Read the Full Video Transcript

Petros Grivas: Hello, I'm Dr. Petros Grivas. I'm a medical oncologist at the Seattle Cancer Care Alliance, associate professor at the University of Washington and Fred Hutch Cancer Research Center. Thanks for joining us in the year today, a webinar. I'm very excited today to be joining by two of the experts and pioneers in the field of genitourinary cancers, and particularly with the work, which is very meaningful in genetics of urothelial carcinoma. Joining me today is Dr. Guru Sonpavde, who is the director of the bladder cancer program at the Dana-Farber Cancer Institute and he's on the faculty at the Harvard Medical School in Boston, Massachusetts. Guru, welcome.

Guru Sonpavde: Thank you, Petros, for the invitation.

Petros Grivas: Of course. And we're also very excited to be joined by Dr. Maria Carlo, who's a medical oncologist and she has a joint appointment in the genitourinary cancer program service and the cancer genetics at Memorial Sloan Kettering. And she's an assistant professor there as well. Maria, welcome.

Maria Carlo: Thank you, Petros.

Petros Grivas: Thank you guys for joining me today. And you know, I was really, really eager to talk to you because we have a lot of immersing data about genetics and germline mutations in urothelial cancer, and CAF contributed meaningfully in the field. Let me start with you, Maria, and the work you have presented in conferences and you have published recently. Can you give us a brief overview of the study? And if you can comment a little bit on the practical considerations of the findings of your work?

Maria Carlo: Sure. So, a lot of our work centers on the basics on who with bladder or urothelial cancers should get genetic testing. And before we can answer that question, there really was not much data on journaling genetic and bladder [inaudible] Then, we knew there was an association with Lynch syndrome. So, my initial research and what I presented at conferences about the prevalence of germline mutations and not only in genes which we expect to find mutations like the Lynch syndrome beings but in other cancer risk being BRCA for example. So, we've looked at a large cohort of patients who participated in the impact study, which essentially is a matched tumor germline next-generation sequencing platform that we offer our PDs and either a standard of care. They have advanced cancer to look for, for example, mutations in FGFR3 or as part of a protocol to look at expanded number of genes.

So with this platform, we now look at about a hundred germline genes, include BRCA and many other genes, a lot of them DNA damage repair. And our study show about somewhere around 15% of patients have a clinically significant germline mutation in a cancer risk gene. And interestingly, we saw that a lot of patients had mutations in the BRCA gene and other DNA damage repair gene. So that really was something I think was novel to the field. Now, there might be increased prevalence of BRCA mutations in certain populations. So even correcting for that, we've seen an association, particularly BRCA too, with increased risk of bladder cancer. Now, this is findings in one cohort, right, at Memorial Sloan Kettering. So, I don't think it's ready for prime time to say, all patients with BRCA mutations need to be screened for bladder cancer, but it's certainly something that needs to be looked for. And interesting data from other centers, including Dana Farber, show that association with BRCA and other DNA damage repair gene can increase risk of bladder cancer.

Another part of the research is who should get genetic testing. And we try to find associations between groups that have a higher risk of having a germline mutation and what's come out from our data, is that young patients. So, particularly those diagnosed with urothelial cancer before age 45, have a higher prevalence of germline mutation and not unexpectedly upper tract patients, just because there's a much higher risk of Lynch syndrome in those patients. So practical considerations, we've taken based on our data and data from other centers, we tend to offer germline genetic testing to young patients. So those diagnosed in their twenties, thirties, and forties. Unfortunately, we do see patients with urothelial cancer diagnosed at age 20. We also screen for Lynch syndrome in patients with upper tract cancer. So, I know there's some groups that have been pushing us. We try to do microsatellite instability testing if tissue is available on upper tract cancers, particularly those that are diagnosed with De-novo with upper tract or mismatch repair staining when that's available.

Petros Grivas: Thank you, Maria. A lot of very important work. And I think it's important for the audience to recognize this work and also the importance to look at factors like age of diagnosis. You mentioned a personal history of other cancers and family history of cancers, which in my practice, I find that many patients are struggling with because they don't know the family history. And sometimes they follow up with their family and they say, oh yeah, this cousin had the early onset of the cancer X. So I think it's important to emphasize the skills so that we need to ask the patients for good family history as well. Thank you. And Guru can you tell us a little bit about your work published recently. And if you can comment also on the practical implications as well.

Guru Sonpavde: Thank you, Petros. So our study was a little bit different than the study done by Memorial Sloan Kettering. So, this was a dataset from the Invitae company that does the germline testing for patients that we frequently use at Dana-Farber. So, this was a dataset of 1038 patients with urothelial carcinoma. And so, as you can imagine, patients who have their blood set to Invitae for germline testing are somewhat more selected group than an unselected group that was done at MSKCC. So some are younger with a family history. There was some suspicion that there was something going on at the germ level. So a little bit different, more enriched potentially for germline alterations. So what we found in this data set, it was a pretty big dataset, 1038 patients, 18.6% of patients harbored one or more actionable germline variants. And these were enriched for DNA damage repair variants.

And also we looked at the control data set, which was a data set control for age and gender, and ethnicity, but did not have cancer. And indeed, compared to this cancer-free cohorts, there was enrichment in MSS2, MLH1, BRCA2, and ATM. So, that is the finding from our study, like Maria said, I think that they're not quite sure how to select patients at the moment for germline testing, but in my practice, of course, I am considering it in patients, at least patients who are younger or with the family history of even some cancer, not necessarily a bladder cancer.

Petros Grivas: And that's very interesting because you both mentioned the incidence of germline mutation is much higher than we would expect and we know. And one thing that I tell my patients is the more we dig into that topic, right? The more we perform next-generation sequencing, the more we find the sensitivity goes up. And it sounds like the mutational landscape for germline is, includes of course microsatellite instability, as Maria mentioned in DNA repair, a mutation has been a damaged function mutation. And it's interesting anecdotally, I have seen cases with BRCA2 mutation have very poor prognosis, but we do not have much data, right? Correlating these germline mutations with outcomes of those patients. Is that correct? Maria any comments on this?

Maria Carlo: Right. There's not much data at all. Interestingly though, for mismatch repair, in particular, I think there is definitely emerging data as expected that these patients would do very well with checkpoint inhibitors and in our retrospective reviews, for example, [inaudible] have presented on this. You know, patients with germline mismatch repair mutations tend to respond were small cohorts, but a lot of them don't tend to respond to platinum therapy, but then have near-complete responses or complete responses to checkpoint inhibitors. So I think they may not be necessarily prognostic, but certainly predictive as you expect. I think, unfortunately, the data on things like PARP inhibitors in patients with DNA damage repair mutations hasn't panned out as cleanly. But it's still hard to tell, cause we don't know we've summed up all these mutations. And as we know from the data from prostate and ovarian and maybe that the BRCA2s respond very well but the ATMs don't. And even the BRCA1s respond worse than the BRCA2s, at least in prostate. So, I think that's still something that could be explored further. Prognostically, I don't think we have any data gene-specific germline than worst outcome.

Petros Grivas: And Maria, just to clarify for the audience, that data you quoted by [inaudible] and your group associating DNA damaged function mutation cell response with checkpoint inhibitors.

Maria Carlo: So, that's the difference. That's why I was actually referring to the study looking at microsatellite instability-high patients that they presented in ASCO 2020. But yes, we also do have some studies showing that by [inaudible] that was published in [inaudible] a couple of years ago, showing that there is, it seems to be a higher response and to checkpoint inhibitors and patients with various DNA damage repair alterations. And that study did include both germline and somatic. We've tried to look at the data a little bit more granularly to try to identify if germline per se, versus germline somatic and the numbers just start small for now. So we can't come to any conclusions about germline versus somatic.

Petros Grivas: Very interesting. Thank you, Maria. So there's data, as you mentioned from 2020 with microsatellite instability, MSI high status, and response to checkpoint inhibitors, which is you could expect that there's a significant association and also separate data sets also from Memorial Sloan Kettering Cancer Center with DNA repair gene mutations and potential shares with response to checkpoint inhibitors. And there's also data with response in blood-based chemotherapy with DBR mutations as well. So I think that the question is how we can potentially design clinical trials that we kind of look at this in a perspective way. Guru, any comments for that, the discussion on your end?

Guru Sonpavde: Yeah. In terms of how this information could be used to guide drug development. I think that obviously what makes sense I think is to look at agents like PARP inhibitors in these patients. It has shown activity and benefit in pancreatic cancer, that approach, and many other cancers also ovarian cancer. Retros, you have led a study that looked at Rucaparib in patients with metastatic urothelial cancer, an unselected cohort, as well as a somatic, not germline, somatic DNA damage repair alterations, which unfortunately did not show robust activity, for rucaparib, but that was somatic. So maybe germline, the germline group is a group to look at because potentially the activity might be more robust in this population.

Petros Grivas: Thanks, Guru. And I can remember that we recently published the results of the Atlas trial looking at true Campari, but single agent, as you know, and we had a little bit hard time because there was no confirmed response in an unselected population, regardless of mutations in patients with a platinum or immunotherapy refractory, advanced urothelial cancer. We still have to figure it out. They're all of PARP inhibitors and this negative trial, the Atlas trial was also in the context of the best case study that Professor [inaudible], myself, and others did. And combining the volume of that laboratory sort of about 37% response rate in platinum-refractory disease, second line, however, there was no randomization to see how much of that it was combination versus single-agent. And this circulation was enriched to your point to DDR mutation. So a lot to be learned on that regard and how to utilize PARP inhibitors down the road.

Before we close, I would like to ask both of you, we always advocate for genetic counseling and we don't, we never have enough, right? And I think all cancer centers are trying to recruit more genetic counselors and geneticists. Any practical points about how do you approach patients? Do you refer them as we discussed before if you do germline testing yourselves, at which time point you do that in the patient journey because sometimes patients get overwhelmed? At the same time, you want to address it earlier than later because they could clinical deteriorate. Guru, we'll start with you this time.

Guru Sonpavde: But in the bladder cancer situation, as you know, we don't really have any guidelines to follow, I guess, but I just use my clinical judgment. In terms of bladder cancer patients, as you know, typically they're older in the high sixties, low seventies. So if I see someone younger, younger even than 65 or 60 with bladder cancer, I tend to at least consider them for candidate testing, look at the family history, especially if they have family history or if they had a second cancer in the past. So, I'm just using my judgment at the moment. But I think that more work needs to be done on this issue. I mean, my own feeling is that the field is moving towards germline genetic testing in everyone with at least in advanced cancer, if not all cancers. So I think that's where the field is moving, but we need more data.

Petros Grivas: Thank you, Guru. Maria?

Maria Carlo: Yeah, I definitely agree. That's where the field is moving. So, since there are no clear guidelines and no FDA-approved therapies that would use germline that would alter the first line treatment, I typically talk about genetic testing. If I think it's indicated after the patients have started their first-line treatment, then just to kind of not overwhelm them in the first few visits. But of course, some patients where I have a really high suspicion of something like Lynch syndrome, I, the initial visit, I send them for genetic testing or I have the tumor, honestly, the easiest way is just to do immunohistochemistry on the tumor and you get the results in a couple of days. So, that's if I have a high suspicion of there, or if they're young. We're fortunate that we have the MSK impact protocol that allows for both germline and somatic testing.

And there's, the protocol is designed so there's an education video so the patient does not meet with a genetic counselor beforehand. We can just order tests and then they meet with us or they talk to a genetic counselor on the backend. And if their results are negative, we disclose the results. And most patients don't need to see a counselor. I think that's honestly where the field is moving because it's there's just not enough genetic counselors to meet with every cancer patients. And then in terms of who gets genetic testing. So, I mean like where do I say I recommend genetic testing for everybody young. Again, we define young in our cohort at 45 and below, and we chose that because it's outside the 95% confidence interval, but I acknowledged that's an arbitrary number. And the ideal age we don't know. But certainly, somebody in their fifties and below have a think of it. Family history and to the point of getting a family history, it's very hard.

Most people don't know what kind of cancer their grandparents died from. Many years ago, it's cancer of the stomach. So I tend to, if they say advanced cancer, they don't know, I presume it could be Lynch-related cancer or BRCA related cancer. But for urothelial cancer, almost always, if it's Lynch-related, there is a family history of something. It's rare when it's a new diagnosis of Lynch syndrome that comes as a complete surprise. So usually those you can get a sense from a little bit just family history of any cancer. And again, I don't get into the nitty-gritty details of is it colorectal or stomach or ovarian versus endometrial. Cause I think most people don't know.

Guru Sonpavde: And Maria correct me if I'm wrong, but everyone with microsatellite instability in the [inaudible] needs germline testing. Is that right?

Maria Carlo: Yes. So would strongly, we do everybody who has MSI high, we send for genetic testing. So we looked at our data and of all patients with urothelial cancer who had microsatellite high or microsatellite intermediate and MSI intermediate. So, we define for most score of three to 10 in our emphasize sensor. So, we looked at all those patients and look up all their germline and 40% of patients with urothelial cancer who have MSI high or MSI intermediate tumors have Lynch syndrome. So, it actually was the two out of all the tumors looked at colon, endometrial, gastric, urothelial cancer had the highest hit rates. So interestingly, so for colon cancer, for example, most MSI high tumors are somatic or sporadic and they get an acquired mutation, but for urothelial, it's 40%, that's really high. So if you have a patient with MSI high, you should strongly, strongly counsel them about germline testing.

Petros Grivas: And that's a very important take-home point for the audience. I think to your point Maria and Guru, you have someone with MSI high and intermediate, definitely this 40%, it's a high number that can be germline and so definitely worth sending all those patients for germline testing. And to your point, younger patients and people kind of use a cutoff. What defines young it's arbitrary, as you said, Maria, and relevant, personal history of cancer or family history of cancer. And the question is how much you can get this information in the clinic. Sometimes you can ask the patients to dig a little bit more and come back.

And obviously the implications, we talk about a day, mostly for the patients but there are also many implications for the broader family and the so-called cascade testing that can happen. And this can lead to prevention and screening tests for other family members in an effort to diagnose early or prevent cancers in the broader family. So that's another point that they make in clinic. So, I can discuss with patients and many of them said, oh yeah, I can save my kids or my cousins or my siblings. Then, they are very open to do the genetic testing.

I would like to thank you so much for all the work you do in the field. And I think the data will keep increasing. I agree with you that the field is going towards germline testing in general. And I think we'll probably get there one day across the board. And again, the implications for the families of those patients, that triggered opportunities for research and how we prevent cancers in the families are established in prostate cancer, like [inaudible] and others that look at us asymptomatic carriers of germline mutations in the family of patients with cancer in the control prevention. So thanks again for your time. And I'm looking forward to seeing you in the future and discussing more about this.

Guru Sonpavde: Thank you, Petros.

Maria Carlo: Thank you.

Petros Grivas: And thanks to the audience for the attention.