SOHO Italy 2025 | Optimizing reporting of infection rates in clinical trials of bispecific antibodies in myeloma

Yeah, so, you know, we have a lot of therapies in myeloma, both approved and in clinical trial, and a lot of them have similar targets. Right now, there are four FDA-approved BCMA-targeted therapies. We’re going to have likely six BCMA-targeted entities before the end of the year, many more in clinical trial, and, you know, they all have similar toxicity rates. But we’re trying to tease out which ones may be better or worse in terms of efficacy and toxicity. 

And infection has been the main adverse event we’ve been looking at. The problem is, depending upon how you enumerate the infection risk and rate, when you’re looking at different clinical trials, you can get very, very different numbers, and it can be very hard to tease out what they mean. For example, if you look at the overall incidence across the bispecifics, at our first glance, we saw that linvoseltamab had one of the lowest rates of infection. However, when a subsequent paper came out, it looked at infection rates over time, and they focused on the infection rate per unit time on the drug. That’s what they focused on. Linvoseltamab actually had the highest rate. Now, the reason for this was that in the data sets that were used, the linvoseltamab group had the shortest follow-up, less than around 11 months, whereas some of the others were two, three months, et cetera. And because the majority of the infections that we see are early on and people that are able to stay on drug longer are generally the fitter patients, better responding, better tolerating overall, if you look at just that initial time, you’re going to have a skewing of infection rate. If you just look within the first, you know, six months, you really need to kind of look at it per unit time per dose. 

And the other thing is, do we really need to get more granular and not just say infections per patient per 100 days or per unit time, but grade because having two or three grade one infections is actually far preferable than one grade three or four infection. 

So in reality, what we’re really trying to do with this manuscript is explore a number of potential statistical ways and functional ways to kind of look at infection rate to figure out which therapy may be best for which individual. And we’re not there yet because there’s a lot of ways to look at it. It’s like the classic analogy of an elephant. You know, you have someone feeling the side of an elephant. It feels like a wall, the tusk feels like a spear, the tail feels like a snake, but you need to be able to take a step back and realize that there’s a much bigger picture and many parts to it. So this is really our attempt at a first step of saying, you know, we have one study that says this drug is the lowest rate. The next study says it is the highest rate. Both of these approaches are probably incorrect. Let’s try to figure out a harmonization to approach all of the agents so we can compare apples to apples.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.