Ashish Kamat: Hello everybody and welcome to your today's Bladder Cancer Center of Excellence. I'm Ashish Kamat, urologic oncologist in Houston, Texas. And it's a pleasure to welcome to the forum Professor Daniele Raggi, who's joining us now from the Royal Marsden Hospital, London, UK. I've of course known him since his days in Milan, and we've stayed in touch, and he continues to do excellent work. And today, Daniele, welcome. Teach us everything about HER2 and urothelial carcinoma.

Daniele Raggi: Thank you so much, Ashish, for your kind invitation. So I can start with a brief presentation. So we are talking today about HER2 and urothelial carcinoma. So basically, our current understanding and the future directions.

So why HER2 in urothelial carcinoma? I would say because of the expression. So HER2 expression is quite frequent in urothelial carcinoma, potentially up to 68%. So nearly 70% of the samples studied, including the HER2-low disease. Then we have different positivity rates that can vary by anatomical site. So potentially, for example, we have the idea that the upper tract is probably expressing more HER2 than the typical bladder cancer for the lower tract. Then the stage. So by stage, metastatic disease is highly expressing HER2 compared to muscle-invasive or potentially non-muscle-invasive disease as well. We have differences regarding histologic subtypes as well, because we know that the micropapillary tumor may potentially highly express HER2 compared to typical urothelial carcinoma and compared to some variant histologies as well.

And then across molecular subtypes. So we know that basically, the luminal subtype can definitely highly express HER2 compared to the basal one. Then we have a discordance between, unfortunately, HER2 expression in the primary tumor compared to the metastatic site. We know that up to or potentially more than 50% of metastatic lesions will lose HER2 expression. So once again, another reason to retest the patient potentially at progression, after the first- or second-line treatment. And be cautious about the single-site results, because HER2 expression, especially when you are looking at the immunohistochemistry, can be tricky to interpret. This is the biology. It's a well-known biology of HER2 expression, especially in urothelial carcinoma. So what we really know is that basically, HER2 expression can activate and drive growth, invasion, and resistance to any treatment, especially via using two different pathways.

So the PI3K pathway and the MAP kinase pathway as well. We also know that potentially, HER2 may interact with the anti-tumor immunity, so can potentially affect the response to immune checkpoint inhibitors. And we also know, as said, that HER2 can be a dynamic biomarker. So potentially, its status can change. So once again, a reason to try to retest our patient. What about the testing? So definitions matter. So this is very, very important in terms of heterogeneity. First, we can have some issues in terms of fixations. We can have intra-tumor variability. We know that up to 20% of our HER2-positive samples can contain negative foci. As said, we have differences between primary tumor and metastatic sites, up to 55% of patients.

We also have problems in terms of assay discordance. We know that by using immunohistochemistry, we are looking at the protein, basically at the protein level. We can use in situ hybridization, looking at the genes, or potentially use NGS, looking at specific mutations for HER2. We have nice data collected by the MSK showing that, basically, these assessments in terms of assays are not really overlapping. Then we have scoring rules issues. So within clinical trials, especially registrational clinical trials, we have different criteria for immunohistochemistry. Some of them are adapted from breast, for example, in terms of immunohistochemistry expression, or other trials are based on the upper GI criteria. Then what about genomic testing? We definitely need further studies in order to uncover the real role of ERBB2 genomic testing compared to protein-based immunohistochemistry tests. So to keep the story short, we need standardization in terms of universally accepted HER2 immunohistochemistry testing framework, in order to be reliable and reproducible.

What about composite biomarkers? So this is really interesting. We have co-expression frequently with HER2 and Nectin-4, but both of them are highly expressed in the luminal subtype. We have co-expression frequently with EGFR, so HER1. This is really interesting in order to potentially develop specific ADCs, so specific treatment targets. FGFR3 alteration can be mutually exclusive, but we need more data on that. And KRAS alteration or mutation can often drive anti-HER2 treatment resistance. And immune context is largely stage-dependent. So basically, at very early tumor invasion, we are seeing more immune evasion, and less definitely when we have metastatic disease. So these are just data looking at the correlation between Nectin-4 and HER2 expression taken from the JAVELIN Bladder 100 and Tempus datasets. So the interplay of HER2 with other biomarkers definitely may guide sequential strategies and refine patient selection.

Very, very quickly, mechanism of action of the anti-HER2 ADCs. So we are looking at HER2 as a docking target more than dependent on pathway signaling. This is very important as a concept. We have different payloads, looking at trastuzumab deruxtecan or disitamab vedotin. We have different bystander effects and different immune modulation. But the most important question here is how much HER2 expression is enough for us? It's an unanswered question, unfortunately. Therapeutic landscape. As you know, ADCs are leading the way. We have negative or inconsistent data from trials using TKIs or monoclonal antibodies. We have T-DXd now approved, with the agnostic FDA approval for immunohistochemistry 3+ HER2-high-expressing tumors. We have disitamab vedotin FDA breakthrough therapy designation in HER2-positive urothelial carcinoma. And so we have definitely very, very good data on these ADCs, alone or in combination with immune checkpoint inhibitors.

So just to close, in summary. Please, we need to make an effort to standardize HER2 testing—pre-analytics, scoring, and reporting. We consider biomarkers, and especially HER2, as a docking biomarker for ADCs and not only as a classical oncogene addiction, especially for HER2-low or ultra-low-negative disease. We definitely need more data on localized disease, so non-muscle-invasive or muscle-invasive disease. More effort on composite biomarkers and definitely try to standardize consensus-based reporting frameworks. So my typical messages here are: test, please. Test broadly across patients and disease settings. Retest if possible when the disease is definitely getting worse or evolves. Prioritize trial or ADC-targeted HER2 treatment for HER2-expressing disease—I mean expressing at least 1+ positive immunohistochemistry. And please optimize sequencing through trials, especially with the new upfront EV + pembro combination. Thanks.

Ashish Kamat: Thanks so much, Daniele. You went through a lot of data in the HER2 field and of course, there's a lot more in your publication. But briefly, what are you excited about as far as the data that's come out? The data that's coming out, I'm sure, at ESMO and more to be revealed. If you had to look into your crystal ball, do you think we are finally in an era, because of the ADCs, that HER2-targeted therapies are now going to be more mainstream?

Daniele Raggi: Yeah, thanks for the question. I would say yes. I'm definitely more excited about this new era in terms of ADCs in general for urothelial cancer, because we are changing completely the landscape in terms of treatment. You know that EV-pembro is now first-line. So definitely, we are nearly doubling the efficacy outcomes compared to standard chemotherapy. So what I'm expecting to see, definitely, I would say positive data coming in two weeks' time at ESMO, probably from the Chinese study with disitamab vedotin and toripalimab versus standard chemotherapy, carbo-gem vs gem. So I'm really excited, especially because we don't really know how much these HER2-positive patients will respond to the EV-pembro combination up front. So I'm waiting for these data.

Ashish Kamat: So tell me a little bit about the co-expression of HER2 and Nectin. They're co-expressed, but there's obviously some mechanism behind the relative resistance. They're not pan-resistant, but there's something going on. Shed a little light on that based on your understanding.

Daniele Raggi: Yeah, thanks for the question again. Unfortunately, I don't have the crystal ball, so it's very difficult to give you an answer right now. I would say we have data collected from very large retrospective datasets, basically showing high expression of HER2 that is highly expressed in patients highly expressing Nectin-4. These data, I think, are related more to the luminal subtypes that potentially can express both of them. We will see, I would say, at ESMO in two weeks' time, some data regarding resistance to EV basically related to HER2 high expression. So I would say in the future, if possible, I can foresee probably a kind of combination in terms of treatments. So you can try to target basically Nectin-4-expressing tumors along with targeting HER2-positive disease as well. But what I don't really know is if the real upfront combination of anti-Nectin-4 and anti-HER2 is the right way to go, or maybe wait for a sequence of treatments, selecting basically patients not responding to anti-Nectin-4 agents and then treating them with anti-HER2 agents in this context.

Ashish Kamat: Of course, because we have to balance toxicity with efficacy. One of the things that I've been interested in for now going on 25 years is micropapillary disease. It really has a different biology. It tends to have worse prognosis in the non-muscle-invasive space and, of course, in the metastatic space. Many people, and you too, have done work looking at the expression of HER2 alteration, amplification, et cetera, when it comes to micropapillary. There seems to be a good amount of correlation. Share with us a little bit your views on IHC for HER2 versus recognition of patterns of micropapillary, and where you think it fits in. Is it a developmental pathway, is it a de-differentiation pathway? What do you think about HER2-directed therapy for histology-based micropapillary diagnosis?

Daniele Raggi: Thanks for the question once again. I would say, generally speaking, what we saw in this review is that basically, HER2 correlates more often with high-grade disease and a more aggressive pattern of disease. So once again, metastatic disease, unfortunately, is highly expressing HER2. Muscle-invasive is probably expressing more HER2 compared to non-muscle-invasive disease. But if you look into non-muscle-invasive disease, the highly expressing HER2 tumors are CIS. So once again, the more aggressive and potentially the pT1 more compared to the pTa. So the biology behind HER2 is probably showing more aggressive features.

Micropapillary is, unfortunately, by definition, quite an aggressive disease and difficult to treat as well. This is why we found that micropapillary urothelial carcinoma is highly expressing HER2, probably, I would say, through the differentiation pattern of basically the evolving disease, since the very early beginning of invasion of the disease. Regarding treatment, I would say we are living now in an era where we can potentially deliver directly into the bladder new medication. We have seen the TAR-200 and et cetera. So I would say probably for this specific subpopulation, a really interesting approach by using TAR-200, basically a device potentially exposing for a long period of treatment and bringing the anti-HER2 treatment directly into the bladder, especially for this subset, very, very aggressive.

Ashish Kamat: Yeah, I think over the years, what's happened is we've recognized the biology, but the treatments didn't exist. Then you had the molecules, but the delivery didn't exist. Now you have delivery and molecules, better genomic understanding. Hopefully for our patients, everything is coming together at the right time, at the right pace. Thank you so much for taking the time. It was a pleasure exchanging this with you and a Q&A. And thanks again.

Daniele Raggi: Thank you for having me. Thanks.