As new data emerge in breast cancer, stepping back from the updates and shedding light on the standard practices as they exist in the clinic are crucial approaches to management of the disease.
As new data emerge in breast cancer, stepping back from the updates and shedding light on the standard practices as they exist in the clinic are crucial approaches to management of the disease. For example, in early-stage, HER2-positive disease, small changes have rapidly improved outcomes for patients regardless of disease burden. With de-escalation strategies allowing for fixed-duration therapy in the pre- and postoperative settings, clinicians can provide less toxic treatment options for their patients.
“We’ve come a long way over the past decade [in] treating [patients with] early-stage, HER2positive breast cancer, [and] we’ve learned that we can give therapy in the preoperative setting, [determine] how someone responds to that treatment, and then adapt their treatment after surgery based on that response,” Sara M. Tolaney, MD, MPH, said in an interview with OncologyLive. Tolaney is chief of the Division of Breast Oncology and associate director of the Susan F. Smith Center for Women’s Cancers at Dana-Farber Cancer Institute and an associate professor of medicine at Harvard Medical School, both in Boston, Massachusetts. Tolaney will serve as a faculty member during the 21st Annual School of Breast Oncology® meeting, hosted by Physicians’ Education Resource (PER), LLC. In her interview she noted that the meeting is one way in which clinicians can gain access to physicians who are translating the topline data into practice.
“One thing I love about the School of Breast Oncology meeting is that there are many experts who know the field so well and provide the most up-to-date educational sessions on the [the most recent] data,” Tolaney said. “[Chair] Joyce O’Shaughnessy, [MD], does a phenomenal job bringing tough case discussions to the forefront [of the agenda]. These are cases that we struggle with in clinic, and it’s helpful to get guidance from very thoughtful clinicians who are there. The discussions with the cases are probably one of my favorite parts of the meeting.”
Of focus in early-stage breast cancer are ways to ensure that patients are receiving the right amount of the optimal therapy for their disease. Clinicians are tackling this through 2 avenues, Tolaney said: de-escalation and biomarker-driven approaches.
For example, at the 2023 American Society of Clinical Oncology (ASCO) Annual Meeting, findings from the 3-year follow-up of the PHERGain trial (NCT03161353) showed that the use of PET imaging may be a tool in early identification of patients with HER2-positive, early-stage breast cancer who may have a pathologic complete response (pCR) to neoadjuvant HER2-directed therapies. Using fluorine-18 fluorodeoxyglucose (18F-FDG) PET/CT, investigators traced patients for early metabolic response to de-escalate chemotherapy in the neoadjuvant setting.1
Patients were administered pertuzumab (Perjeta) and trastuzumab (Herceptin) with or without endocrine therapy—letrozole for those who were postmenopausal and tamoxifen for those who were premenopausal—for 2 cycles before undergoing 18F-FDG PET/CT.
If patients demonstrated a response, they continued therapy with no changes. If there was no response, patients proceeded to chemotherapy/antibody therapy of TCHP (docetaxel, carboplatin, trastuzumab, pertuzumab) for 6 cycles. Patients then proceeded to surgery, and those in the TCHP arm went on to pertuzumab/ trastuzumab with or without endocrine therapy, and those in the original therapy arm were assigned based on pCR to either continue therapy (pCR) or to TCHP (no pCR).
Among all evaluable patients in the intentionto-treat population at the time of surgery (n = 267) the 3-year invasive disease-free survival (iDFS) rate was 95.4% (95% CI, 92.8%-98.0%). Additionally, the 3-year event-free survival (EFS) and overall survival (OS) rates in the overall population (n = 285) were 93.5% (95% CI, 90.7%-96.5%) and 98.5% (95% CI, 97.1%-100%), respectively.1 For comparison, in the arm evaluating TCHP for 4 cycles prior to surgery, [among those who received] pertuzumab/trastuzumab with or without endocrine therapy after surgery regardless of response (n = 63) had an iDFS rate of 98.3% (95% CI, 95.1%-100%) at the time of surgery. In evaluable patients from the overall cohort of this regimen (n = 71), the EFS and OS rates were 98.4% (95% CI, 95.3%-100%) and 98.4% (95% CI, 95.3%100%), respectively.1
Another avenue of prognostication comes in the form of an assay, the HER2DX genomic risk score, which has been examined in exploratory analyses of pivotal HER2 breast cancer studies.2,3 The algorithm to develop the score, the aim of which is to stratify patients with earlystage disease into low- and high-risk groupings for recurrence, uses “4 gene expression-based signatures to track adaptive immune cell infiltration, tumor cell proliferation, HER2 amplicon expression and tumor cell luminal differentiation, together with tumor stage (T1 vs T2 vs T3/ T4) and nodal stage (N0 vs N1 vs N2/N3).”2
Tolaney and colleagues published findings of an analysis in The Lancet Oncology using the tool in an analysis of findings from the phase 2 APT trial (NCT00542451). When applying the tool, the authors wrote, “HER2DX risk score as a continuous variable was significantly associated with iDFS (HR per 10-unit increment, 1.24 [95% CI, 1.00-1.52]; P = .047) and recurrence-free interval (HR, 1.45; 95% CI, 1.09-1.93]; P = .011).”
Tolaney: We know this paradigm—our current treatment approach is to give a patient with stage II/ III, HER2-positive breast cancer preoperative treatment. The standard at this time is TCHP for 6 cycles, take them to surgery, and if they have a pCR, then continue the trastuzumab and pertuzumab. They get a complete year of therapy, or if they have residual disease, we give them T-DM1 [ado-trastuzumab emtansine (Kadcyla)] for 14 cycles.
If [we] have patients with stage I [disease]— someone with a small tumor that’s clinically node negative—we often [go straight to] up-front surgery, and then give them adjuvant therapy. [Once] we’ve confirmed that they have a stage I cancer, the choices would be to give weekly paclitaxel and trastuzumab for 12 weeks, followed by 9 months of trastuzumab monotherapy, or to give T-DM1 for 17 doses. Either of those approaches would be reasonable for that patient.
We have learned that we can add pertuzumab to chemotherapy and trastuzumab. Originally, the data focused on the benefits of adding pertuzumab in the preoperative setting. For example, we had data from studies such as NeoSphere [NCT00545688] and TRYPHAENA [NCT00976989] that taught us that adding pertuzumab to chemotherapy and trastuzumab improved pCR. This was really important to understand, and it led to the initial FDA approval of pertuzumab based on improvement in pCR. But in truth, we really wanted to know whether adding pertuzumab would improve survival outcomes. And those data came from the APHINITY trial [NCT01358877], which taught us that adding pertuzumab to chemotherapy/ trastuzumab improved iDFS. Now, with over 8 years of follow-up [data, the regimen] still has not improved OS, and the benefits for adding pertuzumab were restricted to those patients with node-positive disease, with no benefit from adding pertuzumab in a patient with node-negative disease.
We’re fortunate because we have seen multiple new HER2-directed agents come along that have dramatically improved outcomes. We’ve added pertuzumab, we know how to use T-DM1, and we already had trastuzumab. Now that we have this armamentarium, we realize that maybe we can achieve the same efficacy, with decreased toxicity, if we tailor therapy to the individual patient.
For example, when we thought of small HER2positive cancers, we learned we can give an abbreviated amount of chemotherapy—12 weeks of paclitaxel—and achieve great outcomes.
Now we’re wondering whether there are other ways we can tailor therapies. One way we are trying to do this is by [asking]: Can we give a certain regimen in the preoperative setting, and if the patient achieves a pCR, can we say that that was enough therapy? For example, [if a patient receives] THP for approximately 12 weeks and has a pCR, can they do their adjuvant 9 months of HP therapy and achieve good outcomes? In essence, can we de-escalate from TCHP to THP in those who achieve pCR? That’s one such approach in an ongoing trial called CompassHER2-pCR [NCT04266249].
Some [investigators] call them de-escalation trials, others call them optimization trials, but a lot of these trials have been trying to tailor therapy based on response. [Regardless] what we’re trying [to ask] with these trials is whether we can go a step further and tailor therapy by looking at response prior to surgery.
At ASCO [we saw some data] from the PHERGain trial, [which uses] PET response to help us tailor [therapy]. And then I think we’re trying to even move toward biomarker-driven trials to help tailor therapy. There’s a new biomarker [risk score] called HER2Dx, which seems to help us predict who’s going to achieve pCR and who’s going to have good long-term outcomes.
Whether size matters is an important question, and I do think size does still matter. If you look at patients who achieve pCR to HER2-directed therapy, we know that’s a good thing because if there is pCR, there likely [will be] very good long-term outcomes. But we know that not all pCRs are the same.
If you take a patient who started with a big tumor and who had multiple positive lymph nodes, and they achieved pCR—and you compare their outcomes with [those of] someone with a small cancer with no lymph node involvement up front—even though they both had all their cancer gone at the time of surgery, the person who started with higher clinical and anatomic risk has higher risk of recurrence. To me that suggests that biology does matter, but so does size because that tumor was clearly very sensitive to HER2-directed therapy because they achieved pCR. But their clinical anatomic risk did put them at still higher risk of recurrence.
So if we go back to that example, looking at HER2DX, [for] that score, where [investigators] looked at 4 different gene expression signatures, they add tumor size and nodal status. That is key because, again, you use biology, you use clinical risk, and you come up with the best outcome prediction that you can.
Next, the question [of whether] trastuzumab deruxtecan could replace all preoperative chemo-therapy [arises]. There’s a trial that’s looking at replacing ACTHP [doxorubicin hydrochloride, cyclophosphamide, paclitaxel, trastuzumab, pertuzumab] with trastuzumab deruxtecan—so either replacing the whole duration of ACTHP with [the antibody-drug conjugate], or replacing half of it, replacing the AC with 4 cycles of trastuzumab deruxtecan, then following up with THP.
Trastuzumab deruxtecan does seem very promising given what we’ve seen in the metastatic setting. There certainly are some toxicity concerns that come up, particularly the risk of interstitial lung disease. We will have to look at the efficacy data from these trials and balance those with potential toxicity to figure out how [trastuzumab deruxtecan] fits in. But I do think that it has potential to change standard [treatment].
There’s also interest in looking at tyrosine kinase inhibitors in the early-stage setting. For example, CompassHER2 RD [NCT04457596] is looking at adding tucatinib [Tukysa] to T-DM1 with the hope that it will improve diseasefree survival and help prevent CNS [central nervous system] recurrence because it has CNS penetration.
Multidisciplinary care in breast cancer is critical right now. When we see patients with newly diagnosed breast cancer at Dana-Farber, for example, we see them in combination with our surgical oncologist and sometimes a radiation oncologist up front with the medical oncologist.
The reason is that these [treatment] decisions are complicated. Does it make sense for the patient to get systemic therapy before surgery?... We need to have that as a discussion [with the team]. What about additional tests that need to be done before preoperative therapy? Do they need a lymph node biopsy? What if there’s an area on imaging that the surgeon is concerned about and they want a biopsy to understand whether the patient is going to be a breast conservation candidate?
Even more complicated is this whole issue about nodal evaluation at the time of surgery. [For example,] can a patient get away with sentinel node [biopsy] alone? Do they need an x-ray dissection? Will [those results] influence decisions regarding radiation? Could that change systemic therapy recommendations? It’s just so complicated. So it’s really nice to have these multidisciplinary discussions with our colleagues so that we make the right decisions for the individual patient.
We’ve been successful in early-stage, HER2positive disease, where we are curing more and more patients, and we’re doing so in a way where toxicities are being thoughtfully considered. That said, I think we are giving some patients too much therapy and some too little therapy. Where I’d love to see us go is to figure out how to really optimize treatment and personalized care.
Biomarkers will help us get there, but we also need to be bold in our clinical trial designs. We’re learning how to do de-escalation studies, but they’re tricky because you don’t want to undertreat a curable patient and leave them with a recurrence. [These trials] have to be done thoughtfully. Certainly we all always want to get patient advocates involved in such designs as well.
The optimal duration of HER2-directed therapy is still not known. We’ve done multiple trials, but we’re still giving 1 year of trastuzumab-based treatment. Does everyone really need that? Can some patients get away with less? Who’s going to need trastuzumab deruxtecan? [Eventually] we’re going to need to figure that out because, again, there are potential toxicities [and we want to know whether there are those] who can get away with less toxic regimens. And who can get away with no chemotherapy? I do think there are patients who can get away without any chemotherapy at all and just [receive] antibodies. But we don’t have robust enough predictors for that, and we need to get there. I think we’ll continue to see things evolve as we learn how to integrate biomarkers into the early-disease setting.