A panel of experts led by Sara Hurvitz, MD, discuss recent data within the HER2-positive breast cancer treatment landscape.
At an Around the Practice presentation hosted by CancerNetwork®, experts spoke about new data in treatment regimens for HER2-positive breast cancer. The panel was led by Sara Hurvitz, MD, medical director of the University of California Los Angeles (UCLA) Jonsson Comprehensive Cancer Center Clinical Research Unit, co-director of the Santa Monica-UCLA Outpatient Oncology Practices, and director of UCLA’s Breast Cancer Clinical Trials Program.
The other experts on the panel were Neil M. Iyengar, MD, a medical oncologist at Memorial Sloan Kettering Cancer Center in New York; Vijayakrishna Gadi, MD, PhD, professor and director of medical oncology in the Department of Medicine and associate director of translational oncology at the University of Illinois in Chicago; and Reshma L. Mahtani, DO, chief of breast oncology at Miami Cancer Institute at Baptist Health South Florida in Miami.
Hurvitz: We know that HER2-postitive metastatic breast cancer is a rapidly evolving field right now. We have 8 FDA-approved therapies that target HER2, with more to come. [Data for] a new ADC [antibody-drug conjugate] were presented at the [European Society for Medical Oncology] Congress, and I’m excited to discuss how you are implementing the new data and what your standard of care is for HER2-postitive metastatic disease. There was a big data splash looking at the efficacy of T-DXd [trastuzumab deruxtecan; Enhertu] vs T-DM1 [trastuzumab emtansine; Kadcyla] in the DESTINY-Breast03 study [NCT03529110].1
Iyengar: The trial included more than 500 patients. Eligible patients had HER2-postitive unresectable or metastatic breast cancer and had previously been treated with trastuzumab plus a taxane in the metastatic setting. Patients with stable, treated brain [metastases] were eligible for enrollment on this trial. Patients were randomized 1:1 to either T-DXd or T-DM1, and the primary end point was progression-free survival [PFS]. Some important things about the study population to remember as we interpret the data [are that] the median age was 54 years and the majority of patients, close to 60%, were enrolled in Asia. Nearly 50% of the patients were receiving study treatment in the second line and the rest were receiving study treatment in the third line or beyond. All patients had [received] prior trastuzumab [Herceptin] and about 60% enrolled had [received] prior pertuzumab [Perjeta]. This PFS had a follow-up of 15 months, while median PFS was not reached in the T-DXd arm [vs] 6.8 months in the T-DM1 arm. The 12-month PFS rate was 75% in the T-DXd arm vs 34% in the T-DM1 arm, and the HR was 0.28 favoring T-DXd [95% CI, 0.22-0.37; P = 7.8 10–22]. There was an overall survival [OS] benefit reported with a HR of 0.56 [95% CI, 0.36-0.86; P = .007172], and an objective response rate [ORR] of 80% in the T-DXd arm vs 34% in the T-DM1 arm. There were no new safety signals in terms of adverse events and the rate of any-grade interstitial lung disease [was 10.5%].
Hurvitz: When you have a patient in the second-line setting, what are the end points in clinical trials that compel you to change practice? T-DM1 is supported by OS data from the EMILIA trial [NCT00829166], but the DESTINY-Breast03 OS [data are] not mature.2 It has a strong trend, but we don’t know if patients receiving crossover therapy as standard of care will have a substantial OS benefit. Does that sway you away from using it in the second-line setting, or are these efficacy outcomes acceptable for you to switch?
Iyengar: I look at the OS data. As we begin to see our patients with HER2-positive disease living longer, we will wait longer for those data. Most of our patients will receive these treatments at some point. I’m looking at the PFS and ORR, particularly if someone needs more rapid disease control. The response rate was impressive in the DESTINY-Breast01 [NCT03248492] and the DESTINY-Breast03 trials, which is important.3 Quality of life and the safety profile of the drug are important as well, because our patients are living longer and will likely be exposed to all these drugs at some point. Patient selection factors will matter in how we select our agents. Patients may have preferences for regimens that fit better into their lifestyle; it may not just be the data [that play into a choice]. We’re fortunate to have enough available treatments that we can have these nuanced conversations with our patients.
Hurvitz: T-DXd is an ADC. It’s basically trastuzumab, the HER2-targeted monoclonal antibody, linked to a chemotherapy payload that targets topoisomerase 1. In DESTINY-Breast03, this was compared with a different ADC, T-DM1. A question arose about hormone receptor–co-expressing breast cancers. In the forest plot of the DESTINY-Breast03 trial, none of the subgroups shown had less benefit from T-DXd. It’s rare to see a forest plot where every subgroup has benefits from the treatment. The subsets of patients with hormone receptor–positive and –negative breast cancer both seem to benefit. How do you feel about these data based on the subgroup analysis?
Mahtani: I share your [opinion] on that forest plot where you don’t see a differential benefit in ER [estrogen receptor]–positive vs ER-negative, HER2-positive patients. In the earlier setting, when we used the CLEOPATRA regimen [pertuzumab, trastuzumab, and docetaxel; NCT00567190], many of us dropped the taxane [after that induction period] and started hormonal therapy as a maintenance with the [trastuzumab/pertuzumab] backbone.4 There are ongoing trials looking at the benefit of CDK4/6 inhibitors on top of that. The PATINA study [NCT02947685] is looking at that approach, plus or minus a CDK4/6 inhibitor. There’s rationale in considering that approach, but in the setting of the data we looked at, I’d feel comfortable giving a patient with ER-positive, HER2-positive disease trastuzumab deruxtecan, knowing that the ER expression will not differentially affect their benefit.
Gadi: Interstitial lung disease [ILD] is a toxicity that we don’t have a lot of leeway with; the label and guidance from the company are very clear. If you suspect ILD, the recommendation is to hold the drug if it’s grade 1. If it’s grade 2, 3, or 4, the recommendation is to discontinue the drug. You’ll have a patient in front of you, and you’ll make the right choice, but for grade 1 toxicities you may or may not need to start steroids. Hold the drug, let it resolve, and revisit the issue. If the risk-benefit ratio is in favor of restarting, then restart the same dose or a lower dose, whichever the guidance is, based on what you’re observing. It’s not worth crossing that line with this drug.
Hurvitz: [What about] data from the TULIP trial [NCT03262935]?5
Gadi: This [trial involves] a molecule called trastuzumab duocarmazine vs physician’s choice [treatment]. This is another ADC [that] is an alkylator, a distinction from T-DM1—where you have a tubule inhibitor—or [trastuzumab] deruxtecan—where you have a topoisomerase 1 inhibitor. This is a new mechanism of action for the chemotherapy component. Once this drug is internalized and the linker is broken down, it is also a molecule [that] is capable of bystander effects. It kills the cell within, but it also theoretically might be able to attack some cells. It’s similar in terms of its abilities to [trastuzumab] deruxtecan.
In this randomized phase 3 study, physician’s choice had several options. Three of them were chemotherapy plus trastuzumab, and one of them was lapatinib [Tykerb] plus capecitabine. Some options [may cause] hair loss, but it is not a major component. The chemotherapies were vinorelbine, and eribulin, and capecitabine. That’s something that people may notice. This is a 2:1 randomization; [also], the study population was well balanced [and] was about 10% Asian, which is different compared with what we saw on the DESTINY-03 trial.
The PFS was the primary end point, [and it] favored the new molecule. The [trastuzumab] duocarmazine molecule [had a median PFS of 7.0 months] vs physician’s choice at 4.9 months, and the HR was 0.64 for a 36% improvement and a statistically significant P value (P = .002). If you look at the [Kaplan-Meier] curves, you see each one of those landmark points at 12 months. Going onward, the lines stay apart; they don’t meet, and they stay stably apart. In terms of OS—it’s not mature yet [and] not statistically significant—but it tilts in favor toward the new molecule at 20.4 months vs 16.3 months [HR, 0.83; 95% CI, 0.62-1.09; P = .153]. That may emerge as a positive signal in this [setting]. When you look at response to therapy, it’s not a lot different than what you’re seeing with the physician’s choice chemotherapy.
Overall responses were about 28% with the new molecule vs 30% with physician’s choice, and target lesions responded [at rates of] 70% vs 58%. The clinical benefit rates—this combination of [complete responses], [partial responses], and stable disease—were 38.5% vs 32%.
A hint at more activity, patients maintained their responses longer, and PFS was likely different: Based on that, [this trial is] looking positive for an efficacy signal.
The other side of the coin, [though], is the toxicity [with trastuzumab duocarmazine], particularly ocular toxicity. What do you think about that and how we’re going to roll that into our practices? Apart from that, it’s not different from the physician’s choice chemotherapy in terms of toxicity. There were few grade 3 or greater toxicities in terms of meaningful numbers, other than the ocular toxicity, which is a special interest here. That’s 30% of folks with events in the physician’s choice arm vs 78% with the new molecule for ocular toxicity, and the grade 3 rate was 21%. That is a lot of serious dry eye, and the requirement [is] that we’re going to need [an] ocular oncologist involved in this. Pneumonitis is generally low grade, and grade 3 events were around 2.5%, not in the ballpark of where we see [trastuzumab] deruxtecan.
In a polling question, audience members indicated that the DESTINY-Breast03 trial was the most relevant to their practice (poll question 1).
In a polling question, clinicians indicated they would give this patient T-DXd (poll question 2).
Hurvitz: Is this how you would address a patient whose disease has progressed in the lung with no brain metastases after trastuzumab/pertuzumab and docetaxel?
Gadi: It makes sense to me. We have the phase 3 [DESTINY-Breast03] study that shows that this is the choice agent in this context. It’s hard to argue with 22 zeros in a P value in terms of PFS [P = 7.8 10–22]. I can understand some people wanting to wait for the survival data; it’s very reasonable. For the quarter of folks who are sticking with T-DM1 in this patient, that makes sense, too, because there’s a big difference in the tolerability of the agents. We have data showing that when this is used in a slightly later line, it’s still active. If you’re looking at the patient in front of you, you make choices, but those choices are reasonable.
In a polling question, clinicians indicated they would give this patient tucatinib plus trastuzumab plus capecitabine (poll question 3).
Gadi: I agree with the community choice of a tucatinib-based regimen for many reasons. This is the type of patient who was enrolled in [DESTINY-Breast03], and the survival data that we look at speak to this patient. More worrisome is the rate of progression. In the first-line setting, brain metastases are very concerning. Patients like this continue to have more evidence of metastases in the CNS [central nervous system] over time. If you target these 1 or 2 lesions with stereotactic radiosurgery and pick the tucatinib regimen, you might have an opportunity to suppress future metastases in a patient like this. The other option is a very active regimen with trastuzumab deruxtecan. Sticking with the data, I’d go with [tucatinib], and see what’s going on over time.
Hurvitz: Would you use SRS [stereotactic radiosurgery] in a patient like this and follow it with tucatinib-based therapy? Would you use SRS and do T-DXd? Or would you do tucatinib and put off locoregional therapy for the time being?
Mahtani: When you send patients to a radiation oncologist, it’s difficult to get them to not offer SRS. Patients who participate in trials are highly selected, and the strategy that was employed in the HER2CLIMB study [NCT02614794] would have allowed a patient with a brain lesion that was 2 cm or less and didn’t require immediate local therapy to be treated with just tucatinib.6 If it was up to me, I’d like to use only tucatinib, knowing that some of these local therapies are associated with other adverse effects. Now you have a patient who has a brain lesion and you’re not treating it with SRS. Most of these patients will be treated with local therapy. The scenario in this case is something that we’ll see more frequently, because if you look at the KATHERINE data [NCT01772472], if you look at high-risk patients who had residual disease, even using our best drug in that setting, T-DM1, it didn’t affect the incidence of brain metastases for those patients.7 There are ongoing trials looking at that patient with high-risk residual disease. Tucatinib plus T-DM1 in the COMPASS trial [NCT04457596] looks to see if prevention is the best medicine.