Can treatment with inhibitors of cyclindependent kinases 4 and 6 be optimized for patients with breast cancer? This remains an open question 6 years after the first approval of a CDK4/6 inhibitor, palbociclib.
Can treatment with inhibitors of cyclindependent kinases 4 and 6 (CDK4/6is) be optimized for patients with breast cancer? This remains an open question 6 years after the first approval of a CDK4/6i, palbociclib (Ibrance).1 Three CDK4/6is, abemaciclib (Verzenio),2 palbociclib,3,4 and ribociclib (Kisqali),5 are currently approved for advanced or metastatic hormone receptor–positive, HER2-negative breast cancer, either combined with endocrine therapy (ET) in ET-naïve patients or with fulvestrant (Faslodex) in ET-pretreated patients with progressive disease.2,3,4,5
Additional indications include abemaciclib as monotherapy in ET-pretreated patients with progressive disease2 and ribociclib that is combined with fulvestrant in ET-naïve patients.5
“What’s been remarkable about this class of agents is that among the 8 large, randomized trials that have explored the role of these agents in the metastatic setting, there is remarkable congruence of the benefit of this class of medication when added to any endocrine partner, with very consistent hazard ratios across all of the trials,” said Erica L. Mayer, MD, MPH, during an interview with Targeted Therapies in Oncology. Mayer is a senior physician at the Dana-Farber Cancer Institute and assistant professor of medicine at Harvard Medical School, both in Boston, Massachusetts.
New and updated results of trials of these CDK4/6 is were presented at the 2020 San Antonio Breast Cancer Symposium.
Priya Rastogi, MD, an associate professor of medicine at the University of Pittsburgh School of Medicine in Pennsylvania, presented results from the primary outcome analysis of the phase 3 monarchE study (NCT03155997) of abemaciclib plus ET versus ET alone.6
The monarchE trial enrolled patients with hormone receptor–positive/HER2-negative, node-positive, high-risk early breast cancer (EBC; N = 5637) in 1 of 2 cohorts. Cohort 1 included patients with high clinicopathologic risk factors; cohort 2 included patients with a high Ki-67 index (≥20% of tumor cells positive for Ki-67 protein expression; n = 517). High expression of Ki-67, a marker of proliferation, is associated with increased recurrence. Patients were randomly assigned to either abemaciclib 150 mg twice a day for up to 2 years plus standard-of-care ET for 5 to 10 years as clinically indicated (n = 2808) or to ET alone (n = 2829). The primary objective was invasive disease–free survival (IDFS). Key secondary objectives included distant relapse–free survival (DRFS).
At a median follow-up of 19.1 months, there was a statistically significant and clinically meaningful improvement in IDFS for abemaciclib plus ET (HR, 0.713; 95% CI, 0.583-0.871; P = .0009). The 2-year IDFS rate was 92.3% for abemaciclib plus ET versus 89.3% with ET alone. The Kaplan-Meier curves of the 2 arms continued to separate after the 2-year mark.
Abemaciclib produced a statistically significant and clinically meaningful improvement in IDFS for patients with high–Ki-67 tumors (HR, 0.691; 95% CI, 0.519-0.920; P = .0111). The 2-year IDFS rate in these patients was 91.6% with abemaciclib plus ET versus 87.1% with ET alone.
In the entire intent-to-treat (ITT) population, there were fewer DRFS events with abemaciclib plus ET compared with ET alone (HR, 0.687; 95% CI, 0.551- 0.858; P = .0009).6
According to a poster detailing the implications of monarchE results for use of Ki-67 as a biomarker, 7 high Ki-67 expression increased the risk of recurrence even further.
These results suggest that a Ki-67 index of 20% or greater can be used, along with clinicopathologic features such as nodal involvement and tumor size and grade, to identify patients with hormone receptor–positive/HER2-negative EBC at high risk of recurrence.7
Safety was consistent with the known profile of abemaciclib. Over half of the early discontinuations due to adverse effects (AEs) occurred within the first 5 months of treatment, with rates declining over time.6
The ongoing monarchE study will continue until the final assessment of overall survival (OS). When combined with ET, abemaciclib is the first CDK4/6i to demonstrate efficacy and tolerability for patients with hormone receptor–positive/ HER2-negative, node-positive, high-risk EBC.6
Sibylle Loibl, MD, PhD, CEO and chair of the German Breast Group and associate professor at the University of Frankfurt, Germany, presented the first results from PENELOPE-B (NCT01864746).8 The phase 3 study enrolled patients (N = 1250) with high-risk breast cancer defined by postneoadjuvant therapy residual disease and a score on the clinical-pathologic stage, plus estrogen receptor status and grade (CPS±EG) staging system of\ 3 or greater or 2 or more with nodal involvement. The primary end point was IDFS; secondary end points included DRFS, OS, safety, and compliance.
Patients were randomly assigned to 13 cycles of ET plus either daily palbociclib 125 mg for 3 weeks of every 4-week cycle (n = 631) or placebo (n = 619). All 13 cycles were completed by 80.5% of the palbociclib arm and 84.5% of the placebo arm. A Ki-67 index greater than 15% was detected in 25.5% of patients in both groups; approximately 60% of patients had a CPS±EG score of at least 3.
At a median follow-up of 42.8 months, there was no difference in IDFS between the arms (HR, 0.93; 95% CI, 0.74-1.17; P = .525). The 2-year IDFS rate was 88.3% for palbociclib versus 84.0% for the placebo arm; the 3-year IDFS rate was 81.2% versus 77.7%; and the 4-year IDFS rate was 73.0% versus 72.4%.
At an interim OS analysis, there was no difference between the arms (HR, 0.87; 95% CI, 0.61- 1.22; P = .420). The 2-year, 3-year, and 4-year OS rates were similar for palbociclib and placebo.
The results of PENELOPE-B to date do not support the addition of 1 year of palbociclib to ET in this patient population.
Ruth M. O’Regan, MBBCh, BAO, chair of Medicine and the Charles A. Dewey Professor at the University of Rochester, and physician-in-chief of Strong Memorial Hospital, New York, discussed the monarchE and PENELOPE-B presentations. She observed that PENELOPE-B was “quite a different trial compared with the other adjuvant CDK4/6i trials.” It enrolled fewer patients and included a very high-risk population defined by a clinicopathologic score (CPS±EG) rather than the anatomic stage criteria used in monarchE. It is possible that the PENELOPE-B results would have been positive if palbociclib had been given for longer. Noting the similarity between PENELOPE-B’s IDFS curve and typical results in the first-line metastatic setting, O’Regan hypothesized that PENELOPE-B participants may have had occult metastatic disease.
Regarding “who would need adjuvant CDKi therapy, O’Regan explained, “Given the added toxicity and expense, it is crucial to determine which patients are most likely to benefit. We all agree patients with high-risk disease [would benefit], but how should we define high risk?” Commenting on the monarchE results that showed greater benefit of abemaciclib in patients with high–Ki-67 tumors compared with the ITT population, she speculated that patients with biologically high-risk tumors, such as luminal B cancers, benefit more from abemaciclib. In contrast, patients in PENELOPE-B, who were less likely to have a pathologic complete response, may have been more likely to have luminal A tumors. That suggests that CDK4/6i therapy may not be as effective for luminal A cancers. “Obviously, this is not supported from biomarker data in the metastatic setting and would need to be confirmed,” O’Regan said, noting the critical need for robust biomarkers apart from stage.
The question of the best time to administer adjuvant CDK4/6i therapy is unanswered. Pointing to the high frequency of late recurrence in hormone receptor–positive tumors, O’Regan suggested that a CDK4/6i might be more beneficial if given later in treatment, such as after the first year.4
It is not clear how to reconcile the different outcomes of studies of CDK4/6is in the early setting. Mayer said this has been the topic of significant discussion within the breast cancer community. Possible reasons for the discrepant findings include differences in patient populations, specific CDK4/6is, discontinuation rates, and length of follow-up.
Debu Tripathy, MD, professor and chair, Department of Breast Medical Oncology, Division of Cancer Medicine, The University of Texas MD Anderson Cancer Center, Houston, presented a poster of exploratory updated OS results from the phase 3 MONALEESA-7 trial (NCT02278120).
Pre- or perimenopausal patients with hormone receptor–positive/HER2-negative advanced breast cancer were randomly assigned to ribociclib (n = 335) or placebo (n = 337) along with goserelin (Zoladex) plus a nonsteroidal aromatase inhibitor or tamoxifen. These patient numbers make this trial even smaller than PENELOPE-B. Median follow-up was 53.5 months, one of the longest follow-ups reported.
Median OS was 58.7 months with ribociclib and 48.0 months with placebo (HR, 0.763; 95% CI, 0.608-0.956). OS values for subgroups were similar to the OS values of the overall population. Most patients in both arms discontinued treatment, primarily for progressive disease, and most received subsequent cancer treatment. Median time to subsequent chemotherapy was longer for the ribociclib arm (50.9 months) compared with placebo (36.8 months; HR, 0.69; 95% CI, 0.56-0.87), as were chemotherapy- free survival and progression-free survival on subsequent therapy.
This exploratory analysis confirms a benefit for continued use of ribociclib in the first-line setting for the study population, even after crossover and subsequent CDK4/6i use in the placebo arm.9
A poster presented evidence that mutation of the estrogen receptor 1 (ESR1) gene is a potential biomarker for response to abemaciclib in patients whose hormone receptor–positive metastatic breast cancer has progressed on palbociclib. In a retrospective analysis of a multicenter patient cohort, sequencing of cell-free DNA revealed the ESR1 mutation in 2 patients. Further exploration of this finding in a larger cohort of patients is under way.10
ESR1 mutations lead to ET resistance and poorer prognosis.11 Thus, monotherapy with lasofoxifene (Fablyn), a third-generation selective estrogen receptor modulator, is being compared against fulvestrant in pre- and postmenopausal women with metastatic breast cancer harboring ESR1 mutations in a phase 2 trial, ELAINE 1 (NCT03781063).
The open-label, single-arm ELAINEII (NCT04432454) trial is assessing the safety of lasofoxifene in combination with abemaciclib. The study is enrolling 24 pre- and postmenopausal women who have metastatic ESR1- mutated breast cancer that has progressed on up to 2 previous hormonal therapies (FIGURE 11).