Transcript

Hello, I'm Dr Hope Rugo from the University of California at San Francisco's Comprehensive Cancer Center, and I'm going to review for you today a few different abstracts that were presented as posters at this year's virtual ASCO 2020.

At this meeting, I presented data based on the BYLieve trial, which is an ongoing trial that looks at patients who had their immediate prior treatment in the cohort A that we presented, had to be an aromatase inhibitor and CDK4/6 inhibitor.

They had to have a PIK3CA mutation in their tumor, and then they were treated in a non-comparative, single-arm trial with fulvestrant and the alpha-specific PI3 kinase inhibitor, alpelisib.

There are other cohorts in BYLieve that have looked at patients treated with fulvestrant and a CDK4/6 inhibitor. That cohort received an aromatase inhibitor and alpelisib as their treatment in cohort B. That trial has not yet reached their 6-month landmark, so we're waiting for that.

Then a third trial where people received chemotherapy and then went onto alpelisib and fulvestrant. That part of the trial, cohort C, is still accruing. The primary end point was progression-free survival at 6 months.

We showed that that met the end point with a way greater than a lower boundary of the 95% confidence interval of 30% at a little over 50%, and that the adverse events appeared to be even more favorably managed and less than we saw in SOLAR-1, indicating that experience makes a difference as well.

What comes up to us, I think, really looking at the BYLieve data, which we also compared to a real-world evidence population, is how does that real-world evidence population do?

In BYLieve, we used the Flatiron database; they have a genomics database. All of the patients in the real-world comparison had known PIK3CA mutations and had received an aromatase inhibitor and a CDK4/6 inhibitor, and were treated with a variety of agents, including single-agent chemotherapy, hormone therapy, even hormone combinations in AI and fulvestrant.

What we showed was, in a variety of different matching paradigms, was that there was still a doubling of progression-free survival with the alpelisib and fulvestrant versus the real-world evidence population of patients.

What was interesting in that real-world evidence analysis from the Flatiron database, which is in the oral presentation of BYLieve available online at the ASCO virtual meeting, is that we saw that the PFS in the patients, regardless of what they got, in the patients who were that the Flatiron database, which is a little under 100 patients, it was actually short. It was in the 3- to 4-month range, a little under 4 months. Really short.

This particular poster, which I think is interesting, looked at PIK3CA mutation status and progression-free survival in patients with metastatic, hormone receptor–positive, HER2-negative breast cancer in clinical trials that have been published.

They did a meta-analysis, and I'm actually the last author on this particular paper. What was done with my colleagues at Novartis and colleagues from the SOLAR-1 steering committee, was just to look at a systematic literature review from articles published from January of 1993 to April of 2019.

Look at all of the trials in a standard, very strict, meta-analysis type approach. What we saw, I won't go through it in detail (I think the poster's quite interesting and worth a quick look) and there's also, of course, a QR code that allows you to download the poster.

What we saw was that, in patients who had hormone receptor–positive, HER2-negative, metastatic breast cancer who received therapies that didn't target PI3 kinase... Of course, as you can imagine, that's the majority of trials. What happened was the trials were done, and then people looked at biomarkers to see how that affected outcomes. So that's where this data is all coming from.

A mutation of PIK3CA almost universally was a negative prognostic factor. It was associated with a significantly shorter progression-free survival.

This has been really reported in many different prior reports, and having a meta-analysis that suggests this is helpful.

There are a couple of studies that didn't suggest it, that PIK3CA didn't really impact outcome, but we've seen that that may be related to the way the mutational analysis was conducted and the tissue samples that were used, since we know that you can acquire PIK3CA mutations over time over the conversion from early to late-stage disease, and as your tumor progresses in the metastatic setting.

I think that this meta-analysis, along with the data that we've seen before, highlights the importance of finding agents that are reasonably well-tolerated and target PIK3CA-mutant tumors effectively.

We're very lucky now to have the alpha-specific PI3K inhibitor alpelisib approved now both in the United States and Europe, albeit in a limited population in Europe, and are learning ways to use this agent with the least amount of toxicity to our patients, maintaining both dose intensity and efficacy.

There's also a large amount of studies going on looking at other agents that target both PIK3CA (we'll see where that goes) but also AKT along this pathway, and combinations as well, which have been really stymied by toxicity.

But I think AKT inhibitors are of great interest, and there are several phase 3 trials looking at that as well. They are analyzing the subset of patients with PIK3CA mutations.