A panel of experts explain how they approach CNS prophylaxis in DLBCL.
John Leonard, MD: Within the last year at meetings, CNS [central nervous system] prophylaxis has been all over the map. I’m going to ask your approach to that, Cyrus. We can critique the data on double-hit lymphoma. There are more data on CNS prophylaxis, but I’m not sure how definitive they are, even though a lot of patients have been studied. So Cyrus—and then I’ll ask others—how do you approach CNS prophylaxis? Do you use the CNS-IPI [CNS International Prognostic Index]? And how do you approach it with intrathecal, with sampling and systemic methotrexate? Cross your fingers and hope? What’s your strategy?
Cyrus Khan, MD: Great question. You’re right: It’s all over the place. But I follow the CNS-IPI. That’s our standard at West Penn Hospital. In case a patient comes in with adrenal gland involvement, kidney involvement, or breast involvement, obviously I don’t really look at the CNS-IPI. Those are really high-risk patients who need serious prophylaxis. As far as the mode, there’s a lot of to and from between using intravenous [IV] methotrexate vs intrathecal methotrexate, and the jury is still out. I do prefer intravenous methotrexate, particularly for young patients who can tolerate it and are high risk for CNS relapse. For older patients who are high risk, but who I don’t think would be able to tolerate IV methotrexate, I certainly do intrathecal. Then the question comes in, “Do you need an Ommaya reservoir, or can you still do the LP [lumbar puncture] between conventional radiology? We’re still doing intrathecal prophylaxis. We go through interventive radiology and give it each time. The jury is out there too about how to approach that. That’s how we approach it, and we do it each cycle, so for 6 cycles.
John Leonard, MD: Greg, anything different at Mayo Clinic, typically?
Greg Nowakowski, MD: Yeah. With the thought that even systemic methotrexate may be not necessarily be as protective as we’ve thought in the past, we’re dialing down a little on the CNS prophylaxis. We still use CNS-IPI, but what we really worry about are the unusual size of testicular lymphoma; adrenal lymphomas, as you mentioned; patients with multiple areas, multiple sets of external disease. In others, we sometimes worry about the potential for kidney damage, which affects the therapy overall, so it’s a very controversial area. John, I would challenge you, as the head of the Alliance for Clinical Trials in Oncology. This is a great topic to study for the cooperative groups. I want it to be the ideal way for prophylaxis in diffuse or B-cell lymphoma for CNS relapse. You can almost imagine an add-on therapy to any induction treatment, trying to evaluate a prospective patient. It would be challenging, but we need to answer this question to move forward.
John Leonard, MD: I appreciate the interest in the area, but I’m concerned about designing a study. My first question would be, is this the most important issue in aggressive lymphoma? We’d all agree that it’s important, particularly for patients affected by it. But at the end of the day, if you want to impact the largest number of patients, and you can do only so many studies in a randomized way, this is probably not the issue to tackle as a primary thing, mainly because most of these patients who have a CNS relapse also have systemic disease that needs to be addressed. I’m sympathetic to the idea, but I’m skeptical that this rises to the top of the list. Honestly, for the number of patients in whom we need to see a difference, and the differences between intrathecal methotrexate and systemic methotrexate, my guess is they’re in the order of a few percentage points, if anything; therefore, you’d need a lot of patients. But I certainly sympathize with and share the interest if we could do it in a feasible way.
I do think we’ve seen some interesting data with ctDNA [circulating tumor DNA] in the CSF [cerebrospinal fluid]. With some of those techniques, we might be able to identify a group of people who are at ultra-high risk or ultra-low risk at some level, and therefore focus and perhaps design a study that would require a smaller number of patients. If we knew, for instance, that the majority of people with a certain level of ctDNA—it’s not really ctDNA, but it’s in the CSF—then we could focus on that group and maybe wouldn’t need a huge number of patients to see a difference. Anyway, it’s a great point. A lot of people are interested in it, and it’s frustrating not to have an answer.
TRANSCRIPT EDITED FOR CLARITY