Key Takeaways from Clinical Trials of Trastuzumab Deruxtecan in Different Populations

Dr. Ian Krop: Folks may remember that back in 2019 there was the DESTINY-Breast01 trial which was a phase two study that looked in the post T-DM1 setting. In fact, these were very post T-DM1, these were very heavily pretreated, median of six prior lines of therapy in the metastatic setting, looking at the then new antibody drug conjugate trastuzumab Deruxtecan. This kind of second generation ADC which was- it differed from T-DM1 in that it had a topoisomerase inhibitor payload as opposed to a microtubule inhibitor payload. And it had this by- it introduced the idea of bystander effect, meaning that because the payload molecule was membrane permeable and it had a cleavable linker, once it got in- once the HER2 positive cell was targeted by the conjugate, the payload could then diffuse out and kill neighboring tumor cells, even if they didn't have a lot of HER2 expression. And in that phase two DESTINY-Breast01 trial it was clear that T-DXd or trastuzumab Deruxtecan had substantial activity, a response rate of about 60%, progression free survival up around the 18- or 19-month range. So clearly active drug. But because- and then that led to the accelerator approval of T-DXd in- at the end of 2019, actually about two years ago exactly. And actually three years ago, time flies. But because it was a single arm phase two trial, it needed confirmation and that was the rationale for doing the DESTINY-Breast02 trial, looking at the post T-DM1 population comparing trastuzumab Deruxtecan versus treatment of physician's choice, which in this trial was either capecitabine with trastuzumab or with lapatinib. And this was the first analysis of this trial and the- and it was a positive trial. T-DXd was superior to the treatment of physician's choice arm, progression free survival was about 18 months with T-DXd versus about eight months with the treatment of physician's choice, hazard ratio was 0.35. So clearly a very positive trial. Overall survival also favored T-DXd which was substantially statistically significant even at this early analysis. Curve separated pretty early in terms of survival. So the difference in survival was manifest even at that one year time point. So all favorable data supporting T-DXd over treatment of physician's choice. We- of course, whenever you're talking about a study of T-DXd, you have to look at interstitial lung disease because that is a known toxicity of this drug. In our study, the rate was about 10%. There were two patients who unfortunately died from T-DXd related ILD, 0.5%. But again, it just reinforces that this is a toxicity that you have to pay attention to, you have to monitor when you're using this drug so that we prevent those serious events and pick it up early. So I think the take home message from the DB02 trial was that it did clearly confirm the efficacy of T-DXd in that third, post third line, post T-DM1 or later setting. But in practice, I don't think it really changes our clinical usage of T-DXd and that's because of the DB03 data, which were also updated in the same session at San Antonio. So DB03 is- was the second line study comparing T-DXd directly head-to-head to T-DM1, which was the standard in the second line prior to that trial. And the data had presented about a year and a half ago at ESMO initially, showing that that was also a very positive trial. T-DXd was far superior in terms of progression free survival to T-DM1. The hazard ratio was 0.28, which I don't think we'd ever seen a hazard ratio that low before. And then that led to the approval of T-DXd in the second line and all the guidelines basically supporting using it in the second line. Sarah Hurvitz at San Antonio this year updated those results, and they become- I think they're even more compelling now that the hazard ratio in her updated results still is around 0.35 or something in that nature. But the PFS was over 28 months with T-DXd. And again, these are still pretreated patient pop- these were mostly second- or third-line patients. And again, just dramatically improved progression free survival with a PFS of, as I said, I think about 28 months. And actually, the duration of- and the response rate was about 80%, which also was very good. But the duration of response, so if you're one of those 80% of patients who gets a response to T-DXd, the median duration of response was 36 months. So, three years, which, again, these are numbers that just would have been kind of, I guess, unimaginable prior to this trial. All, of course, is good news for patients. And again, the reason why I don't think that DB02, post T-DM1 data are relevant is because you really should be giving T-DXd in the second line because of this- the very substantial response rates, and duration of response, and long PFS, the drug should be used in the second line whenever possible. Obviously, if you have a patient who's already had T-DM1 and you're looking for a third line option, then the DB02 data would- clearly shows that T-DXd is very active in that line of therapy as well. But given the strength of DB03, the second line is really where T-DXd should be used. And again, I think that's what the guidelines would support.

Dr. Charles Geyer: And I think the other thing about Sarah's update that was good to see is that while the ILD, with that long therapy, the rates of ILD slowly increase from 10% to 15%, but on this study still no four or five events. So it- there is a sense that that recognition, early monitoring, appropriate therapy can mitigate it. Though there are occasional patients, as you alluded to in yours where that's unfortunately not the case. So remarkable drug, just absolutely remarkable drug.