Ulka Vaishampayan, MD: The other point that I want to bring up is the concern for the way we assess response. Our RECIST [Response Evaluation Criteria in Solid Tumors] criteria are largely designed around chemotherapy type of responses. For immune responses, the criteria may not always completely apply and be able to be used to help us make a decision about the actual response rate in the patient. Clearly, there are occasionally patients with stable disease or progressive disease who also tend to benefit, although not as strongly as the ones with objective response.

Rana R. McKay, MD: We’ve all seen patients like this in our clinics, who have residual masses after they’ve been on therapy for a long, durable period. They may undergo resection in an attempt to make them NED [no evidence of disease] on imaging. You resect out this maybe large, several-centimeter tumor, or what you think may be a tumor, and it ends up being necrotic—all dead tissue. Our response parameters don’t really assess that very well, and I think people who do enter a deep remission, though they may not necessarily have a complete response, can still derive benefit. I think some of that data, from the pembrolizumab/axitinib trial, was presented by Brian Rini, MD, at ASCO [the American Society of Clinical Oncology annual meeting] last year. And I think it’s important to highlight the limitations of our current response assessment criteria.

Ulka Vaishampayan, MD: Are there better response criteria? Radiomics is an up and coming translational correlative imaging technique that may be able to determine responses better, or predict benefit to immune therapy. Are there any other criteria that anybody is using?

Tian Zhang, MD: In the cooperative groups, at least, there’s been a very big push to use iRECIST [immune RECIST] criteria to think about what is true pseudoprogression versus not. And the confirmatory scan for progressive disease is very important for iRECIST criteria. We are incorporating these into our trials of immune checkpoint inhibitors, but I think these criteria still have to be validated.

Ulka Vaishampayan, MD: Yes, agreed. And maybe in these large cooperative group trials, we’ll be able to validate iRECIST as a better response technique.

Matthew T. Campbell, MD, MS: This may come up later, but I completely agreed with some of the statements that were made at ASCO last year. A panel explained that patients who have clear cell cancer with sarcomatoid features tend to benefit in a major way from this combination—with a 60% response rate, and close to 20% CR [complete response] rate. Those patients in particular, who had been the worst off, were now clearly benefitting in a tremendous way from this approach.

Ulka Vaishampayan, MD: Yes, agreed.

Mehmet Asim Bilen, MD: And also, Ulka, I want to point out that, nowadays, we are in the era of advanced imaging. PD-1 [programmed cell death protein 1] PET [positron emission tomography] imaging is coming up, and I think this is very promising. We may be able to follow up on the patient and see the pattern of resistance, as well as determine whether this is immune hot or immune cold. I think knowing this while using advanced PET is really going to be very helpful for us.

Ulka Vaishampayan, MD: Yes, that’s an excellent point. The novel imaging techniques will help us predict responses better. Stay tuned for that.

Transcript Edited for Clarity