In Support of CDK4/6 Inhibitors—A Meta-analysis of Available Randomized Data

The recent development of inhibitors of cyclin dependent kinases 4 and 6 (CDK4/6) is arguably one of the most clinically important discoveries for patients living with hormone receptor (HR)–positive, ERBB2 (formerly HER2)–negative metastatic breast cancer. Currently, there are 3 available CDK4/6 inhibitors: palbociclib, ribociclib, and abemaciclib. These CDK4/6 inhibitors induce cell cycle arrest by preventing the G1 to S phase transition, and they are often administered in combination with endocrine therapy (ET). Li et al¹ performed a meta-analysis of the 9 published randomized clinical trials that compared ET plus either a CDK4/6 inhibitor or placebo. Eight phase 3 clinical trials were included as well as 1 randomized phase 2 study. Phase 1 trials, retrospective studies, and those without overall survival (OS) outcomes were not included in their analysis. The authors sought to analyze the potential benefit of CDK4/6 inhibitors with regard to OS, progression-free survival (PFS), overall response rate, and development of adverse events.

Across all 9 trials, a total of 5043 patients were included in the analysis. In regard to OS, 6 trials were deemed to have adequate OS reporting for inclusion: MONALEESA-2, MONALEESA-3, MONALEESA-7, PALOMA-1, PALOMA-3, and MONARCH 2, which included a total of 2030 patients enrolled in the CDK4/6 inhibitors plus ET arm and 1391 patients in the ET arm. The addition of a CDK4/6 inhibitor was associated with significant benefit on OS (hazard ratio [HR], 1.33; 95% CI, 1.19-1.48; P < .001) across the entire data set. The aforementioned 6 studies, in addition to PALOMA-2, MONARCH 3, and MONARCHplus, were included in the analysis for the additional points of interest. Again, benefit was also associated with the use of the CDK4/6 inhibitors for PFS (HR, 1.84) and overall response rate (odds ratio, 2.02). The associated OS benefit was observed across all prespecified subgroups: line of therapy received, menopausal status, visceral vs bone-only metastases, and age. Grade 3 or 4 adverse events were noted to be higher in the CDK4/6 inhibitor arms.

Each CDK4/6 inhibitor has been prospectively shown to improve PFS in the frontline setting for patients with HR-positive, ERBB2-negative metastatic breast cancer in 4 randomized phase 3 trials. The PALOMA-2, MONALEESA-2 and MONARCH 3 trials randomized previously untreated participants to receive an aromatase inhibitor in combination with either placebo or a CDK4/6 inhibitor (palbociclib, ribociclib, or abemaciclib, respectively). The OS data for these trials are immature at this time. In addition, premenopausal patients were treated in MONALEESA-7 with ET (either tamoxifen or goserelin in combination with an aromatase inhibitor) and either ribociclib or placebo. Again, a significant improvement in PFS was seen, and this was the first CDK4/6 inhibitor trial to demonstrate an OS benefit in the frontline treatment setting.

For patients who had progressed on frontline ET, the PALOMA-3, MONALEESA-3 and MONARCH 2 trials randomized participants to fulvestrant with either placebo or a CDK4/6 inhibitor. Again, a significant PFS benefit was detected for participants who received the CDK4/6 inhibitor, and higher overall response rates were noted. Updated analyses of the data from all 3 trials have confirmed an OS benefit from the addition of a CDK4/6 inhibitor to ET. These agents also appear to increase the time to chemotherapy, a potentially meaningful end point for patients.

The US Food and Drug Administration performed a pooled analysis of all 7 phase 3 trials using CDK4/6 inhibitors.² They examined 4200 participants who received a CDK4/6 inhibitor in combination with an aromatase inhibitor in the frontline setting or with fulvestrant in any setting. Those who received tamoxifen as the endocrine partner or chemotherapy in the metastatic setting were excluded from the analysis. Among participants who received the CDK4/6 inhibitor with an aromatase inhibitor in the frontline setting, the difference in median PFS was 13.1 months (HR, 0.55). For participants who received a CDK4/6 inhibitor with fulvestrant in the second-line setting or beyond, the difference in median PFS was 6.9 months (HR, 0.56). A PFS benefit was associated with participants with progesterone receptor–negative disease, those with a short disease-free interval from adjuvant therapy, de novo metastatic disease, lobular histology, and bone-only metastases, among others.

Additional meta-analyses have previously been published and confirmed similar findings. Messina et al³ performed their analysis on available randomized phase 2 and 3 trials examining ET with or without a CDK4/6 inhibitor. A total of 8 clinical trials were analyzed including 4578 patients. The primary objectives were to compare the association of a CDK4/6 inhibitor with PFS and adverse events in both endocrine-sensitive and endocrine-resistant populations. The addition of a CDK4/6 inhibitor was associated with improved PFS in both populations (HR, 0.55 and 0.51, respectively). Similar responses were observed among participants with visceral vs nonvisceral metastases. Adverse events were more common in the CDK4/6 inhibitor arms.

Li et al⁴ have recently published a meta-analysis using the same 8 randomized clinical trials that are analyzed in Messina et al.³ Again, improved PFS was associated with the addition of the CDK4/6 inhibitor compared with ET alone (HR, 0.55). The benefit was observed among participants who received the CDK4/6 inhibitor in the first-line or second-line settings. The PFS benefit was observed across all prespecified subgroups, including presence (or not) of visceral metastases, premenopausal or perimenopausal status, race/ethnicity, or participants who received chemotherapy in the advanced or metastatic setting. In contrast to the present companion article by Li et al,¹ only 3 clinical trials were included in the OS evaluation (PALOMA-1, PALOMA-3, and MONALEESA-7), and OS benefit appeared to be associated with the addition of a CDK4/6 inhibitor (HR, 0.79).

The search for potential biomarkers to predict responses to CDK4/6 inhibitors is under way. In PALOMA-1, CCND1 amplification and loss of p16 were not found to be predictive of palbociclib benefit. In PALOMA-3, the presence of PIK3CA sequence variants did not predict a lack of benefit from the CDK4/6 inhibitor. In addition, the potential impact of ESR1 sequence variants were retrospectively analyzed in baseline plasma samples from PALOMA-3; their presence did not appear to negate a benefit of the CDK4/6 inhibitor.⁵

Loss of RB1, which encodes for the retinoblastoma protein, has been shown preclinically to confer resistance to CDK4/6 inhibitors and appears to shorten PFS compared with patients who have wildtype RB1.⁶ However, alterations in RB1 have not been shown in a prospective manner to suggest lack of benefit of the CDK4/6 inhibitor compared with ET alone. Finally, in a post hoc analysis of PALOMA-3, archived tumor samples were analyzed among 302 participants, and those with high CCNE1 expression appeared to have a smaller PFS benefit than those with low CCNE1 expression for the addition of palbociclib.⁷ Currently, there are no biomarker data to suggest avoidance of a CDK4/6 inhibitor for a particular group of patients. These data are hypothesis-generating and are not currently recommended to guide clinical care. The present meta-analysis by Li et al¹ did not use biomarkers as a predefined subgroup for assessment.

In summary, Li et al¹ present a meta-analysis of randomized trials that show improvement in PFS, OS, and overall response rates in association with the addition of a CDK4/6 inhibitor to standard ET compared with ET alone. However, adverse events were higher for participants who received a CDK4/6 inhibitor. This meta-analysis is considered along with several other meta-analyses of these same randomized trials, which have largely arrived at similar conclusions. These data support the current clinical practice of discussing the potential benefits of CDK4/6 inhibitors with nearly all patients living with HR-positive, ERBB2-negative metastatic cancer, as recommended by the National Comprehensive Cancer Network guidelines.⁸

Published: October 13, 2020. doi:10.1001/jamanetworkopen.2020.21062

Correction: This article was corrected on November 12, 2020, to fix an error in the author affiliations.

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 Martin JM et al. JAMA Network Open.

Corresponding Author: James M. Martin, MD, Medical Oncology, University Hospitals/Seidman Cancer Center, 11000 Euclid Ave, Cleveland, OH 44106 (james.martin@uhhospitals.org).

Conflict of Interest Disclosures: The project described was supported by Award P30CA006927 from the National Cancer Institute.

Disclaimer: The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Cancer Institute or the National Institutes of Health.