Moving Beyond Medians

Transcript:James R. Berenson, MD: There have been several large randomized clinical trials all comparing a newer agent with Revlimid/dexamethasone (RD) versus a placebo with RD. And, they have shown a consistent prolonging of the progression-free survival among patients receiving the third drug. There have been issues with the trials’ designs in several of them, and they really are not as inclusive as I would like. The shape of the curves does differ in some of them, for some puzzling reasons in some of the cases. For example, the ixazomib (Ninlaro)/Revlimid/dexamethasone (NRD) versus RD trial shows that separation that appears out of nowhere at 9 months and is stable in both curves or down-going. And, then, in the trial done with carfilzomib (ASPIRE), there is a separation in those curves, but they continue to go down. In addition, in both arms, they’ve really tried to treat a pretty good group of patients. There weren’t a lot of high-risk patients. These patients had to be responsive to proteasome inhibitors. So, they didn’t really set the bar very low for responses, they set it very high for enrollment.

Whereas with the randomized trial that was done with elotuzumab/Revlimid/dexamethasone versus RD (ELOQUENT-2), it included a lot of patients that were high risk. And yet, we’re seeing a flattening of the curve, suggesting long-term benefit for the addition of elotuzumab. And, in my own experience, both on and off trial, I am seeing patients who have ongoing responses that are continuing in their myeloma markers, way beyond what I see with RD alone; second year, third year. The numbers just keep getting better, suggesting there’s something different here with the use of the antibody beyond simply another chemotherapy drug, another proteasome inhibitor, or another immunomodulatory agent. There’s something here.

Jatin P. Shah, MD: It’s important when you start looking at our different combinations to start thinking beyond our typical median PFS and median overall response rate. And this is a consistent message that we’re seeing with many of our novel therapies now. So, we historically look at just the median PFS or overall response rate. But, I think things are changing. I think it’s important for our treating physicians now to look beyond just that one line, and think about looking at the actual progression-free survival Kaplan-Meier curves.

There are a lot of differences that we see in terms of what the kinetics of those look like. We need to look at both the hazard ratio, as well as the PFS. I think it’s important to look at both of those together. I don’t think one individually gives you the whole picture for the benefit of each combination. So, it’s not only looking at the combination of both the hazard ratio and the PFS, but it’s also actually looking at the shape of the curves. That adds a lot of value. If you look at two Kaplan-Meier curves that separate, but then come back together, that’s very different—maybe a Kaplan-Meier curve that separates and remains separated—because that starts speaking to long-term disease control, as well as saying, “I’m really getting some long-term disease control and benefit from this therapy for these patients.”

There are a couple of things, when we start looking at these Kaplan-Meier curves with some of our newer therapies, to look at. It’s challenging or interesting, for example, with the TOURMALINE study with ixazomib/lenalidomide/dexamethasone versus lenalidomide/dexamethasone, the progression-free survival curves don’t actually start separating out until 9 months. So, they have a late separation, which is different than what we’ll see with most of our therapies where we’ve seen early separation at 1 month from the two different arms—in the comparator arm or the experimental arm—again, even though there’s a nice 4-month progression-free survival in the final analysis. When you look at those Kaplan-Meier curves, that really adds a lot of perspective and a lot of meaning, in addition to just looking at that PFS and hazard ratio when you start seeing when those progression-free survival curves separate, for example. And also, when they separate and when they come together, or if they don’t come together, I think that’s very important and really adds a lot of value beyond just the bottom line of the median progression-free survival and overall response rate.

And, importantly, the other thing when you think about looking at this is when you look at the median progression-free survival, you’re only looking at one very specific time point. So, I think it’s important when you look at the Kaplan-Meier curves and you look at the hazard ratio, you’re looking globally across the benefit, across the entire duration of the trial or while patients are on study. I think that adds a lot more perspective other than just looking at one very specific time point such as the median PFS, which is important. But, I think looking at the global picture is very helpful for our patients, especially as we start looking at some of these new combinations.

The other the important implication of when you look at median progression-free survival and overall response rates, is the type of relapse. As you start applying this, it’s very helpful when you look, for example, at the combination of elotuzumab/lenalidomide/dexamethasone, you’ll see a very nice response rate, as well as an improvement in the median progression-free survival. But, importantly, what you see is also good long-term disease control. And, why is that important here, especially in somebody, for example, with a more indolent relapse? Importantly, because some of these drugs, like our immuno-stimulatory component of elotuzumab, require time. This is not an active cytotoxic therapy that’s going to lead to disease control within the first 2 weeks, for example. And so, I think that built on the backbone of lenalidomide/dexamethasone, which provides that initial disease control as well— but the nice thing that we see with elotuzumab, for example, because of the immuno-stimulatory component, is that you get to see that long-term disease control—which is the exciting option for patients. It’s not only that it’s early disease control, but long-term disease control as well, and that takes time to train the immune system, to activate the immune system. For those patients with an early relapse, an indolent relapse, or a biochemical relapse, I think that’s an excellent option for patients really to not only get you that disease control early, but late as well. So, that’s an important option for patients when you think about which combinations to use in which patients.

Transcript Edited for Clarity