Read the Full Video Transcript

Shilpa Gupta: Hi, everyone. I'm Shilpa Gupta, GU medical oncologist at the Cleveland Clinic, and really delighted to be joined by Dr. Tom Powles from the UK, who will talk to us about EV-302 data he presented at the ESMO meeting 2024. Hi, Tom.

Thomas Powles: Hey, Shilpa. How are you?

Shilpa Gupta: I'm good. Thanks for joining us, Tom. I know you're very busy.

Thomas Powles: Thanks for inviting me.

Shilpa Gupta: I would love to get your take on what you thought was the reason for doing the PD-L1 and the nectin-4 correlative data in the EV-302, and what can practitioners learn from that and apply?

Thomas Powles: Yeah. Enfortumab vedotin is a nectin-4-targeted ADC. When you combine it with pembrolizumab, it's got 70% response rates in urothelial cancer. It reduces the risk of death compared to chemotherapy by 50%. The forest plot analysis showed it worked across broad subgroups of patients—platinum-eligible, etc., ineligible, whichever biomarker you want to use. But we presented data on PD-L1 using CPS, which didn't look discriminatory. I'm not a great fan of the PD-L1 biomarker in urothelial cancer. And I wasn't surprised that there was no enrichment for response or outcome or better survival associated with PD-L1.

What we did here is we looked at nectin-4 by IHC. We scored it out of 300, the key to nectin-4 in urothelial cancer. And actually, the EV data in urothelial cancer looks better than the EV data in some other cancers, in my opinion. Now, clearly, it's more robust. But that may in part be related to the fact that nectin-4 is almost universally expressed in urothelial cancer. When we looked at our score out of 300—and it's important the way we score it is if all of your cells score a positive, that's out of a hundred, and if they're 1+ it's 100, if they're 2+ it's 200, if it's 3+ it's 300. So if half your cells express it, which is quite a lot, having half the cells, and it's only 1+, that only scores 50 out of 300. And actually, we only had 10% of patients scoring less than 150. So almost all the patients had some expression. There are also complicated bystander effects associated with ADC, so it's not thought that all your cells have to express the biomarkers.

Now, there are other biomarkers that are more complicated—HER2, HER3—the expression is not as high, and the same, to some extent, applies for TROP2. But with nectin-4, and in this study, we looked in EV-302, which is the randomized phase three study of EV-Pembro versus chemotherapy. We showed initially, as I said, that 90% of the patients scored over 150. We then went above and below the median. The median was 285 out of 300, which is really high. And as you can imagine, there wasn't any discriminatory effect. The hazard ratios were all in the 0.4s and 0.5s for PFS and OS. So above and below the median didn't seem to make any difference.

The expression in metastatic tissue and in the primary tissue was about the same, again, in the high 270s, 280s. And then the last piece was we then went into that lower population, because although the median was 275, which is probably too high, so we said, "Okay, what happened to that population below 140, for example?" And even there, although it's, say, 10% of patients, the confidence intervals are wide, the hazard ratio is well below 1. So using immunohistochemistry, the nectin-4 biomarker by IHC didn't seem particularly useful. We then combined it with PD-L1, particularly looking at that nectin-4, PD-L1-low score, and again, the hazard ratios were in the 0.4s and 0.5s for PFS and OS.

So I think, overall, the IHC nectin-4 biomarker doesn't pick a group of patients where chemotherapy clearly does better. And I'm very confident in that statement. The German group have done fabulous work looking at amplification and found nectin-4 amplification associated with response to single-agent EV. Jonathan Rosenberg's group looked at nectin-4 in 103 patients by IHC expressions in monotherapy, and the results there are a bit inconsistent.

As I currently stand, and I've spoken to lots of people around this issue, I don't think at the moment we're trying to find a subgroup of patients who should have chemotherapy and not EV-Pembro with this biomarker. I don't think that's the objective here. I think the objective is, moving forward over the next decade or two, we want to find even better therapies. And I think the way we can do that with this next generation of ADCs—maybe DV—we had a meeting at our Uromigos meeting around the next generation of ADCs, and the question is, we saw 75% response rates for DV-Pembro, and the question is, in those patients with lower levels or low amplification, whatever, can we then go in with new, even better drugs and cure more patients?

And so that's where I see this biomarker piece. And I think we learned from the renal cancer days—we had sunitinib. It was transformative compared to interferon. Everyone agreed. And no one was saying, "Oh, let's go and find the patients that benefited from interferon at the time." I don't think there were very many, to be honest. And so we didn't develop biomarkers for sunitinib. And I think with hindsight, that was a mistake, because when the immune checkpoint inhibitors came along, we then didn't have biomarkers to test. And only now are we finding sunitinib biomarkers.

So I think the biomarker piece—and of course we've got a series of new ADCs coming through, and they may need biomarker direction. T-DXd is FDA approved, and that's biomarker-directed therapy. So I'm a great believer in biomarker development at this point. I think it's extremely important. I don't think it should be selecting patients for platinum-based chemotherapy versus EV, but we should be trying to find for the next generation of drugs an opportunity where EV potentially is less active or new drugs are even more active.

Shilpa Gupta: That's such a great answer, Tom, as always, so thoughtful, and I totally agree with you that for EV-Pembro, we are not really looking for the winners. We know most people will do very well, and our goal is to really do something for patients who may not respond early on and whatever that biomarker will be. So as a practical thing, we are not really recommending people check for PD-L1 or nectin-4, but this just sets the story that there's no need to do that. Is that something you'll agree with? Because the community is very obsessed with checking with PD-L1 and different biomarkers.

Thomas Powles: I wouldn't be doing that. I think that's a bit of a waste of time for EV-Pembro. My experience of EV-Pembro is a huge proportion of patients have deep, durable responses. Many of those responses are ongoing. The things that I would be focusing on are, number one, maximizing the management of toxicity, being really alert for early skin rash, early cessation of therapy, intermittent therapy. So skin rash—stop the therapy, re-challenge with a lower dose once that's settled. That's really important. Ascending peripheral neuropathy cycles five to ten dose interruptions, dose reduction—that's really important.

And then the other piece is which patients are eligible for EV-Pembro at the beginning? And I think the answer to that question is actually very simple. It's actually the vast majority of patients. These issues around is diabetes an exclusion? No, it's not. Well-controlled diabetes, EV-Pembro is very safe. In fact, it's probably safer than patients who don't have diabetes because those patients know how to monitor their blood sugar. A very small proportion of patients get hyperglycemia associated with it. It's an important side effect, but it's not an exclusion. And the same applies with renal function. Renal function's been an exclusion for platinum-based chemotherapy, and in some of the trials, because platinum-based chemotherapy was in that, there were some criteria for creatinine clearance. But EV-Pembro is hepatically excreted—well, EV is. And so I think that these issues are not the focus for EV-Pembro.

Now, clearly exclusion criteria for pembrolizumab. So autoimmune, hepatitis, on methotrexate—you're not going to want to give pembrolizumab. A grade 2 peripheral neuropathy after, for whatever reason, neoadjuvant platinum-based chemotherapy—grade 2 peripheral neuropathy—for those patients, you're going to want to be really cautious. I probably wouldn't treat those patients. But then you wouldn't want to give platinum-based chemotherapy to a patient with a peripheral neuropathy, and you wouldn't want to give avelumab to a patient on methotrexate with autoimmune transaminitis.

And then performance status 2. Well, it's difficult, isn't it? There are some performance status 2 patients—if it's cancer-related bone metastasis—you might want to give them a go with EV-Pembro. But then if they're performance status 2 because of a list of comorbidities and concurrent medications—for those patients, there are some patients you and I both know who just missed the boat with urothelial cancer or have too many comorbidities to have treatment. But that applies for platinum-based chemotherapy as well as EV-Pembro.

So I would be looking at this and saying the biomarkers and baseline characteristics aren't necessarily contraindications for EV-Pembro. The more important issue is that I think the combination is transformative for some patients, and it's about getting the right doses, getting the right dose interruptions, and giving the treatment safely.

Shilpa Gupta: Absolutely, Tom. Basically, we have to practice common-sense medicine, right? Like you just said, if somebody has been a performance status 2 their entire life and now they've got cancer, yeah, be cautious. But I have seen patients turning around after cycle one, and I can tell I do not need a CAT scan to say this patient has had a major response. And there's so many such examples.

I also agree with you that we never made a big deal about a diabetic patient getting cisplatin or carboplatin where we give boatloads of steroids every time. We never really got hung up on that. And so I think we need to really focus on the common sense of giving this agent. And just like we've talked before, you don't need to exclude a hypertensive from anti-VEGF drugs, you don't need to exclude other people from other targeted therapies. So I think this whole noise about, yeah, the toxicities are unique and we are good learners and we will learn and learn to give it the way which is the safest for our patients.

Thomas Powles: There's important education and training around this, and I think that is important.

Shilpa Gupta: Very important, yeah. I think right now the key is to reassure the practitioners that it's okay to hold doses, it's okay not to give it continuously. Just don't withhold it for these presumptions, which may not really mean anything. With that, Tom, that is a great lesson. So we don't really need the strict criteria for using EV-Pembro apart from common-sense criteria, and looking forward to seeing that we are able to do more education for all these global launches that now we'll see the EV-Pembro becoming available in other countries.

Thomas Powles: Beautiful, Shilpa. Lovely to see you.

Shilpa Gupta: Thank you, Tom. Have a great day.

Thomas Powles: See you soon. Bye-bye.