Highlights
- The Phase II CARABELA trial compared 12 months of neoadjuvant abemaciclib plus letrozole to 6 months of standard chemotherapy in high-risk HR+/HER2- breast cancer.
- The primary endpoint of Residual Cancer Burden (RCB) 0-I was achieved in 13% of the abemaciclib/letrozole group compared to 18% in the chemotherapy group.
- While overall similarity was not established, clinical response rates were comparable (78% vs. 71%), and patients with lower proliferation markers (Ki-67 <30% or RS <26) showed similar outcomes between the two arms.
- These findings suggest a potential path for chemotherapy de-escalation in specific genomic and proliferative subgroups of luminal breast cancer.
Introduction: The Evolution of Neoadjuvant Therapy in HR+/HER2- Breast Cancer
Neoadjuvant chemotherapy (NACT) has long been the standard of care for patients with high-risk, hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-) breast cancer (BC). Despite its widespread use, the pathological complete response (pCR) rates for this subtype remain significantly lower than those observed in triple-negative or HER2-positive diseases. This clinical reality has prompted a search for more targeted, less toxic alternatives that leverage the biological drivers of luminal breast cancer.
Cyclin-dependent kinase 4 and 6 (CDK4/6) inhibitors, in combination with endocrine therapy, have revolutionized the management of metastatic HR+/HER2- BC and have recently moved into the adjuvant setting. The CARABELA trial (NCT03254316) sought to evaluate whether the combination of abemaciclib—a potent CDK4/6 inhibitor—and letrozole could serve as a non-inferior neoadjuvant alternative to traditional chemotherapy in patients with highly proliferative tumors.
Study Design and Methodology
The CARABELA trial was a phase II, open-label, randomized study designed to compare the efficacy of neoadjuvant endocrine-targeted therapy against standard chemotherapy. The study enrolled 200 patients with HR+/HER2- stage II-III breast cancer characterized by a Ki-67 index of 20% or higher, signifying a highly proliferative phenotype.
Patient Population and Stratification
The median age of participants was 53 years, with 57% being postmenopausal. The majority of patients had stage II disease (79%) and high proliferation markers, with 77% presenting a Ki-67 index ≥30% and 58% having a Recurrence Score (RS) ≥26 via the Oncotype DX assay. Randomization was 1:1, stratifying patients by menopausal status, TNM stage, and Ki-67 index (<30% vs. ≥30%).
Intervention Arms
The experimental arm received letrozole (2.5 mg/day) plus abemaciclib (150 mg twice daily) for a total of 12 months. The control arm received standard chemotherapy (typically anthracycline and taxane-based regimens) for 6 months. Following the neoadjuvant phase, all patients underwent surgery.
Endpoints
The primary endpoint was the rate of Residual Cancer Burden (RCB) 0-I, a validated prognostic metric that accounts for the primary tumor area, cellularity, and nodal involvement. Secondary endpoints included the clinical response rate (CRR), safety profiles, and the correlation of biomarkers (Ki-67 and RS) with therapeutic response.
Key Findings: Analyzing the Residual Cancer Burden
The primary analysis utilized a Bayesian design to assess the similarity between the two treatment approaches. The results indicated that the endocrine-targeted combination did not meet the statistical criteria for similarity to chemotherapy in this high-risk population.
RCB 0-I Rates
In the intent-to-treat population, the RCB 0-I rate was 13% for the letrozole/abemaciclib arm (95% Credible Interval [CrI]: 7.4% – 20.5%) compared to 18% for the chemotherapy arm (95% CrI: 11.5% – 26.4%). While the chemotherapy arm numerically outperformed the targeted therapy arm in achieving minimal residual disease, the difference was not statistically significant in a traditional frequentist sense, yet the Bayesian analysis failed to support the hypothesis of equivalent efficacy.
Clinical Response and Surgery
Interestingly, the clinical response rates (CRR)—measured by physical examination or imaging—were slightly higher in the abemaciclib/letrozole group at 78%, compared to 71% in the chemotherapy group (P=0.26). This suggests that while abemaciclib/letrozole may not achieve the same depth of pathological response (RCB 0-I), it is highly effective at inducing clinical tumor shrinkage, which is a critical goal of neoadjuvant therapy for surgical planning.
The Role of Proliferative and Genomic Markers
A crucial aspect of the CARABELA trial was the subgroup analysis based on Ki-67 and Recurrence Score (RS). These markers proved to be significant predictors of which patients might benefit most from each approach.
Highly Proliferative Tumors (Ki-67 ≥30% or RS ≥26)
In tumors with very high proliferation indices, chemotherapy showed a trend toward superior outcomes. The RCB 0-I rate for this subgroup was 23% with chemotherapy versus 17% with letrozole/abemaciclib (P=0.52). While the study was not powered to show definitive differences in these subgroups, the trend reinforces the current clinical practice of using chemotherapy for the most aggressive luminal B-like tumors.
Moderately Proliferative Tumors (Ki-67 <30% or RS <26)
Conversely, in patients with lower proliferation (Ki-67 <30%) or lower genomic risk (RS <26), the RCB 0-I rates were remarkably similar between the two arms. This finding is of high clinical importance, as it suggests that for a substantial portion of the HR+/HER2- population, endocrine therapy combined with a CDK4/6 inhibitor can achieve pathological outcomes comparable to chemotherapy without the associated systemic toxicities.
Expert Commentary and Clinical Implications
The CARABELA trial adds to a growing body of evidence, including the NeoMONARCH and monarchE trials, regarding the utility of abemaciclib in early-stage breast cancer. The primary challenge in the neoadjuvant setting for HR+ disease is that pathological response does not always correlate perfectly with long-term survival, unlike in HER2+ or triple-negative subtypes.
The failure to show similarity in the overall cohort highlights that for the truly “high-risk” patient—those with very high Ki-67 or genomic scores—chemotherapy remains the backbone of neoadjuvant treatment. However, the 12-month duration of the abemaciclib/letrozole regimen in this trial is longer than standard neoadjuvant endocrine therapy (usually 4-6 months), which may have allowed for more sustained cell-cycle arrest but did not necessarily translate into higher rates of pathological clearance.
Clinicians should consider these results as a validation of personalized therapy. For a patient who is a poor candidate for chemotherapy due to comorbidities or preference, and whose tumor has a moderate Ki-67 or RS, neoadjuvant abemaciclib plus letrozole offers a highly effective clinical alternative.
Conclusion
The CARABELA trial demonstrates that while 12 months of neoadjuvant letrozole plus abemaciclib may not replace chemotherapy for the entire spectrum of highly proliferative HR+/HER2- breast cancer, it is a potent and effective regimen. The study underscores the necessity of genomic and proliferative profiling to guide neoadjuvant decisions. Future research should focus on identifying the specific molecular signatures that predict a deep pathological response to CDK4/6 inhibition, potentially allowing for a more nuanced “chemotherapy-free” pathway for selected patients.
Funding and ClinicalTrials.gov
The CARABELA trial was supported by Eli Lilly and Company and the Spanish Breast Cancer Research Group (GEICAM). ClinicalTrials.gov Identifier: NCT03254316.
References
Martín M, Guerrero-Zotano AL, Pérez-López ME, et al. Neoadjuvant Abemaciclib plus Letrozole vs. Chemotherapy in patients with HR+/HER2- Highly Proliferative Breast Cancer. Clin Cancer Res. 2025 Dec 12. doi: 10.1158/1078-0432.CCR-25-2435. PMID: 41385615.