Extended follow-up with pembrolizumab/chemotherapy for patients with previously untreated PD-L1-positive metastatic triple-negative breast cancer (mTNBC)

DR TRAINA: Hi there. I’m Dr Tiffany Traina, a breast medical oncologist at Memorial Sloan Kettering Cancer Center in New York, and I’m so happy to be speaking today in the Oncology Today program about optimizing the management of metastatic BRCA-negative, triple-negative breast cancer.

And to kick this off I’m going to start with the first-line setting to really establish KEYNOTE-355, which supported the use of immunotherapy in this first-line setting for patients who have PD-L1-positive breast cancer. Here, this study, as you’ll recall, recruited women in the first-line setting with triple-negative breast cancer, and they were randomized to either standard-of-care chemotherapy with taxane or gemcitabine/carbo versus chemotherapy with pembrolizumab. And patients were randomized in 2:1 fashion, and the primary endpoint was progression-free survival in the PD-L1-positive subset, but all patients were included on this study. And secondary endpoint was overall survival.

And as you recall, these data were really practice changing for patients who had PD-L1-positive tumors with a CPS of 10 or greater. The addition of pembrolizumab to chemotherapy significantly improved progression-free survival, with median PFS of upwards of almost 10 months as compared to 5.6 months with standard-of-care chemotherapy.

When we look at the forest plot of subgroups deriving benefit, you can see most groups really derived benefit from adding pembrolizumab, although I’ve highlighted for you an area that really remains a high area of unmet need, and these are the patients who have short disease-free intervals from their adjuvant treatment. Those women that recur within a year unfortunately are doing rather poorly and do not appear to derive additional great benefit from pembrolizumab over chemotherapy.

In addition to the PFS benefit, we have seen a significant overall survival benefit with the addition of pembrolizumab, a small numerical improvement in the intention-to-treat population, as I’m showing here, but more significant benefit in OS in the subset of patients that had PD-L1-positive tumors, defined as CPS greater than 10. And so this is really demonstrated almost a 30% improvement in overall survival, and establishes pembrolizumab plus chemotherapy as the first-line standard of care for patients with PD-L1-positive tumors. Now, those patients who have PD-L1-negative tumors and are BRCA wild type really remain having standard-of-care chemotherapy with taxanes or platinums as that first-line option.

ASCENT trial: Survival advantage with sacituzumab govitecan versus physician’s choice of chemotherapy for patients with relapsed/refractory TNBC

DR TRAINA: We’re going to spend the remainder of our time together really talking about antibody-drug conjugates in the second-line setting and later, but I will say there are a lot of exciting studies looking at moving up those ADCs into the first-line space, and I look forward to seeing emerging data there.

So the first ADC I’d like to talk about is sacituzumab govitecan, and this was the first antibody-drug conjugate that we have approved in triple-negative breast cancer. Sacituzumab is an IGG1 antibody targeting TROP2. There is a cleavable linker, as shown here, hydrazone, and the payload for sacituzumab is a camptothecin, specifically a topoisomerase I inhibitor, the metabolite of irinotecan, SN-38.

The really landmark data that we had from a randomized Phase III study was this ASCENT trial, and this was the randomized Phase III study of sacituzumab versus treatment of physician’s choice chemotherapy. Patients had at least 2 prior lines of chemo and no limit, no upper limit, to how many prior lines they had seen, so really a heavily pretreated population. More than 500 women were recruited. And the primary endpoint for this study was progression-free survival, with all the typical secondary endpoints, as shown here.

I’ll mention the treatment of physician’s choice options for patients included capecitabine, eribulin, gemcitabine, or vinorelbine, so really the typical standard-of-care chemo options we had for patients who had all likely already seen taxane-based therapy.

So what I’m sharing here are the updated analyses from ASCENT with longer follow-up, as demonstrated in the gray box on the right. So here progression-free survival, consistent with longer follow-up, sacituzumab prolonged progression-free survival by 60% as compared to standard-of-care chemotherapy. You really see how poorly — standard-of-care chemo offered a median PFS of only 1.7 months, and sacituzumab having a median PFS of almost 6 months, and this was really upheld with longer follow-up.

We have also seen really substantial benefits in overall survival that have been durable with longer follow-up. So here about a 50% improvement in overall survival, median OS with sacituzumab measuring just over a year. Unfortunately, standard-of-care median overall survival is only about 6 months.

Treatment-associated side effects with sacituzumab govitecan

DR TRAINA: To balance efficacy, we always want to be considerate of toxicity and safety. And as I’m showing here, from the original publication in the New England Journal, some of the more common adverse events to be aware of that were Grade 3 or higher were neutropenia, with a Grade 3 event of about 34%, although fortunately febrile neutropenia is uncommon, only about 6% Grade 3 or greater. And this is something that can be mitigated with the introduction of growth factor, often a long-acting growth factor incorporated on day 8 of the cycle.

Diarrhea is also a bit of a concern and something for awareness here. Grade 3 diarrhea has been consistent in the sacituzumab trials of about 10%. All-grade diarrhea about 60%. And this is also something that requires patient education, awareness and intervention as needed. And there are very clear guidelines in the package insert in the protocols.

I will say that treatment-related discontinuation in the ASCENT study was relatively low at less than 5% with sacituzumab and actually lower with sacituzumab than with standard-of-care chemotherapy.

Activity of sacituzumab govitecan irrespective of TROP2 expression level and in patients with brain metastases

DR TRAINA: So there are several subgroups of interest that have been explored in ASCENT and in some other exploratory analyses as well. And one big question that comes up often is whether TROP2 expression is relevant or necessary as a biomarker for benefit from sacituzumab. The short answer is no. TROP2 expression is not necessary for use of sacituzumab. And we can see in the curves here that regardless of level of TROP2 expression sacituzumab always outperforms standard-of-care chemotherapy. So there is no need for sacituzumab TROP2 expression as a companion biomarker here.

Another area of interest in triple-negative breast cancer are patients with brain metastases. And I will call out a reminder that the ASCENT study included patients that had stable treated brain mets. And analysis, as shown here, included that subset of patients in the overall analysis, and you can see the benefits really maintained what was seen in that planned study — first primary study population, where about a 60% improvement in progression-free survival was observed incorporating those patients that had stable treated brain mets.

Since that time we have also just recently seen a publication exclusively looking at the subset of patients with brain mets. So this was about 60 patients, just over 10% of the overall study population, and in terms of progression-free survival sacituzumab did outperform standard-of-care chemotherapy, even in those patients with brain mets. Unfortunately, this is still an area that needs improvement. These patients have much shorter PFS than the population without brain metastases. We’ve also recently seen a publication showing that SN-38, that payload of sacituzumab, does cross the blood-brain barrier, so I think emerging data that saci has activity in the brain met-positive population.

Ongoing Phase III studies evaluating sacituzumab govitecan in earlier lines of treatment (eg, ASCENT-03, ASCENT-04)

DR TRAINA: So I’ll call your attention to a couple of exciting ongoing trials with sacituzumab. ASCENT-03 and ASCENT-04 are kind of partner trials. They are both looking at moving sacituzumab up into the first-line setting of patients with metastatic triple-negative breast cancer. ASCENT-03 is for either PD-L1-negative patients or patients who have PD-L1-positive tumors but previously saw immunotherapy, perhaps in the early-stage setting, perhaps those patients that had short disease-free intervals.

So ASCENT-03 is really a PD-L1-negative or not a candidate for immunotherapy first-line trial. Patients are randomized to sacituzumab versus standard-of-care chemo, which is either a taxane or gem/carbo combination. Crossover is permitted in this trial, which does call into question challenges around ever seeing an overall survival difference, but primary endpoint here is progression-free survival.

ASCENT-04 is the companion first-line trial for patients who have PD-L1-positive tumors with a CPS of 10 or greater. The randomization is sacituzumab plus pembrolizumab or standard of care plus pembrolizumab. Prior exposure to checkpoint inhibitor in early-stage disease is permitted. Many of these patients have likely seen the KEYNOTE regimen in the neoadjuvant setting. And again, primary endpoint is progression-free survival, and crossover is permitted. So please do keep an eye out for these clinical trials, and recruitment is ongoing.

Efficacy and tolerability of trastuzumab deruxtecan in patients with HER2-low and HER2-ultralow metastatic breast cancer

DR TRAINA: So we’re going to switch gears and talk about another antibody-drug conjugate that’s been quite exciting, trastuzumab deruxtecan. And here we’re going to focus just in the HER2-low population. So trastuzumab deruxtecan is another IGG1 antibody. It is targeting HER2. It also has a cleavable linker, and it also has a payload that has a topoisomerase inhibition mechanism of action.

What I’m sharing on the right-hand side of the slide here is much of the work that has been reported about trastuzumab deruxtecan in the HER2-low population has focused on hormone receptor-positive disease, but there is a component of patients that have hormone receptor-negative disease that also are classified as HER2-low. And that reflects about a third of the population that otherwise would be considered triple negative, so a relevant large body of patients that can be candidates for trastuzumab deruxtecan.

DESTINY-Breast04 is the large, randomized study that tested T-DXd in patients with HER2-low breast cancer. Patients were able to have up to 2 prior lines of therapy for metastatic disease, and they were randomized to T-DXd or capecitabine, eribulin, gemcitabine or taxane. Now I will remind you that this was largely a study for hormone receptor-positive patients. 89% of the population had hormone receptor-positive disease, but there was a small subset, about 11%, that had hormone receptor-negative breast cancer. And I will share the primary endpoint for this trial was PFS in the hormone receptor-positive patients, but I’ll share the exploratory analysis in those otherwise considered triple-negative, HER2-low patients.

So here are the DB-04 results, and I’ve split this out by hormone receptor-positive in the column on the left and the exploratory subset of triple-negative, HER2-low patients on the right. The PFS of this study overall was really favoring trastuzumab deruxtecan, with a 50% improvement. And in the exploratory analysis for hormone receptor-negative patients you can see T-DXd was associated with a median PFS of about 8 and a half months. In contrast, standard-of-care chemotherapy offered only about a 3-month median PFS.

When you look at the overall survival endpoint in the exploratory analysis trastuzumab deruxtecan outperformed standard-of-care chemotherapy. Again, about a 50% improvement in OS, with a median overall survival of about a year and a half versus only 8 months with standard chemotherapy. And response rates also much higher with T-DXd than with standard-of-care chemo. So although this is an exploratory small subset, about 60 patients, I think the data are consistent with what was seen with the larger overall study population and does offer a nice opportunity for our patients with triple-negative, although HER2-low breast cancer.

The PFS forest plot here is for the overall study population, so a lot of these patients are really hormone receptor positive. But at a glance you’ll see really across subsets T-DXd outperforms standard-of-care chemotherapy.

And I do want to take a moment to just address the brain metastases question because this is, as I alluded to before, clearly an issue for our patients with TNBC. There is a large body of data for activity of trastuzumab deruxtecan in patients who have HER2-positive brain metastases. And I’m sharing with you some data just out of ESMO last year looking at a pooled analysis from trastuzumab deruxtecan in a series of DESTINY-Breast trials, DB-01, DB-02 and DB-03. And you can see on the left, CNS-PFS in patients with treated stable brain mets, it looks like trastuzumab deruxtecan outperformed the comparator arms, really showing activity there in the CNS. On the right, an even more dramatic difference in patients who had CNS disease that was untreated or active, so T-DXd clearly demonstrating activity and CNS penetration. Again, the data I’m showing you is in the HER2-positive population, but we’re seeing more and more data consistent with this.

So looking at toxicity from DESTINY-Breast04, I’m highlighting specifically some of the adverse events of special interest, primarily higher rates of nausea of any grade of about 76%, some emesis, about 40%. Neutropenia not so much of an issue as compared to treatment of physician’s choice. But I really do want to highlight the concern around ILD, or pneumonitis. We have now seen with built-in monitoring in our protocols and early intervention and withholding of drug that the ILD rates with trastuzumab deruxtecan are now coming in at around 10% to 15%. Here, in DESTINY-Breast04, it was around 12% all grade. And when we look at Grade 5 events, fortunately very uncommon now, less than 1% in DESTINY-Breast04, so that’s reassuring.

So there are some new data that I think are hypothesis generating and certainly will generate some discussion, and that was looking at expanding T-DXd, looking at it in the first-line setting and also looking at it in the HER2-ultralow population, which was defined as some element of staining, so a little bit more than zero but not enough to qualify — to classify as HER2 1+.

DESTINY-Breast06 was just presented. Really exciting data. It is limited to the hormone receptor-positive population, but I just want this on your radar. In this PFS primary endpoint in the HER2-low population T-DXd seemed to outperform standard-of-care chemotherapy, with a hazard ratio of 0.62 and about a 5-month delta in PFS. And if we look at specifically the ultralow population, those 0+, you can see that both in PFS and overall survival T-DXd seems to outperform standard-of-care chemotherapy.

So lots of caveats here, right? This is a first-line study. It’s hormone receptor positive only, but it is suggesting some activity in the HER2-ultralows, and I would love to see additional data in this hormone receptor-negative, HER2-ultralow population, so we’ll have to stay tuned for that.

There are ongoing trials of T-DXd in combination for patients with hormone receptor-negative, HER2-low breast cancer, and I’ve highlighted some of these on the right. There are combination trials of T-DXd with capecitabine, T-DXd with paclitaxel and durvalumab and even T-DXd and capivasertib, all having hormone receptor-negative cohorts. So stay tuned in this space also.

We just saw at ASCO some real-world evidence around sequencing of ADCs. So this is a big question that we all have. We’re missing prospective data, how to sequence these agents, these ADCs like sacituzumab and T-DXd. And I think a big takeaway from this particular report out and others is that there is activity with ADC number 2, whichever one that might be. It does appear that the first ADC you use seems to have a longer and more durable benefit. And we do see practice variation of some providers sequencing 1 ADC immediately after the next and others using an intermediate therapy, a traditional cytotoxic, maybe a protocol therapy, and then coming to their second ADC. So I think we really do need to support these prospective studies that are going to look at mechanisms of resistance, whether it is resistance to the target or resistance to the payload, to give us really better data-informed decisions about how we sequence our ADCs.

Alright. Moving on to a little more data that we have seen from a platform study, the BEGONIA trial, which looks at first-line metastatic triple-negative breast cancer and has multiple treatment arms of exploration. What I’m going to share with you are 2 of the antibody-drug conjugate arms in combination with immunotherapy.

So the first we’ll look at is T-DXd plus durvalumab in the first-line setting. And here you can see a really impressive waterfall plot. Primary endpoint here in BEGONIA is often looking at safety, but we get signals of efficacy, so overall response rates of about 60% in about 60 patients treated in this particular cohort. And this was not limited to patients with PD-L1-positive tumors. This is all-comer first-line metastatic triple-negative breast cancer.

I will balance the efficacy with safety. So rather high discontinuation rates of about 17%. About 50% of patients required some kind of dose modification or delay, and there were 8 cases of ILD and 1 Grade 5 ILD; a bit concerning when we think about combination of checkpoint inhibitor that can also cause pneumonitis.

Activity and ongoing investigations of other novel agents and strategies for mTNBC

DR TRAINA: So I want to talk about another antibody-drug conjugate, dato-DXd. And this is an agent that has an antibody targeting TROP2 with a similar payload that we’ve seen with trastuzumab deruxtecan, that same topoisomerase exatecan derivative.

So dato-DXd has been studied in kind of early-phase trials, the TROPION-PanTumor01 study. What I’m sharing with you here is the subset of patients that had breast cancer, and these were patients that were heavily pretreated. 71% had 3 or more prior lines. Many of these had already seen sacituzumab, another anti-TROP2 antibody. But you can see signals of benefit here in waterfall plot, and a very different adverse event profile. So we’re not seeing high rates of ILD. Rather, we see stomatitis, about 13% Grade 3 or greater, and a bit of ocular issues that need to be monitored.

So dato-DXd has been looked at on the BEGONIA study in combination with durvalumab in the first-line setting, and here I think we see an even more impressive waterfall plot with about an 80% response rate. Again, this is regardless of PD-L1 expression you’re seeing benefit in the combination of dato and durva, so really encouraging. And again, when we look at a deeper dive into adverse events you see that the most common adverse events are really GI related and of lower grade, but stomatitis was the most common adverse event that led to dato dose reductions. And built into the dato-DXd trials now are very clear prophylaxis guidelines around steroid oral care that helps to mitigate that stomatitis. Fortunately, ILD and pneumonitis rates were much lower at only about 5%.

So certainly keep an eye on dato-DXd. I’m happy to say that there is a first-line trial in the metastatic setting looking at dato versus investigator’s choice chemotherapy. This study is now fully accrued globally, and we are eagerly awaiting results.

So some other data that’s on the horizon that I want to share with you is another antibody-drug conjugate that we just saw presented at ASCO. This is enfortumab vedotin. This is an antibody-drug conjugate that targets nectin-4, and it has an MMAE payload. So any time we’re talking about ADCs you always want to ask yourself what is the target, what is the payload, to really define how we differentiate these different antibody-drug conjugates.

So EV was studied in 2 cohorts, a hormone receptor-positive breast cancer cohort and a triple-negative breast cancer cohort. And the primary endpoint was overall response rate. And what I’m sharing here is just the data around the triple-negative breast cancer cohort of about 42 patients. These folks had seen 2 prior lines of chemotherapy on average. The benefit rate was about 60%, you can see that on the waterfall plot here, and an overall response rate of about 20%.

So the study actually did not meet its prespecified response rate primary endpoint, although we do see signal of activity here. Median progression-free survival of about 3 and a half months or so in the third-line or later setting, a median overall survival of just over a year. So I’m happy to see that there’s some activity here. A lot of work to still be done on exploring biomarkers and subset analyses.

In terms of adverse events with this ADC, fatigue was one of the more common adverse events, true in the triple-negative population, as well as the hormone receptor-positive population. Rash and other dermatologic adverse events were common, and upper GI nausea, about 40%, all grade. Fortunately, Grade 3 or higher adverse events were uncommon.

Two other report outs that I do want to share from ASCO with you. One was the TBCRC 048 trial, which has looked at expanding the role of olaparib outside of germline BRCA only mutation carriers, and the final analysis was just reported out on an expanded subset of patients who had germline PALB2 mutations. 26 patients in this cohort but quite high response rates, a 75% overall response rate in germline PALB2 carriers, and clinical benefit rate was quite high at 83%. The other cohort of interest were those patients with somatic BRCA mutations, so here 30 patients, and there was a response rate of about 40%, and clinical benefit rate of about 50%.

So I think we’ve now seen these data confirmed with olaparib. There are similar series with talazoparib in looking at patients who have somatic BRCA mutations, as well, and so I think it’s an encouraging space to look at potentially expanding the role of PARP inhibitors.

Some unfortunate data to share with you is from CAPItello-290. So we know we have capivasertib available to us, FDA approved for hormone receptor-positive breast cancer in patients who have AKT mutations or PI3 kinase mutations or PTEN loss. Well, CAPItello-290 looked at capivasertib in metastatic triple-negative breast cancer based on exciting Phase II data, and this trial enrolled women and randomized them to paclitaxel with or without capivasertib. What we just saw from a press release earlier this summer is that this trial did not meet its primary endpoint in terms of improving efficacy outcomes in the intention-to-treat population or even in the subgroup of patients with tumors that had specific biomarkers that would have been thought to be predictive, like AKT, PI3 kinase, or PTEN loss. So unfortunately, CAPItello-290 is a negative trial in triple-negative breast cancer, and we await seeing those data a little bit more carefully.

And then my final slide is looking at the role of AR in triple-negative breast cancer. TBCRC 058 is open and recruiting. We presented a trials in progress at ASCO. And this has cohorts both for patients with traditional triple-negative breast cancer, defined as ER/PR 0, but also allows for ER-low tumors, so patients whose tumor stain 10% or less and have AR positivity of 10% or greater. Patients are randomized to enzalutamide, the AR antagonist, or enzalutamide with mifepristone versus standard-of-care chemotherapy. And this trial is recruiting patients being treated in the first-line setting right now or second-line setting, soon to expand to even include third-line setting. So keep an eye out for a TBCR site close to you. I’m happy to help with enrollment.

So with that I’ll thank you for your attention.