Background: ESR1m lead to constitutive (estrogen-independent) activation of estrogen receptor alpha and are rare (<5%) at initial diagnosis of mBC. At disease progression following first-line (1L) treatment with an aromatase inhibitor (AI) plus CDK4/6i, ESR1m are detectable in the tumors of ∼40% of patients and are recognized as the most common mechanism of acquired resistance to AI treatment. Current guidelines recommend ESR1m testing at the time of 1L treatment progression for informing treatment decisions. This retrospective cohort analysis aimed to characterize second-line (2L) treatment use and clinical outcomes (real-world progression-free survival [rwPFS] and overall survival [rwOS]) in mBC patients with a positive ESR1m test after 1L treatment with AI+CDK4/6i.
Methods: Adult patients with a confirmed diagnosis of HR-positive/HER2-negative mBC after February 2015 were identified from electronic health records (EHR) in the US Flatiron Health database. Inclusion criteria included 1L treatment with AI+CDK4/6i (initiated before December 31, 2024), a positive ESR1m test before initiation of 2L therapy, and 2L therapy regimen adherent to NCCN Guidelines. Patients were followed until June 30, 2025. Descriptive statistics were used to summarize demographics, baseline clinical characteristics, and treatment patterns. rwOS and rwPFS were estimated using Kaplan-Meier survival analysis. The start of 2L therapy was the index date. Multivariable Cox proportional hazards regression was used to adjust for prespecified confounders, including age at 2L start, ECOG PS, metastatic site, and 1L treatment duration. Sensitivity analyses were conducted by duration of 1L AI+CDK4/6i. Similar analyses in a separate large US EHR database are ongoing and will be reported.
Results: A total of 226 patients with a positive ESR1m test were included; median age at diagnosis was 63 years, and most were White (74.1%) with ECOG PS ≤1 (71.2%). Liver metastases were common (51.8%), and 25.2% of patients had bone-only metastasis. 2L treatments included ET+CDK4/6i (32.3%), elacestrant monotherapy (22.6%), PI3K/AKT/mTOR inhibitor+ET (19.0%), chemotherapy/ADC (11.9%), fulvestrant monotherapy (11.1%), or other (3.1%).In the overall population, median rwPFS was 4.8 (95% CI 4.0-5.6) months and median rwOS was 24.9 (95% CI 18.8-30.9) months; results varied by treatment (Table). Sensitivity analyses in patients with ≥6 months of 1L AI+CDK4/6i showed similar findings (Table).
| All patients |
rwPFS median (95% CI), months |
rwPFS hazard ratio (95% CI) |
rwOS median (95% CI), months |
rwOS hazard ratio (95% CI) |
| All patients (n=226) |
4.8 (4.0-5.6) |
— |
24.9 (18.8-30.9) |
— |
| Fulvestrant (n=25) (reference) |
3.0 (2.8-6.3) |
1.00 |
17.4 (11.8-30.9) |
1.00 |
| ET+CDK4/6i (n=73) |
5.2 (3.6-6.1) |
0.54 (0.31-0.95) |
39.1 (19.79-NE) |
0.47 (0.24-0.93) |
| Elacestrant monotherapy (n=51) |
5.3 (4.4-7.5) |
0.63 (0.34-1.16) |
17.8 (16.18-NE) |
0.92 (0.45-1.89) |
| PI3K/AKT/mTOR inhibitor+ET (n=43) |
5.7 (4.0-10.2) |
0.61 (0.33-1.14) |
26.7 (15.12-NE) |
0.63 (0.29-1.34) |
| Chemotherapy/ADC (n=27) |
4.8 (2.8-6.7) |
0.58 (0.30-1.14) |
17.6 (13.84-NE) |
0.53 (0.24-1.21) |
| Other (n=7) |
2.5 (0.92-NE) |
0.75 (0.28-1.96) |
7.1 (0.92-NE) |
2.58 (0.90-7.40) |
| Duration of 1L AI+CDK4/6i ≥6 months |
|
|
|
|
| All patients (n=194) |
4.6 (3.7-5.6) |
— |
23.5 (17.8-30.9) |
— |
| Fulvestrant (n=21) (reference) |
2.8 (1.61-NE) |
1.00 |
19.1 (11.84-NE) |
1.00 |
| ET+CDK4/6i (n=52) |
4.6 (2.7-5.9) |
0.75 (0.39-1.42) |
39.1 (19.79-NE) |
0.43 (0.20-0.90) |
| Elacestrant monotherapy (n=49) |
5.7 (4.5-7.5) |
0.64 (0.32-1.24) |
17.8 (16.18-NE) |
0.76 (0.35-1.65) |
| PI3K/AKT/mTOR inhibitor+ET (n=42) |
5.4 (4.0-10.2) |
0.64 (0.32-1.25) |
26.7 (15.12-NE) |
0.48 (0.21-1.12) |
| Chemotherapy/ADC (n=24) |
3.5 (2.8-7.6) |
0.54 (0.26-1.13) |
17.6 (13.84-NE) |
0.48 (0.19-1.20) |
| Other (n=6) |
4.3 (2.07-NE) |
0.63 (0.22-1.81) |
19.2 (3.52-NE) |
1.71 (0.52-5.55) |
Conclusion: In patients with ESR1m HR-positive mBC, the most common real-world 2L treatments were a rechallenge ET+CDK4/6i, elacestrant, and PI3K/AKT/mTOR inhibitor+ET. ESR1m defines a difficult-to-treat subgroup, with rwPFS outcomes highlighting the need for more effective strategies that address resistance and improve patient outcomes.
VIEW THE PUBLICATION
VIEW THE POSTER