You’ve been looking at advanced breast cancer and specifically using everolimus as part of the therapy. Could you explain what you were trying to do in the BOLERO-1/TRIO-019 study?
Absolutely. This is phase III double blind randomised study that enrolled 719 patients and we were aiming to look at whether or not everolimus improves the progression free survival in patients treated with trastuzumab and paclitaxel. It was a 2:1 randomisation, everyone received trastuzumab and paclitaxel and two-thirds received everolimus, one third placebo. This was the first line metastatic setting, HER2 positive breast cancer.
And the theory behind using everolimus was what?
We know that not all disease responds to trastuzumab. There are patients whose disease will be resistant up front and patients whose disease will acquire resistance. One of the pathways of resistance that develops may be the PI3 kinase/ATK/mTOR pathway which can be up-regulated via a variety of different mechanisms. Everolimus inhibits the endpoint of that pathway, mTOR, and has been shown pre-clinically and in in vitro and in vivo models, as well as in phase I and II studies, to be active in HER2 positive breast cancer.
So a complementary activity to paclitaxel and trastuzumab?
Certainly to trastuzumab, it may help block that escape mechanism that cells develop to trastuzumab.
So what did you do?
We randomised patients to receive paclitaxel trastuzumab with or without everolimus, two-thirds received everolimus in this clinical trial. There were two primary endpoints of this study: first, progression free survival by investigator assessment in the overall population but then we were very interested in looking at the benefit, or lack thereof, in terms of progression free survival in the hormone receptor negative subpopulation.
And you didn’t give endocrine therapy, did you?
We did not give endocrine therapy to patients who had hormone receptor positive disease; there were a variety of reasons for this. First of all, some studies have been done that have shown relatively low response rates and poor progression free survival when AIs have been used with trastuzumab. Secondly, there was one study, it’s rather an old study, but there was one study that was done that seemed to indicate in the adjuvant setting that when you add endocrine therapy to chemotherapy you may diminish the benefit of the chemotherapy or endocrine therapy. So that wasn’t done in this clinical trial.
So you were looking for progression free survival, what did you find?
In the overall patient population there was no difference between patients receiving everolimus or not. The progression free survival was 14.95 months in patients treated with everolimus, 14.49 months in placebo treated patients. So that was not statistically significantly different. The hormone receptor negative population we saw an interesting finding. There was a 7 month improvement in progression free survival, however, it was not statistically significant according to our pre-specified p-value. We set a real high bar for this study. When you have two endpoints that are both primary you have to split your p-value and we heavily weighted it towards the full population. So we had to show a p-value of 0.0044 in order to be statistically significant and what we saw was a p-value of 0.0049 associated with that 0.66 hazard ratio.
I’m not sure what to make of that but very interesting.
But the overall result was neutral it seems, but what about toxicity?
Toxicity, in terms of adverse events we saw similar adverse events that were seen in a similar trial, BOLERO-3, which also combined chemotherapy with trastuzumab and everolimus. However, we did see more patients who had on treatment adverse event related deaths, so 3.6% of patients in the everolimus arm versus none in the other arm. This was concerning but, as we looked at the data, tended to occur in the first 15 months of the study, tended to occur at sites that had less experience with everolimus and managing its toxicities. A letter did go out that educated investigators about the importance of proactively managing AEs related to everolimus and subsequent to that letter there was only one more on treatment AE related death. So it’s very important that when you’re giving everolimus with chemotherapy that the management of the adverse events be taken quite seriously.
Now, there are a couple of things to ask, then. One might be should there be another study in hormone receptor negative patients but the other is what are the overall lessons coming out of this for doctors?
Certainly. I think that there should be studies in the hormone receptor positive subpopulation and, in fact, there are at least two studies in the US looking at PI3 kinase pathway inhibition in combination with trastuzumab and endocrine therapy. Because remember none of our patients received endocrine therapy and that would be a natural escape pathway for those cells which may be why we did not see so much benefit with everolimus. In the hormone receptor negative subpopulation I’m personally quite intrigued and do think that a prospective study to evaluate this in a powered study that didn’t have the sort of splitting of the p-values that we had would be useful.
Do you think there are any clinical messages right now for doctors?
At this point we have not changed the standard of care for first line metastatic HER2 positive breast cancer. The lessons are primarily hypothesis provoking at this point and future studies need to be done before practices change.
So a brief message for doctors to take home would be what?
When treating HER2 positive breast cancer we have multiple treatment options available right now. Currently the use of everolimus in the first line setting is not supported by the data.