Study selection
A total of 251 potentially relevant manuscripts and 4 additional abstracts were sorted out by using the search string mentioned before. Of these, after reviewing the titles and abstracts, 241 manuscripts were excluded. We then performed a full-text review for the remaining 14 articles, 10 of which were excluded for non-conformity with the present inclusion criteria. Eventually, 4 articles from 4 trials were considered eligible for the meta-analysis. The Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) flowchart is shown in Fig. 1.
Characteristics of studies
Finally, our study involved 4 clinical trials published between February 2022 and December 2022, focusing on different endocrine treatment regimens of HR+/HER2- advanced breast cancer, and including a total of 1,290 patients (Table 1). Oral SERDs arms include elacestrant (EMERALD), camizestrant75mg/camizestrant150mg (SERENA-2), amcenestrant (AMEERA-3), giredestrant (acelELA). Control arms included fulvestrant, anastrozole, letrozole, exemestane, tamoxifen. All trials compared oral SERDs to standard-of-care ET in patients with HR+/HER2- aBC after progression on ≥ 1 line of ET.
Table 1
Characteristics of eligible studies in the meta-analysis
Study | EMERALD | SERENA-2 | AMEERA-3 | acelELA |
First author | Francois-Clement Bidard | Mafalda Oliveira | Sara M. Tolaney | Miguel Martin |
Year of publication | 2022 | 2022 | 2022 | 2022 |
Phase | Ⅲ | Ⅱ | Ⅱ | Ⅱ |
Patients, n | 477 | 220 | 290 | 303 |
Patients | Men or postmenopausal woman | Postmenopausal woman | Men or woman (any menopausal status) | Men or woman (any menopausal status) |
Oral SERDs regimen / dose | Elacestrant / 400mg | Camizestrant / 75mg (A)*/ 150mg (B)* | Amcenestrant / 400mg | Giredestrant / 30mg |
Standard-of-care ET | SOC | Fulvestrant | TPC | PCET |
ESR1m, n | 228 | 68 | 120 | 90 |
Prior CDK4/6i, % | Required, 100 | Permitted, 49.6% | Permitted, 79.7 | Permitted, 42 |
Allowed Prior Fulvestrant | Yes | NO | Yes | Yes |
HR | 0.70 | 0.58 (75mg)/ 0.67 (150mg) | 1.051 | 0.81 |
95%CI | 0.55–0.88 | 0.41–0.81 (75mg) / 0.48–0.92 (150mg) | 0.789-1.40 | 0.60–1.01 |
Abbreviations: SOC: standard-of-care; TPC: treatment of physician’ choice; PCET: physician choice of endocrine monotherapy |
* SERENA-2 was divided into two cohorts because the comparisons were between two doses of camizestrant 75mg and 150mg. |
Progression-free survival
In the whole population, patients with HR+/HER2- advanced breast cancer treated with oral SERDs significantly improved PFS compared to those treated with standard-of-care ET (HR: 0.75, 95% CI: 0.62–0.91, p = 0.004; I2: 52%, p = 0.08; Fig. 2A). For enrolled patients with disease progression following previous use of CDK4/6 inhibitors, oral SERDs regimen was significantly better than standard-of-care ET (HR: 0.75, 95% CI: 0.64–0.87, p = 0.0002; I2: 48%, p = 0.10; Fig. 2B). In HR+/HER2- ESR1m aBC, the two treatment regimens compared, namely oral SERDs resulted in a better PFS versus standard-of-care ET (HR: 0.58, 95% CI: 0.47–0.71, p < 0.00001; I2: 42%, p = 0.14; Fig. 2C). For enrolled patients with ESR1 mutations, arms of oral SERDs were significantly better than arms of fulvestrant (HR: 0.47, 95% CI: 0.36–0.62, p < 0.00001; I2 : 0%, p = 0.41; Fig. 2D). For patients who had previously failed treatment with fulvestrant, oral SERDs as monotherapy was significantly superior to standard-of-care ET (HR : 0.67, 95% CI: 0.47–0.95, p = 0.02; I2 : 0%, p = 0.93; Fig. 3A). In patients with visceral disease, arms of oral SERDs was significantly better than arms of standard-of-care ET (HR: 0.60, 95% CI: 0.48–0.74, p < 0.00001; I2: 33%, p = 0.22; Fig. 3B). Arms of oral SERDs were significantly better than arms of fulvestrant (HR: 0.65, 95% CI: 0.54–0.78, p < 0.00001; I2: 0%, p = 0.76; Fig. 3C).
Safety
Adverse events (AE) of Grade 3 or higher were more frequently in oral SERDs regimen than in standard-of-care ET (HR: 1.40, 95% CI: 1.03–1.90, p = 0.03; I2: 0%, p = 0.99; Fig. 4). The proportion of treatment-emergent adverse events (TEAEs) leading to discontinuation was 6.3% (Elacestrant ) vs. 4.4% (SOC) in EMERALD's two treatment arms; The most common adverse event was nausea. The proportion of drug discontinuation caused by TEAEs in the three treatment groups of SERENA-2 was 14.9% (camizestrant 75mg), 21.9% (camizestrant 150mg), 4.1% (standard-of-care ET), respectively; Common adverse events were photopsia and sinus bradycardia. In AMEERA-3, the proportion of TRAE ≥ Grade 3 was 4.9% in the experimental arm and 0.7% in the control arm; The most common adverse event was nausea. In acelELA, the incidence of AE ≥ Grade 3 was 12% (Giredestrant) vs. 8.6% (PCET); The most common adverse event was hepatotoxicity.
Bias assessment
In all included in the trials, the overall risk of bias was low (Supplement 1 Fig. 1). Since these trials were conducted with an open-label design, performance bias that did not affect the results may exist. There was no obvious publication bias (Supplement 1 Fig. 2–3).