Immunotherapy is showing great promise in many cancer tumor types as medical experts continue to expand their studies in the field.
In an interview with CURE, Jason Luke, M.D., assistant professor of medicine, University of Chicago, discussed the biology of immunotherapy, similarities across tumor types, the importance of biomarkers and what still needs to happen in the future.
What are some of the common questions in this field?
In the field of immunotherapeutics, the difference between cancers isn’t really the relevant point – we don’t think. It’s more the difference between the immune system response to cancer. So we have to sort of take a step back and rethink the way we thought about cancer therapeutics in the past.
Shared features across different tumor types can include the things like mutations, which we had commonly thought about, but we tend to think of them in a different way than we did in the past. So instead of being focused specifically on mutations like RAS or BRAF – it’s more the number of mutations than the consequent downstream number of neoantigens generated that different tumor types may or may not have. Then the quality of the baseline immune response in the patient can really dictate how well these drugs can then work.
That can be measured a number of different ways – PD-L1 assay, gene expression profiling and detecting the diversity of the T-cell receptor. There are a number of technologies that are evolving that are going to help us understand how to apply immunotherapy more broadly, not so much in a cancer-specific way in terms of disease histologies, but rather in an individual cancer sort of way aiming at the quality of the immune response that they already have at baseline.
Do you think one day, perhaps, we will have immunotherapy to treat almost all tumor types?
Yes. That’s our overriding hope, and I think the future probably won’t be immunotherapy alone for many patients. There seem to be sort of three main phenotypes in patients – one is with a very robust response and they respond well to PD-1 antibodies, those patients who sort of have an intermediate response where you can get some activity and then some patients are like no immune response.
Those patients with immunotherapy on its own probably aren’t going to have much activity. But almost certainly, even in those patients, combining immunotherapy with chemotherapy or targeted therapies or other therapies will enhance the benefit and hopefully generate long-term, potentially cures in some patients.
We see immunotherapy, particularly right now in melanoma and lung cancer, and it’s working really well, but then we look at some other tumor types and there is difficulty. What makes immunotherapy work so well in some situations? Where do we need to go to get it to work in others?
Well, the paradigm really is thinking about what the baseline immune response of the patient is and in some tumors we see, in most patients, a very robust anti-tumor immune response. You mention melanoma and lung cancer, and there are hypotheses out there that part of the reason that those baseline immune responses are so robust is a number of neoantigens or mutations that are different between the cancer and the regular tissue are enough that the immune system will notice that.
Other diseases have less of that. Other diseases also seem to have other specific cancer histologies, and seem to have sort of unique properties that make them immunosuppressive, so pancreas cancer is one of the real tough nuts to crack here; colon cancer has been another one. Microsatellite instability-high colon cancer responds well to immunotherapy, whereas microsatellite stable colon cancer really doesn’t.
We’re only really at the beginning of understanding these molecular differences, and probably between different cancer types, there will be individual issues that we need to overcome to enhance that benefit and bring immunotherapy to patients in each disease.
What do you think will happen or needs to happen within the next five or 10 years?
I think we really have to improve our diagnostics in terms of translational biomarkers and we have to start actually applying them in clinic. So I think it’s really not good enough that we do just PD-L1 screening. We know that that’s not going to tell us exactly who’s going to benefit and who’s not.
We need to start using multiple of these biomarkers and use, what I refer to, as multidimensional analysis to try to figure out what the optimal sequence is. Then we need to start thinking about other ways to use immunotherapy.
Are there other biomarkers being explored?
There are many biomarkers in immunotherapy that are coming forward and so PD-L1 is one, but PD-L1 is only one molecule, so it’s subject to a lot of variability.
We are much more interested in gene expression profiling, looking at a whole host of genes which biologically we would expect in parallel to be up or down that increases the strength of any statistical test quite dramatically.
The mutational load hypothesis is another biomarker that people are trying to get into clinic – meaning could we find cut points where just on a standard assay we could know well there are enough mutations we would expect the immunotherapy would likely work.
Other biomarkers of relevance are the T-cell receptor diversity – meaning does it look like there is a baseline immune response? We need to start doing clinical trials where we get all of this data, so that we can then go back and say, “which of all of these, or if we use two out of three or so on and so forth, is going to give us the greater power to make a difference for an induvial patient?”