Germline Testing for Patients with Genitourinary Malignancies - Maria Carlo

February 5, 2020

Together with Alicia Morgans, Maria Carlo delves into the topic of germline testing for patients with genitourinary malignancies. In a more non-traditional approach to investigating the effects of DNA repair defects on prostate and bladder cancer, Carlo factors in not only oncological outcomes, but also psychological outcomes of her clinical trials.


Biographies:

Maria Carlo, MD, Medical Oncologist, Memorial Sloan Kettering Cancer Center, New York, New York, USA.

Alicia Morgans, MD, MPH, Associate Professor of Medicine in the Division of Hematology/Oncology at the Northwestern University Feinberg School of Medicine in Chicago, Illinois.


Read the Full Video Transcript

Alicia Morgans: Hi. I'm delighted to have here with me today Dr. Maria Carlo, who is a medical oncologist at Memorial Sloan Kettering, where she has a focus in genetics. Thank you so much for talking with me today.

Maria Carlo: Oh, thank you for having me.

Alicia Morgans: Wonderful. So, you do a lot of work in genetics, germline testing, and understanding germline in patients across, actually, all GU malignancies, which is really important. And I'd love to hear some highlights in prostate, bladder, and kidney cancer just for viewers to really get a snippet of what's going on in germline testing for patients with GU malignancies.

Let's start with the prostate. I think we have had multiple segments where we've talked about germline DNA repair defects, for example, in advance settings or microsatellite instability, but in the earlier phases of disease, I think it's also important for us to understand the implications of DNA repair defects, for example, or other aberrations that could affect outcome, and there's a lot less information in that space. Can you tell us a little bit about it?

Maria Carlo: You're right, yes. And that's one of the reasons we were interested in looking at the space. We know the effect of BRCA2 mutations, for example, in the advance setting. They're associated with higher risk disease, with worse outcomes. We know the implications for a patient and their family members. But in the active surveillance clinic, for example, we were left with the question, well, what if you find somebody with a BRCA2 mutation or BRCA1, is that person a good candidate for active surveillance? There wasn't really anything in the literature, so together with our colleagues in urology and actually in psychiatry, we came up with a trial where we're going to offer men who are currently undergoing active surveillance, these are men with Gleason Score 6, for example, and offer them germline testing, which otherwise they wouldn't have been offered testing because they don't qualify, they don't have high-risk disease or necessarily family history, and then measure outcomes.

We're measuring several outcomes. First, what's the rate of reclassification when they have another biopsy? There are some preliminary data out of Johns Hopkins suggesting that, for example, BRCA2-positive men have a higher rate of grade reclassification while on active surveillance. But this was really a small cohort retrospective, so we're looking at it prospectively. We're looking at other mutations, so BRCA1, ATM, does that affect outcomes, for example.

Then we're looking at the pathology itself for men who go on to have a second biopsy or even a prostatectomy while on active surveillance. What does the prostate cancer look like for germline mutation carriers versus non-carriers?

And then finally we're looking at patients' reported outcomes in terms of their anxiety levels. For example, we're offering this germline testing, but we're not using it to necessarily direct them towards stopping active surveillance or continuing it. So, if you have an ATM mutation while on active surveillance, does the patient get more anxious? Did they choose to undergo a prostatectomy when otherwise they wouldn't? And we're also asking the urologist, does this change how you think about your patient on active surveillance? So, we're hoping to recruit about 600 patients and then see what the outcomes are.

Alicia Morgans: That is so interesting, because I think so many times in medicine, particularly in oncology, we are very focused on the biology. We're very focused on the data that we can gather, but the data does not necessarily always include behavioral data. So, a really intriguing aspect of that study to understand from a patient perspective and a clinician perspective, how the germline data affects their anxiety and their decision-making. Very, very interesting.

You've also done a fair amount of work in bladder cancer, and this is a field that is about to explode, I think. Tell us a little bit about that.

Maria Carlo: In bladder cancer, we're at a very different point. In prostate cancer, we know what genes increase the risk of prostate cancer. In bladder cancer, there really was nothing out there until very recently. We knew that in Lynch syndrome there's a higher rate of upper tract urothelial carcinoma but not much else. So, we had the good luck that with our MSK-IMPACT cohort we sequenced both the tumor and the germline, and we went back and looked at our first about 500 to 600 patients who had IMPACT. And then in a de-identified manner, because some of these patients had not formally consented to have the results back, so neither us nor the patient would know these results, we looked at the prevalence of germline mutations in patients with mostly advanced urothelial carcinoma, both bladder and upper tract, and we found a surprisingly high-rate of germline pathogenic mutation. About 14% of the entire cohort had a mutation.

Surprisingly, the highest rate was actually in BRCA1 and 2, so we compared to controls, and we were like, well, is this just our population in New York that has a higher prevalence of BRCA1 or 2 mutations? But then we compared the rate of BRCA mutations in urothelial cancer versus controls without cancer and found that still that held for BRCA2, even controlling for different populations.

This is still very preliminary. We don't know what it means. We still don't know what it means for treatment or outcomes, but just having knowledge that there might be a higher prevalence and there might be an association with the risk I think is interesting and can lead to other studies.

Alicia Morgans: Absolutely, and ultimately it could lead to differences down the line in screening protocols and other things when you have families who have BRCA2, for example.

Maria Carlo: Right.

Alicia Morgans: The one other thing that I think is important to mention is that PARP inhibitor studies have been tried in urothelial cancer and, unfortunately, at this point have demonstrated some futility. But I think the most important thing there is to recognize that those failed trials are actually in non-selected populations, and so if we selected patients, we may see a very different result. So, I think everyone should be on the lookout for potential future studies, perhaps with selected populations, and certainly for more information on germline genetics from your group and from others in bladder cancer.

Maria Carlo: Another interesting thing in that study too is that we looked at rates of loss of heterozygosity in the tumor, and there was actually quite a high rate in patients with DNA damage repair mutations. About 40% had either loss of heterozygosity or a second somatic hit in the tumor ...

Alicia Morgans: Wow.

Maria Carlo: ... so that also points towards some role in pathogenicity.

Alicia Morgans: Absolutely. Well, moving on, on our journey through GU malignancies, tell me a little bit about non-clear cell RCC.

Maria Carlo: Right. Non-clear cell RCC, especially in the advance setting, can be a very difficult disease to treat. Usually, it affects younger individuals. So, the same idea in our bladder cohort, we wanted to look at what's the prevalence of mutations in patients with non-clear along with clear cell kidney cancer. So, using our same MSK-IMPACT cohort, we looked back this time at patients, all were patients who had actively consented to have their germline so we could match clinical outcomes to the germline genetics, and the surprising part was that we saw that a high percentage of patients with non-clear cell without any family history of kidney cancer, without early-onset, had a high rate of mutation, specifically in FH, or fumarate hydratase.

Now, this has been known for a long time that it's associated with a syndrome called HLRCC, or hereditary leiomyomatosis and renal cell carcinoma, but those patients had usually been identified because they had a family history, because they had early-onset, or, for example, women have a history of large uterine fibroids.

But looking at a cohort, all-comers, that we were in pre-selecting, we found that the rate in patients with non-clear cell kidney cancer was quite high, over 10%, and a lot of these cases were not identified by pathologists as an FH-deficient renal cell carcinoma. Sometimes they're able, but if you, for example, have a patient who presents with metastatic unclassified kidney cancer, you don't think to screen for germline genetics, and in the metastasis, the pathologists have a hard time identifying what the exact origin is. But if you stain for FH, you find that the tumor is lacking FH, but this was done retrospectively after we told them this patient had an FH germline mutation that we wouldn't have picked up otherwise.

So, in my practice I do, for all patients with non-clear cell advanced kidney cancer, I am recommending genetic screening. There are no official kidney cancer guidelines, although the American College of Genetics does recommend people with non-clear cell be referred to a genetic counselor just to explore, for example, if they have tuberous sclerosis or something, some other syndromes. But I'm offering germline genetics to all my advanced non-clear cell kidney cancer patients just because of this high rate of mutations that you wouldn't pick up by clinical features.

Alicia Morgans: And it could have implications for family members.

Maria Carlo: Certainly. I have a kidney cancer clinic and a genetics clinic, and in my genetics clinic I do see a lot of the sons and daughters of patients who have advanced FH-deficient kidney cancer, and then we screen. We screen for kidney tumors. We screen for leiomyomas of the skin and for uterine fibroids.

Alicia Morgans: So, it sounds like a work in progress in terms of the guidelines, in terms of recommendations?

Maria Carlo: Definitely, and I think there is a need for guidelines, especially now that there's more information available to help the clinicians find out who they should be referring to a genetic counselor or offering genetic testing.

Alicia Morgans: With those implications for families, I think that's going to be really important.

Maria Carlo: Definitely.

Alicia Morgans: Okay. Well, do you have any overarching themes for the audience, because I think germline testing, although it's somewhat challenging for some of us to think about implementing in our urologic oncology clinics, is something that is not just touching prostate anymore and will increasingly be touching the rest of the patients that we treat. Any recommendations or considerations?

Maria Carlo: I think setting systems up in your clinic for counseling, and then after results come back, what the implications of the results are important because sometimes we can send off testing but then when you get a result you need somebody to help explain those results. We've done efforts to do rapid testing in clinic with streamlined education through videos. I think that's the future when we're going to be offering lots of germline or somatic testing for patients.

Alicia Morgans: Absolutely, and clinicians can feel stressed too when they don't know how to understand the data that they see. So, engaging with companies, whoever is the company that is your choice to do these testing and using one maybe that has either chatbots or an online interface or phone interface to counsel the patient and maybe the clinician is going to be important, I think.

Maria Carlo: Definitely.

Alicia Morgans: Great. Well, thank you so much for your expertise and your time today.

Maria Carlo: Thank you.