Notable Treatment Advances in HER2+ MBC

Vijayakrishna Gadi, MD, PhD: I’m going to keep this moving. ESMO [European Society for Medical Oncology Congress 2021] was exciting. There were a lot of neat papers and manuscripts. You know it’s important when The New York Times also picks it up. The public press has also spoken. I’m going to highlight some of these, but I’ll also query with you guys of others you thought were important.

The first one is designated LBA1, or late-breaking abstract No. 1. It was trastuzumab deruxtecan [Enhertu] vs trastuzumab emtansine [T-DM1] in patients with metastatic HER2 [human epidermal growth factor receptor 2]-positive disease. It’s obviously a randomized phase 3 trial called the DESTINY-Breast03 trial. The top line with this data was a dramatic improvement in PFS [progression-free survival]. If you count the zeros in the p-value [probability value], there were 22 of them. The last time we had 22 zeros in the p-value was CHOP [cyclophosphamide, doxorubicin, vincristine, prednisone] vs R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, prednisone], so you have to go way back to find a drug that was that dramatically better for PFS.

Interestingly, OS [overall survival] is very immature data, with neither arm reaching a median at this point, but enough of a difference that there’s a noncredible statistically important change. We’ll follow that data over time as more events come along. It’s clearly looking good. On the other hand, it was also nice to see the toxicity tables very consistent with what we’ve observed before, with nothing scaring you. If anything, the main scary toxicity is the pulmonary toxicity, and there are no grade 4s or grade 5s, with overall hovering around a 10% rate for these toxicities related to pulmonary. That’s consistent with what we saw in DESTINY-Breast01. This is probably game-changing. We’ll get some ideas from other folks here in just a minute in terms of that statement.

Another trial that is related to DESTINY-Breast03 was obviously DESTINY-Breast01. We saw some updated survival results. With more events having taken place, 50% of the patients had expired, so we’ve got a more confident estimate of this. We’re looking at a median OS of almost 2 and a half years in that trial. Keep in mind that this is a group of patients who had seen on average 5 to 6 lines of therapy before. This is after nothing else worked. Here’s a molecule that comes along and keeps these folks able to survive. Duration of response was also very meaningful and different in this nonrandomized single-arm study.

Another molecule was late-breaking abstract No. 15, the TULIP trial. This is another antibody-drug conjugate [ADC], trastuzumab duocarmazine. It’s a randomized study looking at this therapy vs physician’s choice. The physician’s choice was 1 of 3 arms containing chemotherapy and trastuzumab [Herceptin] and a fourth arm where patients were able to do lapatinib [Tykerb] and capecitabine [Xeloda], so no alopecia in that arm either. When you looked at these combinations vs the new molecule, the new molecule had a meaningful difference in the PFS. The p-value met the threshold for statistical significance. Although, if you look at the curves, they spread apart, but they both go downhill, so none of the long tail that you’re seeing with trastuzumab deruxtecan.

Backing up a little, deruxtecan is obviously a topo [topoisomerase] 1 inhibitor and mechanistically very different from duocarmazine, which is a very recent generation alkylating agent that’s connected to trastuzumab. Both of them have this bystander killing effect available to them, so once they enter the cell and get unlinked from the trastuzumab monoclonal, the chemotherapy component can not only kill the cell within, but also get into nearest neighbors and kill those as well. This idea that maybe cancers that aren’t super consistent in their HER2 expression or maybe even have HER2-low expression in some pockets might also be dealt well with both of these molecules.

The toxicity data with the new agent, duocarmazine, caught my eye—no pun intended. There was a lot of ocular toxicity there, with 20% or 30% of folks with serious ocular toxicities. In breast cancer, we don’t do a lot of this, but engaging ocular oncologists and ophthalmologists into our practices to help take care of patients would be a new paradigm for us. Some other oncologists are already dealing with this in other diseases, like myeloma, but that would be an issue here. The last point about this new molecule is that there isn’t a lot of pulmonary toxicity. There was a couple percentage points of pneumonitis associated with it, which is something we’ve got to pay attention to, but not nearly the rate of what we’re seeing with the deruxtecan. Those are the newish molecules and some new data. Mylin and Neil, did you guys see anything that caught your eye at ESMO as well in terms of recent data? I’ll start with you, Neil.

Neil M. Iyengar, MD: No, that was great, VK. That was very comprehensive. Those are the key points from ESMO. I’ll just point out that sequencing is going to be really interesting, because if we think about DESTINY-Breast03, the comparison of trastuzumab deruxtecan with T-DM1, about half of these patients were in the second-line setting and the remainder beyond. And certainly, in DESTINY-Breast01, this is a more heavily pretreated population. If we think about where we’re going to put trastuzumab deruxtecan, we do have data in DESTINY-Breast03 in the second-line setting, but we also have quite a bit of data in later lines.

Similarly, the TULIP trial has a very heavily pretreated population as well, but very few patients in TULIP previously received trastuzumab deruxtecan. When we start thinking about the activity of antibody-drug conjugates after prior exposure to antibody-drug conjugates, we can look at the T-DM1 experience followed by one of the newer ADCs. But how we sequence these new-generation ADCs is something we’ll learn in clinical practice. With forthcoming trials, the landscape is likely to continue changing. My eagle-eye takeaway from some of these new data is that we have some new treatment options. Several of these treatment options are moving up into the earlier-line setting. We’re going to have to consider patient and disease factors in terms of how we sequence these therapies.

The last thing I’ll say is we can briefly cover the TUXEDO trial, which you mentioned earlier, VK. TUXEDO was a single-arm phase 2 study that looked at trastuzumab deruxtecan in patients with brain metastases. These were active brain metastases. Essentially, it was a Simon 2-stage design, and the results from the first stage were presented. The trial is fully accrued. The bottom line is that there was activity. There was CNS [central nervous system] activity response to trastuzumab deruxtecan and enough to allow for progression into the second stage of the phase 2 trial. But the numbers were fairly small. There was a response in 5 out of 6 patients in that first stage, which is reassuring, but we’ll need to see more data to confirm that kind of activity moving forward. Those were my big takeaways from ESMO. Mylin, I’ll hand it over to you.

Mylin A. Torres, MD: I agree with everything that’s been said. You all have done a comprehensive job of reviewing all of the great work that was presented at ESMO. The only thing that I’d add is that there’s the HER2CLIMB-04 study that is currently enrolling, and it’s a great addition and a natural extension of the work that has already been done looking at the combination of tucatinib [Tukysa] with trastuzumab deruxtecan. It’s an open-label phase 2 trial that’s currently enrolling patients who have brain metastasis and looking into whether those 2 agents in combination may have more efficacy than what was seen in HER2CLIMB. The benefit of this is potentially not having to use capecitabine as a backbone to the regimen. Ian Krop is the lead investigator on this trial. When we have those data, we’re going to learn a ton from it. Hopefully it will be of some clinical relevance as we see patients with more advanced disease.

Transcript edited for clarity.