CLL: Prognostic Markers

William Wierda, MD PhD: There was a prognostic model for ibrutinib-based therapy that was reported by the NCI [National Cancer Institute] group. I wonder if 1 of us can comment on that model and the 4 factors. It’s a 4-factor model looking at outcomes, both progression-free and overall survival, for patients treated with ibrutinib. Maybe someone can comment on that model, what the factors are, and whether we can extrapolate and apply it to other covalent irreversible inhibitors of BTK and expect similar correlations. Is anybody familiar with that model?

John Allan, MD: I can speak to it a little. You may have to correct me on all the models. I know TP53 disruption, deletion, or mutation was 1. I believe unmutated IGHV is another. I want to say age was 1. The fourth could have been relapse disease. In that paper, they actually took that factor out because in frontline treatment, it doesn’t matter if you’ve been treated or not. Ultimately, whether or not you’re treated out of it, those 3 factors I previously spoke to were still prognostic and predictive for an outcome on a frontline therapy with ibrutinib.

These scoring systems are useful and helpful, but they’re somewhat intuitive, because they identify people that we already know are high-risk with TP53 disruption. In the CLL-IPI [Chronic Lymphocytic Leukemia International Prognostic Index], you automatically become high risk and get 4 points just by having a TP53 disruption. I don’t believe this is weighted, but many times TP53 disruption is associated with an unmutated IGHV. Therefore, all of a sudden, these things stack together and help prognosticate. I find them useful in terms of time to first treatment, because while it predicts for some outcome after therapy, the CLL-IPI and the NCI scoring system can predict the time to first treatment as well, which is very important to calculate for your patient and then provide some discussion on what the future may hold in the next 2 or 3 years to prepare them for this eventual reality that will come to fruition.

That’s how I like to use these. We need to study them in these different approaches. Now that we’re talking about combinations and things, every scoring system that we do with a single type of treatment has to be almost reinvented, because the biology and outcomes may change. These scoring symptoms are nice and useful things we can calculate. But in my clinical practice, I’m not using them or putting them in my notes because, for the most part, they’re intuitive. They identify that patient with high-risk deletion 17p, and you’re treating them the same way just by having that 1 little piece of information. That’s how I’m using it. I don’t know if people are different, though.

Jennifer Woyach, MD: I do think that the NHLBI [National Heart, Lung, and Blood Institute] [model] is typically for people who are going on treatment. That 1 didn’t focus so much on time to treatment. It was TP53, beta-2 microglobulin, LDH [lactate dehydrogenase], and whether they had been previously treated.

William Wierda, MD PhD: I find them useful because it gives me a gauge, but they’re very intuitive, as you said. We know 17p and mutated TP53 is a high-risk group.

TRANSCRIPT EDITED FOR CLARITY