Joyce O’Shaughnessy, MD:
Bardia and colleagues¹² presented an interesting update from the EMERALD trial, which I think will help us better understand who will ultimately benefit from the oral selective estrogen receptor degrader (SERD) elacestrant. This randomized, active-controlled phase III trial enrolled 478 patients with HR‑positive/HER2‑negative metastatic breast cancer who had received a previous CDK4/6 inhibitor. Patients received either elacestrant or investigator’s choice of fulvestrant or an AI as standard of care; most patients received fulvestrant. The coprimary endpoints were PFS in the intention-to-treat population and PFS in the patients who had an ESR1 mutation. 

Joyce O’Shaughnessy, MD:
Per the study design, all of the patients had a previous CDK4/6 inhibitor. Most—approximately 70% in each arm—had visceral disease. Approximately 50% of the patients had ESR1 mutations.

Joyce O’Shaughnessy, MD:
The investigators previously reported the primary analysis, which showed a statistically significant improvement in PFS with elacestrant compared with standard ET in both the intention-to-treat and in the patients with ESR1 mutations, which is very encouraging.¹³ The median PFS values were only modestly impacted by elacestrant vs standard of care, but there is a tail on the PFS curve indicating that a greater percentage of patients had prolonged benefit from elacestrant, both in the intention-to-treat population and in the ESR1‑mutated population. It is clear that elacestrant is benefiting patients with more ET‑sensitive disease and that the ESR1 mutation is conferring an ET‑sensitive phenotype to those breast cancers. Using targeted endocrine therapies that inhibit this mutant estrogen receptor can really benefit these patients. We can select patients based on the presence of an ESR1 mutation, but not all patients with this mutation benefit, so how can we otherwise identify and enrich for patients whose breast cancers are ET sensitive and who will benefit from elacestrant?

To try to start addressing this question, the analysis presented this year at SABCS focused on the duration of the antecedent CDK4/6 inhibitor therapy and its impact on PFS. The authors divided the study population based on duration of previous CDK4/6 inhibition of ≥6 months, ≥12 months, or ≥18 months. Recall that the median PFS is approximately 2 years in most of the first‑line CDK4/6 trials.¹⁴ The bottom line of this analysis is that patients who had received the antecedent CDK4/6 inhibitor for a longer time had greater benefit from ET in general—either standard of care or elacestrant. For example, patients who received the CDK4/6 inhibitor for ≥18 months had the longest median PFS for both elacestrant and standard of care. For the patients who received elacestrant, a PFS of 5.5 months and a 12‑month PFS rate of 27% vs 8% with standard of care are quite clinically meaningful.

Joyce O’Shaughnessy, MD:
Examining a similar analysis of the patients with ESR1 mutation, the results are even more impressive. Patients who received CDK4/6 inhibition for ≥12 months had a median PFS of 8.6 months, with 35% of the patients still progression free at 1 year compared with only approximately 8% of those receiving standard of care. These are very interesting data and, again, support the idea that an ESR1 mutation indicates potential for ET sensitivity if we can give patients an ET that works against the mutant estrogen receptor.

Joyce O’Shaughnessy, MD:
No new safety signals emerged in this analysis. Most AEs were low grade, and there was a low rate of discontinuation because of toxicity. 

Sara A. Hurvitz, MD:
SERENA‑2 was a phase II clinical trial looking at various doses of another oral SERD, camizestrant, compared with fulvestrant in 240 postmenopausal patients with ER‑positive, HER2-negative advanced breast cancer. This study originally had 4 treatment arms, but the 300 mg dosing of camizestrant was stopped early. The primary results that were presented evaluated camizestrant at 75 mg or 150 mg, compared with fulvestrant at 500 mg.¹⁶ Patients could have measurable or nonmeasurable disease, and they had to be candidates for fulvestrant monotherapy. They must have progressed on ≥1 line of ET but had to have had ≤1 line of ET or chemotherapy for advanced disease. No previous fulvestrant or oral SERD was allowed. Investigator‑assessed PFS was the primary endpoint and secondary endpoints included response and safety.

Sara A. Hurvitz, MD:
Looking at the baseline characteristics, of interest, the patients in the fulvestrant arm had a higher rate of ESR1 mutations—47.9%—compared with only 29.7% in the camizestrant 75 mg arm and 35.6% in the camizestrant 150 mg arm. Quite a few patients—nearly one third in the camizestrant 75 mg arm and almost one half in the fulvestrant arm—had liver metastases. As you can see, the majority of patients had received an AI in the adjuvant setting or later, and one half had received a previous CDK4/6 inhibitor.

Sara A. Hurvitz, MD:
The investigator assessment of PFS showed that each dose of camizestrant was associated with a significantly improved PFS compared with fulvestrant. The median PFS was 7.2 months and 7.7 months with camizestrant 75 mg and 150 mg, respectively, compared with 3.7 months with fulvestrant—a 4-month difference that is a clinically meaningful improvement. 

Sara A. Hurvitz, MD:
The investigators also reported PFS in prespecified subgroups. These numbers are very small, so it is important not to overstate conclusions from the results here. Patients who had previous CDK4/6 inhibitors tended to have a shorter PFS, but the trend is still in favor of camizestrant over fulvestrant. For patients with evidence of an ESR1 mutation at baseline, the median PFS was 2.2 months with fulvestrant and much higher with camizestrant: 6.3 months with the 75 mg dose and 9.2 months with the 150 mg dose. We are beginning to see a consistent story emerge that patients with an ESR1 mutation are the ones who benefit most from the use of an oral SERD. 

Sara A. Hurvitz, MD:
The ORRs and clinical benefit rates associated with camizestrant were numerically higher than those with fulvestrant, although in this small trial statistical significance was not seen. 

Sara A. Hurvitz, MD:
The safety of this drug does appear to be overall tolerable, but there is an eye disorder called photopsia that is noted by patients receiving camizestrant. It is often described like seeing a flashing light when entering a bright room. It tends to be mild, with no grade 3/4 cases, but it is a unique AE. Sinus bradycardia was also noted in approximately one quarter of patients with camizestrant at 150 mg, but no grade ≥3 events. 

Joyce O’Shaughnessy, MD:
EMBER is a phase Ia/b noncomparative trial with the oral SERD imlunestrant and abemaciclib with or without an AI.¹⁷ Patients were randomized to the 2 trial arms for enrollment purposes, but not for comparison between the treatments. The trial enrolled 85 patients with metastatic HR‑positive/HER2‑negative advanced breast cancer. No previous CDK4/6 inhibitor therapy was allowed.  

Joyce O’Shaughnessy, MD:
As the primary endpoint, the recommended phase II dose was determined to be 400 mg of imlunestrant with abemaciclib, which is the same dose that is used as monotherapy. Safety of this combination was very good. The pattern of toxicity reported here appears to be driven by the gastrointestinal toxicity of abemaciclib, but toxicity was manageable. At the recommended phase II dose, dose reductions of imlunestrant, abemaciclib, and both agents occurred in 1%, 29%, and 6% of patients, respectively. The regimen was quite tolerable, with the toxicity essentially what you would expect with abemaciclib.

Joyce O’Shaughnessy, MD:
The efficacy data did not convince me that adding the AI to imlunestrant and abemaciclib was beneficial. I think the trial will conclude that AI probably does not add to the ORR. The clinical benefit rate, which I think is probably more clinically relevant than ORR and more important in the natural history of the disease, is very similar in both treatment arms; and 12‑month PFS is the same in both arms at 80%. It does not appear that adding the AI gives us any additional benefit.

Sara A. Hurvitz, MD:
The RIGHT Choice trial was a fascinating look at how a CDK4/6 inhibitor compares with multiagent chemotherapy in the highest‑risk breast cancer patients. A lot of clinicians are still uncomfortable using ET and a CDK4/6 inhibitor in a young patient with aggressive disease, such as one with symptomatic visceral metastases. This study directly addressed how well ribociclib with ET would work in comparison with multiagent chemotherapy in patients with aggressive disease.²¹

Patients enrolled on this trial had to be pre- or perimenopausal, so these are younger women. The ER had to be >10% expressed in the tumor and these patients had aggressive disease—such as symptomatic visceral metastases, rapid disease progression or impending visceral compromise—or markedly symptomatic nonvisceral disease. The majority of patients enrolled were treatment naive and had de novo metastatic breast cancer—these were not patients who relapsed quickly after ET. Patients had to have an Eastern Cooperative Oncology Group performance score of ≤2 and bilirubin ≤1.5 the upper limit of normal. The primary endpoint was locally assessed PFS.  
 

Sara A. Hurvitz, MD:
PFS was significantly longer with ribociclib plus ET, with the median PFS nearly doubled at 24 months compared with combination chemotherapy at 12.3 months. These results are quite interesting. Even overlapping PFS curves would have been encouraging to me, but to see how well ribociclib performed in this trial was very exciting. 

Sara A. Hurvitz, MD:
The time to treatment failure was much shorter for patients treated with combination chemotherapy, at a median of 8.5 months vs 18.6 months with ribociclib.

Sara A. Hurvitz, MD:
I also found the response rates to be encouraging: approximately two thirds of patients treated with ribociclib had a tumor response, which compared favorably with chemotherapy. The clinical benefit rate appeared to be better with ribociclib. The median time to response was quite good—4.9 months with ribociclib—which was a little bit longer than the 3.2 months seen with combination chemotherapy but not statistically significant. I think these data underscore the idea that it is reasonable to treat patients who are treatment naive with ribociclib, even if they have aggressive symptomatic disease. 

Sara A. Hurvitz, MD:
I think it is important to call out the high rate of dose reductions seen with patients treated with combination chemotherapy, with 20% of these patients having ≥3 dose reductions. No patients receiving a CDK4/6 inhibitor had ≥3 dose reductions. The rate of treatment‑related serious AEs was also higher with combination chemotherapy, including grade 3/4 events, and more patients receiving chemotherapy discontinued treatment because of toxicity. It is likely that these dose reductions are affecting the benefit of chemotherapy. 

Sara A. Hurvitz, MD:
It is not surprising that ribociclib is associated with neutropenia and leukopenia. There were other grade 3/4 events as well. For example, grade 3/4 anemia and increased alanine aminotransferase (ALT) or aspartate aminotransferase (AST) appear relatively comparable in the 2 treatment arms. 

Joyce O’Shaughnessy, MD:
CAPItello-291 is a randomized, double-blinded phase III trial of the first oral AKT inhibitor capivasertib. This is an agent that I am really looking forward to having in my practice. This trial enrolled 708 patients with HR‑positive/HER2‑negative metastatic breast cancer and randomized them to receive fulvestrant and either placebo or capivasertib.²⁴ Capivasertib was dosed at 400 mg twice daily, 4 days on and 3 days off, to manage toxicity, mainly diarrhea. The trial required that more than one half of the patients enrolled had received a previous CDK4/6 inhibitor. The dual primary endpoints are investigator-assessed PFS in the intention-to-treat population and in the subset of patients with an AKT pathway‑altered tumor. Those alterations include a PIK3CA, AKT1, or PTEN alteration.  

Joyce O’Shaughnessy, MD:
Approximately 40% of the patients in each treatment arm had an AKT pathway alteration. Most of the genomic alterations were PIK3CA mutations. Approximately 5% had an AKT1 mutation and approximately 5% to 6% had a loss of PTEN. Approximately 70% of patients in each arm received a previous CDK4/6 inhibitor and two thirds had visceral metastasis.

Joyce O’Shaughnessy, MD:
In the intention-to-treat population, the PFS was significantly improved with capivasertib and fulvestrant, with a median PFS of 7.2 months vs 3.6 months with placebo and fulvestrant. This improvement was statistically significant, with a hazard ratio of 0.60.

Joyce O’Shaughnessy, MD:
PFS outcomes were similar in the patients with AKT pathway alterations: a median PFS of 7.3 months in the capivasertib arm and 3.1 months in the placebo arm. The hazard ratio was 0.50, which was slightly better than that in the overall population. The investigators also reported that an exploratory analysis showed improved PFS with capivasertib in the nonaltered population, with a hazard ratio of 0.70. The CIs did not cross 1 on any of those hazard ratios. There are other mutations that activate the AKT pathway besides those analyzed here, but it appears that even in the nonaltered population we may still see benefit from capivasertib.

Joyce O’Shaughnessy, MD:
In the subgroup analysis of the overall population, it is very encouraging to see capivasertib was favored in both patients with liver metastases and those with previous use of CDK4/6 inhibitors, with very good hazard ratios of approximately 0.6. Patients with liver metastases in particular need better treatment options, so these results are very encouraging to see.

Joyce O’Shaughnessy, MD:
Investigator-assessed response rates were also better with capivasertib compared with placebo. With capivasertib, the ORR was nearly double that with placebo in the overall population and approximately triple that in the AKT pathway‑altered subgroup.

Joyce O’Shaughnessy, MD:
The OS data have only reached 28% maturity in the overall population, but they are trending toward a benefit with capivasertib. The hazard ratio is 0.74 and the CI does not cross 1. Longer follow-up is needed, but these early survival data are promising. 

Joyce O’Shaughnessy, MD:
Toxicity with capivasertib was generally quite manageable. Approximately 20% of patients did have a dose reduction with capivasertib.

Joyce O’Shaughnessy, MD:
Much of the need for a dose reduction of capivasertib is to manage diarrhea. Generally, patients experience some diarrhea during the 4 days they are on capivasertib, and then it improves during the 3 days they are off the drug, so it is somewhat managed. Patients experienced some hyperglycemia, but the incidence is much lower with this agent than it is with alpelisib.²⁵ There was also some rash seen with the capivasertib, but overall, the toxicity appears manageable

Sara A. Hurvitz, MD:
The open-label, randomized phase II PACE clinical trial evaluated the benefit of continuing a CDK4/6 inhibitor, with ET and/or an immune-based therapy, avelumab, in patients whose disease has progressed on a CDK4/6 inhibitor.²⁷ Enrolled patients had ≤2 previous lines of therapy (≤1 chemotherapy) for metastatic disease and had been on ET with stable disease for ≥6 months, but had progressed on a CDK4/6 inhibitor. No previous fulvestrant was allowed. The patients were randomized 2:1:1 to receive fulvestrant and palbociclib, fulvestrant and palbociclib with avelumab, or fulvestrant alone. The primary endpoint evaluated PFS for fulvestrant plus palbociclib vs fulvestrant alone. 

It is important to note that the majority of patients entered this study from a palbociclib‑based therapy. Fewer than 10% of patients enrolled in this clinical trial had disease progression on ribociclib or abemaciclib. Thus, PACE is really addressing the questions of whether you can swap out the endocrine partner in patients whose disease is progressing on palbociclib plus an AI, and whether adding an immune checkpoint inhibitor benefits patients.  

Sara A. Hurvitz, MD:
The median PFS was no different for patients who had palbociclib plus fulvestrant vs fulvestrant alone (4.6 months vs 4.8 months, respectively). Thus, adding palbociclib to fulvestrant in patients who had recently progressed on palbociclib with an AI does not benefit patients; they could do just as well with fulvestrant alone. 

However, an interesting signal was observed with continued palbociclib and the addition of both fulvestrant and the immune checkpoint inhibitor, avelumab. In these patients, there may be a trend toward a longer PFS, with a median PFS of 8.1 months compared with 4.6 months with the fulvestrant and palbociclib combination. This is a small phase II study and baseline characteristics might account for this trend. These data are not practice changing at this point, but they are certainly intriguing.

Sara A. Hurvitz, MD:
An exploratory subgroup analysis looked at PFS by baseline mutation. Patients whose tumors had a PIK3CA mutation or an ESR1 mutation tended to benefit more from the fulvestrant and palbociclib. 

Sara A. Hurvitz, MD:
The ORR appears to be similar in the fulvestrant and fulvestrant plus palbociclib arms, but there appears to be a trend toward improvement by adding the immune checkpoint inhibitor.

Sara A. Hurvitz, MD:
Not unexpectedly, adding an immune checkpoint inhibitor to the fulvestrant plus palbociclib combination was associated with a higher rate of AEs. 

Sara A. Hurvitz, MD:
There were higher rates of immune‑related AEs with the addition of an immune checkpoint inhibitor, including ALT and AST increase, rash, and colitis. 

Joyce O’Shaughnessy, MD:
The randomized phase III TROPiCS‑02 trial of sacituzumab govitecan (the Trop-2‑directed TOPO‑1 inhibitor payload antibody–drug conjugate) vs single‑agent chemotherapy of physician’s choice as third-line or later therapy for HR‑positive/HER2‑negative metastatic breast cancer patients demonstrated a longer PFS and OS with sacituzumab govitecan.^30,31

This presentation reported an exploratory analysis of Trop-2 expression and efficacy.³²  

Joyce O’Shaughnessy, MD:
The patients in this trial are a very heavily pretreated population with advanced disease. All had ≥2 previous lines of chemotherapy and most had 3 or 4 previous lines of chemotherapy for metastatic disease. Nearly all have visceral disease and have received a previous CDK4/6 inhibitor. Approximately 95% of patients with evaluable tissue samples had measurable Trop-2 expression.

Joyce O’Shaughnessy, MD:
PFS was improved with sacituzumab govitecan vs single‑agent chemotherapy of physician’s choice regardless of Trop-2 expression, like what was observed in the intention-to-treat population. The ASCENT trial of sacituzumab govitecan in triple-negative breast cancer also found that patients benefited from sacituzumab govitecan regardless of Trop-2 expression.³³ Because all patients show a PFS benefit with sacituzumab compared with chemotherapy, there is no need to evaluate Trop-2 expression in patients with HR-positive/HER2-negative disease. 

Joyce O’Shaughnessy, MD:
Similar to PFS, sacituzumab significantly improved OS compared with chemotherapy at all evaluated Trop-2 expression levels. Sacituzumab govitecan is now included in the NCCN guideline for HR-positive breast cancer and recently received approval from the FDA for patients with unresectable locally advanced or metastatic HR-positive/HER2-negative (IHC 0, IHC 1+, or IHC 2+/ISH-) breast cancer who have received endocrine-based therapy and at least 2 additional systemic therapies in the metastatic setting.^34,35

Joyce O’Shaughnessy, MD:
There were no new safety signals in this analysis. The major AEs seen with sacituzumab are myelosuppression, fatigue, and diarrhea, similar to those seen in triple‑negative breast cancer patients.