番茄社区

Chromosome 1 abnormalities (C1As) are among the most common recurrent chromosomal aberrations in multiple myeloma, and a post-hoc analysis of the E1A11 trial assessed the impact of C1As on newly diagnosed multiple myeloma patients treated with a proteasome inhibitor, an immunomodulatory drug, and dexamethasone. were presented at the recent American Society of Hematology (ASH) virtual meeting.

番茄社区 has brought together three expert leaders in their field: moderator , is joined by , and , for a virtual roundtable discussion; and this first of four exclusive episodes focuses on the background and results of this subgroup analysis.

Following is a transcript of their remarks:

Prasad: All right, I'm back. I'm Dr. Vinay Prasad here at University of California, San Francisco. I'm here for a roundtable discussion with two experts in multiple myeloma. I'm joined by Aaron Goodman, he's known as "Papa Heme" -- he's at the University of California, San Diego. And I'm also joined by Dr. Manni Mohyuddin, he's an assistant professor at the Huntsman Cancer Institute in Utah, and a well-known expert in multiple myeloma. Gentlemen, thank you so much for joining me to talk through these abstracts.

Mohyuddin: Thanks.

Goodman: Thanks for having us.

Prasad: Let's start with the . This is a provocative abstract that was presented at ASH. This is the "Impact of Chromosome 1 Abnormalities Among Patients with Newly Diagnosed Multiple Myeloma." It's my favorite kind of analysis, a subgroup analysis from the Endurance ECOG-ACRIN E1A11 trial.

Manni, I wonder if you might start by giving us some historical perspective about high-risk cytogenetics multiple myeloma. When should we be thinking beyond VRd [bortezomib (Velcade), lenalidomide (Revlimid), and dexamethasone] -- what's the history in this space?

Mohyuddin: Totally. So, as you all know VRd has been the historical standard of care for treatment of multiple myeloma. It still is a vitally recognized standard of care. Our definition of high-risk cytogenetics and in high-risk myeloma in general has been evolving. There are some classically well-recognized, high-risk cytogenetic abnormalities, such as deletion 17p, translocation (4;14), translocation (14;16), (14;20).

You know, gain1q has been a little controversial. It's not formally recognized as part of our R-ISS [Revised International Staging System] high-risk staging. But previous data seems to suggest that it was a big factor when it was associated with other abnormalities. But I think recently based on some emerging data, the consensus of the field has been moving towards, yes, gain1q even without other coexisting cytogenetic abnormalities, is a high-risk feature.

The other relevant question, a big controversial debate in myeloma, is KRd [carfilzomib (Kyprolis), lenalidomide, and dexamethasone] versus VRd. And I think that it's a topic that divides a lot of us. And just a historical perspective, KRd based on non-randomized studies seems to have done better than VRd, if you compare it historically to how people have done with VRd. So lots of people before this trial was run, and especially for high-risk patients, they were like, KRd is better than VRd.

The reason why this trial excluded patients with most high-risk cytogenetics was that there was a parallel trial running that was Elo [elotuzumab (Empliciti)]-VRd versus VRd. That trial ended up being negative; that trial recruited patients exclusively with high-risk abnormalities. In this trial, the only cytogenetic high-risk abnormalities that were recruited were translocation (4;14), which is a well-known high-risk feature. And then gain1q abnormalities.

So everybody was expecting, for myeloma as a whole, everybody was expecting KRd to be better. So I think people have had a little bit of a tough time accepting that in this trial, KRd was not better than VRd. And I think people still hold onto this, that high-risk data genetics were excluded. And so we don't know for high-risk cytogenetics that KRd is better than VRd.

And I'd like to sort of think about the biological possibility of the hypothesis. And I think as we're going to discuss today, these subgroup analyses are very interesting. And there's a lot of questions about biological possibility, statistical power, and just holding onto our preconceived notions.

Prasad: That's a great summary for where the field is at, and the only thing I love more than post-hoc subgroup analyses are historical controlled studies in multiple myeloma. So let the record state that. And when you're talking about 1q gain, three copies as gain and four copies or more as amplification.

Mohyuddin: That is true.

Prasad: Why don't you give us a brief summary of what this trial showed? And then I want to jump to Aaron for his response to what he thinks. Not this trial, what this post-hoc subgroup analysis of one chromosome 1 abnormality shows.

Mohyuddin: Yeah, totally. So I will jump into that, but before I do, one very important fact which I want to make sure everybody sort of captures is that (4;14), which is high-risk, was included in this trial. And (4;14) based on the original manuscript that was published, there was no statistical difference. If you look at the forest plot, the outcomes were slightly better with VRd, not statistically significant. So that is why we're not seeing a (4;14) subset analysis, I'd argue, because with (4;14), KRd was no better.

So with this particular subgroup analysis, they are looking at 1q abnormality specifically. And, one of the first things they're doing is they're comparing 1q abnormalities overall versus patients without 1q abnormalities, regardless of whether they got VRd or KRd. And the data would seem to suggest that PFS [progression-free survival] was inferior for patients who had 1q abnormalities. And there was a trend towards worse OS [overall survival], but that was not statistically significant.

Then they sort of go further into the types of chromosome 1 abnormalities. So they look at deletion 1p, and then they look at gain1q versus Amp1q. And they showed that for gain1q, three copies, there was a benefit for KRd compared to VRd. But if you look for Amp1q, the overall survival trended to be better with VRd compared to KRd. So there's really in my mind, there's no way to biologically explain that. And when I run into that situation, I'm like, are we seeing statistical noise?

But I'd love to hear what Aaron has to say about this.

Prasad: Aaron, what do you think, does the story make sense to you? What do you think about chromosome 1 abnormalities?

Goodman: No. So, it doesn't make sense. I want to just start with one little story. Like a year and a half ago, right when I was getting on Twitter, a KOL [key opinion leader] in myeloma posted that they used KRd for induction not if there's three, but if there's four of these 1qs. Now we go to this study that actually tried to look at it post-hoc analysis. And as you said, with three -- three, not four -- they found that PFS was no better with KRd, but overall survival was better with KRd.

Prasad: And what does that tell you? Is that classic treatment effect?

Goodman: It doesn't seem like it, yeah ...

Prasad: Seems the opposite. It seems like it's selection bias.

Goodman: And then when they looked at four, there was no difference. But actually the hazard ratio started to actually favor RVd for four. So the bottom line is, I mean this data, they were looking for something and that abnormality does seem to do worse in general, I agree. But I don't think there's anything ... You know, I'm a clinician, and when I look at this, it was interesting, but there's nothing that I can use to make any sort of treatment decision. And I don't think anyone should be looking at this and go, well my patient has three I'm gonna use KRd, they have four, I'm gonna use RVd. I just don't think that's applicable to the clinic.

Mohyuddin: And I do think that the presenter, in his defense, was very balanced. And I don't think that they were trying to sell this agenda. I think they tried to present the data as is, and the data is messy. I think all of us can conclude that.

So I think this sort of reinforces people's notions they came in with. If I want to prescribe VRd, I will look at this and be like, hey there's no convincing evidence that KRd is better. And I look at (4;14), another high-risk feature, and from this study, KRd was no better than VRd for (4;14). And I think people who sort of have an ideological inclination towards using KRd would look at this data and find some reinforcement.

Goodman: So this doesn't even matter anymore, right? We got quads now for induction. So this...

Prasad: We're coming to that! That's the next one.

Well I guess closing thoughts on this topic, I would say this: If you believe something in medicine that a newer drug is superior to an older established standard-of-care, I think you need to prove to me under what circumstances that is true. That means you pick whatever cohort you want, you run a randomized trial, you show me the newer drug is better.

That never happened with KRd. We jumped on the bandwagon in the absence of that information. When we finally did run a randomized controlled trial, it surprised many of us. Because the older established standard-of-care VRd was just as good, if not better, because it has a better safety profile and it's been around longer and it's certainly more affordable.

Now we do the post-hoc subgroup analyses. These analyses are only, at best, hypothesis-generating. They're not good enough to hang your hat on. What you find here is something that as you both point out, doesn't make a lot of biological sense. There's not the concordance between PFS and OS. It doesn't make sense, three copies, four copies.

And for those reasons, I would take it with a huge grain of salt. I wouldn't change my practice based on these results. And I'm sticking to, you can mark it down for right now, I'm sticking to VRd until proven otherwise.

So thank you, gentlemen.

Watch episode 2: Any Role for Maintenance Ixazomib in Multiple Myeloma?

Watch episode 3: Isatuximab Plus RVd Induction in Newly Diagnosed Multiple Myeloma

Watch episode 4: Updated MajesTEC-1 Results Continue to Show Teclistamab's Promise

Comment