CME
Physicians: Maximum of 0.75 AMA PRA Category 1 Credit™
Released: August 14, 2024
Expiration: February 13, 2025
DESTINY-Breast06: Study Design
Kevin Kalinsky, MD, MS, FASCO:
The most highly anticipated breast cancer study at ASCO 2024 was the DESTINY‑Breast06 trial. This global, open-label, randomized, phase III trial enrolled patients with HR-positive/HER2-low or HER2-ultralow MBC with progressive disease following at least 2 previous lines of endocrine therapy ± targeted therapy.1 Patients treated with previous chemotherapy for advanced disease were excluded. The patients were randomized to T-DXd or physician’s choice of chemotherapy with options of capecitabine, paclitaxel, or nab‑paclitaxel.
The patients enrolled on this trial were primarily HER2‑low (immunohistochemistry [IHC]1+ or 2+ and ISH negative; n = 708). But it also included a population of patients who were HER2‑ultralow (IHC between 0-1; n = 152).
The primary endpoint for this study was PFS in the HER2‑low population, per blinded independent central review.
Sara Hurvitz, MD:
In some ways, DESTINY‑Breast06 was a follow up to the DESTINY‑Breast04 trial. DESTINY-Breast04 was conducted in patients with HER2-low MBC who had received 1 to 2 prior lines of chemotherapy, and if they had HR-positive disease, had also exhausted endocrine therapy options. DESTINY‑Breast04 showed that T-DXd was more effective than single‑agent chemotherapy in terms of both PFS and overall survival (OS).2
DESTINY‑Breast06 was done in the setting of HR-positive MBC that was chemotherapy-naive.1 So essentially, this study was testing whether T‑DXd would be beneficial after exhausting endocrine therapy but before receiving chemotherapy for patients with HR‑positive MBC. The study also allowed a subset of patients with ultra-low expression of HER2. HER2 IHC 0 expression is subdivided into complete lack of HER2 expression or less than 10% of cells having HER2 expression. Patients in the latter category were allowed on the trial. Similar to DESTINY‑Breast04, DESTINY-Breast06 was a randomized study comparing T‑DXd to treatment of physician’s choice chemotherapy. This is an important feature of this study because many clinicians, including myself, after exhausting endocrine therapy will turn to capecitabine given that it’s an oral chemotherapy. So, I think this was a smart design to compare T‑DXd to capecitabine or a taxane, and 59.8% of patients did receive capecitabine in this trial.
DESTINY-Breast06: Baseline Characteristics
Kevin Kalinsky, MD, MS, FASCO:
Approximately 85% of the patients enrolled on this trial had visceral metastases including liver metastases in approximately 2 of every 3 patients.1
DESTINY-Breast06: Prior Treatment
Kevin Kalinsky, MD, MS, FASCO:
In the intention-to-treat (ITT) population, the median number of lines of endocrine therapy was 2 with a range from 1 to 4 or 1 to 5 in the 2 arms of this study.1 Here I would like to point out that fewer than 10% of patients had been on frontline endocrine therapy plus CDK4/6 inhibition for less than 6 months, a population of patients that appear to be very endocrine‑resistant.
In terms of the prior endocrine therapy, the vast majority had prior CDK4/6 inhibitor. In the neoadjuvant/adjuvant setting, close to two thirds of patients had prior endocrine therapy and approximately half had prior cytotoxic chemotherapy.
DESTINY-Breast06: Efficacy
Kevin Kalinsky, MD, MS, FASCO:
The trial met its primary endpoint in the HER2‑low population with a significant improvement in median PFS with T‑DXd vs physician’s choice chemotherapy of 13.2 months compared with 8.1 months, respectively (HR: 0.62; 95% CI: 0.51-0.74; P <.0001).1
In terms of the secondary endpoint of OS, there was a trend towards improvement in the 12‑month OS rate, but this is a premature analysis. It is also worth noting that T‑DXd is approved as subsequent therapy for patients with HER2‑low disease and that approximately 20% of patients on the control arm received T-DXd at the time of crossover.
Focusing on the HER2‑ultralow population (n = 152), the difference in median PFS really seemed to replicate what was seen for the HER2‑low patient population. Because it is a smaller population, the confidence interval is wider for this analysis, but it does suggest that even for the HER2‑ultralow population that there could be benefit with T-DXd compared with chemotherapy.
Sara Hurvitz, MD:
I find the improvement in PFS with T-DXd in patients who had HER2-ultralow disease to be interesting because we are unsure at present how much HER2 expression is needed to see a benefit with T‑DXd. Here, there appeared to be no difference in outcomes in patients with HER2-ultralow disease vs the ITT population.
DESTINY-Breast06: Antitumor Activity
Kevin Kalinsky, MD, MS, FASCO:
Similar to what has been reported in other breast cancer studies, we can see some nice responses with T‑DXd. In the ITT population, the ORR is almost 60%—mostly partial responses compared with about a 30% partial response rate with chemotherapy.1 So, if you have a patient who has an endocrine‑resistant tumor and you really need to get a response, T-DXd is a very active drug.
DESTINY-Breast06: TEAEs and AEs of Special Interest
Kevin Kalinsky, MD, MS, FASCO:
Consistent with previous experience, nausea and vomiting were common treatment-related adverse events (AEs) associated with T-DXd.1 I typically use at least triplet antinausea prophylaxis for my patients being treated with T-DXd. Approximately half of patients experience some alopecia, which is an important topic of discussion.
Approximately 10% of patients had treatment-related interstitial lung disease and pneumonitis with T-DXd, mostly grade 1/2 but 3 patients died from this toxicity. It is important for us to remember and monitor for interstitial lung disease and pneumonitis and if a patient has a grade 2 or higher pneumonitis, you have to discontinue the T-DXd.
DESTINY-Breast06: Clinical Implications
Kevin Kalinsky, MD, MS, FASCO:
I would just reiterate that, for me, in my practice, the DESTINY-Breast06 therapeutic strategy with T-DXd is for patients who have endocrine-resistant disease. For patients with HR-positive MBC, treatment is a marathon and we want to continue well-tolerated regimens for as long as we can. Once you have a patient with a tumor that has progressed through multiple lines of endocrine-based therapy, this is when I would think about utilizing T‑DXd. I also think that it is appropriate to utilize capecitabine in patients with a low volume disease and slow pace of progression who are no longer endocrine sensitive. If you have a patient who really needs a response, for example with visceral metastases or where you are worried about the pace of disease progression, now we can use T‑DXd earlier.
The challenge becomes the HER2-ultralow population, because most pathology reports are not reporting that level of HER2 expression. How are pathologists going to define HER2-ultralow? The work is being done, it may just need to be a conversation with the pathologist and defined on the reports. It also begs the question of whether we need to be utilizing HER2 as a biomarker at all. The way that HER2 was originally evaluated and developed was to determine positivity, not so much the spectrum of expression. So the question becomes: Do we even need to evaluate HER2 or can we just utilize this agent a bit more freely? Of course, based upon payer coverage right now, I do think that we will need to continue to demonstrate either HER2‑low or -ultralow disease.
Sara Hurvitz, MD:
I agree with you regarding how I will use T-DXd in this setting in my practice. The ORRs with T‑DXd are quite high, almost 60%, and they are higher than with chemotherapy. So for a patient who has heavy disease burden in need of a response or who has an aggressive pace of disease, I would be more likely to turn to T‑DXd than to capecitabine or a taxane, given the PFS data and the ORR. However, if I have a patient with bone‑only metastases or lower disease burden, I might still opt for capecitabine and then go to T‑DXd, which is administered intravenously and has a greater potential for AEs, as the next line after capecitabine in those patients.
I think these data do provide evidence that T‑DXd is effective even in the chemotherapy‑naive setting and in patients with HR-positive/HER2- ultralow disease. What this trial does not answer is whether using T‑DXd in the frontline chemotherapy-naive setting after exhausting endocrine therapy rather than in the second- or third‑line setting actually benefits patients in terms of OS. There was no crossover built into this trial and so it remains to be seen whether using it in the chemotherapy‑naive setting is benefiting patients long term as opposed to using a line of capecitabine and then switching to T‑DXd at the time of progression. I am eager to see OS data once mature, and also to see the number of patients in the control arm who went on to receive T‑DXd, with the hope of better understanding whether there is truly a survival benefit to using T‑DXd earlier.
postMONARCH: Study Design
Kevin Kalinsky, MD, MS, FASCO:
In the setting of HR-positive metastatic disease, an important question is whether there is benefit for continuing a CDK4/6 inhibitor in the metastatic setting after progression on frontline CDK4/6 inhibition plus an AI. Previously, we saw in the randomized phase II MAINTAIN trial that patients with HR-positive/HER2-negative MBC who had primarily received prior palbociclib together with endocrine therapy and switched to a different endocrine therapy with ribociclib at disease progression had improved median PFS compared with those patients who switched to a different endocrine therapy and placebo.3 However, in the randomized phase II PACE trial, patients with HR-positive/HER2-negative MBC with progression on prior AI plus palbociclib were randomized to fulvestrant alone vs fulvestrant plus palbociclib and failed to show a PFS benefit with the continuation of palbociclib.4
The global, double-blind, randomized phase III postMONARCH trial enrolled patients with progressive disease on a first-line CDK4/6 inhibitor plus an AI or with recurrence following an adjuvant CDK4/6 inhibitor plus endocrine therapy.5
The patients in this study were randomly assigned to fulvestrant plus abemaciclib vs fulvestrant plus placebo and the primary endpoint was PFS per investigator assessment.
postMONARCH: Baseline Characteristics
Kevin Kalinsky, MD, MS, FASCO:
Approximately 60% of patients had visceral metastases, approximately 40% to 50% of had a tumor with an ESR1 mutation, and about half of the patients had an alteration in the PI3K/AKT/PTEN pathway. This is clearly representing a patient population with disease features that suggest resistance to standard therapy.
postMONARCH: Prior Treatment History at Baseline
Kevin Kalinsky, MD, MS, FASCO:
Almost all patients received their prior CDK4/6 inhibitor in the advanced setting; there were only a few patients with tumors that had progressed on adjuvant CDK4/6 inhibition. Of importance, patients were not allowed to have prior chemotherapy in the advanced disease setting, so this is a truly second‑line study. Approximately 60% of patients received prior palbociclib, about one third prior ribociclib, and 8% received prior abemaciclib.
Approximately one quarter of patients were on their prior CDK4/6 inhibitor for less than 12 months.
Sara Hurvitz, MD:
I think it is important to emphasize that very few patients had received abemaciclib in the frontline setting, so this study does not address the question of whether there is a benefit to continuing abemaciclib in the second‑line setting after progression on abemaciclib in the first-line setting. It was essentially restricted to palbociclib or ribociclib being used in the frontline setting.
postMONARCH: PFS
Kevin Kalinsky, MD, MS, FASCO:
The trial met its primary endpoint. The addition of abemaciclib to fulvestrant significantly improved investigator‑assessed PFS vs placebo plus fulvestrant (HR: 0.73; 95% CI: 0.57-0.95; P = .02).5 The median PFS was 6.0 months with abemaciclib plus fulvestrant vs 5.3 months with placebo plus fulvestrant. The 6‑month PFS rate was 50% in the abemaciclib arm vs 37% in the placebo arm. By blinded independent central review, the median PFS with abemaciclib plus fulvestrant was even better at 12.9 months vs 5.6 months with placebo plus fulvestrant for an HR of 0.55. Thus, compared to other recently reported post‑CDK4/6 inhibitor trials, the placebo arm of fulvestrant alone did better than what we have seen previously.
Sara Hurvitz, MD:
Although the investigator-assessed PFS with abemaciclib/fulvestrant was significantly improved compared to fulvestrant alone, the absolute improvement was quite small, about 0.7 months. Of interest, when looking at the PFS by blinded independent central review, patients appeared to do better, and there was a greater difference between the 2 treatment arms.
postMONARCH: PFS in Key Subgroups
Kevin Kalinsky, MD, MS, FASCO:
In the subgroup analysis, there appeared to be a greater magnitude of benefit for those patients who were on their CDK4/6 inhibitor for ≥12 months with a median PFS of 7.0 months vs 5.4 months with abemaciclib plus fulvestrant vs placebo plus fulvestrant, respectively. Although there was not a statistically significant difference between those who did or did not have visceral metastasis, those who had visceral metastasis did worse with a median PFS of 5.4 months vs 3.7 months (HR: 0.87; 95% CI: 0.64-1.17) in comparison with those patients without visceral metastasis where the median PFS was 11.1 months vs 5.6 months (HR: 0.53; 95% CI: 0.34-0.83) with abemaciclib plus fulvestrant vs placebo plus fulvestrant, respectively.
Approximately 85% of patients had evaluable circulating tumor DNA using the GuardantINFINITY assay with the rates of ESR1 mutations and PI3K pathway alterations being as expected at around 40% to 50%. In the exploratory analyses of these patients, PFS benefit was reported with the addition of abemaciclib regardless of the presence or absence of these mutations.
Regarding the investigator-assessed subgroup analysis by previous CDK4/6 inhibitor, the majority of patients had prior palbociclib with a HR of 0.62 (95% CI: 0.44-0.86). For patients who received previous ribociclib, the HR was 1.01 (95% CI: 0.67-1.51) with a wide confidence interval and fewer than 100 events, but the interaction P value was not statistically significant. At the end of the day, it is hypothesis generating and we will see from other soon to be reported studies whether it makes a difference in terms of what the prior CDK4/6 inhibitor was.
Sara Hurvitz, MD:
An important point to note is that although the PFS primary endpoint was met, when looking at the Forest plot of different patient subgroups, the patients who seemed to benefit the most were those who received palbociclib in the first‑line setting and switched upon progression (HR: 0.46; 95%CI: 0.30-0.72). In contrast, those previously treated with ribociclib (HR: 0.73; 95% CI: 0.43-1.23) or abemaciclib (HR: 0.59; 95% CI: 0.17-2.09) had wide confidence intervals overlapping the null point. This may be due to smaller numbers of patients, but it gives me less confidence in recommending abemaciclib/fulvestrant in patients coming off abemaciclib/AI or ribociclib/AI.
postMONARCH: Safety
Kevin Kalinsky, MD, MS, FASCO:
There were no surprises in terms of safety. In the abemaciclib arm, the rate of grade 3 or higher diarrhea was 4% and grade 3 or higher neutropenia was 25%. Dose reductions due to AEs were more common with abemaciclib vs placebo (30% vs 3%), and the discontinuation rate was low (6% vs 0%).5
postMONARCH: Clinical Implications
Kevin Kalinsky, MD, MS, FASCO:
I think that this therapeutic approach of continuing a CDK4/6 inhibitor is an option for some of our patients, regardless of the presence or absence of ESR1 mutations and PI3K pathway alterations. Approximately 30% of patients in this study did not have an ESR1 mutation or an alteration in the PI3K pathway. We do have elacestrant available for patients with ESR1 mutations, and we have alpelisib and capivasertib for those with PIK3CA and PIK3CA/AKT1/PTEN mutations, respectively, though I am increasingly using capivasertib because of its more tolerable safety profile. In clinical practice, however, not all patients have a low hemoglobin A1C or controlled hyperglycemia and many are not optimal candidates for these targeted therapies, so continuing a CDK4/6 inhibitor does provide another option for those patients. In particular, I would think about this approach as a strategy for patients who were on their CDK4/6 inhibitor for at least 12 months, and those who did not have visceral metastasis. These are the groups of patients that saw the greatest magnitude of benefit numerically. Also, I would consider this approach for patients who have a PIK3CA mutation or a PI3K pathway alteration where I am concerned about their tolerance of a PI3K or AKT pathway inhibitor.
Sara Hurvitz, MD:
In my practice, I would consider using abemaciclib/fulvestrant in the second‑line setting in patients who had been on palbociclib and an AI in the frontline setting. I would not use this combination in a patient who started with abemaciclib, and I probably would not use it in a patient who started with ribociclib. I would also select patients based on the length of time that they benefited from their initial CDK4/6 inhibitor. Those who had benefited less than a year, I probably would not trust utilizing another CDK4/6 inhibitor—abemaciclib—in that setting and instead would turn to a PI3K pathway inhibitor.
This study was conducted only in patients who were in the second‑line setting, and did not allow patients who had an intervening therapy with everolimus, capivasertib or alpelisib in the second‑line setting. So if I were to use abemaciclib in this manner, I most likely would use it in the second‑line setting and would look at a targeted therapy in the third-line setting.
SACI-IO HR+: Study Design
Kevin Kalinsky, MD, MS, FASCO:
SACI-IO HR+ was a randomized, investigator‑initiated phase II trial in patients with HR‑positive/HER2‑negative advanced or MBC, similar to the population in the DESTINY‑Breast06 trial. The study enrolled patients who had received at least 1 prior line of endocrine therapy and no more than 1 previous line of chemotherapy. Patients were randomly assigned to sacituzumab govitecan with or without pembrolizumab. The primary endpoint was PFS for the ITT population.6
Sara Hurvitz, MD:
The SACI‑IO HR+ trial was asking whether adding pembrolizumab to sacituzumab govitecan improves outcomes for patients with HR‑positive/HER2‑negative advanced breast cancer. Of note, they did not allow patients who had had a prior topoisomerase 1‑inhibiting ADC, such as T‑DXd.6
SACI-IO HR+: Baseline Characteristics
Sara Hurvitz, MD:
This was a smaller but decent sized study with 110 patients. I also think it is notable that they allowed both PD‑L1-positive and PD-L1-negative disease and fewer than 10% of the patients enrolled had a combined positive score (CPS) of 10 or greater; approximately 38% of patients had a CPS of 1 or greater.6
SACI-IO HR+: PFS in ITT (Primary Endpoint)
Kevin Kalinsky, MD, MS, FASCO:
Overall, there was no benefit with the addition of pembrolizumab to sacituzumab govitecan compared with sacituzumab govitecan alone, with a median PFS of 8.12 months vs 6.22 months, which was not significant (P = .37).6
SACI-IO HR+: PFS by PD-L1 IHC Status
Kevin Kalinsky, MD, MS, FASCO:
PFS by PD‑L1 positivity per CPS was evaluated and, numerically speaking, there was a greater magnitude of benefit for those who had a CPS of 1 or greater, though this was not statistically significant.6
Sara Hurvitz, MD:
I was somewhat disappointed that there were not more patients with PD‑L1 CPS of 10 or greater (n = 7; 7%) as this is the CPS cutoff that signified benefit from pembrolizumab in triple-negative breast cancer (TNBC).6 It would have been interesting to better understand whether a higher CPS was associated with benefit from adding pembrolizumab to sacituzumab govitecan.
SACI-IO HR+: OS and Response
Sara Hurvitz, MD:
The OS, as well, was not significantly different and neither were ORRs or clinical benefit rates, but the duration of response did appear to be trending longer with pembrolizumab.6
SACI-IO HR+: TEAEs and TRAEs
Kevin Kalinsky, MD, MS, FASCO:
There were no surprises in terms of toxicity, with the typical side effects that we see with sacituzumab govitecan including neutropenia, diarrhea, nausea, and alopecia.
SACI-IO HR+: irAEs Attributed to Pembrolizumab
Kevin Kalinsky, MD, MS, FASCO:
The addition of pembrolizumab added in some immunotherapy-related toxicities that were not unexpected.6
SACI-IO HR+: Clinical Implications
Kevin Kalinsky, MD, MS, FASCO:
My takeaway from this study is that this is not standard of care or changing clinical practice. Maybe there is still a population that we could identify that may benefit from adding an immune checkpoint inhibitor to an ADC in the HR‑positive setting, but we are not there yet.
Sara Hurvitz, MD:
At the end of the day, we still do not know whether adding an immune checkpoint inhibitor to sacituzumab is beneficial in estrogen receptor (ER)-positive disease. At the European Society of Medical Oncology Congress in 2023, we did see a couple of neoadjuvant studies in HR‑positive/HER2‑negative breast cancer indicating better pathologic complete responses (pCRs) by adding an immune checkpoint inhibitor to chemotherapy.7,8 Although intriguing, those were initial results without long-term follow-up and it is still unclear whether this is going to be a better option for patients. We will have to wait for larger phase III studies.
DESTINY-Breast07: Study Design
Sara Hurvitz, MD:
DESTINY‑Breast07 was a randomized, open‑label, 2‑part phase IB/II trial evaluating novel combinations with T‑DXd. This trial randomized patients to 1 of 3 arms: T‑DXd alone, T‑DXd with pertuzumab, and T‑DXd with durvalumab. Patients had to have previously untreated HER2‑positive advanced or MBC.9
This study was designed to get safety and early efficacy data primarily in patients receiving T‑DXd with pertuzumab. These data set the stage for the DESTINY‑Breast09 trial (NCT04784715), which is an ongoing phase III study evaluating T‑DXd with or without pertuzumab vs the CLEOPATRA10 regimen of docetaxel, trastuzumab, and pertuzumab in the frontline setting for HER2-positive MBC. This analysis presented the data of T-DXd monotherapy vs in combination with pertuzumab.9
DESTINY-Breast07: Baseline Characteristics
Sara Hurvitz, MD:
I think it’s interesting that almost two thirds of the patients enrolled had de novo MBC. We are seeing now in many studies a rising proportion of patients with de novo disease, likely owing to our very effective therapy available in the curative intent setting. It’s also notable that 10% to 15% of patients in both arms had previously received pertuzumab.9
DESTINY-Breast07: Response and PFS
Sara Hurvitz, MD:
The ORR for T‑DXd alone was 76.0% and slightly higher with the addition of pertuzumab at 84.0%. The median duration of response was not estimable for either arm. The 12‑month PFS rate for T-DXd monotherapy was 80.8% and 89.4% in the combination arm.9 So although we are not seeing that adding pertuzumab to T‑DXd dramatically increases the efficacy results, these data do support evaluating whether this combination is beneficial in a larger study.
DESTINY-Breast07: Adverse Events in >20% of Patients
Sara Hurvitz, MD:
With the combination of T-DXd and pertuzumab, we do see higher rates of anemia and diarrhea compared to T-DXd alone (40% vs 24%, and 62% vs 35%, respectively).9
DESTINY-Breast07: Clinical Implications
Sara Hurvitz, MD:
We await the results of the phase III DESTINY-Breast09 trial to better understand the benefit of adding pertuzumab to T-DXd in this setting. The DESTINY-Breast07 results show slightly higher rates of anemia and diarrhea, including grade 3 or higher. I think any small or incremental benefit that we see by adding pertuzumab to T‑DXd has to be balanced against the resulting increase in toxicity.
Back
