Round 2—Can Institutions, Payers, and Clinicians Pivot to Adopt Anti-Drug Conjugate (ADC)-Based Regimens in Everyday Practice?

Return to the Series Home Here. 

Tiffany A. Traina, MD: Round 2—here we're going to have Sonia back up, and we're going to talk about, as clinical trials are validating this ADC approach, can institutions, payers, and clinicians' guidelines all pivot fast enough to get this care to our patients? 

Sonya Reid, MD, MPH: So, Tiffany did a great job at priming us for this question. I think that when she mentioned earlier the data that showed us that so many of our patients in the real-world setting are not getting to second-line therapies, when we think about the current first-line standard of care, we know that that's pembrolizumab plus chemo, as Seth mentioned. But again, as he highlighted, the median PFS is 9.7 months with a median OS of 23 months, which is great. But again, what we saw from ASCENT-04 was that we had an 11.2-month median PFS. We see superior PFS; of course, we're not comparing across trials, but just to put it in context. 

More importantly, we have spoken about how well these ADCs work. With the current landscape, our patients have to wait until the second line or beyond to be able to receive these agents because again, we see that 49% of our patients from, again, a real-world setting may be a little higher than what we see in the academic center. When you look at this data from the Flatiron database, what we're seeing is in the real-world setting where patients receive care; 80% of our patients are receiving care in the community. We see that 49% of patients did not receive second-line therapy. Then, more disheartening, 34% of patients die before receiving second-line therapy in the real-world setting. I think this just argues to the point of really for us advocating, getting our best drugs up front to our patients, knowing the data that we saw, which is immunotherapy plus ADC, which has shown us that that has improved activity. 

This graph is really just showing the same data, but again, just showing the OS. The real-world OS curves here. I think when you look here—only showing the data in reverse—essentially, but what you see from the median OS here, real-world setting, again, that's 11.3 months, which is even more disheartening compared to what we saw in a clinical trial setting, which was 23 months. We see that the numbers are even lower in the real-world setting. But again, one of the things we have to think about is the cost from a side effect standpoint of combining an antibody drug conjugate with IO. I think that's one thing. Of course, we always think about efficacy, but how does that affect our patients? I think what was good to see from ASCENT-04 is that the safety profile seemed to be very consistent with the known safety of each of the drugs, not anything we can't manage. Neutropenia GI side effects were really the top side effects that kind of trickled to the top. There were no new safety signals, which I think was also important to mention. But, also, we saw low rates of AEs leading to treatment discontinuation or dose reduction, which I think was quite good to see at least from this early safety analysis. Back to you, Seth. 

Seth Wander, MD, PhD: Okay. So, in terms of getting these drugs into our clinical practice, I think there are various factors we need to think about. First is the toxicity experience, as you just mentioned. I think that is worth discussing. While there are no new safety signals, I think there may be some evidence that there are at least increased rates of lower-grade toxicity when we're combining ADCs with IO. Luckily, it doesn't seem like there are higher rates of immune-related toxicities than we can see from Keytruda or pembrolizumab. I think this is an accurate depiction of what our current therapeutic approach is for metastatic triple-negative breast cancer, which, as we're talking about today, could change at any minute. But as of today, if you have a patient with metastatic TNBC and they are PD-L1+ and they've had a reasonable treatment-free interval from adjuvant therapy, these patients are going on to the KEYNOTE-355 regimen. 

If they are less than 12 months disease-free. I think it depends a little bit on what they got in the adjuvant setting. Of course, these patients now who have rapid progression on KEYNOTE-522 really present a unique clinical challenge for us. I'm going to give it to my opponent here that, absolutely, in that scenario, I would feel a little bit more compelled to go right to the ADC. If we could include the pembrolizumab, that's great. I would argue we don't know much about how much mileage we can get from immune therapy in the setting of near-term recurrence on KEYNOTE-522. I'm curious to hear what your approach is today in the clinic. Most of us would've been given sacituzumab monotherapy. Would you now try to give the pembrolizumab? Do we know anything about how much benefit? But for the purposes of this, I think if you have a recurrence within 12 months from adjuvant chemo without immune therapy, then certainly going to a more aggressive version of KEYNOTE-55 with, for example, doublet chemotherapy plus pembrolizumab for PD-L1+ patients, is reasonable if the patient has a BRCA mutation. 

We can obviously work in PARP inhibitor therapy hereafter, typically, first-line treatment if the patient's PD-L1 is negative. Again, I think this is an important area of unmet need. If they're BRCA wild type and PD-L1−, we're giving them the same drugs we've been giving them for decades, and it's kind of the dealer's choice of cytotoxic chemotherapy. If they have a BRCA mutation, again, we're going to think about a PARP inhibitor therapy. Again, in the second line, for those patients who are progressing on frontline chemo or chemo with IO, we would typically deploy sacituzumab. If we take this ASCENT-04 approach, that shifts this entire algorithm on its head. I think it is worth having a discussion here in the next few minutes about what we would do in the setting of progression on frontline sacituzumab and pembrolizumab. Then, I think that, in the third line and beyond, this is, again, a real area of clinical development and an unmet need. We have many older conventional cytotoxic chemotherapies that we could use other checkpoint inhibitors or checkpoint inhibitors that could be deployed for a PD-L1−TMB high or MSI-high patient. Although again, in my experience, these patients who are PD-L1− with high TMB don't necessarily always demonstrate huge amounts of clinical benefit. 

I have here a couple of the toxicity tables to look at, but I actually think the version that Sonya had is a little bit more helpful. We can visualize the relative frequencies, but this is from the original ASCENT trial. Those of us who spend a lot of time in the clinic are used to the toxicities that we typically run into with sacituzumab. I think the way I describe these drugs to patients, with the exception of T-DM1, which is sort of on its own as an ADC, but thinking about trastuzumab, deruxtecan, Dato-DXD, sacituzumab-govitecan, they are somewhere between sort of your targeted therapy, hormonal therapy, and doublet cytotoxic chemotherapy, right? They're more toxic than our easiest therapies, but they may not be as toxic as doublet cytotoxic chemotherapy regimens. We can see that here with fairly frequent low-grade toxicities, including fatigue, GI-related issues, cytopenias, and some grade three toxicity in the second-line sacituzumab population. 

Remember, here these patients have already had prior cytotoxic chemotherapy, so they may have less bone marrow reserve, and they may be coming into treatment a little bit more compromised than the patients that Sonya just showed us from ASCENT-04. Here's doing the thing that Tiffany said we can't do, which I completely agree with, putting trials next to each other and comparing across KEYNOTE-355 vs ASCENT-04. I do think, globally speaking, it's true that there's not a huge new signal for toxicity. I actually think, going back to Sonya's slide, this is a little bit easier way to visualize it. I will argue, because I have to, that on the blue side, if you count all grade toxicity, there are more frequent toxicities with SG and pembrolizumab. Most of them are not high grade, I'll give you that. But there are higher rates of cytopenia, and for example, diarrhea and nausea when we're giving the SG with the pembrolizumab. 

In terms of the second part of this debate, thinking about clinical access and deployment may put some additional strain on our teams in terms of managing toxicity. I think we would have to be ready for that in our clinical practices. Should we move here? So again, to argue on the side of chemo, we have positive survival data. Currently, I don't think we have a great approach to second-line therapy for patients with SG + pembrolizumab. So, we don't really know what the overall PFS one, PFS two would be here. I have to argue that there's more toxicity numerically for the combo of ADC and pembrolizumab, at least with low-grade toxicity in the first-line setting.