Disclaimer: This video transcript has not been proofread or edited and may contain errors.
Dr. Aditya Bardia:
Welcome to the ASCO Post Roundtable Series on Novel Therapies for Advanced Breast Cancer. I'm Aditya Bardia, breast medical oncologist at Massachusetts General Hospital, Harvard Medical School. Joining me today are two of my esteemed colleagues. Dr. Wolff?
Dr. Antonio Wolff:
Hi everybody. Great to be here with you today. My name is Antonio Wolff. I am a breast cancer medical oncologist at Johns Hopkins in Baltimore.
And Dr. Tolaney.
Dr. Sara Tolaney:
Hi everyone, I'm Sara Tolaney. I'm a breast medical oncologist at Dana-Farber Cancer Institute and really looking forward to the discussions this morning.
Dr. Aditya Bardia:
So today we'll be discussing recent updates in the treatment of advanced breast cancer and how to integrate these new developments into three patient scenarios. Our last installment will focus on treatment of relapsed/refractory triple-negative metastatic breast cancer. So let's get going. I'll start with a patient story.
This is patient SD. SD is a 35-year-old female with metastatic PD-L1–positive triple-negative breast cancer, HER2 IHC 1+. She experienced disease progression on first-line therapy with carboplatin plus pembrolizumab, has excellent performance status, no organ dysfunction, had germline BRCA status evaluated, which was negative. So the next question is, what should be the optimal management in this setting for this patient with PD-L1–positive, HER2 IHC 1+ metastatic TNBC who's had disease progression on one line of therapy? Sara, how would you approach this scenario and what therapy would you consider next?
Yeah, it's nice that we have so many choices now. A few years ago, really all we had in the pretreated metastatic triple-negative breast cancer setting was using sequential single agent chemotherapy. But now we actually have two ADCs to choose from in this second-line setting, which is really nice. So this patient has HER2-low metastatic triple-negative disease. So we'd be thinking about either using sacituzumab or thinking about T-DXd here. My personal preference for choosing between these two would probably be going to sacituzumab as second-line therapy really because the data behind sacituzumab comes from ASCENT, which Aditya, you were certainly very involved in leading. But really as a phase III study that established that sacituzumab significantly improves both progression-free survival as well as, in essence, almost doubles overall survival. With T-DXd, we have data from DESTINY-Breast04 however, it was only 58 patients in that trial that was a small subset of patients with metastatic triple-negative disease.
So the study was not powered to look at outcomes in the triple-negative subgroup. And while the data is very impressive in that setting, really improving PFS and OS and having a 50% objective response rate, very consistent with what was seen in the hormone receptor–positive setting, I think the challenge is it is a very small dataset. And so because of that I tend to go with the level one evidence suggesting to use sacituzumab second-line from ASCENT and then thinking about T-DXd subsequently in this particular patient. But again, nice to have these choices and I don't think there's a wrong or a right here. We certainly don't have enough data, no comparison data for example, between the two ADCs and not large phase III data for T-DXd in the phase III setting to even attempt reasonable cross-trial comparisons.
And I think also to point out that the T-DXd was studied a bit earlier than sacituzumab was in a sense. And so I think one has to be cautious when trying to do these cross-trial comparisons as we all do when we're not supposed to be. But sacituzumab in essence was a third-line and beyond setting. There were patients who were treated in the second line, but there were patients who had to have had early relapse. So that's different than T-DXd , which was one to two prior lines of chemotherapy. So again, very hard to juxtapose the data and say one agent is better than the other.
That's a good point, especially related to median PFS because that can be dependent on lines of therapy and if you look at hazard ratio, the hazard ratio on ASCENT was very similar to DESTINY-Breast04, so that's an excellent point. Antonio, do you have a similar approach? Any thoughts on this case?
Yeah, no I do and I'm going to reminisce a little bit and even go back before we had the availability and access to the antibody-drug conjugates because I think it has always been incredibly frustrating to figure out exactly which is going to be the next single-agent chemotherapy regiment we are going to use. I mean we know that triple-negative is negative for the markers we usually test for. What we don't know is what is it positive for? And we have been crazily surprised patients who go on to receive second-line traditional single-agent chemotherapy with rapid progression and then we switch to a third line and then we see enormous activities. It's almost like playing the lottery. Having said that, clinical trial evidence is clinical trial evidence and we absolutely have sufficient evidence at this point to move very quickly the antibody-drug conjugates to the second- and third-line settings because the drugs are more active, on average, than traditional chemothermal agent chemotherapy, on average.
And the challenge I think Sara summarized quite well, the evidence we have a little more evidence with sacituzumab govitecan, a little less evidence thus far with trastuzumab deruxtecan. I think it's just a matter of time before we have more of these evidence. I think the study is looking for biomarkers, predictive. I would say positive predictors are always more difficult. Negative predictors of response are always easier to develop. So that's what I hope to see in the near future. But a key question for which we don't have a lot of data now is can we adequately sequence ADC1 followed by ADC2? What is the optimal sequence? Who are the patients more likely to benefit? Does duration on the ADC1 versus ADC2, does that make a difference? And this is work that some of you, you especially Aditya, have looked into this recently in your poster presentation at ASCO.
So I think there's a lot to be learned here. And then the other question which is just thinking about clinical practice, some patients have toxicities and maybe I want to turn back to both of you in terms of one, Aditya, with trastuzumab deruxtecan, any evidence for the patients who had dose reduction after initiation at full dose? Do we see maintenance of clinical benefits? Meaning should we maybe start at full dose but very quickly dose reduce if we see toxicity without fear of losing clinical activity, patients are always fearful of having dose reduction. So what have you seen in your data?
Good question. So this was specifically evaluated in the ASCENT trial looking at patients who had a dose reduction versus not. And even in patients who had dose reductions, the benefit was maintained. It's tough to compare in the same trial patients who have dose reduction versus not because of the time bias, but at least what you can see is those who had dose reductions still had benefit. So that is reassuring. The other thing that was seen in the ASCENT trial was patients who are elderly or more than 65, if you will; in that subset, also the benefit with SG was maintained, they just require more dose reductions. So clinically I feel very comfortable reducing the dose if a patient is having toxicity because in the metastatic setting, our goal is twofold. We want to prolong survival but also want to maintain quality of life.
So that same vein for with an older patient who may look great in terms of labs but may have kind of a narrower band in terms of tolerating toxicity, is there a scenario where you may start a lower dose of saci-govi?
I've generally not done that, but I guess you could make a case if you're really concerned about a patient to start at a lower dose and then you can increase the dose. I think that's where the art of oncology comes in. You have to ultimately see what would work best for the patient. Sara, your thoughts?
Yeah, it is tricky. I actually had a patient just a few weeks ago request to be started at a lower dose. I think she had had so much trouble with her prior line of treatment that she was anxious about a new line of treatment and just said, "Can we just start low and you can always bump it up." And she did and has been doing well and I think it is not an unreasonable approach. I think I also just try to think about other things I can do to prevent toxicity. So with sacituzumab for example, if someone's had a lot of neutropenia on their prior line of therapy and maybe required growth factor or not, that's someone I probably would start on prophylactic growth factor because the rate of neutropenia is just so high with sacituzumab. For others, if they've not had issues with neutropenia on previous lines, then I just wait and see what happens and initiate growth factor as needed.
Certainly thinking about loperamide as needed is really important to instruct patients on as well. And so any supportive measures that we can provide for these patients I think are really critical. And similarly for T-DXd , really thinking about antiemetics is quite critical. Now we're using three-drug antiemetic prophylaxis with T-DXd , which has been really helpful and I find works great. Olanzapine for that late nausea that sometimes happen with T-DXd also is wonderful. So I think we've learned a lot about how to help patients through these treatments.
Sara, thoughts on pneumonitis with T-DXd and you had a very nice review article talking about management of pneumonitis related to T-DXd and I think it had 3S or the 4S approach.
5S, I think the idea here was just it's really important to be aware of this as a potential toxicity for T-DXd. We've seen 10% to 15% of patients who get T-DXd in the metastatic setting do develop pneumonitis. Thankfully as we've been giving this drug more and more the rate of grade 5 ILD, so people dying from ILD has gone down, which is very, very important. But still that isn't zero. And so I do think we all need to be very well aware of this. We need to let patients know that they do need to let us know right away if they develop any signs and symptoms of pneumonitis. So I do tend to image patients more frequently if they're on T-DXd than I would if they were on other treatments. I will say I am scanning people every 6 to 9 weeks if they're on T-DXd . In the DESTINY studies, they were scanned every 6 weeks.
I think the thought here is if they developed any ground-glass changes on imaging, at least you'd catch it early and maybe when they had asymptomatic ILD by scanning so frequently. In truth we don't really know that because we've not had data looking at less frequent imaging to know if it would be different. But that being said, I am a little anxious about it so I do tend to scan people quite regularly if they do develop grade 1 pneumonitis. Very important to hold the drug and it's up to the physician to initiate steroids or not in grade 1 pneumonitis. I will be honest, I typically do, as I'm just trying to get that pneumonitis to resolve as quickly as possible. And then I typically will reimage them at around the 4-week mark and see if they've resolved on imaging. But I'll be honest, I do think pneumonitis on imaging takes a while to resolve and I'm not usually getting it to resolve at that 4-week mark and usually then I have to reimage again.
And then you do have to dose modify if it takes longer than 4 weeks to resolve when you restart. But it's hard because you're getting anxious while you're keeping someone off drug for that pneumonitis to resolve. So it's a challenge for us right now and with grade 2, so symptomatic ILD, you do have to discontinue drug, which is quite hard because these patients are often benefiting from drug. And so to take away a drug where they're having benefit isn't our usual with grade 2 pneumonitis with other drugs we do tend to treat and then retry usually with dose modification like with pembrolizumab, everolimus, things like that, we’re used retrying but not with this agent at this point in time. So I think we have a lot to learn with pneumonitis. We don't have great predictors, we don't have perfect monitoring tools and hopefully we'll improve over time.
Well that's a great point and maybe initially it was because of the concern of mortality with grade 5 events seen with T-DXd with has been used more. Maybe there's a subset of patients with grade 2 where it can be used again.
I'll be interested to see. I do think this really needs to be further studied. I wonder if we'll have some change on PFTs or something that's going to tell us it's safe or not safe to re-challenge. We really need to understand this more. So I think a very good point that that is a critical question for us right now.
Before we wrap up, any additional thoughts Antonio, related to these therapies or upcoming directions?
No, I think this is a great summary and I learned a lot from you guys today. I think the key issue for us is what we discussed at the very beginning. We know that these cancers are negative for the usual markers and the challenges to figure out what the markers might be. So learning about the development of somatic changes and starting with blood and if you don't see anything blood go after tissue. Because sometimes you have a population of patients that may be biomarker negative in blood, but you could find something interesting in tissue that could guide you. We have data, this patient had the germline BRCA negative profile or normal profile, and so the use of PARP inhibitors would not be an option. But we know that patients who develop somatic mutations in BRCA1 do have clinical benefit from drugs like olaparib.
This is actually known the NCCN guideline on the basis of the study led by Nadine Tung, TBCRC 048. A lot to be learned. I think there is a number of markers out there. We need to learn more what to do with checkpoint inhibitors and even in terms of maintenance, if the patient is having disease responsiveness upfront or continuing after disease progression, which we at this point would not do, we would stop. Whether there are scenarios where some patients who haven't developed major toxicities could benefit from continuing a checkpoint inhibitor. I think it remains to be seen. It's a big black box right now.
Thank you, Antonio. You bring up good point about there's interest in both novel therapies as well as novel biomarkers. Great. So I'll summarize with key clinical takeaways.
At this time, pembrolizumab based therapy is the preferred first-line therapy for patients with metastatic triple-negative breast cancer that is PD-L1–positive. In the second line and plus setting, sacituzumab govitecan is the preferred therapy for patients with metastatic triple-negative breast cancer regardless of PD-L1 expression or trop2 expression. Similarly, trastuzumab deruxtecan is approved for HER2 IHC 1+, 2+ plus metastatic breast cancer, both in hormone receptor–positive and triple-negative breast cancer setting. At this time, the role of ADC after ADC is unclear, but subject to ongoing investigation in registry studies and clinical trials.
This brings us to the end of this case. Please see the other segments for further discussion about the latest data in metastatic breast cancer or visit ASCOpost.com. Thank you for joining today.