Predictive Biomarkers and Response to Immunotherapies in Urothelial Cancers

Padmanee Sharma, MD, PhD

Today we are speaking with Padmanee Sharma, MD, PhD, a physician scientist who specializes in immunotherapy for cancer at the University of Texas MD Anderson Cancer Center in Houston. Dr. Sharma recently gave a talk at the 2016 American Society of Clinical Oncology (ASCO) Genitourinary Cancers Symposium, held January 7–9 in San Francisco, on predictive biomarkers and response to immune checkpoint inhibitors for urothelial tumors, which include tumors of the bladder, ureters, and the renal pelvis.

- Interviewed by Anna Azvolinsky 

Cancer Network: First, what are some of the immunotherapies currently being developed for these types of tumors?

Dr. Sharma: The immune checkpoint therapies that are being developed currently are those that block the cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4) inhibitory pathway, as well as the programmed death 1 (PD-1) and programmed death ligand 1 (PD-L1) inhibitory pathways. These therapies are very important because they have shown that by blocking the inhibitory pathways on T cells, which are the soldiers of the immune system, you can actually get very robust anti-tumor immune responses that lead to regression and, in some patients, even complete disappearance of disease. Since we are blocking pathways on T cells and not tumor cells, tumor types such as melanoma, lung cancer, and kidney cancer have been explored, and they have shown really great results. These immunotherapies are now being explored for urothelial cancers like bladder cancer.

Cancer Network: Do we have clinical evidence that some of these urothelial tumors are likely to respond to immunotherapy?

Dr. Sharma: We did one of the first clinical trials at the time, in 2006, with the anti–CTLA-4 antibody ipilimumab and showed that in 12 patients with early-stage bladder cancer, 3 of the patients’ tumors completely went away. They had no tumors at the time of surgery. That was a very early clinical trial. More recently, researchers have been doing clinical trials with anti–PD-1 and anti–PD-L1 antibodies. At the ASCO Genitourinary Cancers Symposium, we heard about the anti–PD-L1 antibody from Genentech/Roche and that there are patients who are having really dramatic responses to it, and I think we will hear more data in the next set of upcoming scientific meetings. Clearly there is evidence that patients with these types of tumors are responding to these immunotherapy agents.

Cancer Network: As far as candidate predictive biomarkers of response, are there any leads for anti–PD-1 and anti–PD-L1 antibodies?

Dr. Sharma: There are intensive investigations to look for predictive biomarkers, and I think we have to be really thoughtful in this arena because predictive biomarkers are important; if you can predict which patient will respond, then you can choose the correct patients for treatment and avoid giving the treatment to patients who are not going to respond. So, that helps us from an economic standpoint in that we are not getting extra cost incurred because we are not giving the treatment to everyone, we are picking the right patient.

We saw the importance of predictive biomarkers in genomic medicine; since you can predict, for example, that a patient has a BRAF mutation, they will get the drug that targets that mutation. Only those patients with the mutation receive that particular therapy. However, the immune system is very dynamic and cannot fit into the same box as genomic medicine. The immune system changes on a daily basis. The immune response we measure today is not the same immune response we see tomorrow. And that makes it difficult to come up with a predictive biomarker. We want to be careful and not restrict the therapy from patients who could potentially benefit. We want to give patients the chance of the best clinical benefit possible. These therapies to date do not have a predictive biomarker that show that only some patients should get the therapy.

But, I should point out that some groups have explored the idea that PD-L1 expression either on tumors or immune cells could be a predictive biomarker; unfortunately, to date, all the clinical trials indicate that even if you do or don’t have expression of PD-L1, both can potentially have responses to the PD-L1 and PD-1 antibodies. So, PD-L1 is not a predictive biomarker. It seems to be a prognostic biomarker, which means that it can indicate which patients may do well overall. We are still looking for what may be predictive. At the end of the day, because of the dynamic nature of the immune response, what we will see is that we could monitor patients’ immune responses over time, longitudinally, and then maybe make a decision of which immunotherapy to give at a particular time.

Cancer Network: Lastly, are there clinical trials that are being planned that take into account biomarkers for urothelial tumors for some of these immunotherapies?

Dr. Sharma: I do think that researchers will try to look into how to integrate multiple biomarkers, to try to make decisions about appropriate monotherapy or combination strategies. I think those clinical trials will be done, but they are further in the future and just being planned.

Cancer Network: Thank you so much for joining us today, Dr. Sharma.

Dr. Sharma: Thank you!