CME
Physicians: Maximum of 1.00 AMA PRA Category 1 Credit™
Released: August 18, 2023
Expiration: August 17, 2024
TROPiCS-02: Sacituzumab Govitecan vs CT for Previously Treated HR-Positive/HER2-Negative ABC—Final OS Analysis
Sara Tolaney, MD, MPH:
Next, we will move onto studies in patients with advanced breast cancer. We have seen a growing number of studies and approvals for antibody–drug conjugates (ADC) in the advanced setting. This year at ASCO, my colleagues and I16 presented the final OS data from the phase III TROPiCS-02 study of sacituzumab govitecan in patients with previously treated, HR-positive/HER2-negative advanced breast cancer.
Sacituzumab govitecan is a TROP-2–directed ADC that delivers a payload of SN-38, a derivative of irinotecan, with a high ratio of drug to antibody.]17 We have seen very robust activity with this agent. The FDA has approved sacituzumab govitecan for patients with unresectable locally advanced or metastatic TNBC previously treated with ≥2 prior systemic therapies (≥1 for metastatic disease).18 Based on results of the TROPiCS-02 trial, the FDA also approved a new indication for patients with unresectable locally advanced or metastatic HR-positive/HER2-negative breast cancer previously treated with ET and ≥2 additional systemic therapies for metastatic disease. HER2-negative status was defined as having an IHC of 0, 1+, or 2+/ISH-.
TROPiCS-02 enrolled 543 patients who had progressed after treatment with ET, CDK4/6 inhibition, and a taxane in any setting, and who had been previously treated with 2-4 prior lines of chemotherapy for metastatic disease.16 Patients were randomized to receive sacituzumab govitecan or physician’s choice of chemotherapy. The primary endpoint was PFS by blinded independent central review and secondary endpoints included OS, response, patient-reported outcomes, and safety. We also included an exploratory endpoint to look at OS by HER2 status determined by IHC.
TROPiCS-02: Baseline Characteristics
Sara Tolaney, MD, MPH:
The baseline characteristics of the study population confirm that this is a heavily pretreated group of patients.16 In both arms, the median number of prior lines of chemotherapy for metastatic disease was 3, with 56-58% having received ≥3 prior lines.
TROPiCS-02: Updated PFS (by BICR)
Sara Tolaney, MD, MPH:
The primary endpoint was previously reported, showing a significant improvement in PFS among patients treated with sacituzumab govitecan vs chemotherapy.19 This update includes the final PFS data, confirming an improvement from a median of 4.0 months with chemotherapy to 5.5 months with sacituzumab govitecan, yielding a hazard ratio of 0.65 (95% CI: 0.53-0.81; nominal P = .0001).16
TROPiCS-02: Updated OS
Sara Tolaney, MD, MPH:
This final analysis shows that sacituzumab govitecan significantly improved median OS vs chemotherapy (14.5 vs 11.2 months, respectively), yielding a hazard ratio of 0.79 (95% CI: 0.65-0.95; nominal P = .0133).16 With a longer median follow-up of 13 months, we continue to see benefit with sacituzumab govitecan in this final analysis.
TROPiCS-02: PFS and OS by TROP-2 Expression Level
Sara Tolaney, MD, MPH:
We also looked at outcomes stratified by biomarkers. First, we looked at TROP-2 expression, which was centrally tested in archival tumor tissue.16
Regardless of TROP-2 expression levels, patients had better PFS and OS outcomes with sacituzumab govitecan vs standard chemotherapy. Among those with lower TROP-2 expression, defined by an H score <100, the median OS was 14.9 months with sacituzumab govitecan vs 11.3 months with chemotherapy. Patients with higher TROP-2 expression (H score ≥100) had similar median OS values of 14.4 months vs 11.2 months, respectively.
TROPiCS-02: PFS and OS by HER2 IHC Status
Sara Tolaney, MD, MPH:
We also looked at outcomes by HER2 expression, which was tested locally by IHC.16 Again, sacituzumab govitecan improved outcomes vs chemotherapy in subgroups defined by HER2 expression, including those with tumors that were HER2 low or HER2 IHC of 0.
TROPiCS-02: Updated Response
Sara Tolaney, MD, MPH:
The updated response data continue to show that patients receiving sacituzumab govitecan have a higher objective response rate (21%) than those receiving standard chemotherapy (14%).16
TROPiCS-02: TEAEs
Sara Tolaney, MD, MPH:
The safety profile for sacituzumab govitecan is well established. As expected, neutropenia was one of the major toxicities in TROPiCS-02, with grade ≥3 events seen in 52% of patients treated with sacituzumab govitecan.16 Other notable toxicities with sacituzumab govitecan were diarrhea, with grade ≥3 events in 10% of patients, and low-grade alopecia, which occurred in 48%. There were no instances of grade ≥3 alopecia.
TROPiCS-02: Clinical Implications
Sara Tolaney, MD, MPH:
Overall, the TROPiCS-02 study showed a 35% reduction in the risk of progression or death and a 21% reduction in the risk of death with sacituzumab govitecan compared with standard chemotherapy in this heavily pretreated population with HR-positive/HER2-negative advanced breast cancer. Those benefits have remained consistent with longer follow-up, and no new safety signals have emerged.
I believe these results confirm that sacituzumab govitecan should be a standard option for patients with pretreated HR-positive/HER2-negative advanced breast cancer.
T-DXd Pooled Analysis by Age: Background
Erika Hamilton, MD:
Continuing our discussion of ADCs, we will now review results of a pooled analysis evaluating the efficacy and safety of trastuzumab deruxtecan by age in patients with HER2-positive advanced breast cancer.20
Trastuzumab deruxtecan comprises a HER2-directed antibody conjugated to a topoisomerase I inhibitor.21 The FDA has approved trastuzumab deruxtecan for 2 indications in breast cancer.
The first indication is for adults with unresectable or metastatic HER2-positive breast cancer who had received a prior anti–HER2-based regimen. This anti–HER2-based regimen could have been received either in the metastatic setting or in the neoadjuvant or adjuvant settings; with the latter, the patient must have experienced disease recurrence during or within 6 months of completing treatment.
The second indication is for adults with unresectable or metastatic HER2-low breast cancer, where HER2-low status is defined as IHC 1+ or IHC 2+/ISH- per an FDA-approved test, who had received a prior chemotherapy in the metastatic setting or who experienced disease recurrence during or within 6 months of completing adjuvant treatment.
Trastuzumab deruxtecan initially received an accelerated approval for HER2-positive disease based on results of the phase II DESTINY-Breast01 trial (DS8201-A-U201) and additional safety data from the DS8201-A-J101 trial.22 Regular approval for the current indication was granted in 2022 based on results of the phase III DESTINY-Breast03 trial.23 The approval in HER2-low disease was based on the phase III DESTINY-Breast04 trial.24 Despite these approvals, data are lacking on the efficacy and particularly the safety of trastuzumab deruxtecan in older patients with breast cancer.
The analysis presented at ASCO 2023 pooled data from patients with HER2-positive advanced breast cancer treated with trastuzumab deruxtecan at 5.4 mg/kg Q3W from DESTINY-Breast01, DESTINY-Breast03, and the phase III DESTINY-Breast02 trial.20 DESTINY-Breast01 was a single-arm trial evaluating trastuzumab deruxtecan in 253 patients previously treated with T-DM1, of whom 184 were treated at the 5.4-mg/kg dose. DESTINY-Breast02 randomized 608 patients previously treated with T-DM1 to either trastuzumab deruxtecan or the physician’s choice of trastuzumab/capecitabine or lapatinib/capecitabine. DESTINY-Breast03 randomized 524 patients previously treated with trastuzumab plus taxane who had disease recurrence within 6 months to either trastuzumab deruxtecan or T-DM1.
This pooled analysis compared the efficacy and safety of trastuzumab deruxtecan in patients aged younger than 65 years vs 65 years or older enrolled on these 3 trials, with exploratory analyses in the small subset of patients aged 75 years or older.
T-DXd Pooled Analysis by Age: Baseline Characteristics
Erika Hamilton, MD:
In the pooled analysis of 1315 patients, 78.3% to 81.2% were aged younger than 65 years, with 18.8% to 21.7% aged 65 years or older.20 Only 3.0% to 4.0% were aged 75 years or older.
The older population was generally healthier than we would expect among real-world patients because they had to meet the eligibility criteria for enrollment on these trials. That being said, when we look at the large group of patients who received trastuzumab deruxtecan, we can see that performance status was somewhat worse in those aged 65 years or older. Comorbidities were also more common in older vs younger patients across all treatment groups, particularly vascular disorders (54.4%-63.2% vs 25.2%-26.2%, respectively), hypertension (49.1%-63.2% vs 17.0%-18.3%), mild renal impairment (49.1%-57.9% vs 30.5%-37.4%), and moderate renal impairment (21.1%-36.8% vs 1.9%-5.2%). The rates for these comorbidities were even higher among those aged 75 years or older.
T-DXd Pooled Analysis by Age: Efficacy
Erika Hamilton, MD:
As shown here, the efficacy of trastuzumab deruxtecan was similar between those aged younger than 65 years vs 65 years or older in terms of median PFS, median OS, and ORR.20
T-DXd Pooled Analysis by Age: Safety
Erika Hamilton, MD:
Turning to safety outcomes, essentially all patients regardless of age experienced an AE related to trastuzumab deruxtecan.20 Grade ≥3 AEs related to trastuzumab deruxtecan were numerically more common among those aged 65 years or older vs younger than 65 years (54.2% vs 43.6%, respectively), as were AEs leading to discontinuation (23.7% vs 15.0%). Of concern, there was a numerical increase in drug-related AEs leading to death in older vs younger patients (1.7% vs 0.6%).
T-DXd Pooled Analysis by Age: ILD
Erika Hamilton, MD:
The investigators looked further at rates of drug-related ILD/pneumonitis, a potentially life-threatening toxicity that can occur with trastuzumab deruxtecan.20 Adjudicated ILD/pneumonitis was more common among patients aged 65 years or older vs younger than 65 years treated with trastuzumab deruxtecan (17.5% vs 11.8%, respectively). Most of these events were grade 1/2, and the overall increases were largely driven by higher rates of grade 2 events in older patients. Grade 2 events occurred in 7.2% of those aged younger than65 years, 11.3% aged 65 years or older, and 15.2% aged 75 years or older.
There was 1 grade 5 event (0.6%) and 6 grade 5 events (0.9%) in the older vs younger patients, respectively.
T-DXd Pooled Analysis by Age: Clinical Implications
Erika Hamilton, MD:
This is an important analysis because we generally lack data on older patients, who are underrepresented in clinical trials relative to their actual cancer incidence.25 The results may be particularly useful for shared decision-making with older patients eligible for trastuzumab deruxtecan.
Tolerability is extremely important, and the higher rate of grade 2 ILD reported among older vs younger patients is concerning. Once ILD has reached grade 2, we must initiate steroid therapy and permanently discontinue trastuzumab deruxtecan. Even if efficacy is similar between age groups—as was found in this analysis—the patient must be on the drug to benefit. These findings further stress the great importance of proactive monitoring, management, and patient education on signs and symptoms of ILD among our older patients initiating trastuzumab deruxtecan.
HER3-DXd in HER2-Negative MBC: Study Design
Erika Hamilton, MD:
The next study we will discuss is evaluating an investigational ADC, patritumab deruxtecan (HER3-DXd). HER3-DXd uses the same linker and the same payload as trastuzumab deruxtecan but is conjugated to an anti-HER3 antibody rather than an anti-HER2 antibody. At ASCO 2022, Krop and colleagues26 reported good activity of HER3-DXd in patients with TNBC, HR-positive breast cancer, and HER2-positive breast cancer.
At ASCO 2023, my colleagues and I27 presented results from the first part of a phase II trial evaluating HER3-DXd in patients with HER2 negative cancers. There will also be another cohort that will look at patients with HER2-positive metastatic disease after T-DXd. Our report at ASCO 2023 presented results from the HER2-negative cohort. The primary endpoints are response and 6-month PFS.
One question we hoped to address is whether there is a minimum threshold of HER3 expression needed for HER3-DXd to be active. The degree of antigen expression needed for an ADC to work is an open question in the field right now. As we saw in Dr Tolaney’s discussion of TROPiCS-02, we do not select for TROP-2 expression before treating with sacituzumab govitecan. By contrast, we do select for HER2 expression before treating with trastuzumab deruxtecan. This agent was initially used for patients with high HER2 levels but is now given to patients with both high and low HER2 expression.
HER3-DXd in HER2-Negative MBC: Baseline Characteristics
Erika Hamilton, MD:
The patients in this analysis were heavily pretreated, with a median of 3 prior therapies in the metastatic setting.27 Of the 48 patients with tumor characteristics available, 19 had TNBC and the remainder had estrogen receptor (ER)–positive disease.
This population had a high level of HER3 expression overall, with 64% of patients having high HER3 expression (≥75%), 28% with HER3 expression 25 74%, and only 4 patients (8.5%) with HER3 <25%.
HER3-DXd in HER2-Negative MBC: Response by Receptor Expression
Erika Hamilton, MD:
The ORR for the full population was 35%.27 Considering these patients were so heavily pretreated—essentially in the fourth line setting, on average—this outcome seems quite meaningful to me.
Looking at the response rates across the breast cancer subtypes, the ORR was approximately 40% for patients with ER-positive disease and approximately 20% for those with TNBC. Of importance, response was independent of HER3 expression, with responses observed among patients with HER3-high and HER3-low disease. There is no signal that patients need to have high HER3 expression to benefit from HER3-DXd.
HER3-DXd in HER2-Negative MBC: Safety and Dosing
Erika Hamilton, MD:
Nausea, fatigue, and diarrhea were the most common AEs associated with HER3-DXd.27 Unlike some early trastuzumab deruxtecan studies where we did not use prophylactic antiemetics, we have now developed protocols using prophylactic drugs to help patients better tolerate ADC therapy. In this analysis, just approximately one half of patients had any grade nausea, and grade ≥3 nausea was very infrequent, seen in less than 3% of patients. Fewer than one half of patients experienced any-grade fatigue or diarrhea.
As our understanding has grown for this class of drugs, I think we have learned to do a better job managing these AEs, and this regimen was pretty well tolerated.
HER3-DXd in HER2-Negative MBC: Clinical Implications
Erika Hamilton, MD:
Moving forward, this study is now enrolling for an expansion cohort of patients with either TNBC or HR-positive disease and any HER3 expression level.27 This cohort will also include patients who have already been treated with a different ADC, such as trastuzumab deruxtecan or sacituzumab govitecan, to help address the question of sequencing and the activity of an ADC after another ADC. We are also enrolling in Part Z, which will include patients with HER2 positive metastatic breast cancer with prior trastuzumab deruxtecan, again to address the question of sequencing ADCs.
Sara Tolaney, MD, MPH:
Although these are small numbers in a single-arm phase II study, I find these results pretty exciting because they are showing real antitumor activity with HER3-DXd in this HER2-negative metastatic setting.
Approach to Therapy for HR-Positive/HER2-Negative MBC: Move to Personalization
Sara Tolaney, MD, MPH:
Where do ADCs fit into our treatment algorithm for patients with HR-positive, HER2-negative metastatic disease? For these patients, we first try CDK4/6 inhibitors, followed by ET and targeted therapies. For patients who progress on those, I move on to chemotherapy—usually something like capecitabine.
Beyond that, I want to know if the patient has low or no HER2 expression. For a patient with HER2-low disease, I would next try trastuzumab deruxtecan. If they have no HER2 expression, I would choose sacituzumab govitecan. If a patient with HER2-low disease has already had trastuzumab deruxtecan, I would then choose sacituzumab govitecan.
As more ADCs enter the treatment landscape, the challenge is becoming how and when we use each one. We are waiting to see how both HER3-DXd and datopotamab deruxtecan may fit into our treatment decisions. We must wait to see more data with HER3-DXd, and we are expecting results within the next year on datopotamab deruxtecan in the population with HR-positive metastatic disease from the TROPION-Breast01 trial.
We need more data on sequencing these agents, particularly those with the same payload. Can ADCs with the same payload but different targets be used sequentially? It is an exciting time with all these new drugs, but we have a lot to learn about how best to use them.
PADA-1 Update: Study Design
Erika Hamilton, MD:
ESR1 mutations are rare in primary breast cancers and at first recurrence of metastatic disease but may be acquired in up to 40% of patients with treatment-refractory HR-positive/HER2-negative metastatic breast cancer.28 These alterations confer resistance to AIs, but tumors harboring an ESR1 mutation still have some sensitivity to fulvestrant.
The phase III PADA-1 trial investigated whether patient outcomes improve when we take clinical action before clinical progression in response to early detection of ESR1 mutations in ctDNA analysis of blood samples (bESR1mut).29 Previously, PADA-1 demonstrated that median PFS was doubled when patients with HR-positive/HER2-negative metastatic breast cancer and rising bESR1mut on first-line AI plus palbociclib were switched to fulvestrant plus palbociclib vs continuing treatment. After a median follow-up of 26.0 months, the median PFS was 11.9 months vs 5.7 months, respectively, with a hazard ratio of 0.61 (95% CI: 0.43-0.86; P = .0040).
We should note that PADA-1 defined early detection of ESR1 mutations as either newly emergent in those without baseline mutations or increasing levels in those with baseline ESR1 mutations. Patients had bESR1mut centrally assessed at inclusion, after the first cycle and then every 2 cycles.
At ASCO 2023, Cabel and colleagues30 presented an updated analysis of PADA-1 after a longer follow-up.
PADA-1 Update: PFS From Randomization (Primary Endpoint)
Erika Hamilton, MD:
The PFS benefit was maintained in this updated analysis of PADA-1.30 After a median follow-up of 28.2 months, patients who switched to fulvestrant plus palbociclib had a significantly longer median PFS of 12.8 months vs 5.8 months in those who continued on AI plus palbociclib despite rising bESR1mut, yielding a hazard ratio of 0.54 (95% CI: 0.38-0.75).
This analysis also compared outcomes by bESR1mut status at 2 months after randomization. Among the 163 patients with ctDNA results available at 2 months, no bESR1mut was detected in 68.2% switched to fulvestrant plus palbociclib arm vs 32.1% of the AI plus palbociclib arm. Those with no detectable bESR1mut at 2 months after randomization had a significantly lower risk of progression or death vs those with a mutant allele frequency >0% (P <.001).
These results indicate that bESR1mut status at 2 months after randomization was an independent prognostic factor for PFS. By contrast, neither ESR1 mutation type nor ESR1 polyclonal status was found to be associated with PFS.
PADA-1 Update: PFS2 From Randomization
Erika Hamilton, MD:
The updated analysis also reported on time to second progression or death (PFS2).30 The median PFS2 was slightly more than doubled among those who switched vs continued treatment (29.4 vs 14.0 months, respectively). The hazard ratio for PFS2 was 0.37 (95% CI: 0.24-0.56), clearly favoring the strategy of switching to fulvestrant plus palbociclib at early detection of rising bESR1mut.
PADA-1 Update: Clinical Implications
Erika Hamilton, MD:
This trial is notable as the first study in breast cancer to demonstrate benefit with switching therapy based on a genomic assay before clinical progression. Furthermore, the assay was blood based, rather than requiring a more invasive biopsy. This updated analysis showed that the PFS benefit with switching vs continuing therapy was maintained with longer follow-up. Switching to fulvestrant plus palbociclib at early detection of rising bESR1mut was also beneficial for median PFS2. ESR1 mutation status has become vital for selecting the optimal ET, and this dynamic assessment further improves our understanding of its clinical utility.
SONIA: Study Design
Sara Tolaney, MD, MPH:
In the setting of HR-positive metastatic disease, an important question is whether a CDK4/6 inhibitor should be given as the first-line or second-line treatment. Delaying the CDK4/6 inhibitor would reduce a patient’s exposure to the drug and associated toxicity over their lifetime. Would this delaying strategy yield a long-term outcome similar to giving CDK4/6 inhibition in the first-line setting?
The SONIA trial was designed to address this tricky question. SONIA is a very clever investigator-initiated, randomized phase III study, in which patients with HR-positive metastatic breast cancer received either an AI plus a CDK4/6 inhibitor or the AI alone up front.31 Upon progression, the patients switched to the other treatment: If you started with an AI plus CDK4/6 inhibitor, you would switch to ET with fulvestrant alone. Patients who received an AI alone up front would switch to fulvestrant plus the CDK4/6 inhibitor as second-line treatment.
The study was focused on longer-term outcomes, with a primary endpoint of PFS2. It was originally designed as a superiority study, not a noninferiority study—an important consideration affecting the clinical implications of SONIA’s results, which we will discuss shortly. Secondary endpoints included OS, quality of life, and cost effectiveness.
SONIA: Baseline Characteristics
Sara Tolaney, MD, MPH:
The patients in this trial were either newly diagnosed or had a disease free–interval lasting longer than 12 months after adjuvant therapy because they were starting treatment with an AI.31 Approximately 40% of patients had prior neoadjuvant or adjuvant chemotherapy, and approximately one half had prior ET in the adjuvant setting.
On study, 91% received palbociclib as their CDK4/6 inhibitor, with 8% receiving ribociclib and the small remainder receiving abemaciclib.
SONIA: PFS1
Sara Tolaney, MD, MPH:
We have seen data on median PFS for AIs with and without CDK4/6 inhibition before in this setting. The results for SONIA were consistent with these prior data. The addition of a CDK4/6 inhibitor to first-line AI prolonged the median PFS from 16.1 months to 24.7 months with a hazard ratio of 0.59 (95% CI: 0.51-0.69; P <.0001).31
Erika Hamilton, MD:
What was interesting to me is that patients who received the AI alone did a little bit better than I would have thought. Most large trials reported a slightly shorter median PFS of under 15 months, but SONIA reported a slightly longer median PFS of 16.1 months.32,33,34
SONIA: PFS2 (Primary Endpoint)
Sara Tolaney, MD, MPH:
The main question was about the time from registration to progression after the second-line regimen.31 The median PFS2 was pretty similar between the treatment arms: 31.0 months among patients who received CDK4/6 inhibition in the first-line setting vs 26.8 months in the second-line setting, yielding a hazard ratio of 0.87 (95% CI 0.74-1.03). PFS2 outcomes were consistent across all analyzed patient subgroups, with no significant differences seen between the 2 treatment arms.
Recall that this study was designed for superiority, which it did not meet. The investigators interpreted the results to say that PFS2 outcomes did not differ between patients who received CDK4/6 inhibition in the first-line vs second-line setting.
SONIA: Overall Survival
Sara Tolaney, MD, MPH:
OS was also similar between the treatment arms, with a median OS of 45.9 months with first-line AI plus CDK4/6 inhibition vs 53.7 months with first-line AI alone.31 The numerical trend suggests that receiving CDK4/6 inhibition in the second-line setting leads to longer survival, but I would not say that. The hazard ratio for OS is 0.98 with a broad 95% CI of 0.80-1.20 (2-sided P = .83).
SONIA: Clinical Implications
Sara Tolaney, MD, MPH:
This study did not show superiority of one CDK4/6 inhibitor sequencing strategy over the other. Although SONIA was not designed to confirm noninferiority, the similar outcomes between treatment arms led the investigators to conclude that holding CDK4/6 inhibition until the second-line setting could be an acceptable option. Furthermore, delaying use of a CDK4/6 inhibitor until the second-line setting resulted in less exposure and therefore less toxicity.
However, there are a few issues to consider. First, the vast majority of patients enrolled in this trial received palbociclib. Different CDK4/6 inhibitors may have different efficacy, and we do not know if the investigators would have seen similar results with a more potent CDK4/6 inhibitor, such as abemaciclib or ribociclib. Administering those more potent CDK4/6 inhibitors in an earlier setting could potentially matter more.
Erika Hamilton, MD:
I consider this to be a very important point. Palbociclib is the one CDK4/6 inhibitor that has not shown an OS benefit, which has already raised the question of whether palbociclib is inferior compared with other CDK4/6 inhibitors. Before we had OS data, I used palbociclib almost exclusively in the first line, but now I often choose other CDK4/6 inhibitors because of the lack of OS benefit demonstrated by PALOMA-1 and PALOMA-2.35,36 I do not find the OS results for SONIA especially surprising.
As I mentioned earlier, the outcomes in the control arm were a little better than we would have anticipated based on other studies. If the control outcomes here were more consistent with previous findings, we might have seen a larger separation between the OS curves.
Sara Tolaney, MD, MPH:
Another issue is the study design. SONIA was not designed to evaluate noninferiority, yet that is how the investigators framed their conclusions. I think one must be cautious to suggest noninferiority when the trial was not statistically designed to demonstrate that.
Erika Hamilton, MD:
I agree with Dr. Tolaney’s point about this trial being designed originally to show superiority. The investigators’ conclusions sounded like more of a noninferiority analysis, which raises questions.
In addition, this was a relatively small study, with only approximately 500 patients in each treatment arm. Although there was a trend toward better survival among patients who received first-line CDK4/6 inhibition, this is a small dataset in the context of the body of evidence for CDK4/6 inhibitors.
Sara Tolaney, MD, MPH:
Another issue is that endocrine monotherapy after CDK4/6 inhibition is not our current standard of care. In this setting, fulvestrant is usually combined with a targeted agent, not administered as monotherapy. This trial does not tell us what we might expect for patients who receive first-line CDK4/6 inhibition followed by a targeted agent in combination with ET.
Given these issues, I am not ready to use SONIA to change my practice. I will still use ET and CDK4/6 inhibition in the first-line setting.
I do think this was a very important and provocative study addressing an important question. I believe there probably are some patients for whom we could give an AI alone up front. The problem is that we need a biomarker to help us identify who those patients are, and we are not there yet.
Erika Hamilton, MD:
We should also note that SONIA was conducted in the Netherlands, which has a universal healthcare system. When presenting these results, the investigators made the provocative statement that administering the CDK4/6 inhibitor for a shorter time by delaying it until second line allowed this trial to pay for itself. This is not relevant to the United States, however, and many of us had questions about the outcomes and design of this trial.
In countries where the cost-benefit of cancer care is strictly regulated, the SONIA results may change practice. In countries like the United States, I do not think they will change how we treat all patients. These data may lend support to tailoring your approach for a patient for whom a first line CDK4/6 inhibitor may not be the best option—for example, an older patient or someone with comorbidities or problems with infection. For me, however, this is not going to change my practice across the board.
PALMIRA: Study Design
Sara Tolaney, MD, MPH:
Our final discussion will focus on a study addressing the question of continuing CDK4/6 inhibition beyond progression in advanced breast cancer. The small, randomized phase II MAINTAIN trial reported a PFS benefit with ribociclib after progression on CDK4/6 inhibition.37 However, we saw the opposite in the phase II PACE trial, where palbociclib plus ET did not provide any survival benefit after progression on a CDK4/6 inhibitor.38
The PALMIRA trial is also evaluating palbociclib after progression on palbociclib. This phase II trial enrolled patients with HR-positive/HER2-negative advanced breast cancer who progressed after upfront ET plus palbociclib and randomized them to switch to a different endocrine backbone, alone with or without continued palbociclib.39 The primary endpoint was PFS.
PALMIRA: Baseline Characteristics and Patient Disposition
Sara Tolaney, MD, MPH:
The baseline characteristics show that approximately 84% of patients received first-line palbociclib for more than 12 months.39 This is consistent with clinical judgment where we would consider continuing a CDK4/6 inhibitor in patients who did well on their first line of therapy with a CDK4/6 inhibitor.
PALMIRA: Investigator-Assessed PFS (Primary Endpoint)
Sara Tolaney, MD, MPH:
Nonetheless, continuing CDK4/6 inhibition beyond progression did not lead to any difference in PFS.39 Median PFS was 4.9 months in patients who continued palbociclib after progression vs 3.6 months in those who did not, yielding a hazard ratio of 0.84 (95% CI: 0.66-1.07; 2-sided P = .149). This outcome was consistent across the majority of patient subgroups.
These data are consistent with the PACE trial, where the median PFS was 4.6 months with palbociclib.38
PALMIRA: Clinical Implications
Sara Tolaney, MD, MPH:
Unfortunately, the PALMIRA and PACE trials both show us that continuing palbociclib after progression on palbociclib does not lead to significant survival benefit for patients.
MAINTAIN showed positive results with ribociclib but was a small study. At this point, I would not use palbociclib post-palbociclib, but I might consider using ribociclib post-palbociclib in select patients.
Erika Hamilton, MD:
We have to consider the cost of CDK4/6 inhibitors in terms of both finances and issues such as fatigue, monitoring counts, neutropenia, etc. We know that treatment related AEs are higher on CDK4/6 inhibitors combined with ET vs ET alone.
The PALMIRA results support my conclusion not to continue palbociclib after progression on palbociclib. However, I think it is less clear whether switching to a different CDK4/6 inhibitor after palbociclib could yield benefit, especially as data now indicate that palbociclib may not be the most potent CDK4/6 inhibitor. Could patients benefit from receiving ribociclib, as in MAINTAIN, or even abemaciclib in the second line? I think this remains an unanswered question. Treating a patient with either ribociclib or abemaciclib beyond progression on CDK4/6 inhibition may be a possibility.
Other types of agents may offer other options for combinations with ET in second-line treatment of HR-positive disease. For example, the AKT inhibitor capivasertib has shown PFS benefit in combination with fulvestrant.40 There are also multiple new endocrine backbones, ranging from selective estrogen receptor degraders to proteolysis-targeting chimeras. This is a treatment area that is evolving quickly.
Approach to Newly Diagnosed HR-Positive/HER2-Negative MBC
Sara Tolaney, MD, MPH:
With the ongoing rapid evolution of the treatment landscape, it is critical to put these data in context to determine treatment approaches for newly diagnosed advanced breast cancer. In an excellent abstract discussion at ASCO 2023, Dr Daniel Stover pulled together many of these elements into a very nice algorithm for HR-positive, HER2-negative metastatic disease. 41
When selecting first-line therapy, we would choose to use a CDK4/6 inhibitor with ET for most patients. Our preference is to use ribociclib or abemaciclib because we have not seen an OS benefit with palbociclib. That being said, palbociclib may be appropriate in the upfront setting for some patients given its tolerability profile. Other options to consider for select patients who are not candidates for CDK4/6 inhibition might be ET alone or for patients in visceral crisis needing a rapid response, chemotherapy.42
Once a patient progresses on ET plus CDK4/6 inhibition, treatment selection becomes more challenging because there are many options. I think it is critical upon progression to collect ctDNA and do a genomic test to assess what mutations are present at the time of progression.
If the patient has a PIK3CA mutation, their disease may be somewhat resistant to ET. I like to give a PI3K or AKT inhibitor in combination with ET. The main option right now is alpelisib with fulvestrant. The oral AKT inhibitor capivasertib is under FDA review, and I hope we will see its approval later this year, making this potentially an option for patients with PI3KCA, AKT, or PTEN alteration. That would give us a choice of either alpelisib or capivasertib with fulvestrant for patients with PI3KCA mutation. Assuming capivasertib is approved, I think many physicians would choose capivasertib over alpelisib, given its more favorable toxicity profile with lower incidence of hyperglycemia and rash.
If the patient has an ESR1 mutation, we can consider treatment with elacestrant. The concern with elacestrant is that its median PFS was short, only approximately 4 months, in ESR1-mutant disease.43 However, at ASCO 2023, we saw a nice subgroup analysis focused on patients who had done well on their up-front CDK4/6 inhibitor. Patients with ESR1 mutations who received an up-front CDK4/6 inhibitor for ≥12 months had a prolonged PFS of more than 8 months with elacestrant.44
Elacestrant might be a good choice for a patient who has disease that is PI3KCA wild type and ESR1 mutant, and who had a durable benefit with a CDK4/6 inhibitor.
If the patient has a germline BRCA or PALB2 mutation, I would consider PARP inhibition.
For patients who do not fall into any of those categories, I would generally give fulvestrant and everolimus. I do not typically use fulvestrant alone.
Beyond this algorithm, I might consider very select patients for fulvestrant plus ribociclib or abemaciclib in the second-line setting. They would need to have been on their first-line CDK4/6 inhibitor for a very long time (ie, 2-3 years) and experienced only the smallest bit of progression—for example, maybe one bone lesion looking a little bit different. For this patient, I might try a continuation strategy such as switching the endocrine backbone in combination with ribociclib or abemaciclib.
I hope we will see more data soon to help us better understand this type of continuation strategy. The phase III postMONARCH trial is looking at the use of abemaciclib and fulvestrant in the post-CDK4/6 setting, either as a continuation of abemaciclib or a switch from another CDK4/6 inhibitor (NCT05169567). The results from this trial will be very important to understand the utility of continuing CDK4/6 inhibition into the second-line setting.
Back
