4th EUTROC, Liverpool, UK
Highlights from the fourth EUTROC meeting
Professor John Green – University of Liverpool, UK
This was the fourth of these meetings that the network which I’m part of has run. We did an interesting smaller one just for half a day in Valencia last year but this was a 1½ day meeting. There were two main themes to this meeting, one was the question of molecular definition of ovarian cancer - the fact that it’s not one disease but broken down into the different subtypes characterised primarily by a mutational profiling on to which there is correlation between the pathological assessment of the different subtypes and the molecular basis. We’ve identified high grade serous cancers as the most common type and possibly the most aggressive; low grade serous tumours, mucinous tumours, endometrioid tumours and mixed tumours. The second theme was about heterogeneity and by that I mean largely the heterogeneity within individual tumours which we’ve recognised in broad terms you can get on histopathological assessment but with multiple sampling of the same tumour, either within the same primary or between multiple metastatic deposits. We’ve shown that there is considerable heterogeneity in the mutational profile and also, presumably, therefore in the function of the drivers in the different parts of the tumour and that, of course, makes it very difficult to have a new classification of these tumours and also to try and determine the appropriate targeted therapies that we are developing now for this disease.
Could you discuss the logistics of biobanking in clinical trials?
Biobanking issues are with us now, they are a major headache, there are difficulties with organising international studies. Because of the fact of these different sub-types that break down each of the different types of ovarian cancer into what are actually quite rare tumours so therefore we need international collaboration in order to run the trials, to get sufficient power to interpret them. That means not just sharing patient data but also having an effective biobanking system. Now once the tissue has been extracted and is in the DNA or RNA, there is no problem about transferring it across boundaries but any tissues such as blood or samples taken from tumour biopsies, they are subject to different ethical regulations in the different countries. So there are the logistic aspects of biobanking and the question of the logistics and the handling and the sample preservation and so on. That’s always going to be an issue, it’s always going to be a session at these meetings, but what we’re hoping to achieve, and I think we did, was a much greater understanding of the biology of the disease. We had two international presenters, both of whom gave first class presentations and contributed immensely to the discussions. One was David Huntsman from the Department of Pathology in Vancouver in British Columbia; the other international representative we had was David Bowtell from the Peter MacCallum Institute in Melbourne. He set up a 2,500 patient bank about ten years ago based on different Australian centres; he went on to become Director of the Peter MacCallum Institute and he currently remains one of the most quoted experts on the molecular biology of ovarian cancer and he gave the keynote lecture.
What were some of the on-going trials and new agents discussed at the meeting?
I think there were two areas that were highlighted at the meeting and I’m not trying to say these cover all the possible targets but there’s a lot of focus on the PI3 kinase pathway and particularly on AKT inhibitors. There were some in depth presentations about how we might target these areas and how we should build on biomarkers, functional biomarkers, from imaging to laboratory biomarker analyses to assess the progress of these trials. One of the themes related to that was the question that now, particularly if you’re treating relapsed disease, it is not satisfactory just to refer back to the original biopsy of the cancer which may have been six months or two years before the time at which you are treating relapsed disease. So you need an up to date biopsy, either an open biopsy or a CT-guided biopsy to get tissue so you know the molecular status of the tumour at the time that you introduce your novel targeted therapy.
The other new drug that was highlighted recently in a paper in The Journal of Clinical Oncology coming from the Karolinska Institute is a small molecule that will restore the folded structure of either mutant or wild-type p53. Now the p53 tumour suppressor gene has been used to characterise a range of tumours, it’s present in probably 99-100% of high grade serous cancers, but it’s been very difficult to actually use that as a potential target. But this molecule offers that possibility so we’re looking at a phase II study in conjunction with combination chemotherapy.
In terms of technologies, the two technologies that were highlighted were looking at circulating tumour cells which, according to some of the earlier assays, we could not find or could only find in a very small proportion of ovarian cancer cells. But with refinement of the technology we’ve found that they’re present in up to 25% of ovarian tumours and that responds either immediately after treatment or six months later to very low levels. So it’s showing promise as a marker of response to treatment.
But perhaps even more interesting, though, was a presentation from the Cambridge group about looking at cell-free DNA. This is circulating DNA in the plasma and there was preliminary data showing that whereas the current blood markers of response to anti-cancer treatment, like CA-125, take about three to four weeks to show the beginnings of a response and to be useful as an endpoint, assessing cell-free DNA, looking at markers such as p53 mutations, for example, you can get evidence of a response within two or three days. If you compare that with conventional CT scan imaging which takes often six weeks to two months, that allows us to know whether drug treatments are actually working earlier rather than later and secondly it means that we may be able to stop or close drug trials early if we know that the targeted therapy isn’t working.
What key messages do you want to give to viewers about the meeting?
To think in terms of ovarian cancer as being a range of different diseases and try as best you can to tailor the treatment towards these groups of diagnoses, that is sometimes called stratified medicine. We’re not yet at the personalised medicine stage, which is tailoring it to the specific characteristics of the individual cancer but we are heading in that direction. Secondly, I think being prepared to re-biopsy in cases of relapsed disease. Thirdly, the take home message is that there are lots of possible areas to target these cancers, lots of promising drugs that over the next few years will be… hopefully some of them will actually come into clinical practice.
One more point I would make is going back to the technologies. At the moment we rely on light microscopic diagnosis of ovarian cancer which may help us distinguish between clear cell, endometrioid, high grade serous tumours and that has served us very well for the last thirty or forty years. But I think with the advances and reduction in the cost of novel sequencing technologies, within the next couple of years we’re going to have profiles of individual tumours which we will add to the conventional histopathological diagnosis. So therefore the assessment of a primary patient is likely to depend on the light microscopic diagnosis but additional information about the molecular biology, almost certainly p53, probably BRCA1 and BRCA2 alterations and maybe alterations in the PI3 kinase pathway. That will be just as useful, perhaps, as the CT scan imaging which shows us the extent of the tumour but currently gives us no idea about the functional or behavioural characteristics of that cancer, particularly whether it’s likely to respond to platinum or any other anti-cancer agent.
