Alicia Morgans: Hi, my name is Alicia Morgans, and I'm a GU medical oncologist and Associate Professor of Medicine at Northwestern University in Chicago in the United States. I'm so excited to have here with me today, a friend and colleague, Dr. Cora Sternberg, who is a Professor of Medicine and the Clinical Director of the Englander Institute for Precision Medicine at Weill Cornell. Thank you so much for being here with me today, Dr. Sternberg.

Cora Sternberg: Alicia, it's a pleasure for me to be here with you today, even though it's on Zoom. I'd like to talk first about the IPATential150 study that has been presented already at ESMO, but we're talking about more about the biomarkers in this study. This study, we know that the PI3K/AKT pathway is dysregulated when there's functional PTEN loss, and then AKT becomes hyperactivated. And this is one of the mechanisms of resistance to abiraterone, the activation of the PI3K/AKT pathway.

So, what we have is a drug called ipatasertib, which is a selective ATP competitive small molecule that inhibits all three isoforms of AKT. We know that PTEN is functionally lost in some 40% to 50% of patients with metastatic CRPC. And what led up to the Phase III IPATential150 study, was a study in which patients were evaluated in a smaller Phase II study, and they saw that patients with metastatic CRPC who had PTEN loss had improved radiologic progression-free survival when ipatasertib was added to abiraterone.

This was a larger Phase III study. There were 1,100 patients. These were patients with metastatic CRPC, asymptomatic or mildly symptomatic, and they weren't stratified initially for a tumor PTEN loss, but that was a stratification factor, and also, another stratification factor was whether they had had prior docetaxel in the hormone-sensitive prostate cancer setting. But this was basically first-line treatment for metastatic CRPC. Patients were also stratified by whether or not they had liver or lung metastasis.

What was presented already last year at ESMO, was that patients who had an event with the combination arm that occurred in 48%, and it occurred in 59% of patients that had placebo and abiraterone. And that was statistically significant. But if we looked at the whole entire population, not just the PTEN-loss population, there was no statistically significant difference in radiologic progression-free survival.

This study was done looking at PTEN loss by immunohistochemistry, and it was defined as PTEN loss as having at least 50% of the viable cells with no cytoplasmic staining on the VENTANA assay. This was just a definition. What we're looking at in this abstract is, what is the impact of different immunohistochemistry PTEN-loss status cutoffs on treatment efficacy? What is the concordance between PTEN loss by immunohistochemistry and by next-generation sequencing? And how does PTEN-loss status by next-generation sequencing impact the outcomes? We also look at genomic alterations in the PI3K/AKT pathway.

What we saw, clearly, was that with that cutoff of 50%, we had radiologic progression-free survival in favor of the ipatasertib arm, but if you looked at 60% or 70% or 80 or 90%, 100%, the results were better and better and better and more in favor of the ipatasertib and abiraterone arm. Patients who had less than 50% PTEN loss didn't have any benefit and had actually done just as well with abiraterone alone.

Then we looked at the next-generation sequencing, which we were able to do in approximately 70% of the cases. There we found that 40% had PTEN loss and 60% were the wild type. It's always thought that the PTEN loss occurs in about 50% of patients with CRPC, but in this case, we found the 40% had PTEN loss. We looked at the immunohistochemistry and the next-generation sequencing, and overall, there was an agreement in about 86%. In patients who have PTEN loss by next-generation sequencing, 91% actually had the loss by immunohistochemistry. But if we looked at the ones with the immunohistochemistry PTEN loss, 77% had PTEN loss by next-generation sequencing. So, it's not exactly equal and there may be other mechanisms for PTEN loss.

If we looked at the improvement in radiologic progression-free survival in patients with PTEN loss, it really is dramatically different. Patients with ipatasertib and abiraterone had a median radiologic progression-free survival of 19 months, versus 14 months with placebo and abiraterone. And this was statistically significant with a hazard ratio of 0.65. If we looked at patients who had radiologic progression-free survival and looked at the PIK3CA/AKT PTEN alterations by NGS, it becomes even wider. The curves and the hazard ratio here is 0.63.

We looked at PTEN loss and prognosis in the patients in the placebo and abiraterone arm, and you can see that those patients generally had a poor prognosis, patients with PI3K/AKT loss, whether we looked by immunohistochemistry or by NGS status, especially the ones with PIK3CA/AKT PTEN loss. There was a difference in patients in the placebo arm with a poor prognosis.

This abstract basically shows that PTEN status in metastatic CRPC by immunohistochemistry and next-generation sequencing demonstrated fairly good concordance. A proportion of the advanced prostate cancers without PTEN loss have other PI3K/AKT pathway alterations, and these alterations are associated with poor prognosis. We saw consistent radiologic progression-free survival benefit in patients at different cutoffs, and the 50% cutoff was huge, but if we lose 60% or 70% or 80% cutoff, we had even more dramatic results in favor of the combination. The improved radiologic progression-free survival with the combination was also demonstrated in patients with NGS, with the PIK3/AKT PTEN alterations. So, I think these data support the use of ipatasertib and abiraterone as a treatment for first-line metastatic CRPC in patients with PI3K/AKT PTEN pathway alterations. Thank you very much.

Alicia Morgans: Thank you for going through that with us, Cora. I do have a quick question about that though. If you were making the clinical decision on a patient who might have one of these alterations, what cutoff would you want to see to try to choose those patients who might benefit most? Would you use the 50% cutoff? Would you aim for something a little bit higher? How would you make that choice?

Cora Sternberg: That's a great question. The 50% cutoff was chosen arbitrarily and I don't know if that is really the best cutoff. I personally think that, if using next-generation sequencing, that we've seen much more dramatic differences in the two arms when we use that. So, I don't know if that is feasible in the future. I don't know whether or not this drug will be approved with this indication, and if it will be approved, it'll depend upon the eventual overall survival benefit, or lack thereof, probably, whichever happens in the future. And I don't know if it can be approved within an NGS. NGS was done with Foundation Medicine. I don't know if that's readily available for everyone.

This was a clinical trial, so I think what it does show, is it's one of the first molecularly-oriented clinical trials in trying to look at a molecular marker and to target that. So I think it's very interesting from that point of view. I don't know what's actually going to happen with this in the future and what cutoff will be used or whether it can be said to use NGS as a cutoff.

Alicia Morgans: I think that's a great point though, and I think there are definitely ongoing studies. It will be very interesting to see how studies that are ongoing or studies that are in development choose that cutoff, and whether they use IHC or whether they use NGS, as so many of us are using next-generation sequencing now to understand whether our patients have DNA repair defect mutations, like other studies that you've been able to participate in that are really these personalized, targeted-type approaches. So, doing as few assays as possible, maybe really practical, ultimately, in our clinics.

But as you said, there may be losses that are not PTEN loss on NGS, but other ways that the pathway could be inhibited that we don't have the activity of PTEN, even though it is itself still intact if we do NGS. So, lots to learn and I really appreciate you putting this together. If you had to give a summary or an overarching statement to the listeners on this particular biomarker analysis and on the IPATential150 study in general, what would that be?

Cora Sternberg: I think we need to say that analysis of the more stringent biomarkers associated with activation of this PI3K/AKT pathway further supports the use of ipatasertib and abiraterone in the treatment of metastatic CRPC in those patients who have the PI3K/AKT pathway alterations. This is a subtype with a very poor prognosis, so I think that this is one of the first biomarker-oriented studies. Next-generation sequencing will become less and less expensive and has become less and less expensive. In the future, I don't know how broadly available it will be for the rest of the world. In the United States, I think it could become something that would be approved and accepted, but we'll have to see what happens.

Alicia Morgans: Great. Well, thank you so much for your time and for sharing your expertise with us today.

Cora Sternberg: Thank you.