Tumour-infiltrating immune cells as classifiers and predictors of outcome

Bookmark and Share
Published: 26 Jun 2014
Views: 2811
Dr Jérôme Galon - Laboratory of Integrative Cancer Immunology, Paris, France

Dr Galon talks to ecancertv at WIN 2014 about tumour-infiltrating immune cells as classifiers and predictors of outcome.

I was invited to talk about the immunoscore and the role of immune cells against cancer. My lab has been focussing on trying to really analyse deeply the tumour microenvironment and to really understand what are the important cells within the tumours that either promote or control tumour progression.

Could you tell us about the research you presented, starting with the reasoning behind it?

The reasoning behind it was first, when I started more than a decade ago, the field of tumour immunology was not at all recognised. Most of the studies were focussing on tumour parameters, tumour genotype and I thought that the immune system was important as well. So I’ve tried to analyse the complexity of the tumours in total, so not only the tumour cell complexity but also the tumour microenvironment complexity that is the complexity of the immune system infiltrating the tumours. So that was with the idea that patients having long term survival versus those having a very short term survival of cancer, one of the reasons could be because of the immune system of the patient, of the immune response of the patients. It turns out that indeed subsets of immune cells are extremely important to predict long term survivors in the patients, the survival. These are the cytotoxic T-cells, the memory T-cells with a memory capacity, so we have shown that over the last ten years.

Were you just looking at the cells or cytokines?

No, we looked at every parameter as related to the immune response, so not only the cell types, the immune cell subpopulations, but also the receptors that they are expressing, the factors that are secreted by those cells or that are present within the microenvironment, so cytokines, chemokines, all sorts of immune related features and some of them are very important.

What cells or cytokines are important?

The main cells which really predict the survival of the patients are the adaptive immune cells, especially the T-cells, the cytotoxic T-cells, the memory T-cells and recently we also have shown the importance of the T follicular helper cells and the B-cells. So all those adaptive immune cells are very important; together with other immune related molecules like the TH1 type of orientation, so interferon gamma, TbAg transcription factor, the gamma cytokines, the cytotoxic molecules such as granulysin, perforin, granzyme are very important and some chemokines are also very important.

How did you take that information and create the immunoscore?

What we did was really to have a very deep analysis of the tumour microenvironment and we took the most powerful cells to predict patient survival and we put them into the immunoscore assay. The immunoscore assay is a very simple test but still very powerful to predict patient survival and to classify cancer patients. So it’s basically the densities of cytotoxic and memory T-cells infiltrating tumours, taking also into account not only the density of these cells but also their location in different tumour regions. Basically with this information of the density of these cytotoxic and memory T-cells we give this immunoscore result that allows us to classify cancer patients.

You mentioned it was a simple test, how would it be performed?

It can be done in a routine pathology lab setting, so regular paraffin tissue from tissue tumours embedded in paraffin and then it’s two simple stains with the two antibodies specific for the cytotoxic and memory T-cells. Then it’s a full quantification of those cells, so it’s an image-based analysis that is done with digital pathology. It’s not a subjective selection of a particular field, it’s an objective quantification of the whole immune cells infiltrating the tumour.

How have you used the test so far?

We have tested it in multiple cohorts of patients with colon cancer, with rectal cancer, with breast cancer, different cohorts at all cancer stages, so from stage 1, stage 2 to stage 4. The immunoscore is powerful at any stage, at all stages, even the earliest tumour or the latest tumours, and we have ongoing studies in colon cancer, we have a consortium that I initiated, an international consortium with 23 centres in the world, currently testing this immunoscore. Also we have a national prospective study in France ongoing involving six hospitals in France.

How will the prognostic value of the immunoscore be tested?

It’s going to be tested based on the disease free survival and overall survival of the patients. What we have shown in the past and what others have shown also is that the density of the cytotoxic and memory T-cells, and especially the immunoscore, that is the most powerful way to quantify them; it’s extremely powerful in terms of classifying cancer patients and allowing their prognosis.

What about the inter-patient variability?

The inter-patient variability, it means that some have a very high immunoscore, so strong T-cell priming, a lot of cytotoxic effector cells, a lot of memory T-cells and those patients will do well compared to the patients not having this T-cell infiltration. What we have also shown more recently is that in fact initially the tumours, the density of the infiltrating immune cytotoxic and memory T-cells is very high and with an immunosuppressing mechanism these densities are decreasing over time and when they are decreasing the tumour is progressing. So we have a relationship between the density of these cells, so between the immunoscore, and the tumour progression and invasion parameters.

How might the research you presented influence clinical practice now or in the future?

Today there is not a single test to evaluate the immune system of a cancer patient, nothing is done in routine and hopefully the immunoscore would be the first standardised reproducible powerful immune test that could be introduced into the cancer classification. So this would be an immediate benefit. The second is that it may completely change the picture and change the way patients are treated. For example, there are patients at, let’s say, a late stage, having stage 3 or very aggressive chemotherapies. We know that some of these patients with a very high immunoscore will do very well but probably not because of the treatment they are receiving, probably just because they have these very strong memory T-cells to start with. Vice versa, there are patients with immunoscore zero, so very low T-cells, memory T-cells in their tumours and those do very bad even though they receive aggressive chemotherapy. So chemotherapy is clearly not efficient in those patients. We could also take other examples at early stage tumours, now we can predict very high risk of tumour recurrence at early stage tumour with a very low immunoscore and so it would be wise to start new clinical trials addressing the question of are there any benefits of chemotherapy or other treatment, like immunotherapy, in patients with these early stage tumours.