Moving Beyond Chemotherapy in Metastatic Triple-Negative Breast Cancer (mTNBC): Are We Clinically Ready? Round 3

Return to the series here.

Kelly McCann: And with that, we're going to go to Round 3, which is, Are ADCs Interchangeable? And we're going to focus on the three that have similar payload topoisomerase 1 inhibitors.

Jason Mouabbi: All right, thank you. All right. So here we're talking about ADCs. So, we're no longer comparing ADTs to chemotherapy. We're just focusing on ADCs themselves. Thank you, Seth. You made my argument very easy. So are ADCs interchangeable? And we're focusing on the ADCs we have today. And the answer is yes, and we have some data to prove that.

Seth Wander: I said they're more similar than different, but still different.

Jason Mouabbi: Okay, more similar. But he did say that if two of them married each other and had a kid, it would be one of them. All right. So the idea is remember that ADCs, they're not that different in chemo from one point, is that it's a better delivery of chemotherapy. ADCs do have chemotherapy attached to them. You're delivering them better. Again, I'm going to use the description you've used so many times, and I love it. It's like a heat-guided missile. So, it's going, and it's looking for a target to deploy that chemotherapy instead of doing it conventionally, throwing it in the bloodstream, letting it go everywhere, hoping, crossing your fingers, that's going to kill the cancer more than the host. So it's a much better way of delivering chemotherapy.

Yet their payloads are the same. All the payloads we have for our antibody-drug conjugates are topoisomerase 1 inhibitors. Because of that, at the end of the day, when you deliver that payload, the delivery mechanism can be different, but you're delivering the same payload that has the same mechanism of action. That's why when you look at different studies, when they try to see, are they that different? You can clearly see from this meta-analysis and the forest plot that there is no single winner among the ADCs. All of them do about the same ones that are being delivered. And that speaks to, yes, the payload, we need a better payload. We need a payload that targets different mechanisms of action, but at this point in time, they seem to be interchangeable, and this is some data to prove it.

The other data to prove it, and this is the nail in the coffin, is that for physicians, they feel that those ADCs are also the same because the way they're prescribing those ADCs, it depends on their comfort. It doesn't matter which ADC is chosen. For them, it's still ADC versus chemo in their minds. It's not like, “okay,” but once they commit to ADCs, they're comfortable choosing anyone. And this is the one that Kelly showed earlier about the confidence in using ADCs. Again, you can see ADCs, not different ADCs. And you can tell that the more seasoned the doctors are in their practice, the more comfortable they are prescribing ADCs. The newer they are, the less confident they are.

Also, when you look at doctors who treat multiple cancers versus those who are pure oncologists, or, let's say, breast medical oncologists, you can see the confidence changes, but this is again talking about ADCs versus chemotherapy, and they group the ADCs together. In their minds, they don't think about different ADCs, side effects, or efficacy. They group the ADCs together and try to compare them because, in their mind, they are also interchangeable. All right, back to you, sir.

Seth Wander: Jason did try to use my own words against me. I said they were more similar than they are different. Now, my wife and I are of average height, and both have brown eyes. My daughter was born with blue eyes and is in the 99th percentile for height by the same logic that Dato-DXd is the child of sacituzumab and T-DXd with quite distinct properties. All right. So this is organic chemistry sort of coming back to haunt all of us. We thought we would never see it again when we were in college. The point here is that yes, there are overlapping elements within these constructs, whether it be the linker, the payload, or the antibody, but there are distinct elements even amongst the three that are most widely deployed today. And as I said, in the future, the next generation, the second, third, fourth generation ADCs will be totally different in terms of linker antibody and payload. So right now we're dealing with the highest degree of similarity we're ever going to get, and there are still important structural toxicity differences between these drugs.

Okay. Speaking of ophthalmology, for those of you who don't have dry eye at the moment, you can try to look at this diagram here, which shows you that the toxicities are, again, very distinct across all these different drugs. And look at this long list of all these different toxicities, and I'm not going to go through all of them in detail, but even though they have structural similarities, there are enough differences to provoke distinct hematologic adverse event rates, distinct rates of pneumonitis, distinct rates of ocular toxicity, distinct rates of rash, cardiac issues, et cetera. So even with those small structural differences, we have very, very different, I think, clinical experiences. And we've been talking about that, who needs preemptive GCSF? Who needs a baseline eye exam? Et cetera.

And then here, to bring the point home, I think the strongest is this concept of sequential ADC therapy. So we'll talk about this probably in our last discussion. This is the biggest area of unmet need, I think, an ADC sort of deployment right now. Can we rechallenge patients who have failed one ADC with another, given their distinct structural and biologic properties? And the answer is, sometimes. Okay. Now, Jason's going to correctly make the point that across the board, most of the time when you rechallenge with an ADC, the second ADC doesn't do as well, but there are important exceptions to that rule. And you can see that here. So on the left-hand side, you have ADC number one. On the right-hand side within each of these graphs, you have ADC number two. Some of them are given sequentially back-to-back. Some of them are given nonsequentially, and you can see an intervening therapy in green.

Now, we're going from SG to T-DXd, and from T-DSd to SG. We're looking at it in ER+, HER2-negative, and in triple-negative disease. And this is from our own institution... Or no, sorry, this is from Laura Huppert's institution at UCSF. We have similar data from MGH from Rachel Abelman. The outcomes here are similar. On average, ADC1 has a longer duration than ADC2, but there are examples, and you can see them throughout these charts, where the duration of treatment on the butterfly plot to the left is shorter, significantly shorter than the duration of treatment on the butterfly plot to the right. Saying that in some instances, patients may fail sacituzumab or T-DXd, and yet when you rechallenge them with the alternative, they can do much better and have longer durations.

And we're still working on identifying molecular biological features that can predict that. This was all progression. They're not stopping for toxicity. So, you can't say they had a short duration of toxicity, switch to the other. These are clinical progression events, yet they're doing better on the second one for reasons that aren't entirely clear. And there's no Dato-DXd in the mix here yet. We haven't generated that data. This is just saci and T-DXd. So as you add more ADCs to the mix, you're going to have a combinatorially more complex amount of analyses to do to look at ADC 1 to 2, to look at ADC1 to 3, to look at 2 to 3, to look at 3 to 2, sequential, non-sequential, et cetera. So those of us working in ADC research will be busy for a long time. And for me, I'm going to watch and read those papers and look forward to those results. All right.

Kelly McCann: All right. So, in the Round 3 discussion, the clinician is evaluating multiple ADC datasets. What trial features should guide interpretation?

Jason Mouabbi: So, I would say it's so hard to do a cross-trial analysis because the patient populations are different. Some studies, for example, excluded patients who progressed within the first six months in the adjuvant setting, or those with TROPION-Breast. We always know the patient: the sooner they recur after an earlier setting, the more aggressive their biology is, the less likely they are to respond to the subsequent line of therapy, or the shorter the duration they can stay on that therapy. So those studies are different. We really cannot do cross-trial analysis. And it's important to look at the patient population and what they selected. What were their comorbidities? How many prior lines of therapy were allowed? And this disease-free interval from the early setting.

Because of that, you sometimes see that each study, by itself, showed better PFS. But you cannot compare them. You cannot say one study showed results above, like, let's say, six months, and the other study showed results below six months, so that means the drug is better. That's not the case. This is a different population. And whenever you deal with a different population, you cannot do those comparisons. They're unfair, and they won't show you the full picture. So, it's always important anytime you see new data, and you want to do a cross-trial analysis, they're highly discouraged because of that. Look at the patient population, and then you can see the differences that account for the response difference.

Seth Wander: I agree. I think there are very distinct inclusion criteria for these studies. The control arms are different. The extent of your crossover point is different. I think you have to take each piece of data where it is, look at your patient, consider their toxicity profile, and account for their comorbidities to make the right choice. I think what you're realizing is we have data to support a lot of these. We didn't talk much about T-DXd today, but we also have data for T-DXd in the HER2-low triple-negative cohort. So I think for a board question, they can't ask you to choose between the three of them because I think-

Jason Mouabbi: No.

Seth Wander: You could argue for all of them; they're all reasonable. You have to make the best choice based on how your patient fits the inclusion criteria, any other medical issues they have, and the side effects you're worried about.

Kelly McCann: I think this next question is interesting in light of how the NCCN guidelines kind of didn't give datopotamab drugs that category 1 preferred label. How should clinicians weigh data maturity and safety experience? Do you think that's what was playing a role there?

Seth Wander: Yeah.

Jason Mouabbi: Go ahead. Go ahead.

Seth Wander: I think it's always hard to know. Certainly, we've had a lot more experience in triple-negative disease, for example, with sacituzumab with or without pembrolizumab. That I think may be part of it. Also, of course, we have overall survival data to see how it shakes out. And to Jason's point, when you have trials that have very distinct control arms, you had doublet chemo allowed on the one, only single agent on the other, crossover on the one, no crossover on the other, that probably comes into their determinations as well in terms of how confident they are in the results at the cut point that they have them.

Jason Mouabbi: And I just want to include one more thing because I agree with what you said, that we shouldn't be penalizing a study because it's not going to show overall survival, because those studies, like the ASCEND studies, allowed crossover, which we should applaud companies that allow crossover and offer the drug for free. Remember, those studies are international studies. In the US, we are blessed to be able to provide those drugs to our patients, even when they progress despite the standard of care. A lot of people in the world don't even have access to those drugs. So, a study that designs a way for patients to get that drug as a crossover after they progress is huge, but it comes at a high cost because it lacks a survival advantage when you use that approach.

And the other study did not include a crossover. So, you're eventually going to see maybe a survival advantage in one and not in the other, but you shouldn't penalize a study that did the right thing for the patients just because they don't see the survival advantage. I want to please caution: what's going to happen if we, as clinicians, start favoring a study that shows a survival advantage over one that didn't, even though the other allowed crossover? Guess what's going to happen? Nobody's going to allow crossover in the future. Who's going to suffer? Our patients. So, we shouldn't be penalizing study because of overall survival if, by design, it was done to avoid detection. So that's very important.

Kelly McCann: The last question, I think, is incredibly important. Why might ADC choice differ between patients? In your practice, do you present multiple medications if they're candidates for that, or do you give your recommendation? What do you do?

Seth Wander: I try to present all the reasonable options, I think. And I might editorialize about why I think one is better for this person than the other. It could be, again, because of their age, or because they live far away, or because it's a Q3 versus a 2-on, 1-off schedule. I mean, there are many reasons. There are also pressures from the institution from infusion capacity and other elements. And you also be honest with patients to say, "Well, we have this drug that's been around for longer, and we have more experience with this drug. We have this drug that's newer."

And I don't think it's ever wrong to verbalize your thought process to the patient and say, "This is where my interpretation of the data is. This is some of the reasons I might choose A versus B." And whether it's ER+ disease, HER2-positive disease, triple-negative disease, I'm sort of putting all options on the table and saying, "We have choice one, two, three, four. This is why I would do A, B, or C." I'm going to say, "I wouldn't do this one, or I would do this one," and let them decide.

Jason Mouabbi: I think it's very important to have options, because different patients may prefer different drugs for various reasons. They can be trivial for us, but very important for them. So, I can tell you, some of my patients, I have one patient who refused to wear her contact lenses. She was like, "I'm not going back to glasses. Absolutely not. I'm not going back to glasses." So then the answer is clear here. Thankfully, I have a choice. I have another choice that I can give her. I have some patients who do not like the GCSF injections. Those patients got it when they got their KEYNOTE regimen, and they were like, "Okay, they've caused so much bone pain. This antihistamine you tell me to take twice a day makes me feel like a zombie all day long." I don't want it. I don't want to get a drug that can give me neutropenia, and I have to take those shots." Then you have an option.

So it's good to have options because patient will choose a drug sometimes because of their side effect. Remember, we're not curing patients in the metastatic setting yet. Quality of life, how patients feel, and their role function are very important. And I spend a lot of time talking about that. Our patients are different. Some of them are moms, some are grandmothers, and some are still leading very active lifestyles. They travel a lot. Some of them have hobbies, such as hiking and doing activities. And you need to tailor it, give them a treatment that fits their lifestyle and their role functioning, because at the end of the day, we don't just want them to survive longer. We want them to continue doing the stuff they like. And that's why it's always good to have options. A different option will better fit the patient.

Kelly McCann: Great.