The study is a phase III randomised trial for patients with locally advanced breast cancer who are receiving radiation therapy to chest wall, breast, with inclusion of the internal mammary lymph nodes. The primary endpoint is major cardiac events, which has not been reached statistically yet. Other endpoints include local regional control. Earlier today I presented on health-related quality of life outcomes, which is our earliest endpoint.
What was the study design?
The study design, it was powered for the endpoint of major cardiovascular events which, again, is the endpoint that we consider most important. The physicians and patient advocates all agree this would be the most important primary endpoint and for that we needed to accrue 1,238 patients which we successfully did, which is something that’s quite an undertaking and an accomplishment on its own. That’s a huge number of patients. Unlike prostate cancer, breast cancer was treated with protons less frequently. The first patient with locally advanced breast cancer was only treated in 2010, for prostate cancer it was in the 1970s. There was a study by Darby that came out showing that any dose of radiation with no threshold to the heart, in terms of mean dose to the heart, could lead to a cardiac event later in life. So these things coming together allowed us to accrue women and randomise them to photon radiation or regular radiation or proton radiation.
What were the results of this study?
The results I presented today were on health-related quality of life. The study was actually over-powered for this endpoint because it was powered for the primary endpoint. We looked at several health-related quality of life metrics, patient-reported outcomes and largely these were the same for both groups, photons and protons. There were a few differences and they were all in favour of protons but, for the most part, the outcomes were the same.
What is the clinical significance of these results?
We know that patient-reported outcomes are extremely important – our patients can tell us things without the bias that physicians might have. So the things that we learned from this are that mainly patients were pleased with both treatments and they did well with both treatments. There were a few factors, there were five facet [??] factors which included willingness to treat or recommend treatment again, and these favoured protons. These factors were probably a little bit more influenced by bias, or could have been. The patients obviously couldn’t be blinded to what type of treatment they were receiving so we do have to interpret it in the context of an unblinded study but they favoured protons. Then there was another finding that you’ll see in the abstract where shortness of breath zero, meaning no shortness of breath, versus any, significantly favoured proton therapy. However, the group decided after looking at the results that it would be best to do something called a multiplicity analysis to ensure that this was not a false positive. With that you assign a stricter p-value, and doing the multiplicity analysis made the shortness of breath outcome non-significant. So while that’s the most appropriate thing to do, it doesn’t mean that that finding can’t be hypothesis forming or investigated further but we wouldn’t state that that was a finding that we feel is a true positive and of clinical significance or of statistical significance with the multiplicity correction.
What is next for this study?
Again, exciting that patients did so well. There’s a lot of retrospective series on photons and protons and this puts to rest some of the concerns of those studies. The next steps are to look at local regional control and major cardiovascular events which will be the main primary endpoints; the major cardiac event is the primary endpoint. So looking at those endpoints will take probably 3-5 years and in the meantime we plan to do a lot of other studies with all the data that we’ve collected from these patients who were willing to enrol on our study and give us so much information. So there’s a huge database, 1,239 patients, there are many investigators interested in doing additional studies which will be performed in the interim.
