EVC: Welcome to this ecancer expert discussion on pancreatic cancer. I’m Eric van Cutsem, I work as a GI oncologist in Leuven, Belgium, and I’m happy to be joined by Talia Golan from Sheba in Israel. Pancreatic cancer is a difficult disease, isn’t it Talia? It’s a difficult disease; we have made some progresses but the progress remains very modest. The good news is that we have seen with chemotherapy in patients with metastatic disease a slight increase in the median survival with the two standard regimens, gem/nab-paclitaxel and FOLFIRINOX, but really we need to find better treatments in this setting. Therefore the data that were presented here at this meeting, at ESMO GI in Barcelona, and also recently at ASCO, they are very important in this setting. They really give some hope for some patients; there was a lot of press recently also on the data and now with the updates that you presented here it’s important. It may be good that you situate a little bit, Talia, the field of BRCA in relationship to colorectal cancer and explain to us also some aspects on PARP inhibition in pancreatic cancer.
TG: What we’ve learned over the last decade in germline BRCA mutated cancer, the first thing was the prevalence wasn’t very clear to everyone. Originally we all thought that it would be about 4-5% and from the data that we got from the pre-screening section of the POLO study, which was over 3,000 patients, we have a prevalence of around 7-8% and that’s in an unselected pancreatic cancer patient with metastatic disease. This is twelve different countries worldwide and we were all surprised to see certain countries in Asia, for example, with around a 5-6% prevalence, certain countries have a higher prevalence as well than that. That’s really been a learning experience for all of us. Important also to note that there were some subsets of patient populations that maybe hadn’t been identified enough that harbour these germline BRCA mutations, for example the African-American population that we saw also from this that had a much higher prevalence than expected. It’s a small cohort so we have to be careful to over-emphasise that but it just once again shows that we don’t really know when it comes to mass screening what’s a true prevalence.
What else do we know about BRCA patients? Is this a predictive biomarker, is this a prognostic biomarker? This is still evolving. When one looks at all the data together it definitely seems as more a predictive biomarker. So they seem to have very similar overall survival and response to treatment, however, if you give them a direct DNA damaging therapeutic, for example platinum agents, they have an extended overall survival. That really laid the basis, this was really in 2014 when we sat together designing the study design of the POLO, it was a really special meeting. We looked at the data and we thought to ourselves can we speak about maintenance in pancreatic cancer? This is a devastating disease, median overall survival is less than a year, how can you even discuss maintenance in a subset like that? But we had this retrospective data that really supported that if we can get them platinum we can really control their disease and we can start modulating that process. Then if we come in with a maintenance strategy, what are we trying to do with maintenance? We’re trying to switch from chemotoxic chemotherapeutics that have very severe debilitating side effects over the course of time, specifically platinum neuropathy and the immunosuppressive components of those treatments. We thought, yes, we can try giving a PARP inhibitor which we know the synthetic lethality in tumours that have some HRR defect, like germline BRCA mutations.
There were several bold thoughts going right into that design, one was what is the real prevalence, is this a biomarker that makes sense to develop further in this huge global randomised study. The second thing was is maintenance really going to be an option, is this the kind of lingo we can start speaking about in pancreatic cancer? So what were your thoughts on these things?
EVC: Yes, in the beginning many of us also in the steering committee involved in the design discussions were a bit sceptical, not about the concept but about feasibility for reasons that you just outlined – the maintenance in pancreatic cancer because in general in pancreatic cancer we give chemotherapy until progression. But now with the data with FOLFIRINOX we know that we can induce some responses or some tumour control and that some patients just have to stop their chemotherapy in this situation because of neuropathy or because they just can’t stand it anymore because of fatigue after several cycles of FOLFIRINOX. So, yes, the design, looking backwards to a couple of years ago, was quite innovative. It was very risky to do that trial. Of course everybody involved in the trial and everybody also in the scientific community is happy for the patients that we continued the efforts. It was a big effort with screening more than 3,000 patients, as you said, in this setting, to have a lower number of patients involved in the trial in this situation. But screening 3,000 patients to find these patients with BRCA mutations and then especially also patients not progressing because we wanted to see what is really the best situation of a PARP inhibitor in this setting – patients responding or without progression – so tumour control for at least four months in this trial. This was a challenge but it’s good for the field of pancreatic cancer that we could continue because it’s the first really targeted agent in pancreatic cancer in an enriched population that has a major impact in this setting.
But that means that there will be strategic discussions in the management of pancreatic cancer because if you and I see a patient with pancreatic cancer, a new patient, we are not always aware of the BRCA status, that there are some patients who know it from their familial history but there are also some patients who don’t know it. As you said, around 7% of patients have a germline BRCA mutation. So the strategy and the question is do we determine up front in all patients with pancreatic cancer a germline BRCA. How would you do that in Israel?
TG: I just want to go back to one of the other things that you said that’s just getting me thinking is that what we also tried to do with the study, that we had this flexibility in the design and that’s how we feel with our patients as well. Some of our patients really need to get only four months and then try to switch to the maintenance but some need eight months or ten months. This approach and the way the study design was designed actually facilitated that. That also means that going back to the clinic we can do a similar strategy so we can really learn the art of that moment, of when is the right time to stop the chemo and start a drug that has a better toxicity profile in the setting. That’s just one of the things I wanted to mention. The second thing is that it’s true that we are definitely having difficulties thinking how we go ahead from a health perspective in screening patients when it comes to having a small biomarker, a smallish percentage of a biomarker. We’ve learned so many different lessons from other cancers. This is not the first time we’re speaking about germline BRCA so the field has gone so far ahead in ovarian cancer and breast cancer and now we’re catching up in pancreatic cancer and I think prostate cancer is coming soon as well. So it’s not like we have to reinvent the wheel, we just have to learn from our
EVC: That’s great but I don’t think that 7% is a small group, that’s a major problem. We do that in other diseases and other cancers, also in some GI cancers now with the data on BRAF mutant colorectal cancer that’s also 8-10% of the patients. There we determine that but here, to my ideas, an extra complexity is that the germline testing, usually many genetic labs are not set up to give the results soon and fast, within a couple of days, because that’s what you need. Because, at least in some countries, including my country, when we see a new pancreatic cancer patient we don’t start systematically with a platinum based therapy, we often start with gem/nab-paclitaxel in these patients while the retrospective data that you and others published on germline BRCA mutant patients, that they benefit more from platinum based therapies. Also now with this olaparib data the strategy is really we should determine in the beginning of the disease whether they are germline.
TG: I think that we’ve got the flexibility here. Countries that really do start with FOLFIRINOX or some platinum-based chemo up front but it’s true that there are institutions and countries that don’t. But in those scenarios if you get your BRCA testing one or two months downline and if your patient is positive and is not doing so well you could switch in a platinum for one of those drugs or change regimens and then try and reintroduce olaparib at a later time. So it’s not the setting of a clinical trial so I don’t think it’s going to be… I think we can still maintain a similar strategy.
EVC: That’s probably correct but it’s not going to be valid for all patients because if we would start with gem/nab-paclitaxel some patients with a fast progression, they may not make it to the platinum based therapy in this setting. So I completely agree that there is much more flexibility than in the clinical trial where the criteria were strict but there will be some challenges and some discussions.
TG: Yes, I think so but turnaround times, even for the germline BRCA, can be even two weeks. Even if then you make your change I don’t think that’s going to be the issue. Maybe the fact that it’s a germline test that may have different health issues or a different approach or different ethical concerns for the patients as well that we need to address each country as its policies. But putting the emphasis on that, that knowledge is empowering us to do probably what we hope is better medicine for those family members as well.
EVC: I completely agree and if we do germline testing we will find some other mutations – PALB, ATM. Do you think we should do trials or can we expand the data of the POLO study in this group of patients? PALB and germline PALB and ATM are probably less frequent or is for sure less frequent than BRCA1 and 2. But the question is, and that’s a question that will come up and that clinicians will call you or me – ‘What do you do? I’ve got here a patient, he is a germline PALB mutation, do we give that patient also olaparib in this setting?’ Do you have any specific ideas on that?
TG: Definitely that is the next question going forward for us in the field because we’ve seen the efficacy in the germline BRCA mutation population but now going forward how can we expand that therapeutic subtype beyond just those specific germline BRCA mutations. We have to be clear about our definitions and the alterations you mentioned are well recognised alterations to the DNA damage repair pathway. How that will respond to PARP needs to be tested in a clinical trial. Specifically for me, I’ve been quite lucky because I’ve got a clinical trial open now for several years in the BRCA-like population. So we’re really assessing that and the field going forward there’s going to be more of these clinical trials around the world so that we can really support putting patients on those trials.
EVC: From your experience, because you have studied that topic a lot, do you have any idea on how many germline BRCA negative patients have a somatic BRCA?
TG: It’s about 2-3% at most, it’s much smaller.
EVC: Do you think olaparib should work also in this setting?
TG: In some of them they do and in some of them they don’t. They are small groups so my data is based on a very small number. Some of them respond very similar and some don’t. So we have to think about what actually is happening in the tumours – there’s the germline BRCA mutation but is the tumour actually behaving like a BRCA tumour? That also comes to that when are we looking at the tumour, at what stage of the patient’s disease, how much treatment have we given him, how many mechanisms of resistance are evolving or have evolved as well? So those are things that we can all be thoughtful about going forward as we develop the drugs.
EVC: So these data really are pivotal, are practice-changing for the strategy in pancreatic cancer. That’s why the study was published in The New England Journal of Medicine but it’s only the beginning of our story, that’s what we hope. As we just were speaking many questions, we hope that there will be other drugs also going in that direction of individualising the management in patients with pancreatic cancer. That’s an important message so therefore at the ESMO GI meeting in Barcelona we had a nice discussion, you gave a nice talk also on setting the scene and explaining the meaning of BRCA. Because we have to educate the community and oncologists to understand that much better. GI oncologists are not yet so familiar with the concept of germline BRCA so therefore that’s very important. I want to thank you for joining this and discussing and thank you also ecancer for giving us the opportunity to speak to you.