IMPAKT Breast Cancer Conference 2013

CD4 T cells in extensively infiltrated breast cancer signal organised immune response and predict survival

Dr Karen Willard-Gallo - Université Libre de Bruxelles, Brussels, Belgium

Karen, thanks for coming in because this whole idea of immunology being important in cancer is crucial and you are looking at things like lymphocytes, you’re looking at arms of the immune system attacking the cancer tumour. Can you tell me what you’ve been doing in this study that you’re releasing right here?

What we did was we asked can we get an image of the immune response in the tumour and without manipulation and can we simply from that learn why some patients respond better in long-term follow up and others do not. Is there an immunologically based reason for that?

Normally speaking, how do you get images of the immune system in action?

So what we did was we took the tumours within two hours from the operating room and we homogenised them mechanically without using anything that would alter their phenotype on the surface. We isolated them and we extracted RNA which we were then able to use PCR techniques to look at genome-wide expression of the RNA and we also used flow cytometry to look at proteins on the surface and identify changes that were occurring.

So the logic of looking for RNA rather than anything else in the genetic superstructure is what?

It was because we wanted to see what were the changes in expression relative to normal cells and we couldn’t do this by protein for the entire expression profile in these cells, we could only do it by RNA because of the sensitivity of the technique.

So the RNA is actually the whole immune process actually happening before your eyes.

Right, there are genes that are transcribed in any cell and then there are genes that are transcribed because it’s a lymphocyte and then there are genes that are transcribed when the lymphocyte is activated and it’s making an immune response and it depends on its given function as to which genes are transcribed. So we asked the simple question of in breast cancer what are the important elements? What we found, I think, is actually very interesting. We found a very strong signal for a given chemokine which is a molecule that is chemoattractant to lymphocytes and this was being produced by a subset of cells of helper CD4-positive lymphocytes. This particular subset is a new subset that was discovered in the year 2000 and it’s called follicular helper T-cells. These cells have never been described in cancer before in solid tumours, other than lymphomas of which they’re involved because they’re in the lymph node. We found that when we detect this signal in breast cancer and that these cells are present, we have a highly organised mini lymph node next to the tumour bed. This lymph node, mini lymph node, called a tertiary lymphoid structure, has a T-cell zone which is primarily CD4 cells but some CD8s. And when you have this, the more you have the more CD8s you have in the tumour killing the tumour cells. And it has a B-cell zone with active germinal centres and the TFH cells are present there and this construction of this organised response correlates very specifically with a good prognosis. So we derived a signature for these highly specialised cells, these follicular helper T-cells, of eight genes and then we interrogated two populations of a thousand patients looking at sections of whole tumour. We found that in patients where the patients had not been treated other than surgically and we had ten year follow up, if they had… the higher the signal for these cells that we could detect, the better the outcome.

Right, so if you see those cells in action then you know the patients could do better. What are the clinical implications of that?

Well, the clinical implications are that these cells are a very minor population but they’re very important and they seem to regulate and these structures are very important. These structures are involved in allograft rejection of kidneys and hearts so they function to reject the graft. So can they function to reject the tumour? Can we enhance them and the immune response in patients who have either nascent structures or well-established structures, and that’s about a third of patients. And in the two-thirds of patients that do not, how can we induce them and do this at the time that the tumour is still present? So re-think therapy and maybe treat while the tumour as a source of antigen is still there so that the one wonder drug for the patient that can control maybe a lifetime is their educated immune response.

So this fulfils the sort of dream of being able to harness the body’s natural ability to fight the cancer. Do you think, however, one of the uses for this would be to decide not to use tough therapies in patients who have lots of these cells around or do you really think that this pathway could be harnessed and you could develop agents to enhance it or at least not to destroy it?

Well the dream would be, yes, to not destroy it by some of the chemotherapies which may because they are very immunosuppressive and to encourage it in patients where you can detect it. To enhance it in those that have a borderline response and to initiate it in patients that don’t, this is obviously the dream. If having this represents a real anti-tumour immune response, even though the tumour is there and it’s a failure, it’s the memory of those particular antigens on the tumour that may allow a given patient to supress any residual tumour cells’ lifetime.

And how much has this been achieved by your group because of this sort of ability to take a snap-shot of what’s happening in the tumour immunologically, rather than taking tumour, expanding it and perhaps getting all mixed up later?

I think because we asked a question of just tell us what’s there rather than trying to look at a given response, and because we made a huge effort, it was an effort by surgeons and pathologists and medical oncologists at our institute to do this work, and because of that I think we’re seeing what is real, not what is a mouse model, not what is an in vitro expansion. We’re seeing what’s actually happening in the patient.

And you’ve identified two arms of this process, do you think there could be other pathways that you have yet to discover that could be equally as illuminating?

Oh for sure. The immune system is complex and one arm of it is certainly not the answer. But what I do think that’s very interesting is that when you have these structures that are like a mini lymph node you have a complete immune response adjacent to the tumour and perhaps this is more important than immune responses in the draining lymph node.

But isn’t that very complexity the very issue for the busy cancer doctor because experience has shown that simple drugs, simple therapies and simple approaches like chemotherapy and surgery work and understanding the immune system is really difficult.

It is very difficult but they don’t work necessarily; they work for some patients but there are still failures and there are still women with breast cancer that relapse after ten, twenty, thirty years. And some of this has got to be that their immune system is handling it to a certain point and then there’s immune failure. So we need to understand a woman who is 35 years old and who has two small children, how do we educate her immune system to her tumour so that she can maintain that suppression lifelong and live a long and full life.

Now that’s an exciting prospect but what about day-to-day medicine today? Is there any message for this for doctors and their patients right now? Can you say anything that could be practically useful in the clinic?

I think that the one thing we have to really think about is taking the tumour out quickly the best idea. That, in fact, if we can identify in the core biopsy whether a person has these responses or not, perhaps we need to use the tumour as a way of vaccinating the patient to their own tumour because the tumour is heterogeneous, it’s a variety of mutations in different cells and if we really want the immune system to control it, the immune system needs to see all of that variety. So maybe it’s beneficial and that’s why many more trials are going to pre-operative chemotherapy. Maybe the efficiency of this is also the immune arm. And to do it with drugs that are not immunosuppressive, that at one time killed the tumour but flood the system with antigen and make a vaccine in the patient. In this case the immune system would do the job of identifying all the variants.

It’s exciting to think that this powerful system could be working to fight breast cancer; do you have any educated guesses about how patients and doctors might steer their therapy towards more harnessing these natural processes even before we discover all the intricacies of what’s going on?

I think we need more and more neoadjuvant trials and window trials where we can look at the ability of a given drug to change the tumour from the core biopsy to the tumour at surgery. This will teach us which of these things are having an effect and eliciting a response. And we need to identify the parameters of the response that are important. In my opinion it’s a global response that’s needed and not just CTL killing, not just innate NK cells killing but we need to induce memory. There are some exciting things going on in chronic viral infections that lead to tumours. You know that 20% of cancers are caused by viruses and some of these studies have shown, for example in skin, that resident memory cells remain there lifelong, they don’t recirculate and they provide immunity to certain viruses that can come in through the skin. I think that memory and generating memory is extremely important.

And to sum up the message coming here from Brussels, from your presentation at the meeting here, what would that short summary be?

The short summary would be identifying that these particularly specialised cells that are called follicular helper T-cells are absolutely identified with these structures, these mini lymph nodes, and that when you have that you have a good response: long term disease free survival or pathological complete response to neoadjuvant chemotherapy and so it’s a good prognostic.