Special Report


An Immunotherapeutic Approach to Glioblastoma: ImmunoCellular Therapeutics' CEO Andrew Gengos

Andrew Gengos Los Angeles-based ImmunoCellular Therapeutics Ltd. is on the verge of a pivotal Phase 3 trial with its dendritic cell immunotherapy for brain cancer. It's an extraordinary and exciting time for CEO Andrew Gengos, who is a veteran biotech executive. In this interview with The Life Sciences Report, Gengos discusses his firm's recently reported Phase 2 data in glioblastoma, as well as the company's strategy for the upcoming registration trial.

Management Q&A: View From the Top

The Life Sciences Report: You and your team at ImmunoCellular Therapeutics Ltd. (IMUC:NYSE.MKT) are studying ICT-107 (autologous dendritic cells pulsed with immunogenic peptides from six tumor-associated antigens) as an immunotherapy targeting glioblastoma multiforme (GBM), a very serious and very tough indication. This is your most advanced program, and the study is in newly diagnosed patients who have had surgical resection and chemo-radiation, but are not as worn down immunologically as patients treated over a prolonged period. Do you think you'll have a more durable result in this population?

Andrew Gengos: We hope so. Obviously, we're speculating, but we have two things in mind when we're selecting an indication like newly diagnosed glioblastoma. The first is that, when you're talking about an immunotherapy that utilizes the patient's immune system to attack cancer, it's important the patient be able to mount an immune response. If a patient is immune-depleted by chemotherapy, you have to question whether that's the best setting to test an immunotherapy in. Newly diagnosed glioblastoma patients, despite standard-of-care chemo-radiation treatment, are still typically able to mount an immune response following administration of our immunotherapy. The second key point is that there are few, if any, cell-based therapies approved in the marketplace.

"We are using the immunotherapy to control and potentially eliminate residual disease."

In the case of using a dendritic cell (DC) or some other immune cell to attack a tumor, you need to walk before you can run. Choosing a setting where you have a residual tumor to attack and clear, rather than a big, bulky tumor, makes more sense. In the case of newly diagnosed glioblastoma, we require the tumor to have been fully, or near fully, resected, and that the patients have gone through chemo and radiation therapy. We are using the immunotherapy to control and potentially eliminate residual disease. That's the logic of how you get to glioblastoma.

There are other cancer indications that might share these features. If we are fortunate enough to further develop this technology platform, we would target some of these, but starting with glioblastoma made sense to us.

TLSR: Andrew, I understand ICT-107 can be frozen, and the cells will keep. Could they be a maintenance therapy, which you might give a surviving patient once a year?

AG: That's a hypothesis we want to test, after registration, if we have a successful Phase 3. In the Phase 3 design, as well as in the Phase 2, there are elements of a primary combination therapy as well as a maintenance therapy.

Let me elaborate a little bit. After surgery and recovery, the GBM patient goes through concomitant chemo and radiation therapy. In total, going from diagnosis through the end of chemotherapy and radiation takes 10–15 weeks. That's when we start our treatment. We start out pretty aggressively, with four doses, one per week for the first month, in what we call an induction phase. We hope to mount a significant immune response in that first month. Thereafter, in the Phase 3, we'll give patients a maintenance dose once a month throughout the first year, as long as they don't progress. Progression terminates further treatment with ICT-107 or placebo.

"When we focused on the A2 patients we saw a significant progression-free survival advantage."

After that first year, we go to a more long-term maintenance schedule, perhaps once every six months. Will there be a surveillance approach with ICT-107 in a long-term maintenance usage? We think there might be. We have elements of that built into the Phase 3 trial, as we did in the Phase 2. To get at that question specifically, we'd probably design another clinical trial.

TLSR: Back on Nov. 20, you presented data from the Phase 2 trial at the Society for Neuro-Oncology annual meeting in San Antonio. This trial was a 124-patient, randomized, double-blind, multicenter, multinational study. You did not meet your p-value for statistical significance of your primary endpoint in the overall population, but you seem to have found a subset of patients who overexpressed the HLA-A2 antigen. My understanding is that these patients mounted a definitive immune response with administration of the ICT-107. Were you able to correlate that immune response to the primary endpoint of overall survival in this HLA-A2-positive (HLA-A2+) population of patients?

AG: The simple answer is yes. The Phase 2 trial enrolled both HLA-A1 and HLA-A2+ patients; ICT-107 has six antigens engineered into it, two of which are HLA-A1-specific and four of which are HLA-A2-specific. In a sense, it was designed to treat both populations. In the Phase 2 data for the 124 intent-to-treat population, we saw a numeric advantage in overall survival, but it was not statistically significant. But we did see a statistically significant advantage in progression-free survival, which was a secondary endpoint.

When we looked more closely at the data, we found the HLA-A2+ population seemed to respond better to ICT-107. We have a hypothesis as to why HLA-A1 group didn't respond as well, but we have no definitive answer to that question. When we focused on the A2 patients we saw a significant progression-free survival advantage. We also saw what could be a clinically meaningful overall survival advantage, although it wasn't statistically significant. We think the A2 population is the right population for the Phase 3 trial.

TLSR: Please address the immune response. Was it correlated to the endpoints?

AG: We found, in general, that in those patients who mounted what we deemed to be an immune response, it was associated with overall survival. I choose that word "associated" carefully, rather than "correlated." "Correlate" implies a statistical, linear relationship, and "associate" is more of a non-linear relationship. We can definitively say that immune-responding patients were associated with a better overall survival in the trial when the immune response was measured by enzyme-linked immunospot (ELISPOT) on peripheral blood.

"We learned a lot from both analyzing the results directly and also looking at the immunology."

We have some concerns about the ELISPOT analysis in general because, essentially, we're looking at the presence of antigen-specific T cells in the circulating blood. You would hope that any antigen-specific immunotherapy would create a population of antigen-specific killer T cells, but we don't particularly need them to be in the blood. We need them to be at the site of the tumor. That would be a better test, but we don't have the technology to do that today, so we use the proxy of circulating T cells in our ELISPOT analysis.

TLSR: Andrew, ImmunoCellular shares did not react favorably to the Phase 2 results. But the purpose of a Phase 2 is to find the right population for the pivotal or registration trial, and you have designed your upcoming Phase 3 to include only the HLA-A2+ population. What else did you learn that informs the design of your Phase 3?

AG: We learned a lot. We learned that an important and large subpopulation of patients is the right one to test. The HLA-A2+ population was about 60% of all the patients we treated in that trial. HLA-A2+ patients are probably about 50% of the U.S. and European GBM population, so that's a significant population to take into Phase 3.

In conjunction with some of the immune response information, we triangulated and realized we might be better off with more dosing. We have made changes in frequency of dosing, and we hope to see a longer overall survival for those patients.

"By getting the FDA to review the protocol and statistical plan in advance, we hope our probability of being turned down for a poor trial design is zero."

In addition, we know that a group of patients, based on the status of the MGMT gene, may not get any benefit from temozolomide, a chemotherapy. The drug has been shown to have an effect in about one-third of patients, whose MGMT gene is methylated. Two-thirds of patients, unfortunately, have an unmethylated gene status. The methylated MGMT gene is the silenced gene, and is related to DNA repair. Normally, DNA repair is a good thing, but in a tumor, DNA repair is bad because most chemotherapies and radiation work by damaging the DNA. If the tumor cell can repair its DNA, the patient has a poorer prognosis. The unmethylated gene is bad as a prognostic variable for chemotherapy in this population, and that group you would expect to live about a year. The methylated silenced gene is good as a prognostic variable for chemotherapy, and that patient population is expected to live roughly two years. It's a very important and very powerful distinguisher of survival based on chemotherapy.

Of course, ICT-107 is an immunotherapy, not a chemotherapy, but the status of the MGMT gene is something we want to know. In our Phase 2, when we looked at ELISPOT, we saw no difference in the ability of the patient receiving ICT-107 to mount an immune response based on their MGMT status. That's a good thing, because a DNA repair gene should have nothing to do with the patient's ability to mount an immune response. So we are putting both the MGMT methylated and unmethylated patients into Phase 3. The outcome results from the Phase 2 trial support that, and the immunological results, which don't distinguish between those two groups, support the fact that either one of them should be able to mount an immune response.

In the Phase 2, and prospectively in the Phase 3, when patients progress in their disease, they come off treatment, because we are targeting newly diagnosed glioblastoma, not recurrent glioblastoma. In our Phase 3, we're going to determine progression in a much more robust way, using a centralized approach to analysis of the MRI scan. Also, we're going to use brand-new criteria for determining progression called immunotherapy response assessment for neuro-oncology (iRANO). It's the most up-to-date way to determine cancer progression in the brain when treating patients in an immunotherapy trial.

"We think if we have a successful Phase 3, we could expect registration."

Finally, we had a curious finding in the immune response data. We found that some portion of patients in our control arm mounted an immune response, compared to the baseline T-cell population in their bloodstream. Our placebo—our control—was an activated dendritic cell without antigens. ICT-107 is an activated dendritic cell with the six antigens. There is some indication that our control might have been active in some way in the immune system, and perhaps had some modest effect on the outcomes in our control-arm patients. In Phase 3, we are going to use less immunologically active monocytes. We're going to change the nature of the placebo group and hopefully separate the signal from the noise by doing so.

TLSR: And thereby, hopefully, get a more definitively significant p-value at the end of the study?

AG: Exactly. While these changes are all forward-looking and speculative, we're very excited by the opportunity to implement them because that's what you learn by doing a Phase 2. Companies that skip Phase 2 don't have this opportunity. We learned a lot from both analyzing the results directly and also looking at the immunology, so we're quite excited. We have no idea what the potential magnitude of these changes will be on survival, but there's a strong basis and scientific rationale for making the improvements.

TLSR: Your upcoming trial will have about 400 patients. Back in mid-August, you and the FDA agreed to a trial design for a special protocol assessment (SPA) for this Phase 3. Does that SPA mean if you meet the endpoints the FDA will approve ICT-107? And does it mean you won't have to do a confirmatory Phase 3 trial?

AG: Many times, when the FDA grants an SPA, people hear what you just said, which is if we run the Phase 3 and the primary endpoint is met, we will get registration. That's not an unreasonable interpretation, but the FDA never commits to registering anything until it has seen all the data from the trial. What an SPA means in my mind—and this is just my personal interpretation—is that if we do have a successful trial, the FDA is going to look at the efficacy data and not question the design or the statistical plan of the trial.

"We can freeze our doses, inventory them and store them essentially forever."

There was a very interesting article released over the summer by some FDA staffers, including the very influential Dr. Richard Pazdur, who is director of the Office of Hematology and Oncology Products (OHOP) at FDA. In this paper, the FDA looked at 15 oncology drugs that were submitted for registration but were turned down. Five of those 15 were turned down because of flaws in trial design. That's an astonishing statistic in my mind, because it says a third of these applications were turned down for something avoidable. By getting the FDA to review the protocol and statistical plan in advance, we hope our probability of being turned down for a poor trial design is zero.

TLSR: What about having to run a confirmatory trial? Does the SPA protect you from that expense and time if you meet the agreed-upon endpoints?

AG: When we went to the FDA in September 2014 to talk about the Phase 2 results and the Phase 3 design, the agency said in writing that the Phase 2 in combination with a successful Phase 3 would be sufficient basis for registration. That has also been repeated in the SPA communications. We think if we have a successful Phase 3, we could expect registration. The FDA will always reserve the right to look at the data for efficacy and safety.

TLSR: Will the FDA allow you to include only the Phase 2 HLA-A2+ patient data in with your Phase 3? Or must you include all the patients from the Phase 2?

AG: I think from a safety standpoint, the FDA will include all 124 patients from the Phase 2, because even an HLA-A1 patient is a reasonable subject as it relates to safety. From an efficacy standpoint, I think the FDA has indicated that the Phase 2 results in combination with the Phase 3 would be the basis for registration.

TLSR: We normally think of a question about resistance being applied to drugs, but is there any evidence that GBM tumors have become resistant to the antigens being presented by the dendritic cells?

AG: We don't have definitive evidence of that, but there is a theoretical argument for it. If you choose a single antigen to target, there is a basis of knowledge that says the tumor cells, being quite smart, can down-regulate the presentation of that antigen. The hope is that by using multiple antigens, the tumor cell would be incapable of down-regulating them all.

TLSR: When is the first Phase 3 patient going to be dosed?

AG: Patients are being screened at a few open sites, only in the U.S. right now. From diagnosis we go to randomization, which is when we treat the patient with the first dose of immunotherapy. That can be 12–13 weeks from diagnosis. We will screen a patient, make the ICT-107 for that patient, and then wait until after chemotherapy and radiation. We hope to treat the first patient with ICT-107, or placebo, in the next several weeks.

TLSR: According to clinicaltrials.gov (NCT02546102), you have final data collection for your Phase 3 scheduled for December 2019. If all goes well, when could you expect approval of ICT-107?

AG: I'd rather not speculate, because I'd have to lump a number of assumptions on top of one another. We assume that we will fully enroll the 400-plus patients in roughly two years. Obviously, that enrollment curve is going to determine when we get sufficient events—number of patient deaths—to stop the trial, analyze the data and write up the biologics license application (BLA). After a two-year enrollment period, we think that is probably two to three years beyond the last patient in. Then, usually the fastest you can analyze data of that magnitude is three to six months, and then usually it's another at least six months to file a BLA. But there is just a big variance based on a lot of assumptions.

TLSR: On Dec. 1, you announced an agreement with the European Organisation of Research and Treatment of Cancer (EORTC), which is the largest European cancer cooperative group. That opens up a lot of access and should assist you in recruiting and enrolling patients, shouldn't it?

AG: The answer is yes. I would also mention that on Dec. 7 we announced an agreement with Alliance Foundation Trials (AFT), through which we will have access to a network of 10,000 cancer specialists, clinics and centers across the U.S. These collaborations will give us access and help ensure we get our Phase 3 enrolled.

TLSR: Andrew, I know many people are thinking about the now-bankrupt Dendreon Corp. and its therapeutic vaccine, Provenge (sipuleucel-T), for prostate cancer. My impression is that your share price has been depressed by the comparison of your autologous dendritic cell technology platform to Dendreon's. There must be a textbook of priceless lessons that came out of the Dendreon/Provenge story. Could you tell me what you learned from that? Investors are especially interested in the cost-of-goods-sold (COGS) line on the income statement, which was very unfavorable for Dendreon.

AG: At a very high level, Dendreon did us all a favor. It proved definitively that a dendritic cell immunotherapy can be an efficacious treatment in the cancer setting. As an aside, Dendreon didn't even call its product a dendritic cell; it was an antigen-presenting cell because it wasn't clear whether the cells were fully mature dendritic cells.

"A long-lived surveillance by antigen-specific T cells in a residual tumor setting sounds like a good idea to me."

Dendreon also proved the FDA would approve an autologous cell-based therapy. Then it proved it could make the therapy and deliver it. It was made, however, in a complex way, because Dendreon had to get a fresh bag of cells from the patient to the factory, get a single dose manufactured, and then get the cells back to the patient in a very short amount of time. What it did not prove—and this is what I think was Dendreon's Achilles' heel—was that it could make a sufficient return for investors on that program vis-à-vis what else was available in the marketplace at the time.

TLSR: How is ImmunoCellular differentiated from the Dendreon model?

AG: We are using a technology that's 10 or 15 years more advanced. We've changed the unit of manufacture from the dose to the patient, and this has had a dramatic effect on COGS. We have also changed the delivery and storage characteristics of the product. These are important distinctions.

The regimen for Provenge was three doses, and the company had to do apheresis (separation of cellular components from the blood) on the patient three sequential times to get three bags of white blood cells to do three manufacturing runs so it could deliver three doses over the dosing regimen period. If it had our manufacturing process, it would only have to do that once. In our Phase 2, we were making 20-plus doses from each run for each patient, on average. The big difference is we can freeze our doses, inventory them and store them essentially forever. With our technology, Dendreon's COGS would have been a third of what they were, and the company would not have had to do the fire drill of getting the doses from the factory, to the patient, in a fresh, unfrozen form. These are two critical improvements.

TLSR: You gave investors news in mid-November about a new technology developed in the lab of Nobel laureate David Baltimore at the California Institute of Technology (Caltech). Your press release said this "Stem-to-T-cell" platform will include a collaboration between ImmunoCellular and Cassian Yee at the University of Texas MD Anderson Cancer Center. This has generated some excitement within the investment community. Would you tell me a bit about this program?

AG: We're excited about this collaboration because it's yet another way to create an antigen-specific killer T-cell population in a patient to target the treatment of disease. This new stem-to-T-cell technology does just that, but in a different way from our DC immunotherapies and other approaches being investigated.

"We are excited about stem-to-T-cell technology. It could be a real solution."

Stem-to-T-cell is an autologous treatment built from the patient's own cells, as is our dendritic cell technology. The starting cell is a hematopoietic stem cell (HSC), which is the stem cell that creates essentially all the cellular components in the blood, including immune system cells. We take HSCs from the patient and use gene therapy outside the patient's body to introduce, via a lentivirus, the DNA sequence that encodes a particular T-cell receptor (TCR). When we put that hematopoietic stem cell back into the patient, it does what all stem cells do—it divides into itself and into a daughter cell. Dr. Baltimore and his colleagues at Caltech discovered that the daughter cell is pre-programmed to be an antigen-specific killer T-cell that's exhibiting the particular TCR we engineer into it. That antigen-specific killer T-cell presumably will have some effect on a patient's tumor.

The fact that the patient will generate these cells, which continue to divide with the same T-cell receptors potentially for a lifetime, means we might have some opportunity to provide a long-term surveillance and treatment by the immune system of any residual cancer cells, or any potential regeneration of a cancer.

TLSR: Where is this program now in the timeline of development?

AG: It's a preclinical program; it's mostly on paper right now. The big advantage is that it would be a long-lived therapy, and it is a potential one-time treatment for the cancer. Dr. Yee is an expert in identifying and isolating cytotoxic T cells, and the first piece that he is assisting us with at MD Anderson is the discovery work to find the right T-cell receptor, or at least what we think might be an optimal T-cell receptor, to use in preclinical testing—and then hopefully clinical testing.

TLSR: What antigen is the TCR targeting?

AG: We have chosen the antigen that the T-cell receptor will correspond to, but we have not disclosed the nature of that antigen. That's proprietary.

But we are making good progress with Dr. Yee to identify potential T-cell receptors and their DNA sequences. We're very excited about moving on to the next steps, which will be to encode that sequence into an HSC, see if we can generate the daughter cells, see if those daughter cells are potent against human cancer cell lines and potentially animal models, and then hopefully get to human testing and an investigational new drug (IND) application filing in as short a course as we can.

TLSR: This stem-to-T-cell technology could obviously have tremendous potential in hematologic cancers—leukemias and lymphomas. But could it have potential in solid tumors?

AG: Yes. Again, I don't know whether the immunotherapy technology in a general sense is potent enough to take on a bulky tumor, but a long-lived surveillance by antigen-specific T cells in a residual tumor setting sounds like a good idea to me. Of course, that residual tumor setting could be a solid micro-metastatic tumor.

TLSR: Could this technology displace chimeric antigen receptor (CAR) T-cell and T-cell receptor projects being developed by other companies?

AG: I think "displace" is a pretty aggressive word: No doubt each technology has pros and cons. CAR T is delivered as a big bolus all at once, and it can have on-target, off-tumor toxicity. There are going to be safety questions about those therapies. To date, the most effective CAR T seems to be targeting CD19 in B cells. My understanding is that the good news there is that every B cell has CD19, both healthy and cancerous, so if you knock out the B cells, you're going to treat the cancer, and humans can live without B cells for a while. That's a perfect target for CAR T.

But in some other applications, where CAR Ts have targeted things other than CD19, the on-target, off-tumor toxicity, particularly in the heart or the brain, has caused death or severe adverse effects. I think CAR T will have the challenge of toxicity to deal with, as well as the challenge of not having longevity. Those T cells will peter out pretty quickly, and would have to be readministered. That's why we are excited about stem-to-T-cell technology. It could be a real solution.

TLSR: Thank you, Andrew.

Andrew Gengos joined ImmunoCellular Therapeutics as president and CEO in December 2012. Spanning more than 20 years in the life science industry, his experience includes executive leadership positions in both large and emerging companies. He has broad expertise in corporate strategy, business development and transactions, including mergers and acquisitions, financing, operations, commercial planning and healthcare policy. Gengos was most recently the president and CEO of Neuraltus Pharmaceuticals, where he led implementation of the company's clinical, regulatory, fundraising and business development strategies while operating the company on a virtual business model. Previously, he served for more than seven years with Amgen where, as vice president, strategy and corporate development, he managed the company's worldwide in-and-outbound business development activities, including acquisitions, licensing, spinouts, divestitures, corporate venture capital investments and alliance management. Before joining Amgen, Gengos was vice president, CFO and chief business officer of Dynavax Technologies, where he led the company's business functions, including finance and accounting, fundraising, budgeting, planning and business development. Earlier in his career, Gengos served as vice president, strategy at Chiron Corp. and as senior engagement manager at McKinsey & Co. Gengos holds a master's degree in business administration from the UCLA Anderson School of Management, and a bachelor's degree in chemical engineering from the Massachusetts Institute of Technology.

1) Dr. George S. Mack conducted this interview for Streetwise Reports LLC, publisher of The Gold Report, The Energy Report and The Life Sciences Report, and provides services to Streetwise Reports as an independent contractor. He owns, or his family owns, shares of the company mentioned in this interview: None.
2) ImmunoCellular Therapeutics Ltd. paid Streetwise Reports to conduct, produce and distribute this interview.
3) Andrew Gengos had final approval of the content and is wholly responsible for the validity of the statements. Opinions expressed are the opinions of Mr. Gengos and not of Streetwise Reports or its officers.
4) The interview does not constitute investment advice. Each reader is encouraged to consult with his or her individual financial professional and any action a reader takes as a result of information presented here is his or her own responsibility. By opening this page, each reader accepts and agrees to Streetwise Reports' terms of use and full legal disclaimer.
5) From time to time, Streetwise Reports LLC and its directors, officers, employees or members of their families, as well as persons interviewed for articles and interviews on the site, may have a long or short position in securities mentioned. Directors, officers, employees or members of their families are prohibited from making purchases and/or sales of those securities in the open market or otherwise during the up-to-four-week interval from the time of the interview until after it publishes.