Chad Tang: Thank you.
Leslie Ballas: You are here at GU ASCO presenting the K-COMPASS study. Can you explain to us what this is?
Chad Tang: Yes. Thank you for the invitation to talk. Yes, the K-COMPASS study is looking at our cohort of patients who are on a prospective trial treated with radiation therapy without systemic therapy for oligometastatic clear cell renal cell carcinoma. And it evaluated the prognostic ability of circulating tumor DNA using the Myriad test and KIM-1, a soluble protein marker for kidney cancer and kidney function. In addition to clinical factors such as number of mets, prior systemic therapy. It looked at the utility of both ctDNA and KIM-1 in terms of prognosis, and then utilized all the factors and selected the most important to develop a prognostic model, which we call K-COMPASS.
Leslie Ballas: And you looked only at patients with oligometastatic disease. Were there patients with oligoprogression as well included?
Chad Tang: Yes. Patients were all oligometastatic because they were not on any systemic therapy. So they were presenting to us pretty much without systemic therapy with up to five metastases.
Leslie Ballas: How do you use K-COMPASS? Who's using it?
Chad Tang: K-COMPASS is used for patients who are not on systemic therapy and they have oligometastatic disease. And the question that it's asking is how long you remain off systemic therapy if you use radiation therapy? And I think it's really hard for patients and providers to figure out whether they should proceed with systemic therapy, which does have some costs and toxicities, versus potentially radiation without systemic therapy. So the K-COMPASS tool is based on helping patients and providers figure out what choice to do next.
Leslie Ballas: And how accurate is K-COMPASS?
Chad Tang: It has great discrimination. We have a C index of 0.76. And it's very accurate at detecting who can be prolonged off systemic therapy. I would say it's pretty accurate.
Leslie Ballas: Have you done any external validation?
Chad Tang: It's a good question. We have another prospective trial, we call ASTROs, which is enrolling now, which will enroll 144 patients, which we plan for prospective validation. The tricky part of this is you do need ctDNA and KIM-1, which is a little difficult in a lot of retrospective cohorts. But we do plan to prospectively validate.
Leslie Ballas: And tell me, when you created the K-COMPASS tool, I noticed that you dichotomized ctDNA into positive and negative. I'm not ordering KIM-1 on my patients. And so I don't know, is that a positive/negative? It sounds like from the abstract that there's an actual value given. How do you integrate that value?
Chad Tang: KIM-1 is secreted from not only kidney cancer, but also kidneys. So if you just have no kidney cancer in like a normal patient, you still have a KIM-1 value, about 70. So we have KIM-1 as a continuous variable, and in K-COMPASS, it's per log 10 increase of KIM-1. Meanwhile, ctDNA has a natural cut point of being, can you detect it or not? So we utilize that in the K-COMPASS model. But you are correct, if we were to look at amount of ctDNA, that's also prognostic as well. So further studies may look more into the exact value of ctDNA. But the threshold we used we felt made sense.
Leslie Ballas: And is K-COMPASS something that providers should be using, patients themselves should be accessing, who is this for?
Chad Tang: It's going to be open for anybody to use. There will be a website in which anybody can plug in their variables, including ctDNA and KIM-1, if they have it. But it's open for everybody to use. I would always recommend that the clinical decisions are done in conjunction with a provider.
Leslie Ballas: Is this the first time that ctDNA and KIM-1 have been used together on a study?
Chad Tang: Yes, that's a great question. So ctDNA independently has been shown to be prognostic in patients with kidney cancer, and same with KIM-1. But so far those have been different studies in different silos. And one of the most important parts of our abstract is we're showing for the first time in the same patients that they're independently prognostic of each other. So doing both is going to give you more about the patient's outcomes than doing separately.
Leslie Ballas: Congratulations. This is incredible. This tool I think is very powerful and will probably provide patients a great deal of peace of mind. Thank you for talking with us today.
Chad Tang: Great. Thank you so much.