NY-ESO-1 immunotherapy for ovarian cancer
Dr Kunle Odunsi - Roswell Park Cancer Institute, Buffalo, USA
Our group has focussed on trying to understand the mechanisms of interaction between the immune system and ovarian cancer. We made the observation several years ago that patients whose tumours are highly infiltrated by T-cells, especially CD8 positive T-cells, have a better outcome compared with patients with poor infiltration. This prompted us to begin to search for the target of recognition by the T-cells. So what are these T-cells recognising? This led us to search for antigens recognised by the T-cells.
We focussed our efforts on a family of antigens called cancer testis antigens. These are a unique class of antigens that are usually expressed by tumour cells, not by normal cells except the adult male testes. The major antigen that we focussed our efforts on is an antigen called NY-ESO-1 and this became the prototypic antigen for our cancer vaccine trials. So we developed vaccine approaches whereby we wanted to prolong remission in ovarian cancer patients because, as you know, ovarian cancer most patients typically respond very well to initial therapies but ultimately the majority will relapse and once relapse occurs it’s difficult to control.
We focussed initially on patients who were in remission from ovarian cancer, we’d vaccinate them to generate anti-tumour immune responses in order to potentially prolong the remission time or, in fact, completely prevent relapse of disease. So these were our initial trials focussing on this antigen target and our results demonstrated significant promise.
However, it became clear to us that when these patients ultimately relapse the question is what are we going to do for these patients. So we studied mechanisms by which ovarian cancer escapes from immune attack. Some of those mechanisms include the role of the checkpoints PD-1 and PD-L1, we published on this about six years ago in ovarian cancer. Another mechanism is an enzyme called indoleamine 2 3-dioxygenase, IDO. This enzyme catabolises tryptophan and T-cells are very sensitive to tryptophan. When T-cells sense there is no tryptophan in the environment they are not able to proliferate and, in fact, it leads to arrest of their proliferation.
So once we identified these pathways we began to combine with our vaccine strategies in order to overcome any attempt by the tumours to escape from the immune attack. We have ongoing clinical trials where we’re not only vaccinating these patients but also counteracting the escape mechanisms by the tumour by blocking IDO and blocking the checkpoints.
One of our major recent efforts has been a focus on adoptive cellular therapies. Adoptive cellular therapy is able to mediate regression of large established tumours, both in animal models as well as in the clinic. But the approach that we have taken for adoptive cellular therapy protocols is really to re-engineer T-cells, that different ways of engineering T-cells, as many will know, one group of engineered T-cells are called CAR T-cells. But we have focussed our attention on another method which is to use T-cell receptors for engineering the T-cells. So these engineered T-cells are produced by doing leukapheresis in patients, modifying ex vivo the T-cells, generating large numbers, billions and billions of these cells, and putting them back into patients.
But the story doesn’t end there because as we started doing these clinical trials we found that although we can induce large numbers of T-cells when they enter the patient’s body, the frequency of the T-cells begins to decline after a while. In fact, usually by 60-90 days you see very low frequencies. So we raised the hypothesis that is it possible to use young cells, precursor haematopoietic stem cells, for these T-cell engineering approaches. This is what we are currently doing; we’ve shown very clearly in our preclinical models that the younger the precursors are and you engineer the cells using a T-cell receptor you can actually be able to provide long-lived cells that are able to last almost for the entire lifetime of the patient or the mouse model. So we are harnessing this approach in a clinical trial for ovarian cancer patients probably within the next 9-12 months whereby we will take haematopoietic stem cells from the patients, we will re-engineer them ex vivo, put them back into patients and allow these cells to be a continual source and replenish the lymphocyte population in vivo and become a lifelong source of anti-tumour effector cells.
What time frame does this remain effective?
What we find is that in animals where you don’t use the haematopoietic stem cells if you adoptively transfer mature T-cells the T-cells don’t last for long whereas when we adoptively transfer the haematopoietic stem cells that have been engineered to be cognate receptors, the cognate T-cell receptors, not only do they persist for long, as these stem cells mature and differentiate into mature T-cells they now bear dose receptors with the ability to recognise and destroy tumour targets.
In the end what I think is going to happen in the future is that we’re going to have to combine all the different strategies. One the one hand generate very robust anti-tumour immune responses either using our vaccine approach or using the adoptive cellular therapy and, at the same time, counteract tumour escape mechanisms, the mechanisms by which ovarian tumours escape from immune attack. We are already beginning to do our clinical trials by actually segregating, by subsetting patients based on the knowledge of the signature of the immune landscape within the tumour microenvironment. So that’s number one. Number two, we’re also beginning to identify patients based on the mutational pattern. Because of advances in genomic technology it’s now possible to rapidly sequence tumours understanding the mutational landscape. So in my opinion now we’re combining the patient’s mutanome, the mutational landscape, with the patient’s immunome which is the immune landscape and together bring them together in order to deliver powerful treatment strategies for our patients.
How will this best be applied?
We’ve paid very careful attention to this at Roswell Park Cancer Institute, so we’ve built the infrastructure to be able to do this. For genetic engineering you need a GMP infrastructure for rapid manufacture of your [?? 8:17] constructs. So we’ve developed this, we have the infrastructure for cell production, we have all of the technologies come together within our Centre for Immunotherapy that allows us to rapidly move the concepts from our mouse models to the clinic. Because we can rapidly put all of these things together, of course it’s a lot of work, especially with regulatory hurdles, but it is now possible and I envision that as we become even more and more familiar with these approaches it should get easier and easier with time.