Ulka Vaishampayan, MD: What are the options for patients with both VEGF-resistant and I/O [immuno-oncology]—resistant disease? Are there any biomarkers we can use to help guide us? How do we make treatment decisions in this setting, Asim?

Mehmet Asim Bilen, MD: Ulka, this is a very important question, and this keeps coming up when we treat patients or when we get a phone call for a second opinion. “OK, what are we going to do? I gave I/O-I/O. After that I gave cabozantinib. Now what are we going to do?” Long story short, we don’t currently have a biomarker, although we are trying to find 1 and are collecting in IMDC [International Metastatic Renal Cell Carcinoma Database Consortium] and other cohorts. In addition to that, we also try to collect things like serial biopsies and blood work, and then try to come up with a biomarker. However, I think we are still far behind in terms of applying something to the clinical setting.

In my clinical practice, as Matt mentioned, after an immunotherapy-based option, I agree that cabozantinib is 1 of the best-in-class VEGF inhibitors. This is what I do. Then after cabozantinib with progression, Lenvima [lenvatinib]—Afinitor [everolimus] is still a very good combination. I’m pretty sure all you guys see responses like after VEGF failure—of Lenvima and Afinitor still working. This is an academic center. All of us are looking for a clinical trial. There are some agents coming into the scene that we’re going to talk about I soon, and hopefully we will have better agents after I/O and VEGF failures. But currently, we still try to come up with our own best sequence when we treat patients in our clinical practice.

Ulka Vaishampayan, MD: Yeah, I agree with you, Asim. There is a huge unmet need. Even with cabozantinib and lenvatinib and everolimus—I mean, granted, we have these options to help our patients. It’s great to have third-line, fourth-line options in metastatic disease. But clearly we can improve upon them. The best way to do that is by considering patients for clinical trials. Does anybody else have any comments?

Rana R. McKay, MD: Yeah. I think the field is really evolving. When we’re seeing these patients in the clinic, especially patients who may have received an I/O therapy initially, can you treat them with another I/O? Can you challenge them with another I/O? This comes up constantly. If I’m going to give axitinib in the second-line space, and if that patient hasn’t gotten the axitinib-pembrolizumab combination, I have this impulse to want to give them the doublet—if they can receive the doublet, understanding that this is sort of a data-free zone but trying to extrapolate from the frontline data. Or if somebody has never received nivolumab-ipilimumab but there may be post pembrolizumab-axitinib or other I/O-VEGF combination, when you’re using nivolumab from the CheckMate 025 data, are you going to feel a push to want to add ipilimumab? In clinical practice, I think we’re doing this. However, we are lacking prospective data to help guide selection. There are going to be some studies that will be coming out that will be looking at the role of I/O following I/O in a prospective manner. That’s going to really help address an unmet need in the field.

Ulka Vaishampayan, MD: Yes. There is a hope that as you get different biomarkers, depending on the type of progression, you may be able to make a smarter treatment decision.

Matthew T. Campbell, MD, MS: I would just like to add that we’re flying somewhat blind right now, in terms of evolution of these tumors post cabozantinib or post lenvatinib and everolimus We have a long way to go to understand what pathways are being turned on and what we can think about in those settings. It really seems like the earlier agents are not active post cabozantinib or post lenvatinib-everolimus, as you would hope. The other thing I would say—and this is just a shout-out to Rana’s work with the IMDC—their work showed that liver metastases and bone metastases were also independently negative prognostic factors. So those factors are part of what I think about for my patients as well. Where is their site of cancer? What is their volume of cancer?

The IMDC is tough because everything is a categorical variable. It’s either yes or no. When you have a patient who had their nephrectomy 11 months ago and now is progressing, if you start them now versus 2 months later, they’re either 0 or 1. If a patient possibly has very mild anemia, are you really going to give them a point? And so I think you still have to take the patient in front of you and look at their sites of disease, and then use the IMDC to help frame what you’re going to do.

Tian Zhang, MD: I absolutely agree. I want to go back to your point about the refractory-disease patients and how we’re learning about the pathways that are turned on with patients who are resistant to cabozantinib or lenvatinib. If that patient has a great performance status and is still a clinical trial candidate, I agree that we should think about putting that patient on a trial. But I am also thinking about potentially doing another biopsy, sequencing at that point, depending what treatment pressures have evolved and how the tumor has mutated in response to these treatment pressures. Could there be anything else targetable with either an on-label or an off-label treatment? Those are the patients who I’m selecting for biopsy and further sequencing.

Ulka Vaishampayan, MD: That’s a good idea.

Matthew T. Campbell, MD, MS: I really think it’s such an epigenetically driven tumor. We’ve got to get down to the protein level again and see what is actually happening at that level. We’ve been at a higher stream, in terms of always looking at DNA and RNA. We are doing more transcriptomic work, but I think proteomics are going to play an important role for us as well.

Ulka Vaishampayan, MD: Interesting. Before, Matt brought up that brain metastases is 1 area where none of our typical agents are performing very well. That is a setting where multiple patients are failing therapies, and we need better treatments for that.

Tian Zhang, MD: Absolutely. I have quite a few patients who have great systemic control, as you mentioned, Matt, and then have disease progression in the brain. We certainly need to understand more about the microenvironment in the brain and learn more about the mechanisms there. They’re evading the immune system when they have such a great response everywhere else.

Ulka Vaishampayan, MD: Yeah. These are excellent suggestions in regard to considering doing a biopsy, looking at genomics, and obviously looking at any clinical trials you have available for the patients to enroll in.

Transcript Edited for Clarity