Personalised, genomic-guided combination therapy could transform early disease management in ovarian and breast cancer

Prof Brian Leyland-Jones - Avera McKennan Cancer Institute, Sioux-Falls, USA

Medicine has transformed over the last ten years and if we’re looking at where breast cancer was, particularly in 2005 versus where it is today, we have identified something like forty drivers of the disease split across the luminals, the ER positive and the HER2s. So what most people are doing at the moment are detecting one or two aberrations on a particular platform like next generation sequencing and then matching a drug to that under a clinical trial. They’re usually doing this later in the disease, on second or third relapse.

What we are doing, the Avera team, has three characteristics. Number one is that we’re using several platforms, so we use a bare minimum of next generation sequencing and proteomics; we often use RNA-Seq as well and we’re building in now a NanoString platform. The second thing that we try to do in every patient is close down all of the oncogenic drivers. So we’re putting together combinations of drugs individually in each patient to try and kill the tumour by blocking all of the structural alterations. And then the third thing we’re doing is treating very early in the disease course, that every patient stands a better chance of cure if treated early. Also the genomic complexity is less at an early stage so we feel that we can make the most impact by going early, using multiplatforms to try and dissect all of the genomic drivers and then matching those with combinations of targeted therapies.

What were some of the main points of your presentation?

I showed in the presentation how the breast cancer has changed, how it has gone from just a couple of drivers in 2006 into forty drivers. I showed how those forty drivers have potential matches in terms of therapy; I showed the same for lung cancer. Then we presented the results; we’ve sequenced about 300-400 patients but we presented the results of our early ovarian and breast cancer data. So by using matched combinations against multiplatform analysis in breast on our fourth line therapy we’re achieving something like a 40% RECIST response rate. In an ovarian cancer we’re doing even better, we have a 70% objective response rate in second and third line therapy.

Could you elaborate more the breast/ovarian cancer studies you mentioned?

So with this approach using just next generation sequencing and using proteomics to discern the driving networks using that combination, we’ve put together combinations of targeted therapies, very often against PI3 kinase, mTOR, FGFR etc. At this moment in time our breast metastatic data average fourth line of treatment we are achieving a 45% RECIST rate. Our ovarian data are even better – second and third line metastatic serous ovarian cancer, it’s a very serious disease, progresses very rapidly, we have a 70% objective response rate with 20% of those being complete responses.

What are some of the combinations that have been used to date?

So it is the usual culprits. So there’s a lot of PI3 kinase, AKT, mTOR, phosphor S6, a lot of RAS-RAF-MEK-ERK, a lot of cell cycle control which implies treatment with either palbociclib or LEE001. There are a lot of the upstream receptors, so FGFR, HER2, EGFR; depending upon the disease type, say ovarian cancer and basal breast cancer, a lot of DNA repair but many of the common culprits. Then the other thing which we’re seeing is what everyone refers to as the long tail in the curve. So we picked up a RET fusion in breast cancer, an HGF amplification, but these have to be taken very seriously because they are very often driving alterations in the disease. So we’re putting together usually a combination of two of these, two of the targeted therapies. Very often it will be one against the structural alteration, one against the driver mutation and we’re adding that on to state of the art ordinary everyday therapy.

Once an effective combination is found, do you then need to do a phase III trial?

At the moment we’ve done all this under protocol but an early stage in the disease so we’ve worked out a lot of the dosing that’s required. We’ve also worked out some of the efficacy from using different patterns of drugs, now we’re about to enter into our two big neoadjuvant studies where we take all three principles early – multiplatform, going early, combination therapies to tackle all the drivers.