The state of play in ImmunotherapueticsFeb 25, 2014 Tags:
- Drug development
Immuno-oncology is suddenly the most talked about approach for changing the treatment paradigm of cancer, though a few short years ago almost no large biopharma wanted to touch these programs. BIO-Europe Spring® in Turin will feature a headline panel that will assess this new and powerful approach, getting into the important differences between platforms in terms of scientific risk, clinical development strategies and commercial challenges.
As a prelude to what is sure to be an enlightening session, we asked Jeffrey M. Bockman of Defined Health and Axel Hoos of GlaxoSmithKline to discuss some of the key issues of this exciting scientific arena.
Jeffrey M. Bockman (JB): Why right now do you think immunotherapies are finally exciting people? What has changed?
Axel Hoos (AH): My interpretation is that the turn of the field comes from progress in two major areas. The first is the evolution of the science and our improved understanding of mechanisms of the immune system. We have been working on this for decades and learned about some of the key mechanisms we could modulate and translate into drugs. CTLA4 is just one of them. It was one of the first that we understood and could therapeutically target. In one statement, scientific progress in immunology is one key reason why we have success now.
A second reason, which cannot be separated from the first, are the methodological improvements that we have been working on in the field over the last decade. This goes in the direction of recognizing the differences between immunotherapy and chemotherapy. If you think about drug development in oncology, over the last 40 years we have applied a chemotherapy paradigm. This means we also operated under a chemotherapy paradigm for immunotherapy development and we were missing some of the now obvious immunotherapeutic effects. Recent years have brought methodological progress which adapted the paradigm for cancer immunotherapy development. This encompasses, for example, new clinical endpoints or reproducible assay conduct for cellular immune responses to minimize data variability across clinical centers and trials.
JB: I want to circle back to this issue of the endpoints in particular the tail effect. Can you talk a little about that in terms of immunotherapy?
AH: The primary difference between a chemotherapy approach and an immunotherapy approach is that chemotherapy directly targets the tumor, while immunotherapy targets the immune system, and – in a second step – the immune system targets the tumor. This indirect mechanism introduces a time factor that can be quite important in clinical investigation. So when we talk about the delayed effect of therapy, it refers to the activation time of the immune system, which includes generation of adequate numbers of tumor-specific T cells, trafficking and infiltration of those T cells to and into the tumor, and translation of these biologic events into clinical efficacy.
This may lead to two new phenomena. The first one is a delayed effect on the tumor. That means the tumor could get bigger before it shrinks, in part because of lymphocytic infiltration. The second effect is the delayed separation of Kaplan-Meier survival curves, which is downstream of the effect on the tumor. The activation time of the immune system may lead to a period in which one cannot detect the survival difference between the control group and the immunotherapy group. It could be several months before the survival curves part, a period that has to be addressed in the statistical design and evaluation of a clinical trial so the study will not be misinterpreted as a failure.
JB: Can you comment on where the field will go in terms of the currently operating paradigm of stratification of people for various targeted agents? What will go on with immunotherapy in terms of the ability to try and predict who might respond better?
AH: We are not yet in a place where we have a workable biomarker for clinical practice to select benefitting patient groups. However, one can anticipate that the extensive biomarker work that’s being done now will yield predictors of benefit, and will also enable us to select patient populations for clinical trials. We don’t have a clear answer yet what those markers might be. From the second generation of checkpoint modulators beyond CTLA4, we already have seen PD-1 ligand expressed on tumor cells emerge as a promising marker to identify patients with a higher probability of a response. The data on PDL-1 is not yet conclusive and there is more work to be done. And this is really the tip of the iceberg. There are other new parameters that we are looking at that will enable us to more intelligently select patients.
JB: I had also seen publications about various types of signatures that look more akin to something like an Oncotype DX test.
AH: You probably know from the MAGE A3 program GSK is running that there is a gene signature that promises to be valuable in selecting patients. Other parameters such as T cell receptor diversity may predict the ability of the immune system to process a wide spectrum of antigens and may be prognostic. While there is no definitive answer yet, both kinds of parameters – either a gene signature that identifies the patient’s immune status and potentially the responsiveness to therapy, or individual markers of resistance or responsiveness to therapy like PDL1 may prove valuable. I expect that the enormous effort that is now put into cancer immunotherapy – and only really since 2011 – will lead us to relevant biomarkers in the foreseeable future.
JB: I referred earlier to how we were referring to immunotherapy as if it were one thing, and clearly it encompasses a range of approaches from the checkpoint inhibitors to more conventional vaccine based approaches. Clearly, the data coming out right now for the checkpoints are quite robust. Do you see any nearer term, similarly robust data being generated from some of these other types of approaches?
AH: The spectrum is quite wide, you are absolutely right. Immunotherapy today contains several modalities that are mechanistically quite different. Two of them have shown that they are efficacious enough to work as monotherapies and improve patient survival in randomized trials. They are the checkpoint modulators that we spoke about. And there are the cellular therapies, particularly, the dendritic cell therapy Dendreon has developed.
In these two categories there is substantial movement: In these two categories there is substantial movement: a) a large wave of next-generation checkpoint modulators is under investigation following the CTLA-4 proof-of-concept, and b) in the cellular therapy space we see increasing momentum with CAR or TCR technologies. These represent genetically engineered autologous T cells that either carry an affinity matured T cell receptor or a chimeric antigen receptor to respond to intracellular or extracellular targets. Both technologies have shown some transformational data in terms of individual patient benefit in small scale clinical trials.
Of course there are other areas of great clinical interest such as oncolytic viral therapies, cancer vaccines, cytokines or modulators of innate immunity.
I think the next generation of immunotherapy that will be successful beyond checkpoint modulators looks to be cellular therapy. These and other modalities lend themselves for combination therapy, and will be part of the future of immunotherapy.
JB: The idea of combinations has been the cornerstone of oncology. There is no reason to think that will not be true with immunotherapy. This includes the idea of combining immunotherapies with chemotherapies, with radiation, with small molecule targeted agents, as well as with other types of immunotherapies. What are your thoughts about where the field is going and what our current state of the knowledge is around those different types of combinations?
AH: There are 3 main categories of combinations. Let’s start with the combination between checkpoint modulators. We have already seen that such checkpoint modulator combinations in preclinical models can show a variety of synergies. There is only one example currently with human clinical data, which is the CTLA4 plus PD1 combination that was shown at last year’s ASCO. It suggests in a Phase I trial an improvement over each of the two monotherapies. However, we have to be cautious because this data is still early and does not offer a randomized comparison. Randomized data from an ongoing study is expected to answer key questions for this combination. Beyond that, I would expect that other checkpoint modulator combinations can be equally attractive. There is likely a group of patients in any histology that will not respond to either monotherapy or a specific combination of checkpoint modulators. For those patients, one can expect that alternative combinations may lead to the desired benefit. Selection of such patients should be based observed resistance paired with immune biomarkers.
The second piece is to combine across modalities, but still within immunotherapy. I can see an oncolytic virus to be combined with a checkpoint modulator. I could even see a cellular therapy combined with a checkpoint modulator, although there are more logistical and cost constraints that may pose limitations. A central possibility remains the combination of cancer vaccines with checkpoint modulators, a concept the community has long been interested in but that has not been systematically investigated. So these are the obvious categories.
Then there is the third piece. We should not forget that conventional cancer therapies have not been adequately investigated for their secondary effects on the immune system. I anticipate that, as we learn more about what targeted and chemotherapies do to the immune system, we will also be able to combine them intelligently with immune modulators. And that will be another large area of combination, to pair targeted therapies with immunotherapy. It would also address the integration of immunotherapy into these existing standards of care.
JB: I want to circle back to these issues of the hot areas, the checkpoint inhibitors and move a little bit from the science and the clinic to the commercial questions. It’s a space that is so hot and we see a huge number of deals being done across the spectrum from the Bayer early deal with Compugen, and a lot of collaborative deals like the recent Merck deal with Incyte and Amgen. Now there has just been the announcement of Pierre Fabre getting into the PD1 space by doing the deal with Aurigene. So there are a lot of companies leaping into this space. It doesn’t seem likely that there would be the need and the differentiation among six different PD1 agents. I am just curious to the degree to which you can comment, how you see this evolving?
AH: It’s like any area that becomes successful. Many groups or companies that have previously not been active in that space try to enter. That generates a lot of volume. It creates somewhat of a bubble. Still, there’s huge value in this space and at this point it’s no longer realistic to think that oncology will go forward without immunotherapy. So any serious oncology drug development organization will have to put their thoughts around how to play in immunotherapy. This is the reason why we see so much activity.
JB: One thing that is missing from that deal volume are deals with mid to large size entities or large pharma around these vaccine approaches. There are a reasonable number of Phase III vaccines that are chugging along. I guess you could argue that people are in a wait and see mode for those. What is your perspective on large pharmas’ interest in the classic antigen based approaches? Why are they continuing to lag behind, despite the fact as you said before we have Provenge approved? Why the continued hesitancy? Is it just because they haven’t been able to show anything even in randomized data to date?
AH: The scientific community has spent much time trying to find out why the plain antigen delivery approach of vaccines has not been successful yet. One of the potential explanations may be inherent immune suppressive mechanisms. There are two opposing forces: the immune response and immune suppression. The delivery of an antigen-specific vaccine can stimulate an immune response. However, it does not manage the immune suppression in the system to ensure the response is enhanced to a clinically relevant level. The CTLA4 or PD1 approach, which deals with immunosuppression unspecifically has been shown to work. The hypothesis under which many people in the scientific community currently operate – a vaccine that delivers specific target antigens may be enhanced by pairing it with a second modality that reduces immune suppression. These kinds of studies would need to be done systematically to understand what is possible. I think that is something the community needs to do.
JB: I wonder if one of the factors with the gold rush towards checkpoint modulators and the like is the fact that these are antibodies, and that is a sweet spot for pharma. It’s something that they know, it’s easy to use, they know how to study it. Do you think that is part of the enthusiasm? It is such a more straightforward approach certainly than cell therapies.
AH: It is certainly easier for any pharmaceutical organization to develop a modality that they are familiar with than to engage in something completely new. The manufacturing capability, the CMC capability, the supply chain are in place. For antibodies, that is the case. For protein based therapeutics or peptide based therapeutics, that is the case. For cellular therapies, which are the new kid on the block, that is not yet the case. I see a lot of activity in the bio-pharmaceutical industry in terms of trying to change that. There have been some large pharmas and some new biotech organizations that have engaged with cellular therapy, and they are building that capability to make those work. This is a long-term process. There will be probably a lot of refinement necessary for an optimized supply chain with cellular therapies. I expect there to be several generations of cellular therapies which undergo consistent improvement.
JB: What role will partnering play in the continued development of the immunotherapeutic space?
AH: This is a very complex and diverse field that will require strong collaboration across the space. And when I say collaboration I mean two types – pharma companies coming together to co-develop different drugs as partners, or academic and industry partners coming together to advance the science to the best of our ability. I think both of them will be essential for driving and maintaining progress at the speed at which it’s moving right now. We will have certain impasses over time. They can only be overcome if we collaborate. I see a trend of such collaborations increasing.
There are some players in the space that are very collaborative and others that are not that open yet. Progress will depend on how open we are to collaboration. If I look back at the history of this space, the reason why the field tipped over in 2011 is because we collaborated across the all key stakeholders including pharma, biotech, academia and regulators. We only figured out the new immunotherapy paradigm and its differences versus chemotherapy because we did not have prejudice towards collaboration. I would always keep an eye on collaboration as a key factor for success.