DNA Repair and PARP Inhibitor Therapy in Prostate Cancer - Veda Giri, Patrick Pilié, and Arpit Rao
July 23, 2020
Arpit Rao, MD, MBBS, Assistant Professor of Medicine, Division of Hematology, Oncology, and Transplantation, University of Minnesota, Minneapolis, Minnesota, USA.
Patrick G Pilié, MD, Department of Genitourinary Medical Oncology, Division of Cancer Medicine, MD Anderson Cancer Center
Veda N. Giri, MD, Medical Oncologist, Sidney Kimmel Cancer Center, Thomas Jefferson University
Charles J. Ryan, MD, The B.J. Kennedy Chair in Clinical Medical Oncology at the University of Minnesota and Director of the Division of Hematology, Oncology, and Transplantation.
PARP Inhibitors, Prostate Cancer and a Promise Fulfilled
Germline Testing for DNA Repair Mutations in Prostate Cancer: Who, When and How?
What Are the Most Common Genomic Aberrations Seen in DNA Damage Response (DDR) Pathways in Advanced Prostate Cancer?
PARP Inhibitors - A Breakthrough in Targeted Therapies for Prostate Cancer
Charles Ryan: Hello, and welcome to a round table discussion on DNA repair and PARP inhibitor therapy in prostate cancer. I'm Chuck Ryan. I'm a Professor of Medicine, a medical oncologist, and I'm at the University of Minnesota in Minneapolis. And, I'm delighted to be joined by Dr. Veda Giri, who's an Associate Professor at the Kimmel Cancer Center at Thomas Jefferson University, where she is also the Director of the Cancer Risk Assessment and Clinical Cancer Genetics Program. Also, I'm joined by Dr. Patrick Pilié, Assistant professor at MD Anderson Cancer Center, where he's a medical oncologist specializing in DNA repair and advanced prostate cancer. And finally, Dr. Arpit Rao, also an Assistant Professor here at the University of Minnesota with myself, who's a medical oncologist, also specializing in urologic cancer. Welcome to all three of you.
Veda Giri: Thank you.
Patrick Pilié: Thank you.
Arpit Rao: Thanks.
Charles Ryan: Thank you for your pieces that were all excellent. And I want to start with you, Dr. Giri. I want to talk a little bit about this breakthrough of the need for genetic testing in our clinics now. And, this is now something that is no longer aspirational. It's actually something that we owe to our patients to be able to respond to the need for genetic testing. So, if you could just briefly outline for us sort of when should a clinician caring for a patient with prostate cancer think, "I should be doing genetic testing for this patient".
Veda Giri: Sure. That's a great question. Thanks again for having me. This is a wonderful platform to discuss this and these are exciting times for germline testing for men with prostate cancer. One of the key areas, of course, is in the metastatic setting. And just as you said, with the recent FDA approvals for two of their PARP inhibitors, rucaparib and olaparib, in the metastatic setting, particularly after a few lines of therapy, germline testing can actually open the door for additional therapies or, of course, for clinical trials as well. So, one of the key areas in terms of thinking of germline testing is in the metastatic setting. In fact, in the NCCN guidelines, in the two guidelines that address prostate cancer germline testing, the hereditary breast, ovarian, and pancreatic guidelines and the prostate cancer guideline, both of them state that men with metastatic disease should have germline testing.
Beyond that, really, there is also this growing space of thinking about men for germline testing that actually can help earlier stage disease. And, that becomes important in the urology space as well. So for men that have high-risk features on pathology, so a higher stage, higher grade group or Gleason score and also then, strong family history becomes important to consider and do an intake of in terms of family history of prostate cancer that might indicate hereditary prostate cancer, or even hereditary breast and ovarian cancer or Lynch syndrome in a family. So, doing a broad family history intake and deciding which men to pursue for genetic testing, can be really important.
Charles Ryan: I should point out that not all hereditary prostate cancer is related to BRCA1 or 2, correct?
Veda Giri: That's correct. Yes. There's actually a host of genes now that are important to consider for hereditary susceptibility to prostate cancer. So BRCA1 and 2 of course, but also the DNA mismatch repair genes associated with Lynch syndrome, of course, HOXB13, which was the first hereditary prostate cancer gene identified. Many of the other genes that we see reported in DNA repair really become important in terms of therapeutic decision making, how much they lend to prostate cancer risks. Some of that data still needs to be addressed.
Charles Ryan: And Patrick, Dr. Pilié, I want to ask you a quick question about somatic versus germline alterations. What we're seeing in the PARP inhibitor data is that both seem to benefit from PARP inhibitors and discuss really, if you could, sort of the difference for clinicians who haven't had much experience or for patients who are listening, about the difference between somatic and germline and how they may alter the clinical course and how you may approach identifying patients with either one of these alterations.
Patrick Pilié: The germline mutations that Veda already mentioned are hereditary, can be passed on to subsequent generations, and increase the risk potentially of a variety of different cancers sometimes. Interestingly, in prostate cancer, in particular, the distribution of these genes in the somatic or tumor only associated level and germline are relatively equivalent and also relatively unique to prostate cancer amongst the HBOC related cancers, such as pancreas, ovarian, and breast. We found in retrospective and prospective clinical data that both somatic origin variance and germline origin variants are targetable with PARP inhibitors.
Charles Ryan: One of the things you do see in your report is that this is a pathological variant, a pathologic variant, correct? Because it's not uncommon to get a report back from a specimen that says there's a variant of undetermined significance. And, I think a lot of people struggle with what they should do with those results. Can you comment on that?
Patrick Pilié: Absolutely. These variants actually are most commonly found as a variant of unknown significance. These genes such as BRCA2 and ATM are very large genes and variants can occur across the entirety of the gene. And, it's really the functional impact of the [inaudible 00:05:37] variant that is targetable with say a PARP inhibitor. And so, the functional impact of the variant is most important in determining the next therapeutic step. There are published guidelines on interpreting variants, whether they be pathogenic, likely pathogenic, a variant of unknown significance, or benign. There are also multiple online tools and resources for interpreting these variants beyond the platform that you use for sequencing gives you.
Charles Ryan: So, many of the commercially available sequencing platforms will give you a panel of mutations. And what you're saying is that the gene may contain mutations that don't really necessarily show up on the panel. Is that correct?
Patrick Pilié: The panels differ slightly. Some panels use hotspot testing, where they're testing only certain variants in the gene. Now though, that sequencing has become more ubiquitous. Most companies are using multipanel gene testing, that tests the entirety of the genome, the exome of the gene, to look for any variance in that coding sequence. However, a majority of these variants are still going to come back as unknown significance.
Charles Ryan: I see. Okay. So Arpit, Dr. Rao, in your clinic, when you're seeing a patient, how does this work on a practical level? What do you do when you think, "Okay, I want to know if my patient is eligible to receive a PARP inhibitor or similar therapy". Are you sending them for repeat biopsies if they have metastatic disease? Are you digging up the old prostate biopsy specimen? How does it work on a practical level where you are?
Arpit Rao: Yeah. So, Chuck, that's an excellent question. I think that's a question we should be asking more and more. I was talking to a patient yesterday about this. He is a researcher himself, a clinical trialist, has done a separate specialty and the question becomes, what is the number needed to treat, right? So, we have certain markers to guide our decision making and when to get these tests. It'd be great to get it on everybody, but sometimes it's not logistically possible and sometimes it's just not paid for, practical for other reasons. So in the clinic right now, the data is the strongest for second-line therapy with olaparib. So, I think you can make a very strong argument there that after progression on treatment with at least abi or enza, you test them and if they come back positive for HR mutation, then olaparib is a very good option.
Of for third-line, if you think about the data from TRITON2 and the approval for rucaparib, that's after exposure to either abi or enza and a taxane. So, that is another setting where I frequently would test these patients. So, to answer your second question about whether I do repeat biopsy or not, we have a growing amount of data on what to do in this setting. So if you go back to PROfound, they had a 31% specimen failure rate, meaning taking archival tissue, they were only able to meaningful data, genomic data in 69% of the samples. That's an issue. And we've realized that over and over again, even the data comparing circulating tumor DNA based assay versus archival tissue-based assay, especially in non-small cell lung cancer. ctDNA tends to out-perform some cases of archival biopsy. And, it certainly seems like it's very comparable at least in prostate cancer. So, my philosophy is if we're going to order this expensive test, I want to look at the most recent biology of the tumor. So, when possible, we do a repeat biopsy to get the sample.
Charles Ryan: Many of these alterations, in fact, may be acquired over time and it may be the result of the treatment mediated selection pressure. So that leads me to ask you, Dr. Giri, about the question of what Dr. Rao is saying is, "I'm going to wait until the patient is eligible for on-label use of the drug", which is certainly reasonable, but the guidelines are actually saying it's actually reasonable for us to begin testing very early on. In fact, even perhaps when a patient shows up with metastatic disease. If you could just highlight, what are the guidelines saying and how quickly are they changing on this point?
Veda Giri: Yeah. So, the guidelines from the NCCN are consistent about testing men with metastatic disease. As far as the optimal time point of doing that germline testing or initiating that germline testing, is probably more of a case-by-case clinical decision level for patients. So for some men who present with metastatic disease, it might just be too much to process all of the information that goes into making an informed decision for germline testing. However, that actually could also be an optimal time to do that because they are more robust. They don't have the ongoing treatment and frailty setting in from therapy when it can also become very challenging to them introducing germline testing and the hereditary cancer implications associated with that.
I think it's important to bring it up as soon as possible. And a patient may not decide right away. They might need to come back to it and revisit that, understanding that it's not going to change their therapy in the first-line, in terms of metastatic treatment, but it certainly opens the door for when progression happens for additional therapies. It's just important also not to miss this because this then goes back to the clinician to keep this on the top of their priority list, to keep bringing up with the patient because of the hereditary cancer implications, too. There are the treatment implications, but then there are also the familiar implications. So, I think it's a case-by-case basis for patients, whether it's metastatic disease or even in the early-stage disease.
Charles Ryan: Sure. And Patrick, the next question is for you. So, it's almost a challenge for us as a clinician. You test somebody early on and they're negative and then, two years later, their disease is progressing and you're wondering, should I be retesting? And, that of course brings up logistical challenges that have been discussed already.
Patrick Pilié: Right.
Charles Ryan: But, Dr. Rao mentioned ctDNA. And I'm wondering, where do you see the role of ctDNA versus tissue in this conversation?
Patrick Pilié: Definitely. And with the germline testing where you have a relatively static or determined profile, somatic sequencing can reveal different variants as the disease progresses, as treatment pressure exerts its effects on the cancer cells. And determining the next treatment after a progression or new metastases would be the time to obtain a biopsy, whether that be a tissue-based biopsy or liquid biopsy, like CTCs or cell-free DNA. There have been multiple studies I think, as Dr. Rao already mentioned, that have shown that the CTCs and tumor tissue reflect each other relatively well. And, that the CTCs may be actually, potentially more informative and obviously easier access. So if there is tissue amenable to biopsy, we tend to biopsy. Because in addition to looking for variants, also can look for staining of mismatch repair proteins, as well as looking to see if there's any evidence of neuroendocrine differentiation or changes in histology. But if there is not tissue amenable to biopsy, or we need to make a quicker decision, say, for instance, the CTCs are definitely a viable option.
Charles Ryan: Right. I want to transition. I want to talk a little bit about the PARP inhibitors themselves. These are now available to us. Many of us have been working on trials with them for quite some time. Arpit, could you just walk us through the two approvals and the differences between the two drugs, in terms of the clinical states in which they are approved?
Arpit Rao: Yeah, absolutely. So, it's funny. We've been waiting for these PARP approvals to come through for a while and it just so happened, they both happened in May 2020. So, that turned out to be a great month for us who focus on this line of treatment. So, rucaparib beat olaparib by a few days in terms of approval. And, that approval mostly is an accelerated approval based on the TRITON2 data. So, TRITON2 is a study of rucaparib monotherapy in HR mutant patients. And, it showed an impressive overall response rate of, I believe, about 45%. And so, that resulted, and with that went a good medium duration of response. So based on that data, rucaparib got approved. Olaparib has in this setting, eluded more of an evidence space, so to speak.
So olaparib not only derived the approval, not only derive some of the data from TOPARP-B trial, which was done in the UK, looking at patients with HR mutant tumors but also derived data from the PROfound trial, which was the randomized trial of olaparib versus abi or enza as a standard of care treatment in patients after one line of treatment for metastatic CRPC. And in this setting, olaparib hands down, improved median progression-free survival. So, radiographic progression-free survival, some of the other key secondary endpoints, including the median time to pain progression and also showed a trend towards improvement in median overall survival. So, I think olaparib is, the current approval as it is, olaparib can be used after progression on first-line abi or enza, as long as the patients have an HR mutation. And then, rucaparib as I mentioned previously, has to be progression on at least one novel antiandrogen and a taxane and requires the HR mutation to be present.
Charles Ryan: And, Dr. Giri, do you think we're going to hold on to these indications for some time or what's your speculation about the expanded indication, perhaps even into earlier stage disease in the future?
Veda Giri: Yes. I think that we might actually see that. There is one trial that is going on right now in the biochemical recurrent space, with a PARP inhibitor in that setting. So that data, when it's reported out, will help us to see if there is an opportunity for benefit, clinical benefit in terms of in the biochemically recurrent space as well. So, I think that that is a potential area where this really could expand from a therapeutic standpoint as well.
Charles Ryan: Right. It's important to note, I think every time that a new drug enters the pharmacopeia for prostate cancer and patients will ask, and I'm wondering what your response will be. If a patient has a pathogenic BRCA2 alteration, and we know that's sort of the critical driver of their disease, are we going to ever get to a point where patients can discontinue androgen deprivation therapy or where a PARP inhibitor is going to be the sole therapy for them? I think patients are going to ask us those questions.
Veda Giri: Yes. And, I don't know because we know of the [inaudible 00:17:28] of prostate cancer. So, I'm not really sure if we're going to get to this point of discontinuing ADT completely. It's a great question from patients just because of the side effects and everything that they endure being on ADT. But, I think that at some point we might know more about that, but that's a great question.
Charles Ryan: A lot of studies for the future. So Patrick, a question for you. Should a patient with an ATM mutation go on a PARP inhibitor?
Patrick Pilié: That's a great question. The approval for olaparib is inclusive of a variety of homologous recombination repair genes, including ATM amongst others. The approval for rucaparib is specific to BRCA1 or BRCA2. In looking at both the prospective data from olaparib and rucaparib, as well as there's been multiple publications now of retrospective looks at PARP inhibitor use in patients with ATM variants and have found relatively few responses and no real sustained or durable responses in the patients with ATM variants, limited mostly to PSA response in those patients. So, while olaparib has approval for patients with ATM variant, deleterious ATM variants, its use still needs some further investigation. While other genes such as PALB2 and BRIC1, that are more closely related to BRCA2, although occur more rarely in patients with prostate cancer. I think, there's better data to support the use of a PARP inhibitor in those patients.
Charles Ryan: Yeah, I think that's well put. I mean, I would say that with an ATM mutation, we can use them with caution, but probably need much more research. And, there's a lot of excitement now around potentially ATR inhibitors and other approaches to patient ATM mutation. So, we're going to have to watch that space fairly closely. Dr. Rao, I'm going to close with you, and I want you to comment on the safety profile of the PARP inhibitors, and as clinicians using it, what do we need to counsel our patients about and what do we need to be watching out for?
Arpit Rao: Yeah. So great question, Chuck. I've put forward a few key principles for this issue. So, one of our colleagues had mentioned that PARP inhibitors are like chemo light. And so, as we started using these agents more and more in prostate cancer, we have to be ready to accept that chronic side effect profile that is associated with this therapy.
Now, as a drug class, there are considerable overlaps in side effect profiles, but there are some nuances that I'd like to highlight. But before that, just a couple of important points. So, when you start a PARP inhibitor be prepared to interrupt therapy or decrease dose. So contextually, about 45% of patients in the PROfound trial of olaparib had those interruptions or reductions. So, it's very common. However, a very small subset, so only about 18-20% of patients require dose treatment discontinuation because of adverse effects. So, the message being that you can get the majority of your patients through this treatment without stopping it for adverse reactions. So, that's one thing.
In terms of what to expect, there are some common themes here. So, the most common side effect being fatigue and then, some GI side effects like nausea happens in about 70-80% of patients. Vomiting is uncommon. That happens in about 35% of patients, 30-35, depending on which trial you're looking at, diarrhea, constipation, some of those GI effects. And then, myelosuppression being a major issue that comes up and we need to be just prepared for it. So typically, once you start this treatment, we've seen it as starting to occur. So, hemoglobin starting to drop kind of in week three, week four of the first cycle. And then, by the second or third cycle, you start to see neutropenia starting to occur, thrombocytopenia starting to occur. And, most of these myelosuppressive effects do actually promptly improve with treatment interruption. And, I typically recheck CBC every week, once I interrupt treatment. But, every other week is also reasonable. It depends on how quickly they dropped and how far they went, just like in your clinical judgment.
There are some nuances like ondansetron or any 5-HT3 antagonist is really good to manage the nausea that's associated with this treatment. Also, asking the patient to take the drug an hour after the meal really helps us. Now, on the other side, there are some nuances that are particular to each agent. For example, niraparib causes a significantly higher rate of thrombocytopenia, and I'm not sure why that happens on a biological level, but the incidents in GALAHAD, for example, which is a trial of niraparib monotherapy, is about 60% as opposed to 15% for olaparib and about 25 to 30% for rucaparib. So in fact, this is such an important issue that the FDA label for riraparib has weekly platelet count built-in for the first month of treatment. So, that's an important point.
Similarly for rucaparib, just because we've put a lot of patients on these trials of rucaparib therapy, we've often seen drug-induced transaminitis. So a couple of weeks after you start treatment, you start to see ASD and everything go up. Generally, you can treat through it as long as it's grade 1 or grade 2. So, less than three times the upper limit of normal or less than five times the upper limit of normal, but you got to make sure that the patient's not developing drug-induced liver injury. So, daily has either a substantial elevation of transaminases or concurrent elevation of transaminases and bilirubin. So, that's really a dangerous condition and that needs to be identified quickly and it's a diagnosis of exclusion. So, you're going to need a couple of weeks to work through and you may want to stop the drug at that point. Hold it until you can sort out what's what.
Rucaparib also messes with the creatinine excretion in the renal tubules, without affecting kidney function. So you would see creatinine go up, but the kidney function will be still normal. And the way to parse that out is, I mean, you can certainly, practically you can stop the drug for a few days, five half-lives later, creatinine should come back, but none of the other metrics should be abnormal. So, BUN should be fine. Patients should be doing clinically okay. And if you're really worried about it, you can always get serum cystatin C and calculate the GFR based on that. And, that shouldn't be affected.
Charles Ryan: It's a great start.
Arpit Rao: Lastly, yeah. And one more thing about olaparib, because that's probably the one that people are going to use the most. I've been a little bit concerned about the emerging cardiovascular safety data for olaparib and I think that needs to be borne in mind.
So for example, in the PROfound trial, there is a 4.3% incidents of PEs and some of them were actually minor. So, grade 1 or grade 2, but there were a couple of grade 3 events. So, that's worse than 1% or less than 1% in the control arm. So, that's the safety signal that's emerging with monotherapy. With abiraterone and olaparib trials, there was a 10% risk of serious cardiovascular events. So, that's substantially higher than I think in the control arm it was 1% risk with abi, alone. And so some of these side effects were clinically significant, like fatal ischemic strokes, fatal heart failure, MI and congestive heart failure, those kinds of side effects. So, please be careful about the cardiovascular profile, at least with olaparib, at this time.
Charles Ryan: Thank you for that. Very thorough. And, these details are reflected in your piece that's posted on UroToday. Dr. Giri, Dr. Pilié, final words from you? Veda, go head first on your thoughts on moving forward.
Veda Giri: Yes. Like I said at the beginning, I think that this is a really exciting time for germline testing for men with prostate cancer. It's absolutely taking off in the metastatic space and also in the earlier stage setting. I do want to add in one more comment about the somatic testing as we have discussed in terms of also the way to assess for PARP inhibitor eligibility for men. One of the things to keep in mind about this somatic testing is that it can actually uncover germline mutations as well. So, clinicians need to be aware that when they're ordering somatic NGS testing, that if there is a mutation in a cancer risk gene, that let's just say, for example, BRCA2, there are other additional factors to then consider to think about reflexive to germline testing.
And so, your variant allele fraction, suspicious family history. In the metastatic setting, since it's a blanket indication, it might be worth considering paired germline and tumor testing at the same time because of the missed areas that might occur, whether it's somatic or whatnot. And then, that way we can pick up on germline mutations. And so, I think, in the earlier stage setting, some of that will be forthcoming, but I think it's just important for clinicians to keep that in mind, that when they're ordering somatic testing, that they might get information that leads to online confirmation as well. This is a very exciting area that is actually exploding now in this setting.
Charles Ryan: Yeah, for sure Patrick, final thoughts?
Patrick Pilié: I think, this is a really exciting time for both patients and clinicians that treat men with prostate cancer. We're taking our first steps into precision medicine and now have standard of care approved therapies that are targeted to patients with certain variants. However, we still have a lot of work left to do to refine the biomarkers further, have them be more functionally informed and dynamic. As well as a multitude of other DNA damage response inhibitors in development that have exciting potential for patients with prostate cancer.
Charles Ryan: Well, I thank the three of you for joining me and for the pieces that you wrote for UroToday. I'm sure that a lot of clinicians out there will learn from this conversation today, can consult your written work and hopefully, and undoubtedly follow your careers as they progress. And, you help pave the way for determining the role of this new class of drugs in prostate cancer, both in the current state, in CRPC, and perhaps beyond. So, thank you all again.