawback is managed by limiting the dose to a level that produces side effects that can be tolerated and combining cancer drugs with different types of side effects. The therapeutic effects are synergistic while side effects can be managed. The problem with CAR-T cells is that they are so powerful that even at low therapeutic doses they can cause serious toxicity with normal cells even if antigen expression is low. Novartis has reported that they developed a CAR-T against HER-2, which is a well-known breast cancer antigen. However, this antigen is also expressed in lung tissue and even though this expression is low, this product caused unacceptable side effects.
Facts to Know about Clinical Development of CAR-T cells
The key points in the following paragraphs are as follows:
- There is very limited clinical data on CAR-T cells. To date, only two phase 1 trials have been completed.
- Kite, Juno and Novartis have relied on technology in-licensed from NCI and academic centers. They are developers not originators of the technology. None of these companies has completed a clinical trial with their products.
- Each of these three companies have indicated that they expect approval of their products in 2017 based on the expectation of positive results in open label phase 2 trials in small patient groups. Juno may have to change guidance to expected approval in 2018.
- One of the recurring problems with biotechnology companies is that in order to raise capital to run their trials, they have to persuade investors that the clinical development timelines are short. Hence, they jump into registrations trials lacking key information on their drugs. This certainly seems the case here.
- The manufacturing process used is based on autologous living cells and is quite expensive, Current estimates are that the price of therapy will be $250,000 to $350,000.
- The phase 2 trials of Kite and Juno are intended to gain approval for the use of CAR-T as a bridge to stem cell transplants. The latter can cost $300,000 to $1,000,000 over and above the cost of CAR-T therapy.
- I think that in order to routinely justify this enormous expense there will have to be evidence of a long duration of effect which the interim and initial results of the phase 2 trials being conducted by Kite, Juno and Novartis are not likely to provide. This could possibly take years of following patients after completion of the enrollment and treatment.
The development of first generation CAR-T cells has been based on pioneering work at the National Cancer Institute over the past 20 years. Leading academic centers such as a collaboration of Children’s Hospital of Philadelphia (CHOP) and the University of Pennsylvania as well as leading academic cancer centers elsewhere in the country have built on the work at NCI. To my knowledge there have only been two phase 1 clinical trials of CAR-T products. These were one r/r ALL trials conducted by NCI and a second by CHOP/U of P building on the work of NCI. There have been a meaningful number of patients treated with CAR-T products in a non-clinical trial setting, but these are problematical because of widely varying patient selection, dosing intensity and intervals, pre-conditioning regimens, different manufacturing approaches and other issues. While informative, there are drawbacks in trying to rely on results not obtained in a clinical trial setting.
Kite was founded in 2009 and its technology comes from the NCI via CRADAs, the first of which was signed in 2012. Novartis in-licensed technology from CHOP/ U of P in 2011 or so. Juno was only formed in 2013 by venture capitalists who in-licensed technology from other leading academic centers. None of these companies have built a solid, long term foundation of experience and knowledge in the technology which often takes decades.
Kite, Juno and Novartis have never completed a clinical trial with their CAR-T products. They have each taken the results of work done at the institutions whose technology they have licensed and jumped into phase 2 trials. These phase 2 trials are open label (no control group) in relatively small patient populations. In each case, their manufacturing process differs from that used by their licensors. Because they are dealing with living cells, there is the potential that even small changes could produce different outcomes. One of the most frequent reasons for biotechnology clinical trial failures is moving too rapidly into registration trials before companies have adequately determined product characteristics and this seems to be the case here.
Juno and Kite have had to rely on the capital markets to fund their companies and have been enormously successful in raising huge sums of cash. However, in order to entice investors they had to promise short timelines for clinical development and regulatory approval. A more thoughtful, carefully executed clinical trial program could take several years and investors generally respond poorly to such long time lines. Kite has suggested that its product for r/r DLBCL will be approved in early 2017 based on an interim analysis of a phase 2 trial. While I think that this could be possible, it is an extremely aggressive representation. Juno was previously guiding to approval in 2017, but this will very likely be changed to 2018.
I think that autologous living cell therapy has extremely promising potential, but the building of manufacturing expertise and clinical experience is critical and there is much to be learned. It is also very expensive and it has been suggested that the price of therapy could be $250,000 to $350,000. In addition, the phase 2 trials of Kite and Juno (not Novartis) are intended to gain approval for the use of CAR-T as a bridge to stem cell transplants. The latter can cost $300,000 to $1,000,000 over and above the cost of CAR-T therapy. I think that in order to routinely justify this enormous expense there will have to be convincing evidence of a long duration of effect. The interim and initial results of the phase 2 trials being conducted by Kite, Juno and Kite are not likely to provide this information. This will take years of following patients after completion of the enrollment and treatment.
Perspective on Immune-oncology Drug Development
-The Adaptive Immune System
Immune-oncology drug development is based on an understanding of how the immune system responds to cancer and developing drugs that mirror its biological action. The immune system is enormously complicated involving many different types of cells and the cytokines (proteins) which they produce to attack abnormal cells such as cancer. There is an innate immune response which is always present as a first line, general response to all abnormal cells (cancer, bacteria, parasites, etc.) and then an adaptive response that is directed specifically. The innate system mechanisms have drawn some research interest, but most of immune-oncology drug research has focused on the adaptive immune system and particularly on the functions of two key cell groups. These are B-cells that produce antibodies and T-cells.
The most established immune-oncology treatment modality is monoclonal antibodies. Nature designed B-cells to produce antibodies that can detect and bind to specific antigens (molecules) that mark a cancer cell as being abnormal. Antibodies can sometimes act directly on their own to destroy the targeted cancer cell and/ or trigger further complementary responses from other components of the immune system which work in concert.
There are effector B-cells that produce antibodies against a targeted antigen on the cancer cell. There are also memory B-cells which at a repeat encounter with an antigen, can quickly differentiate to mount a faster and stronger immune response than the first time the immune system was exposed to an antigen. This is probably more important in fighting infectious disease than cancer although it might be applicable to a cancer that has gone into remission and then recurs.
In the 1970s scientists discovered techniques that allowed for the broad scale production of a (monoclonal) single antibody. Prior to that production techniques only allowed for the production of polyclonal (several different types) antibodies which were difficult to produce in scale and with consistency. One of the significant advantages of monoclonal antibody technology is the ability to manufacture it in large quantities with a high degree of purity.
Rituxan was the first blockbuster monoclonal antibody drug product. It was introduced in 1997 and targets an antigen (the CD 20 molecular complex) that occurs on the surface of lymphoma and leukemias that affect B-lymphocytes (the white blood cells that produce antibodies). The science lies in discovering an antigen target that characterizes the cancer such as CD20 for Rituxan, developing an antibody specific to that antigen and then using monoclonal antibody production techniques to make the drug in large quantities. Rituxan led to a massive effort to develop other monoclonal antibodies for cancer and has led to some very medically important and hugely successful commercial, blockbuster drugs such as Herceptin (introduced in 1998), Avastin (2004), Erbitux (2009) and many others. Today, this approach is the one that accounts for much of cancer drug research spending.
-The Function of T-cells
Creation of antigen specific T-cells is another and more powerful response of the immune system to a cancer. The attack requires the creation and coordination of subsets of T-cells each with different actions. These are:
- Killer T-cells (aka CD8+ or effector T-cells) have receptors (molecules on the surface) that recognize and can bind to an antigen (or identify a component thereof) expressed by the targeted cancer cell. After binding they release proteins which punch a hole in the cell surface causing it to leak cellular contents and die. Importantly, activated killer T-cells are serial killers that can go on to destroy other abnormal cells. They also rapidly differentiate to create a vast number of clones. Killer T-cells are the hitmen of the T-cell gang.
- Helper T cells (aka CD4+ T-cells) are another sub-set of T cells that are not directly involved in the killing process, but play a major role in activating and directing both killer T cells and effector B-cells through cytokine (proteins which affect other cells) signaling. They are essential to the process that activates antigen specific killer T cells and antigen specific antibodies produced by effector B-cells.
- Memory T cells can be either helper or killer T cells that have become experienced through a previous encounter with an antigen. It is this component of the immune system that makes possible preventive vaccines. At a repeat encounter with an antigen, memory T cells can quickly differentiate to mount a faster and stronger immune response than the first time the immune system was exposed to an antigen.
- Regulatory T cells (aka suppressor T-cells) are still another sub-set. They dampen the responses of immune cells actively involved in killing. They settle the immune response after it has done its work because an overly activated immune system can do great damage to healthy cells in the body.
-Checkpoint Modulators Are the Next Big Thing in Immune-Oncology
Checkpoint modulation drugs are based on an understanding of mechanisms used by nature to regulate the T cell response to cancer. Some T-cell checkpoints (these are cell surface receptors or molecules on T-cells) such as CTLA-4 and PD-1 when activated turn down the T-cell response. However, many cancers have mutated and can produce proteins (ligands) that activate these receptors. This binding has the effect of putting a brake on the T-cell’s response to cancer. There are other checkpoints that when activated rev up the T-cell response. All told, there are tens or perhaps hundreds of these checkpoint modulators and scientists are just in the very early stages of understanding their actions and interactions.
You can think of checkpoints modulation as a thermostat that turns up the heat (T-cell response) when it receives a signal that a stronger response (more heat) is needed and then when the response becomes too intense responds to another signal and turns down the response. Nature has designed this mechanism to prevent the T-cell response from overheating and attacking normal cells which can cause serious, even life threatening, damage to other cells in the body.
Monoclonal antibody drugs which bind to the checkpoints CTLA-4 and PD-1 block cancer ligands from binding to and activating these receptors.) Remember cancer ligand binding to these checkpoints dampens the immune response.) You can think of the anti-CTLA-4 and anti-PD-1 antibodies as taking the brake off the T-cell immune response. There are other checkpoint receptors that target other receptors on the T-cell and have the exact opposite effect of enhancing the T-cell response. The result is increased T-cell activity against cancer in both cases. There are many (tens or even hundreds) checkpoint modulators that potentially may be therapeutic targets.
The first checkpoint modulator to be approved by regulatory agencies was Yervoy (an anti-CTLA-4 antibody) which was introduced in 2011 for the treatment of metastatic melanoma. Bristol-Myers Squibb’s Opdivo and Merck’s Keytruda are antibodies against the checkpoint PD-1; both drugs were approved in 3Q, 2014 for metastatic melanoma. Opdivo and Keytruda have better product characteristics and have gained or will soon gain approval in many solid cancers. All three drugs have quickly become commercial blockbusters. Yervoy had sales of $1.1 billion in 2015 down from $1.3 billion in 2014; this was due to competition from cannibalization from Opdivo and Keytruda. Following their introduction in 3Q, 2014, Opdivo reached sales of $942 million in 2015 and Keytruda reached sales of $586 million. Some Wall Street analysts are projecting that the checkpoint modulator class of drugs (there are numerous other new drugs in development) can achieve as much as $20 billion of sales by 2020.
- Are Engineered Autologous T-Cell Therapy (eACTs) the next Major Immune-Oncology Advance?
For the last two decades, there has been extensive work going on at the NCI and academic centers on engineering T-cells to combine their killing power with the specificity of antibodies. This has the promise to be a transformative therapy and many investors believe that engineered, autologous T-cells will be the next great advance in immune-oncology. Autologous means that the T-cells that are taken from the patient’s body, engineered ex vivo (outside the body) to enhance their cancer killing properties and then returned to the patient. Monoclonal antibodies and checkpoint modulators are drugs that are produced in large biologic manufacturing facilities. The manufacturing process for engineered T-cells requires working throughout with living cells so that the manufacturing process that engineers the living T-cell is the product.
There are currently two types of eACTs that are in development: chimeric antigen receptors or CAR-T and T-cell Receptors or TCR. CAR-T cells have produced some amazing early results in B-cell malignancies (cancers of white blood cells that produce antibodies) such as non-Hodgkin’s lymphoma (NHL), acute lymphocytic leukemia (ALL), chronic lymphocytic leukemia (CLL) and others. Research has on eACTS has been taking place for over 20 years involving very small numbers of patients in experiments conducted at the National Cancer Institute and academic centers. In some cases, there were dramatic responses that appear to be close to cures for hematological cancers in which patients had failed all other treatment options and were close to death.
CAR-T products are farther along in clinical development and are in phase 2 clinical trials that could be the basis for regulatory approval in 2017 so let’s consider the CAR-T technology first. Chimeric antigen receptor engineering involves a tour de force of biotechnology technologies. The process starts with understanding the mechanism of action of several proteins involved in activating T-cells and then developing the nucleic acid (DNA) sequence of the genes that produce these proteins. These gene sequences are all then combined into a construct through recombinant DNA. These constructs can then be inserted into the DNA of T-cells using a viral vector to express the desired proteins.
The DNA insertion step in the process uses techniques pioneered in gene therapy. A viral vector containing the CAR construct is used as a delivery vehicle to transduce (integrate) the construct into the DNA of the T-cell. Then, the mechanisms of gene expression and the resultant protein production natural to the T-cell are used to encode a single chain construct as a new or chimeric function of that T-cell. When this T-cell divides, it then produces identical daughter T-cells.
At one end of the CAR is a target binding domain of an antibody that is specific to the target antigen on the cancer cell surface. This domain extends out of the engineered T cell into the extracellular space, where it can recognize target antigens. The target binding domain consists of a single-chain variable fragment from an antibody that recognizes a particular antigen. In the middle of the CAR that extends through the membrane and into the T-cell cytoplasm are proteins that anchor this binding domain to the cell membrane and also facilitate binding to the antigen target.
There are then at the other end of the chain activating proteins of the CAR that activate (cause to spring into action) the T-cell resulting in direct killing of the cancer cell and proliferation of clones (duplicates) of that T-cell. In addition, T cell activation stimulates the local secretion of cytokines and other molecules that can recruit and activate other anti-tumor immune cells.
These engineered CAR T cells are engineered in the laboratory and expanded until they number in the billions. The expanded population of CAR T cells is then infused into the patient. After the infusion, if all goes as planned, the T cells continue to rapidly multiply in the patient’s body and, with guidance from their engineered receptor, recognize and kill cancer cells that displays the antigen on their surfaces.
As with CARs TCRs use a viral vector encoding genes that are transduced into the T-cell’s DNA. A significant difference is that TCR genes encode two proteins that are designed to bind with specific peptides presented by the major histocompatibility complex (MHC) on the surface of certain cancer cells or antigen presenting cells. The function of MHC molecules is to bind peptide fragments derived from cancer cells and display them on the cell surface for recognition by the appropriate T cells. The TCR protein chains are expressed on the T cell surface where they associate with CD3 proteins, which are natural components of the T cell.
Upon binding of the TCR to the peptide-MHC complex on the cancer cell surface or an antigen presenting cells such as the dendritic cell or macrophage, the CD3 proteins deliver signals that trigger T cell activation, resulting in proliferation of the TCR cells, direct killing of the cancer cell and stimulation of cytokines and other molecules that can recruit and activate additional anti-tumor immune cells.
-CAR-T and TCR Differences
There are three main differences between CAR-Ts and TCRs. The TCRs recognize peptides only in the context of MHC molecules expressed on the surface of the target cell or an antigen presenting cell; it is MHC restricted. MHC molecules are also known as human leukocyte antigens, HLA, proteins or more generally as blood types. The several types of HLA molecules in the human population that are genetically different. This means that the construct of a TCR candidate has to be specifically matched to each HLA type. Put another way, a different TCR would have to be developed for each HLA type. In contrast, CAR target recognition is MHC-unrestricted and effective for any HLA type.
CAR-T products can only be used against cancer antigens that are part of an intact protein on the cancer cell surface. Cancers with this characteristic are about 20% to 30% of all cancers. TCRs have the potential to recognize cancer antigens not only presented directly on the surface of cancer cells but also presented in the interior of cancer cells and by antigen-presenting cells such as dendritic cells and macrophages in the tumor microenvironment and in lymph organs. So TCRs can be effective against all cancers. However for a given cancer, I emphasize that a different TCR must be developed for each HLA or blood type. The CAR-T functions independently of HLA or blood type.
-Medical and Investor Excitement with CAR-T Cells Heightened in 2014
The medical excitement with CAR-T cells increased sharply in 2014 with the publishing of important studies in the major medical journals Lancet and New England Journal of Medicine. Then initial public offerings of two companies focused on CAR-T development in 2014, Kite Pharmaceuticals and Juno Therapeutics, created enormous investor interest. Even though neither company had yet completed a phase 1 clinical trial, they were awarded billion plus market valuations.
CAR-T products had been in research for over 20 years at the NCI and based on the foundation laid by NCI, several academic centers did small investigator sponsored studies. Prominent among these were Children’s Hospital of Pennsylvania, Fred Hutchinson Cancer research Center, Memorial Sloan Kettering and Seattle Children’s Research Institute. Studies from all of these centers consistently had shown activity in individual patients with various acute and chronic B-cell leukemias and lymphomas. However, prior to 2014 there were no published papers for phase 1 type studies that used a consistent regimen, treated patients consecutively and were evaluated on an intent to treat basis.
An article in an October 14, 2014 in Lancet reported on an NCI conducted phase 1 study dose escalation study. This trial enrolled 20 children and young adults (aged 1 to 30) with relapsed or refractory acute lymphoblastic leukemia or non-Hodgkin’s lymphoma; there was no control group. The purpose of the study was to gain insights into dosing, side effects and therapeutic activity. Prior to being given a single infusion of CAR-T cells at one of two dose levels, these 20 patients received a conditioning regimen of chemotherapy (cyclophosphamide and fludaribine). After the dose escalation phase, an expansion cohort was treated at the determined maximum tolerated dose.
The results were extremely encouraging as 14 patients (70%) achieved a complete response rate, 12 patients (60%) an MRD-negative complete response. Ten patients who had an MRD-negative complete response subsequently then underwent hematopoietic stem-cell transplantation (HSCT), and all 10 remained disease at a median follow-up of 10 months.
It is important to understand how severely ill these patients were. More than 80 percent of children who are diagnosed with ALL that arises in B cells, the predominant type of pediatric ALL, can be cured by intensive chemotherapy. For the 20% of patients whose cancers return after intensive chemotherapy or a stem cell transplant (termed relapsed or refractory), the remaining treatment options are close to none. Looked at from this perspective, the results show that CAR-T therapy in this relapsed/ refractory population was lifesaving and the responses were amazing.
As had been seen in earlier investigator studies, infusion of CAR T cells was associated with significant, acute toxicities, including fever, hypokalemia, and transient neurological deficits. Note for perspective that grade 5 toxicities are defined as death related to an adverse event, grade 4 is life-threatening or disabling and grade 3 is a severe and undesirable adverse event. There were three (14%) grade 4 toxicities related to cytokine release syndrome. The most common non-hematological grade 3 adverse events were fever 9 patients (43%) hypokalemia 9 patients (43%) and neutropenia 8 (38%) of 21 patients). These are serious toxicities, but all were fully reversible. While CAT-T cells are very effective, they have a very concerning side effect profile.
Results from a second trial was also published in 2014 in the New England Journal of Medicine. This was a trial at Children’s Hospital of Philadelphia conducted in collaboration with researchers from the University of Pennsylvania. Like the NCI study, this was also in relapsed/ refractory ALL and showed similar spectacular results. A complete response (CR) was seen in 27 (90%) of 30 patients treated. Of these 30 patients, 15 had undergone stem cell transplantation. Nineteen of the 27 patients with complete responses remained in remission, the study authors reported, with 15 of these patients receiving no further therapy and only 4 patients withdrew from the trial to receive other therapy. There was a 6-month event-free survival rate of 67% and an overall survival rate of 78%
Serious side effects occurred as was seen in the NCI study. All patients had cytokine-release syndrome. Severe cytokine-release syndrome, which developed in 27% of the patients, was associated with a higher disease burden before infusion and was effectively treated with the anti–interleukin-6 receptor antibody tocilizumab.
Further substantiating the CAR-T results seen at NCI and Children’s Hospital of Philadelphia, Juno reported in 2014 that its lead product JCAR015 had demonstrated in an ongoing Phase I clinical trial an 89% complete remission rate in 27 evaluable adult patients with relapsed/refractory B cell acute lymphoblastic leukemia. This data comparable to the NCI and CHOP studies in relapsed/ refractory ALL in pediatric and adult patients.
There are three companies that have moved to the front of the pack in CAR-T development. Kite Pharmaceuticals established a CRADA with in 2012 with NCI that has given it access to most of the clinical work done by NCI. Novartis formed an alliance with the University of Pennsylvania in August 2012. Juno has built its technology based on licensing deals with Fred Hutchinson Cancer research Center, Memorial Sloan Kettering and Seattle Children’s Research Institute.
Tagged as CAR-T, chimeric antigen receptor T-cells, engineered autologous T cells, JUNO, Juno Therapeutics, KITE, Kite Pharmaceuticals + Categorized as Company Reports, LinkedIn, Smith On Stocks Blog