Germline and Somatic DNA Repair Defects in Prostate Cancer

Dr. Alan H. Bryce, MD presented “Germline and Somatic DNA Repair Defects in Prostate Cancer” at the 26th Annual Perspectives in Urology: Point-Counterpoint, November 11, 2017 in Scottsdale, AZ

How to cite: Bryce, Alan H. “Germline and Somatic DNA Repair Defects in Prostate Cancer” November 11, 2017. Accessed Dec 2024. https://grandroundsinurology.com/Germ-Line-and-Somatic-DNA-Repair-Defects-in-Prostate-Cancer/

Summary:

Dr. Alan H. Bryce, MD, discusses synthetic lethality regarding BRCA deficient tumors, the role of PARP inhibitors and resistance to them, and the importance of germline testing when treating breast, ovarian, and prostate cancer patients.

Germline and Somatic DNA Repair Defects in Prostate Cancer

Transcript:

So the objectives for this talk then we’re going to talk about the concept of synthetic lethality in the context of BRCA deficient tumors.  What does this really mean once you know that your patient with prostate cancer has a BRCA deficient tumor and the role of PARP inhibitors in that.  We’re going to talk about the basics and start with talking about ovarian cancer, the role of germ-line BRCA status in treatment.  Well talk about the resistance to PARP inhibition some and the concept of reversion mutations.  It gets back to the concept that was drilled into me in fellowship training that the dumbest cancer is still smarter than the smartest oncologist.  

We’ll talk about—we’ll talk about, you know what I took this article out in the interest of time, but I’ll talk about it nevertheless, the potential to detect reversion mutations, cell free DNA is definitely a technology platform that is coming into its own, and why all of this matters.  

All right, so the first set of slides I won’t spend too much time on.  You’ve already heard a lot about this.  BRCA1 and 2 are common in prostate cancer, much more so than you might think, different than breast and ovarian cancer, BRCA2 is somewhat more predominant over BRCA1 and appears to have a greater impact in terms of the relative risk of cancer compared to patients who are germ line wild type.  So BRCA2 patients have perhaps an 8, 9-fold increased risk of developing prostate cancer, BRCA1 about a three to four-fold increased risk, and BRCA2 more so than BRCA1 is clearly associated with higher Gleason score at diagnosis, higher stage, and younger age of diagnosis.  By NCCN guidelines, these patients are considered high risk, and the guidelines talk about PSA screening starting at age 45.  Now, the guidelines have only recently been updated, so this might not be disseminated yet.  

Recognize that the positive predictive value of a PSA greater than 3 for predicting a positive biopsy is not changed, but you’re much more likely to find a high Gleason, an intermediate or high-risk cancer rather than that low Gleason cancer.  The risk of higher intermediate risk, three to four-fold higher in the patient with a BRCA2 deficiency.  

And note that amongst patients with lethal prostate cancer with known germline mutations, BRCA1, 2, or ATM up to 60% of them are going to report a negative family  history, so this is one of the key points we want to take home, we rely a lot on family history on this pre-test probability in order to order genetic testing in the first place, and I think this is the key mistake we are making.  I’m going to show you some more data feeding into this.  We are vastly under-diagnosing the frequency of these mutations, and this gets into what David was talking about earlier where his patient’s sister ends up dying of triple-negative breast cancer and he is asking the question, could I have prevented this had I been looking for the gene mutation.  And, you know, two years ago probably people weren’t talking about this at all, or at least the majority of people weren’t talking about this, but the time has come to change this conversation.  The answer is, yes, we can save lives by identifying high-risk patients early.

Triple-negative breast cancer is a bad disease and the patient  might have died anyway, but nevertheless, we’re going to save some lives.  We’re going to decrease mortality, and the best way to decrease mortality in cancer, as you all know, and I say this as an oncologist, the best way to decrease the mortality of prostate cancer is never to let the patient get to me in the first place.  Right?  It’s diagnose them early, let the urologist treat them, cure them with a prostatectomy and don’t let them advance to the point where they meet the medical oncologist.  

So this was a very important presentation we saw at ASCO this past year, and if you haven’t seen this you really need to go back and look at this paper.  It was in JCO in May, and what this looked at was 1158 patients with localized prostate cancer who had had germline testing done.  They retrospectively examined these patients from multiple institutions and then applied a broader panel of testing beyond what had already been done, and in these 1158 patients, they found 199 patients a high-risk mutation. These are impressive numbers, right?  BRCA1 or 2, ATM, a list and we’ll get into the list a bit more of other high-risk inherited germline mutations, and of those patients, of those 199 patients in whom the high-risk germ-line mutation was found, only 63 of them, 63% of them that is would have been referred for germ line testing if you followed published guidelines, so the point here is that published guidelines are inadequate.  It’s not enough to say do you have family members who were diagnosed.  Do you have, you know, is it your father, is it your brother?  Guidelines like that don’t work if your patient is an orphan, but if your patient is an orphan and he has four children, it still matters.  Right?  

What if you know he is an only child?  You know, you are seeing a woman, not prostate cancer, obviously, and she is an only child, her parents are only children, the number of people in the family is low, and so all of these metrics we use that talk about the number of family members representing a cut-off are inherently flawed.  You can’t use these well in patients who come from small families.  Okay?  

So published guidelines as they currently exist are clearly inadequate, and this is a major issue in cancer screening today and in genetic screening today.  And it’s something that in academia we are actively addressing trying to find out how do we expand the definition, make sure we identify the right patients in the first place?  Talk about all of your young patients out to be tested.  You know, you get a man in his 40s with prostate cancer.  I would say that should be automatic.  You’ve got to send him to the genetic counselor to talk about germ line testing.  In 2017 we can now say this ought to be done.  Right?  

In 2010 we weren’t thinking about it, but in 2017 we ought to be doing this.  So the conclusion of this study is that genetic testing guidelines are not reliable, if you don’t, if you take home one message from anything I say, please take home this one.  Think about germline testing far more often because we will save lives by identifying high-risk patients early.  This is about cure and in medical oncology when I start talking about two-month survival benefit, three-month survival benefit, most people glaze over and say why do we care about a two-month survival benefit?  We’re talking about cure here through early detection.  

All right.  So you already saw this paper, so I won’t spend too much time on it.  This is an important paper because this is looking at men with metastatic disease and looking at this panel of known inheritable cancer risk genes, mutations, and what you see is that the overall rate is far higher than anyone expected.  I mean we were all trained with that kind of baseline knowledge of 1 to 2% of all prostate cancers having a mutation in BRCA, probably we didn’t talk about anything else beyond that.  But here in this metastatic population they start looking at 10 to 12% of all patients with metastatic prostate cancer, have a germ line defect, familial defect, 10 to 12%.  So you talk about the utility of a screening test.  A screening test that is going to detect the screened abnormality in 10 to 12% of patients is good enough for widespread use.  Right?  That’s higher than what you get with mammography, colonoscopy, or any of the standard screening tests we think about.  

This data alone should tell us that every man with metastatic prostate cancer should have germ line testing done.  A 10 to 12% chance you are going to find something that is relevant to his family and ultimately to the patient, and I’m going to go on and show you how this matters to the patient today and how he is going to be treated.  

As you have heard, this isn’t limited to just BRCA1/2, this also impacts other cancers.  So the Lynch Syndrome family of genes, which you typically think of applying to colon cancer, but also has implications for upper tract urothelial cancer, uterine cancer, other  malignancies, are in—carry an increased risk of prostate cancer.  Now, maybe a 6, 8-fold increased risk of prostate cancer over baseline in early prostate cancer.  All right.  So again, 10 to 12% of these patients are going to have some kind of germ line deficiency.  We need to be testing these men.  

So why does this matter in terms of my patients in my clinic when I’m trying to decide how I’m going to treat.  So to talk about this, you know, we back up a little bit, and we focus first and foremost on BRCA.  BRCA 1 and 2 are well-known already because of their impact on the treatment of breast and ovarian cancer, and it is based upon this idea of DNA repair.  Now, DNA repair is an every day process, right?  The fidelity of the replication process is quite good, but nevertheless, there is going to be a regular rate of mutations during replication, and that repair happens through a couple of different mechanisms.  If you have a single strand break, then you have base excision repair that gets activated, and this is a process that ultimately runs or relies in part on a gene called PARP1.  When we have double-strand breaks, here again we have to go through homologous recombination repair, and this relies on a variety of genes including BRCA1/2, RAD51, and others, ATM, these genes that you have seen on the prior slide in terms of inherited familial cancer risk, and this process leads to high fidelity DNA repair that keeps our cells healthy, keeps us from ultimately developing cancer, but at the end of the day all cancer is a genetic disease so by definition a cancer cell that has become malignant has a problem with DNA repair somewhere along the way, and what we have come to recognize is the reliance of this process on BRCA and ultimately on PARP has lent an ability for us to exploit this therapeutically.  

So it is this concept of synthetic lethality, and again we know that these gene mutations are common in breast, ovarian, prostate cancer, but synthetic lethality is a very simple concept that there are two steps required to develop lethality therapeutically or otherwise.  If you have two intact steps, you have cell survival.  If you have one, that’s deficient.  You have cell survival, but if you knock out two steps in a process, then we have synthetic lethality, we have cell death and the two steps we’re talking about in this process are BRCA and PARP.

So patients who come in with BRCA deficiency, germline, have one step in the process towards synthetic lethality, and if we can come in and we can knock out their PARP, then we can induce cell death, and this is what we are trying to exploit, so in the normal process, here is normal PARP function, you have a single strand break, PARP comes in, ends up initiating a cascade, so PARP will bind to the breakpoint here.  Ultimately you have other genes coming in RAD51, other proteins, and typically what will happen, once the PARP comes in, it will recruit ADP ribose polymers, you’ll have high fidelity DNA repair, and the cell will go on as normal, but if we come in here with a PARP inhibitor, what this does is it inhibits the further formation of the complex and traps this complex on the DNA preventing the replication from happening, preventing repair from happening, ultimately leading to DNA instability, double-strand break, and cell death, so the point of the PARP inhibitor is to prevent repair in a cell population that is the cancer cell population where damage is happening at a high frequency.  You are exploiting the normal inherent instability of the cancer genome, preventing it from repairing itself, preventing it from replicating itself, and therefore inducing the apoptotic pathway.  

So you look at this, and you know, cell line studies, this is a typical story in prostate cancer.  We’re all used to this.  Cell line studies looking at PARP inhibitors or BRCA, and synthetic lethality go back to 2005.  Clinical trials started in 2008, and in the initial phase 1 clinical studies we had prostate cancer patients who had very good responses to PARP inhibition, but there’s also ovarian and breast cancer.  So what happens next is a story we’re all used to because there’s a lot of kind of nihilism, lack of interest in prostate cancer, and no studies move forward in prostate cancer, ovarian cancer studies move forward.  Breast cancer studies move forward, and by 2014, you have the FDA approval for a PARP inhibitor in ovarian cancer, and in 2013 you have large phase 3 studies that are organized, breast and ovarian cancer, and the field is moving forward very quickly.  

As far back as 2011 when I first came here to Mayo in Arizona, I approached every company that had a PARP inhibitor with an investigator initiated study concept saying let’s move this forward in prostate cancer, and you get the familiar response of crickets.  Oh, you know, that’s a great concept, sure, sure, let’s talk, okay, but here we are in 2017.  It took until 2015 for the first report to come out looking specifically at prostate cancer, and I mean this is seven years after the phase 1 study had already shown that there was a signal, something we should have moved rapidly on to exploit, and this study came out of Johan Debono’s group at the Royal Marsden in the U.K. very small study, and what they are showing here is that they tested patients for germ line mutations, and here is your BRCCA population, here is your ATM population and here is this longer list of genes associated with homologous recombination, or DNA repair.  And what you see here is a dramatic difference both in progression free survival and in overall survival with the use of olaparib in patients, the blue line, patients who have germ line defects versus red patients who do not, and I mean this—you know you don’t need a hazard ratio here to see that there is a significant difference.  Now, this is a single arm, single institution study.  This is very limited data.  This is not definitive by any measure, but this is a powerful signal, and the point of this was to say, look, we need to move with PARP inhibitors in prostate cancer.  

This led to an initial approval by the FDA for breakthrough status for olaparib in prostate cancer, and what this means is we can get olaparib for our patients.  It varies insurance company by insurance company, we have to go through a process, but if a patient has a BRCA1/2 deficiency, in my experience we have a 50/50 chance that the insurance company is going to let us prescribe this drug.  

In my own practice, we have 30+ patients now that have been treated with olaparib off-label with the insurance company paying for it.  I tell you it is every bit as effective as what you see here.  We are also now in an era where we are having randomized phase 3 studies in prostate cancer, rucaparib, niraparib, talazoparib, olaparib, all of the companies that have these drugs are looking in prostate cancer.  So you are going to hear about these studies, and every study is a bit different, pre-chemo, post-chemo, in combination with other androgen receptor antagonists, etc.  There’s a whole proliferation of these studies, and this is going to transform the field in prostate cancer to be sure.  

Here’s another demonstration of the same, and I think in the interest of time, I’ll keep moving forward, but what this one is showing us is again a cohort of patients with metastatic castration-resistant prostate cancer.  The double dots show you patients where the defect is demonstrated in the germ line versus here, where it is demonstrated just in the tumor.

So this is making the point, the difference between somatic and germ line defect, somatic meaning it is only present in the tumor.  The tumor developed the mutation, it is not present in the patient’s germ line, it is not heritable, has no implications for the family, and in the metastatic patient it is about 50/50.  About half of the patients have germ line and somatic, half of the patients have somatic only.  So the way I typically approach this in my practice is what I’ll do is I’ll go ahead and do somatic testing.  If the tumor doesn’t carry the defect, then you’re sure it is not present in the germ line and it is a full stop.  If I find it in the tumor, then I have a conversation with the patient about moving on to germline testing because of the implications for family.  

So at the end of the day about 20 to 25% of patients with metastatic disease will have somatic mutations in DNA repair genes, and clinical trials are ongoing.  Inherited cancer syndromes are under recognized both in practice and in research leading to delayed diagnosis and decreased cure rates.  

We need to know this.  We need to identify it.  This is how we are going to improve cure rates more than anything I’m going to do in my clinic.  

Guidelines are still evolving, and future treatment paradigms I would say are likely to incorporate germ line and somatic mutation data as we move forward, and I will touch on that a little bit more in my next presentation.  

All right, I think we’re going to do some questions.  I think we’ll take a few questions on that, and then I’m going to move on to my next talk.  Yes?  

1. CRAWFORD:  We’re having a discussion in our clinic about ordering germ line testing.  Let’s say somebody meets all the criteria that they have four family members with prostate, the mother had ovarian at age 46, I mean it’s very strong, what should urologists do if somebody comes in with prostate cancer, or they are worried about it, about testing?  Should they order germ line testing like the Ambry [phonetic] test and Myriad has a test too?  What is the informed decision, you know, you get that before—how do you think that should transpire?  
2. BRYCE:  So really I think the right way to do it, the best way to do it is to involve the genetic counselor right up front.  Now, the problem with that statement is most people don’t have immediate access to a genetic counselor, and we know this.  Even at the Mayo Clinic, we cannot hire enough genetic counselors for the sheer volume of patients who require the service, and nationally this is well-recognized, the number of genetic counselors being trained is less than half that of what is necessary for the growth of the field, I mean it is a very difficult problem right now.  And yet we have to be very careful in ordering germ line testing because of all of the implications.

It’s easy enough for us to talk to the patient in front of us and say, look, this is going to impact your care and why it matters, but the issue of how it impacts the family is one that most people haven’t thought through and not a conversation that can be had in 5 or 10 minutes.  The issue here, and it’s not obvious what you want to do.  The fact of the matter is some people don’t want to know.  But you know, your patient Mr. Smith finds out, maybe his sister doesn’t want to know, but his brother does, and we need to prepare them for that, you know, you can’t just force this information on everybody.  And same issue can arise with children, right?  Which children are you going to tell, and what age are you going to tell them at?  You know, Angelina Jolie and Christina Applegate, they went out and had double mastectomies, maybe they had oophorectomy, we understand this, but do you really want a 15-year-old girl to have to think about that?  There’s a point in time where maybe they are ready for that conversation and a point in life where maybe they are not.  I would submit to you that we can’t template the answer to that.  It really depends on the family, their values, and what not, so the ideal thing for the urologist in practice is to have access to a genetic counselor, and tell your patient, you know, you have a real chance of having an inherited cancer syndrome that has implications for your family and I need you to go meet with a genetic counselor and talk about this.  That is the way I take it.

1. CRAWFORD:  In the pure world, that is the way it should go, but it is a three-month delay to get in our genetic counselor – – a lot of people from out of town that come to see us just like you do, and out of mind, you know, out of sight, out of mind, they forget, they don’t come back, so we’re not sort of seeing it.  My philosophy is like the PSA of 1.5 thing, you do it, if it’s negative, there’s no action really needed.  They can go talk to a genetic counselor, if it is positive, that is when they should go talk to the genetic counselor.  They have already said they wanted it done, but the implications, I mean that is the way that I see it happening because otherwise we are going to still be talking about this five years from now.  
2. BRYCE:  Absolutely.  I mean we have to be practical.  Right?  Having said that, you know, on this particular issue a three-month delay is probably not a problem, truly but no I understand I mean patients travel.  What we are templating and other companies already do this is the genetic counseling conversation could happen over the web for example, and we’re already doing that.  Right?  You don’t have to come in for the appointment.  We’re going to do it over video conference, you can talk to the genetic counselor, and that makes it much easier for people traveling from a distance.  
3. CRAWFORD:  All right, thank you.  
4. BRYCE:  We’ll go into the next talk.  We’ll talk about the therapeutic use of genetic testing, and I’m going to talk about both germline and somatic here in advanced prostate cancer.  

All right, so we’ve got some audience response questions for this one.  Which of these pathways is commonly aberrant in metastatic prostate cancer?  So say more than 20% of cases, androgen receptor signaling, DNA repair, cell cycle regulation, Pi3 kinase pathway, or all of the above?  

[music]

All right, good deal.  I like the retro music selection throughout the week here.  All right, which of the following is true regarding the impact of ARV7, and we’ve heard some about this already some this morning, a response to therapy in metastatic castration-resistant prostate cancer, A ARV7 predicts for response to enzalutamide, but not abiraterone; B, ARV7 predicts for lack of response to chemotherapy and androgen-receptor directed therapy; or C, ARV7 has no impact on the response to chemotherapy?  Which of these is true?

[music]

All right, so we have a distribution of answers.  All right, so we’ll move through this.  All right, so the outline here, some of this is going to be repetitive so I’ll move through it more quickly and spend some more time on other pieces.