Emmanuel Antonarakis: Thank you so much for the invitation, Neeraj.
Neeraj Agarwal: So we really want to learn from you today, and for our colleagues worldwide, how to approach genomic testing for patients with prostate cancer, which patients we should be testing, and what we should be testing, and what kind of test we should be using. These are the challenges we have been facing as a field for the last almost a decade now. There are multiple approvals already in place, mostly in the form of PARP inhibitors, but also immune checkpoint inhibitors. And as we know, genetic testing and use of targeted therapies is very low. Unacceptably low. So I would like to start by asking, Emmanuel, first of all, thank you for being here to share your experience, how do you approach testing of a patient with locally advanced or metastatic prostate cancer when you see them for the first time?
Emmanuel Antonarakis: Neeraj, I think that testing has evolved and the commercial companies and the insurance companies are more willing now to reimburse this and pay for it. So my short answer is, in a medical oncology clinic, it should be virtually every patient that a GU medical oncologist, prostate cancer-oriented medical oncologist sees. I can't think of that many patients that a germline and somatic test should not be ordered. So I think that default should be, for any patient coming to the GU medical oncologist, prostate-oriented medical oncologist, the germline and somatic testing should be done. Now, you asked me a very important question. You asked me about the high-risk localized, locally advanced setting, and then you asked me separately about the metastatic setting. Let's answer that question in reverse, because the metastatic setting, I think, should be the easiest one to answer and I think this is the one where I want the audience and those practitioners out in the community to remember the most.
And that is, in my opinion, every metastatic prostate cancer patient, whether they are hormone-sensitive or castration-resistant, androgen pathway modulation-sensitive or androgen pathway modulation-resistant, those are the new terminologies, should undergo both germline testing and somatic genetic testing. We'll talk about in a minute how that can be achieved. But it should be all of those patients. And it meets the guidelines, it meets the NCCN guidelines, and it meets most other European guidelines as well. Now, when it comes to locally advanced or localized high-risk disease, of course, germline testing is still warranted and recommended by most guidelines, including NCCN and AUA. So the germline should still be done at least for the family implications. However, I like to do the somatic genetic testing as well. Now, the caveat is that we don't have any FDA-labeled indications for a targeted or genomically-selected therapy in the high-risk localized or locally advanced setting at the moment. That might change in the future. The reason I like to do it, however, is because many of those patients will become metachronous metastatic in the future, or sometimes when we obtain a PSMA PET scan, we find actually they are metastatic de novo. And I like to have that information so I can plan my systemic therapy for the first-line metastatic scenario.
Neeraj Agarwal: Very well said. Let's start with the germline testing first. I know many of our community colleagues do not have access to genetic counselors, and it is a challenge for even many academic settings for all patients. So how do you approach? Do you order germline testing first and then set up a genetic counseling visit if they have pathogenic alterations, or reverse?
Emmanuel Antonarakis: I think that based on how many cancer types now are doing germline testing and how many guidelines are endorsing it, and based on the paucity of the genetic counselors, and based on the fact that most medical oncologists actually have enough knowledge of the ABCs of germline testing to perform the pre-test counseling themselves, I think that we should feel comfortable as medical oncologists, or in some cases, urologists, radiation oncologists, even, to prescribe the germline genetic test. These tests are done through a saliva sample or a blood sample. So as we all know, we should remind people we do not need a tissue, tumor specimen for them. And oftentimes you can even send the patient home and you can order the test and the saliva kit will be sent to the patient's home directly from the company, for example, Invitae or others. And the same goes with the blood test. Many of these commercial germline companies will send a mobile phlebotomist to the patient's home. So it has really become easy and streamlined. If the genetic test is negative, which means that there are no family implications for that patients' first-degree relatives, then I do not find much value in the post-test genetic counseling.
Similarly, if there is a variant of unknown significance, most people, including myself, will count that as a negative genetic test, and I will typically not send that patient for genetic counseling either. The only scenario that I am currently strongly recommending formal genetic counseling is the case of a pathogenic or likely pathogenic genetic mutation, especially in the common genes such as BRCA1, BRCA2. In those cases, the patients will benefit, and also they should bring their family with them if they're able or have them join by telephone. It's all about the genetic counselor discussing the family implications and also the reproductive implications. If that patient is of childbearing age or potential, the subsequent risk to his offspring as well. And also just to educate the patient that every first-degree relative, which means a sibling, a parent, or a child, could have a 50% chance of inheriting the same pathogenic genetic mutation.
Neeraj Agarwal: That's such a thorough and yet succinct description of how do you approach germline testing. And by getting the testing first and then looking for a genetic counselor, if they have a pathogenic or likely pathogenic variant, obviously alleviate the burden on genetic counseling. We don't have to look for genetic counselor for everyone then. So now let's come to the somatic testing. Is tissue? Is blood? If it is tissue, what tissue? Primary prostate, metastatic tissue, knowing bone biopsies are so difficult to get. What is your approach when you are thinking about somatic testing?
Emmanuel Antonarakis: So if I am faced with a metastatic androgen pathway modulation-sensitive or naive patient, the patients that we used to call hormone-sensitive or castration-sensitive, oftentimes the easiest and fastest way to do the somatic genomic testing is from the prostate biopsy or the prostatectomy if the patient had a prostatectomy in the past. We have to remember that these patients often have metastatic disease only in the bones. Sometimes this is only detected by a PSMA PET scan. Sometimes if we are "lucky", they might have a juicy lymph node that we can access, or every once in a while they might present with a liver metastasis. This is extremely uncommon in the first-line androgen pathway modulation-sensitive setting. So I would summarize it by saying that 80% of the time in the metastatic androgen pathway modulation-sensitive setting I would use the primary prostate tumor for the somatic sequencing. I want to mention two other things that might alleviate some concerns about this. The first is that the type of genetic alterations that we have molecular targeted therapies against, such as PARP inhibitors and PD-1 inhibitors for the mismatch repair deficiency, those are typically what we call truncal mutations. These are typically found in the primary tumor, they're acquired during the transition from benign or premalignant into malignant, and they are not typically mutations that are acquired under selective pressures of hormone therapy-selective pressures, or chemotherapy-selective pressure.
That's important because there are other cancer types out there, for example, PIK3CA mutations in breast cancer, which can be acquired in third-line, fourth-line, fifth-line disease. In prostate cancer, the homologous recombination gene mutations like BRCA2, BRCA1, and most of the mismatch repair microsatellite instability mutations, they are usually not acquired. They're found in the original tumor. So the physicians can be assured or reassured that they don't have to feel guilty if they do not go after a metastatic biopsy. They can get the information that they need which is actionable for targeted therapies from the prostate gland. Now, not all patients have a prostate biopsy. I'm aware of that. And if a prostate biopsy is not available, then there must have been some tissue that was used to confirm the pathological diagnosis. That tissue might have been a lymph node, as we discussed before, or in some cases it might be a liver metastasis that was masquerading as a cancer of unknown primary, or even I had a patient in my clinic recently with an epidural mass that was diagnosed with meningioma. And in fact, when the epidural mass was biopsied, it was not meningioma, it was metastatic prostate adenocarcinoma. So in those cases, of course, I use whatever metastatic tissue has already been obtained for the clinical diagnosis, the pathological diagnosis, and then I repurpose that for the somatic NGS testing. And then if that's not available either, the vast minority of patients who don't have any pathological diagnosis, then I may use a circulating tumor DNA test. The caveat with the liquid biopsies is that there must be some tumor burden in the patient's body for that ctDNA test to be positive.
If that ctDNA test is ordered two months after the ADT or ARPI or plus or minus docetaxel has been started, PSA is 0.5 already, that patient will not have a circulating tumor DNA burden. So oftentimes you may miss the opportunity to do it. And then one final point, and I won't take up too much time, this applies only to the metastatic hormone-sensitive or androgen pathway modulation-sensitive. If the patient evolves down the line to an androgen pathway modulation-resistant, previously known as castration-resistant, in that scenario, I typically do my best, if possible, to have a new biopsy, especially if it may not need to be a bone biopsy.
Neeraj Agarwal: What kind of mutations are you looking at other than the AR variants, which you reported first time in NEJM almost a decade ago? What other mutations are you looking at for ctDNA in CRPC setting?
Emmanuel Antonarakis: The other mutations that I'm looking for at this point in time, 2026, they are not therapy targets at the moment, they are more prognostic indicators. I'm looking for androgen receptor amplification or mutations. I'm looking for splice variants, and not all of the NGS vendors report the splice variants. I'm looking also for the tumor suppressor genes. As we all know, there are three famous tumor suppressor genes, p53, PTEN, and RB1, and finding a mutation or inactivation of one of those three is not so uncommon in metastatic prostate cancer. But as you might know, Neeraj, we've worked on some manuscripts together, if you have a double or a triple tumor suppressor loss of PTEN, p53, or RB1, those patients have a very poor prognosis, and those patients are also at increased risk of developing a transformation into an anaplastic or neuroendocrine or pure small-cell carcinoma. So I like to have that information because it helps me to guide my patient about the urgency of using, for example, earlier chemotherapy or even the frequency with which I might be scanning those patients. If I have a patient with a double or triple tumor suppressor loss, even in the absence of symptoms, even in the absence of a rising PSA, I may want to get episodic CT scans, bone scans, or PSMA PET scans in those patients.
Neeraj Agarwal: That's such a wonderful comprehensive description of somatic and germline testing in metastatic hormone-sensitive and castration-resistant, now called as androgen pathway modulation-sensitive and androgen pathway modulation-resistant prostate cancer. And we know many of the therapies, as we have discussed earlier during the discussion, PARP inhibitors, combination therapy of PARP inhibitors with ARPIs, and then immune checkpoint inhibitors, which are known to meaningfully improve survival in these patients, they are available, and unless we test, our patients will never receive those treatments. So thank you, Emmanuel, for taking the time to share your wisdom with us.
Emmanuel Antonarakis: Thanks very much once again.