I think that fact that we don’t have any major abstracts on quality of life is really one of our deficiencies in prostate cancer research because we perform quality of life measures frequently in our trials, but these are almost always secondary or exploratory analyses. They tend to get reported on later, in lower level publications, and they tend to be kind of blunt instruments that don’t really tell us much new. But when we really come down to it, there are really only 2 things that matter in this population: survival and quality of life. We could argue that, for a lot of our patients, it’s more about quality of life than survival. When I think about when I have conversations with people, it’s really what we’re focusing on: how can we keep this patient living as well as we can for as long as possible? But it’s really that context: it’s not about living as long as possible at whatever cost. Understanding this, measuring this, and reporting on it is really critical, and I think it should factor into all of our decisions regarding combination therapies, timing of therapies, sequencing of therapies, and all of that. It’s one of the areas we’re still struggling with a little bit. With some of the tools becoming more widely available, with more patients being plugged into the Internet, I think there will be opportunities for us to do this better and on a longitudinal basis. We’re launching a big international registry to really look at patients with advanced metastatic hormone-sensitive prostate cancer (mHSPC) and metastatic castration-resistant prostate cancer (mCRPC), and one of the key goals of this longitudinal collection of patient-reported outcomes is really to understand the patient experience. I think this is going to provide critical insight for the field. I don’t know that it’s going to answer all of the questions, but I think it’s going to frame them for future studies.

That’s a really good point, Dr George. When we look at our data today with regard to moving the goal post, particularly for these 5 new therapies, it was all about overall survival. We could certainly have a very positive benefit on a patient without prolonging his survival, and we are very poor in measuring that. The patients understand it, and we understand it as clinicians, but we’re just not very good at measuring it. It’s an area that, despite the fact that a fair amount of work has gone into it, there are still many opportunities for improvement. With the large registry coming up, the International Registry to Improve Outcomes in Men with Advanced Prostate Cancer (IRONMAN), I think we’re going to be able to gather a little more data about the sequences over time and the patient’s perception of how these therapies are affecting their overall health, which is key. But then we also have to really figure out where these goal posts are. Patients do want to live better, and they want to live longer too. Right now, we’ve made a lot of progress, and we have so much more progress to make. I thought that one of the provocative elements here was to ask where are we going to get these next level of gains. It goes back, I think, to some of the biomarkers we were discussing earlier. I do believe that there are going to be segmented areas, molecularly defined, that we’re going to be able to use in a predictive way to choose our therapies more wisely. I also believe that there are a lot of patients for whom we are not going to be able to use those molecularly targeted therapies because we just don’t have the ability to target everything that we can find and diagnose at the genomic level. Taking a high-level view, yes, we’re going to make more progress with therapy targeted to specific molecular defects. We’re also going to require something beyond that to really be able to affect a large number of patients who have defects that we cannot target effectively today.

I agree with that. But stepping back, when a patient initially presents, and we reference the tremendous heterogeneity in this population of patients, I think by using molecular biomarkers we’re going to be able to separate patients out. For instance, in the patients who really have clinically insignificant cancer that doesn’t need a lot of therapy, that alone will improve their quality of life by avoiding toxicity that we sometimes impose on patients who don’t need it. In contrast, we can identify patients who present with what seems to be either lower or intermediate risk, but in fact, based on their molecular biomarkers, they have more aggressive cancer. If we are more aggressive with them, that may allow us to achieve some survival gains. I think that our ability to sort out patients using not just prostate-specific antigen (PSA) initial staging, but more elegant tools, will help us sort out which patients need more aggressive therapy and which patients really should just be left alone.

I think that the biggest gains that we’ve seen in the last few years in prostate cancer is the earlier use of existing therapies, such as chemotherapy in the hormone-sensitive setting. The magnitude of that survival benefit, as everyone knows from STAMPEDE and CHAARTED, was extremely dramatic. It was between 14 and 20 months as a median survival benefit with early use of docetaxel, a drug that many of us have used for castration-resistant prostate cancer (CRPC) but had modest, 2- to 3-month survival benefits when used in the more advanced stage. I think that the major place we’re going to see survival benefits will be the earlier use of some of our existing therapies: androgen target therapies, maybe radium, maybe immunotherapy, moving earlier into the disease course before patients develop more advanced CRPC. Even in the hormone-sensitive setting in CHAARTED Arm G, which is the abiraterone vs androgen-deprivation therapy (ADT)-alone arm, should be reported some time this year, so we may be seeing even abiraterone in a setting of newly diagnosed mHSPC. I think the issue of quality of life and of segmenting the patients is a very important issue to continue to follow. Segmenting patients based on molecular clinical factors should allow us to, as Dr Gejerman mentioned, not treat a substantial proportion of patients or to treat them less aggressively. We can do a lot of harm with our treatments: they’re expensive, they’re sometimes toxic, and if we give a treatment to a patient who is not going to benefit, then we are not doing him a favor in terms of quality of life. But I do think that the flip side of that is that, as we identify these molecular subsets, whether they are DNA repair pathways, novel AR mutations, or other targets that we haven’t really discussed at great length here, there will be drugs that will work for 5, 10, or 15% of those patients. While we have to use drugs that work across the board as much as possible, I think we will see more and more segmenting of prostate cancer, a very large population, into smaller groups and making those incremental gains in those smaller groups.