FormalPara Key Points

CDK4/6 inhibitors are effective and generally well tolerated in hormone-receptor positive breast cancer.

The addition of ribociclib and abemaciclib to adjuvant endocrine therapy significantly improves outcomes in H+/HER2− early breast cancer.

Details of risk profile, co-morbidities, and tolerability issues are crucial for optimal treatment decisions of individual patients.

1 Introduction

Breast cancer (BC) is the most common cancer in women [1] and the second cause for female cancer-related death [2]. Several subtypes of the disease have been identified, with luminal (hormone receptor-positive, HR+) being the most prevalent one [3]. In both the advanced (aBC) and the early (eBC) stage of HR+ disease, endocrine treatment (ET) has been the cornerstone for outcome improvements of anti-cancer therapy for decades [4]. Several years ago, a new class of drugs, inhibitors of cyclin-dependent kinase 4 and 6 (CDK4/6) demonstrated further improved outcomes when added to ET [5,6,7]. In this review, we discuss the scientific background as well as available clinical data of CDK4/6 inhibitors (CDK4/6i) in eBC.

The importance of cell cycle aberration is obvious for the development of cancer, and carcinogenesis requires several changes in the cell genome. Hanahan and Weinberg established six “hallmarks of cancer,” namely resisting cell death, sustaining proliferative signaling, evading growth suppressors, activating invasion, and metastasis, enabling replicative immortality and inducing angiogenesis [8, 9]. Hartwell et al. demonstrated the importance of checkpoints in the cell cycle and their dysregulation in cancer cells [10].

1.1 Mechanism of Action of CDK4/6 inhibitors (CDK4/6i)

In cells with cyclin-dependent kinase 4/6 (CDK4/6)-dependent proliferation, the cyclin-dependent kinases (CDKs) 4 and 6 have a central role in the cell cycle by regulating the G1 restriction cell-cycle checkpoint [11,12,13]. The G1 checkpoint prevents cells from entering the S phase in case of DNA damage and avoids defect chromosome duplication (Fig. 1). During the G0 and early G1, low levels of D-type cyclins (cyclins D1–D3) are typical and increasingly accumulate to the cyclin D-CDK4/6 complex [13]. CDK4 and CDK6 lead to phosphorylation of the serin/threonine residues of the retinoblastoma protein (pRB) and mediate transition through the G1 checkpoint. Unphosphorylated RB1 binds a complex comprising the transcription factor E2F and suppresses transcription. Through phosphorylation by CDK4/6, RB1 has reduced affinity to E2F and thus does not inhibit cell proliferation anymore [14, 15]. Therefore, targeting CDKs with small molecules was identified as a relevant therapeutic approach [16]. CDK4/6i are small molecules that bind to the ATP cleft of CDK4 and CDK6 and suppress their function [17].

Fig. 1
figure 1

Mechanism of action of CDK4/6 inhibitors [184]. With permission from [178]. Al aromatase inhibitor (anastrozole letrozole exemestane), SERM selective estrogen receptor modulator (tamoxifen (TAM)), SERD selective estrogen receptor down-regulator (fulvestrant), ER estrogen receptor, E estradiol, T testosterone, TKIs tyrosine kinase inhibitors (pyrotinib lapatinib), mAbs monoclonal antibodies (trastuzumab pertuzumab), PI3Ki phosphoinositide 3-kinases inhibitor (alpelisib), AKTi protein-kinase B inhibitor (capivasertib), mTORi mammalian target of rapamycin inhibitor (everolimus), E2F a transcription factor, Rb retinoblastoma protein, HER2 human epidermal growth-factor receptor 2, HDACi HDAC inhibitor (tucidinostat), PARPi PARP inhibitor (olaparib talazoparib), HRD homologous recombination deficiency, CDK4/6 inhibitors cyclin-dependent kinase 4/6 inhibitors (palbociclib ribociclib abemaciclib dalpiciclib)

During the cell cycle, two families of endogenous inhibitory proteins, namely the CDK inhibitor 1-kinase inhibitor family—CIP/KIP family—and the INK4 family, regulate the activity of CDK4/6 and CDK2. The INK4 family comprises p15, p16, p18, and 19, which bind with high affinity to CDK4 and CDK6 and suppress their activity. The members of the CIP/KIP family are p21, p27, and p57, which bind to CDKs, and have different functions [18,19,20,21].

CDK4/6 inhibition requires an intact cyclin D-CDK4/6-RB1 pathway, often impaired in triple negative breast cancers, likely to harbor RB1 alterations [22]. Especially in estrogen receptor positive (ER+) BC, high expression levels of cyclin D1 were shown, which might reflect the dependency on CDK4/6 [23, 24]. Cyclin D1 gene transcription is induced by estrogen within estrogen receptor α (ERα) and leads to cell proliferation in luminal BC cells [25]. Furthermore, cyclin D1 activates ER in the absence of estrogen [26]. These data indicate the relevance of the ER/cyclin D interaction and lead to the development of specific CDK-inhibitors as a therapeutic approach in hormone-receptor positive BC.

1.2 Immunological Effects of CDK4/6i

In addition to direct antiproliferative properties, CDK4/6 inhibition leads to various immunological effects within the cancer cell (Fig. 1). Preclinical and clinical data have shown that there might be synergistic effects between CDK4/6 inhibition and immune checkpoint-inhibitor treatment [27, 28]. CDK4/6i are capable of inducing an increased T-cell inflammatory signature with consecutive upregulation of antigen presentation and T-cell infiltration. Endogenous retroviral elements lead to increased levels of intracellular double-stranded RNA and furthermore activates tumor antigen presentation via type III interferon and upregulation of MHC. Additionally, CDK4/6i reduce the proliferation of regulatory T cells. This leads to reduced activity of the DNA methyltransferase 1 and promotes cytotoxic T-cell-induced clearance of tumor cells [29,30,31]. CDK4/6 inhibition enhances T memory cells and moreover leads to a downregulation of their myc expression [32]. As a result, CDK4/6 inhibition can also induce long-term immunity. Indeed, combination strategies with CDK4/6 inhibition and immune-checkpoint inhibition have already been tested in early trials [27, 33,34,35,36]. However, the combination of CDK4/6i and immune-checkpoint inhibition led to higher toxicities such as pneumonitis and elevation of transaminases in these early trials [27, 35], which is why phase III trials of these combinations have not been started yet.

1.3 Importance of CDK4/6i in Breast Cancer (BC)

The development of CDK4/6i has revolutionized the treatment of HR+ HER2 negative (HER2−) BC. Significantly improved rates of progression-free survival (PFS) as well as overall survival (OS) were demonstrated [37,38,39,40,41,42,43,44,45]. Therefore, this class of drugs was rapidly established as a new standard first-line treatment in combination with ET in advanced BC. Furthermore, the CDK4/6i abemaciclib is now approved for the adjuvant treatment of patients at high risk of relapse [46]. Another CDK4/6i, ribociclib, has demonstrated activity in the adjuvant setting and has been recently approved by the US Food and Drug Administration (FDA) and European Medicines Agency (EMA) [47,48,49].

1.4 Clinically Evaluated CDK4/6i

To date, five CDK4/6i are in use in clinical practice: palbociclib, ribociclib, abemaciclib, dalpiciclib, and tibremciclib. The latter two are solely used in the People’s Republic of China [50, 51]. All these agents differ in their affinity to the CDKs [17].

1.4.1 Palbociclib

Palbociclib binds to both CDK4 and CDK6 with the same affinity [52]. The dosing schedule consists of 125 mg once a day for 3 weeks followed by 1 week of rest [44].

1.4.2 Ribociclib

Ribociclib has the highest binding affinity to CDK4 in relation to CDK6 [17, 53, 54]. The clinically used dosing in the advanced setting is 600 mg daily for 3 weeks, followed by 1 week of rest [53]. The NATALEE trial in eBC tested a dose of 400 mg per day, also scheduled 3 weeks on and 1 week off [47].

1.4.3 Abemaciclib

Abemaciclib binds with lower affinity to CDK4 and CDK6 compared to palbociclib and ribociclib, but also has affinity to CDK 9 and some (low) ability to inhibit CDK2 [17, 55, 56]. Abemaclib is administered continuously at a dose of 150 mg twice a day in both the early and the metastatic setting [40]. Data suggest that abemaciclib is able to pass the brain-blood barrier and might have activity in brain metastasis [57]. Limited clinical evidence suggests that abemaciclib leads to an intracranial clinical benefit rate of 24% with therapeutic concentrations of abemaciclib in tissue of resected brain metastases [58].

1.4.4 Dalpiciclib

Dalpicilib is a novel CDK4/6i approved in China. It has shown the ability to directly phosphorylate RB1 and was only efficient in Rb1-positive cells [59, 60]. The dosing schedule is 150 mg per day for 3 weeks, followed by 1 week off [61].

1.4.5 Tibremciclib

Tibremciclib is another novel CDK4/6i that has shown efficacy in second-line aBC [51, 62].

1.4.6 Other CDK4/6i

Several other CDK4/6i are under clinical evaluation, such as dinaciclib, which has a broad binding spectrum including CDK 2, 3, 4, 6, and 9 [55].

2 CDK4/6i in Advanced BC (aBC)

2.1 First-Line aBC

In the phase III PALOMA-2 trial, palbociclib was evaluated as a component of first-line treatment in the metastatic setting [63]. Patients who had received previous aromatase-inhibitor (AI) treatment in adjuvant or neoadjuvant treatment at least 12 months ago were allowed to be included. This study included 666 postmenopausal patients with no prior treatment in the aBC setting, randomized 2:1 to palbociclib or placebo in combination with the AI letrozole [63]. At a median follow-up of 38 months, PFS was significantly longer in the study cohort than in the control cohort (27.6 vs. 14.5 months (hazard ratio (HR) 0.56; 95% confidence interval (CI): 0.46–0.69; P < 0.0001)) [64]. No difference in OS with a median OS of 53.9 versus 51.2 months (HR 0.96; 95% CI 0.78—1.18; P = 0.34) was observed [65].

In the MONALEESA-2 trial, 668 postmenopausal women with advanced BC were randomized 1:1 to receive ribociclib or placebo in combination with letrozole. Patients who experienced disease recurrence had to have stopped adjuvant ET at least 12 months ago [66]. After a median follow-up of 26.4 months, the median PFS was 25.3 months in the ribociclib group versus 16.0 months in the placebo group (HR 0.568; 95% CI 0.457–0.704; P=9.63x10-8) [42]. At a median duration of follow up of 6.6 years the median OS was significantly improved with 63.9 months versus 51.4 months (HR 0.67; 95% CI 0.63–0.93; P = 0.008)) [67].

Up to date, most data for CDK4/6i in premenopausal women are available with ribociclib. In the MONALEESA-7 trial 726 premenopausal women were randomized 2:1 to receive ribociclib versus placebo as first- or second-line treatment [68]. Chemotherapy was allowed as previous treatment in the advanced setting and had been administered in 14%; 40% of patients had de novo metastatic disease [68]. Median PFS was significantly improved with ribociclib compared to placebo, with 23.8 months compared with 13.0 months (HR 0.55; 95% CI 0.44-0.69; P < 0.0001) [68]. Ribociclib led to an improved median OS with 58.7 months versus 48.0 months (HR 0.76; 95% CI 0.61–0.96) [43]. However, ribociclib is not recommended in combination with tamoxifen because of a higher rate of QTc prolongation.

The MONARCH 3 trial included 493 postmenopausal women with de novo metastatic or advanced BC, and randomized them to an AI (anastrozole or letrozole) in combination with the CDK4/6i abemaciclib or placebo [40]. After a median duration of follow-up of 8.1 years, median PFS was 29.0 months compared to 14.8 months (HR 0.535; 95% CI 0.429–0.668; P < 0.0001) [69]. Median OS was numerically improved from 53.7 to 66.8 months in the abemaciclib group; this difference, however, did not reach statistical significance (HR 0.804; 95% CI 0.637–1.015; P = 0.0664) [69].

The DAWNA-2 trial analyzed the CDK4/6i dalpiciclib versus placebo plus letrozole or anastrozole in 456 patients with advanced BC [61]. Median PFS was significantly improved in the dalpiciclib cohort from 18.2 months to 30.6 months (HR 0.51; 95% CI 0.38–0.69; P < 0.0001) [61]. OS data of this trial are awaited.

2.2 Trials Including Endocrine Pretreatment

The PALOMA-3 trial included a heterogeneous population, including both patients who had experienced disease progression or relapse under previous ET as well as patients with disease relapse within 12 months after completion of adjuvant ET. Also, one previous treatment line of chemotherapy was allowed. In total, 521 patients were randomized 2:1 to fulvestrant plus either palbociclib or placebo [38]. Sensitivity to ET, menopausal status, and the presence of visceral metastasis were used as stratification factors [38]. Out of the patients, 21% were pre- or perimenopausal and 21% were in the first-line metastatic setting; 37% had already received chemotherapy for metastatic disease [38]. After a median follow-up of 8.9 months, median PFS was 9.5 in the palbociclib arm versus 4.6 months in the placebo group (HR 0.46; 95% CI 0.36–0.59, P < 0.0001) [38]. The median OS difference at 73.3 months did not, however, reach statistical significance with 34.8 months versus 28.0 months (HR 0.81; 95% CI 0.65–0.99) in the palbocilib and placebo arms, respectively [45]. Among first-line patients, the median OS was not reached in the palbociclib cohort; in the placebo cohort it was 51.8 months (HR 0.64; 95% CI 0.46–0.88). In the second-line setting, median OS was 39.7 months versus 33.7 months (HR 0.778; 95% CI 0.59–1.04) [70].

Ribociclib in advanced BC after endocrine treatment was tested in the MONALEESA-3 phase III study [39]. The trial included postmenopausal women and men who progressed after prior (neo)adjuvant ET or after a maximum of one line of ET for advanced disease. In total, 726 patients were enrolled, including approximately 20% with de novo metastatic BC and approximately 4% with an early relapse defined as a disease-free interval of ≤ 12 months [39]. The trial reached its primary endpoint with a median PFS of 20.5 months versus 12.8 months (HR 0.593; 95% CI 0.480–0.732; P = 0.001) [39]. The median OS after 56.3 months was 53.7 months on the ribociclib group versus 41.5 months in the control group (HR 0.73; 95% CI 0.59–0.90) [70]. In a subgroup analysis of first-line patients, median OS was 67.6 months versus 51.8 months at a median follow-up of 70.8 months (HR 0.67; 95% CI 0.50–0.90) [71].

The MONARCH 2 trial analyzed the efficacy of abemaciclib in patients who experienced disease progression during first-line ET, during (neo)adjuvant ET, or within 12 months after completion of adjuvant ET [37]. In this study, 669 patients were randomized 2:1 to receive fulvestrant plus abemaciclib or placebo. In the intervention group, abemaciclib yielded a median PFS of 16.4 months compared with 9.3 months (HR 0.553; 95% CI 0.449–0.681; P < 0.001) [37]. After a median follow-up time of 47.7 months, the median OS was 46.7 months versus 37.3 months (HR 0.757; 95% CI 0.606–0.945; P = 0.01) [41].

The DAWNA-1 trial included patients who experienced progression ≥ 6 months after initiating endocrine therapy for recurrent or metastatic disease or relapse or progression on adjuvant endocrine therapy but after the first 2 years or within 12 months from completion of adjuvant ET and analyzed dalpiciclib plus fulvestrant versus placebo plus fulvestrant [72]. The median PFS was 15.7 months compared to 7.2 months (HR = 0.42; 95% CI 0.31–0.58; P < 0.0001), but OS data are not yet available [72].

Tibremciclib was evaluated in the TIFFANY trial, in combination with fulvestrant in patients with HR+ aBC who had progressed on previous endocrine therapy or chemotherapy. Median PFS was markedly prolonged versus placebo (HR = 0.31; 95% CI 0.19–0.48; P < 0.0001), and OS data are awaited [51, 62].

2.3 Comparison of Overall Survival Data of aBC Trials

Although the PFS data are consistent in all the CDK4/6i trials, OS data are divergent. Statistically significant improvements in OS had been reported with ribociclib in the MONALEESA-2, MONALEESA-3, and the MONALEESA-7, and with abemaciclib in the MONARCH-2 study [41,42,43, 71]. The combination of abemaciclib with letrozole as first-line treatment showed a numerically marked but not statistically significant OS benefit [69]. Palbociclib did not improve OS in the trials [45, 65], see also Table 1. The reason for the divergent OS results despite promising PFS results has not been fully clarified: Besides questions regarding statistical power, differential inhibitory activity against different CDKs may have yielded contradictory survival results. In addition, subtle differences between the different trial populations are evident. Recently presented retrospective real-world analyses might support this explanation [73, 74]. Studies with CDK4/6i and fulvestrant comprised heterogenous study populations regarding for example number and type of previous treatment lines such as type of endocrine resistance.

Table 1 Survival outcome in CDK4/6 inhibitor Phase III trials for advanced breast cancer

2.4 Sequential Treatment and Combinations in Metastatic BC

Based upon the trials discussed above, CDK4/6 inhibition in combination with ET presents the preferred first-line treatment in HR+, HER2− advanced BC in the absence of visceral crisis [75,76,77]. The SONIA trial showed that there might be some patients who would benefit from single-agent ET as first-line treatment followed by CDK4/6 inhibition as second-line treatment [78]. Eligible patients for this phase III trial were pre- or postmenopausal women with de novo metastatic BC or disease relapse 12 months from completion of (neo)adjuvant ET. Postponing CDK4/6i to the second treatment line did not negatively impact upon OS [78]. In clinical reality, these results are challenged by the fact that palbociclib is not regarded as the default CDK inhibitor anymore based upon its lack of positive OS data, as well as the fact that approximately 20% of patients are lost from first- to second-line therapy, suggesting a best first approach.

The preferred treatment strategy for patients with visceral crisis is a chemotherapy-based treatment according to current treatment guidelines [75]. However, in the RIGHT Choice trial, ribociclib plus letrozole or anastrozole in combination with goserelin was compared with different chemotherapy combinations such as docetacel plus capecitabin, paclitaxel plus gemcitabine, or capecitabine plus vinorelbine in 222 pre- or perimenopausal patients with clinical aggressive advanced BC. Assigned treatment was continued until progression or inacceptable toxicity. Median PFS was significantly improved with ribociclib plus ET [79]. Moreover, in the ABIGAIL trial the combination of abemaciclib + ET showed improved overall response rate (ORR) compared to induction with paclitaxel in previously untreated HR+HER2− aBC with aggressive disease criteria [80]. However, more data supporting this approach are needed. In addition, response kinetics were similar, suggesting ET as the preferred treatment option in this population as well.

The choice of second-line treatment is nowadays based on molecular biomarkers including mutations or alterations in PIK3CA, PTEN, AKT, ESR1, or germline or somatic BRCA1/2 or germline PALB2 [75, 76, 81,82,83,84]. Fulvestrant monotherapy has been superseded for most patients by these molecularly targeted approaches. In the absence of such mutations, treatment with the mTOR inhibitor everolimus plus ET or CDK4/6i treatment beyond progression present potential endocrine-based treatment strategies [76, 81, 85, 86]. In the phase II MAINTAIN trial, patients progressing on CDK4/6i plus ET switched ET to fulvestrant or exemestane and were randomized to the CDK4/6i ribociclib or placebo [86]. Of the randomized patients, 86.5% had received palbociclib and 11.7% ribociclib in the previous treatment line; 83% of patients switched to fulvestrant as the endocrine backbone upon progression. A significant improvement in PFS was shown in the cohort switching to ribociclib, with a median PFS of 5.29 months compared to 2.27 months in the placebo group (HR 0.57; 95% CI 0.39–0.85; P = 0.006) [86].

The phase II PALMIRA trial analyzing the switch of solely the ET partner and continuation of palbociclib beyond progression did not show a significant benefit [87].

Primary outcomes of the postMONARCH phase III trial were presented at the 2024 ASCO Annual Meeting. In this study, patients progressing on first-line endocrine therapy plus CDKi and on/after adjuvant treatment with a CDK4/6i and ET were included and randomized between abemaciclib plus fulvestrant and placebo plus fulvestrant [85]. Abemaciclib significantly improved PFS (HR = 0.66; 95% CI 0.48–0.91; P = 0.01) [85]. Of the patients, 59% had received palbociclib, 33% ribociclib, and 8% abemaciclib as previous CDK4/6i [85]. The efficacy of the strategy of CDK4/6 inhibition beyond progression might be based on the structural differences and varying binding abilities [17, 56].

Testing an intensified treatment combination the INAVO 120 phase III trial analyzed the combination of the PIK3CA inhibitor inavolisib, palbociclib, and fulvestrant as first-line treatment in a high-risk population of patients with early relapse and tumors harboring activating PIK3CA mutations [88]. Addition of inavolisib, a selective PI3K-inhibitor, yielded a significant improvement in PFS from 7.3 months to 15.0 months (HR 0.43; 95% CI 0.32–0.59; P < 0.0001), respectively [88]. Furthermore, PFS2 and time to chemotherapy were significantly improved in the cohort receiving inavolisib as well. Therefore, this combination treatment presents a potential treatment standard for a patient population at higher risk.

Within the PADA-1 trial, a pioneering strategy to optimize treatment sequence in advanced/metastatic luminal BC was tested [89]. Patients with de novo aBC received palbociclib in combination with an AI and were repeatedly tested for the emergence of ESR1 mutations in ctDNA, a well-established mechanism of resistance. Once ESR1 mutations were detected in the absence of radiological disease progression, participants were randomized to switching the endocrine backbone from AI to fulvestrant or remaining on AI. Palbociclib was continued in both groups. The switching strategy yielded a significant improvement in PFS [89]. Based upon these results, the SERENA 6 trial was initiated to evaluate this strategy with the selective estrogen receptor degrader (SERD) camizestrant instead of fulvestrant [90].

2.5 Combination Treatment in aBC

The phase II monarcHER trial analyzed the effect of abemaciclib plus trastuzumab and fulvestrant, abemaciclib and trastuzumab compared to chemotherapy and trastuzumab in patients with HR+, HER2+ metastatic BC. Abemaciclib, fulvestrant, and trastuzumab led to a significant improved PFS compared to chemotherapy and trastuzumab with 8.3 months versus 5.7 months (HR 0.67; 95% CI 0.45–1.00; P = 0.051) [91]. No statistic significant difference in PFS was detected in patients who received solely abemaciclib and trastuzumab without fulvestrant. There was, however, no trial cohort with the combination of fulvestrant and trastuzumab, which makes the exact contribution of abemaciclib difficult to discern. Furthermore, the combination of abemaciclib and trastuzumab with or without fulvestrant lead to a numerical but not statistically significant improvement in OS [92].

Another trial challenging a CDK4/6i-based treatment in HER2+ stage IV BC is the phase II PATRICIA trial [93, 94]. Cohort A included patients with ER-, HER2+ BC, whereas cohort B and C comprised patients with ER+, HER2+ BC. Patients in cohort C were randomized between palbociclib, ET, and trastuzumab or chemotherapy and trastuzumab/T-Dm1 and showed a significant improvement in PFS with 9.1 versus 7.5 months (HR = 0.52; 95% CI 0.29–0.94; P = 0.031) in both PAM 50 luminal A and B [94].

The DETECT V trial showed comparable PFS and OS with ribociclib plus trastuzumab/pertuzumab versus chemotherapy plus trastuzumab/pertuzumab in patients with HR+, HER2+ stage IV BC [95]. Furthermore, adding ribociclib after chemotherapy to ET plus trastuzumab/pertuzumab as maintenance treatment significantly improved PFS (27.2 vs. 15.6 months; HR 0.52; 95% CI 0.37–0.75; P < 0.001) and OS (median not reached vs. 46.1 months; HR 0.42; 95% CI 0.24–0.74; P = 0.002) [95].

2.6 Mechanisms of Resistance

Although CDK4/6i are highly effective treatment agents, cancer cells will eventually develop various resistance mechanisms to either CDK4/6 inhibition and/or the endocrine partner. O’Leary et al reported changes and clonal evaluation on palbociclib and fulvestrant in the PALOMA-3 trial [96]. Acquired mutations in ESR1 and PIK3CA were found in ctDNA. Despite the fact that RB1 loss was previously prescribed as a mechanism of resistance to CDK4/6 inhibition, only infrequent and subclonal mutations have been reported in the plasma of patients from the PALOMA-3 trial [96]. As outlined, RB1 is required for the activity of CDK4/6i. Mutated RB1 has no ability to suppress the transcription factor family E2F nor any function in the G1 checkpoint anymore [15]. Acquired RB1 mutations have been described as mechanisms of resistance to CDK4/6i treatment [97, 98]. Other mechanisms of resistance previously described in preclinical trials include amplification of CCN1 or CDK6 [99, 100]. In a pooled analysis of the MONALEESA trials investigating ctDNA, alterations in RB1, CDKN2A/2B/2C, ANO1, and high tumor mutational burden (TMB) were associated to decreased effectivity of ribociclib [101]. Further potential mechanisms of resistance have been reported such as activating alterations in AKT1, ERB2, CCNE, FGFR2, RAS, and in the Aurora kinase A (AURKA) [98]. The AURKA holds an important role in cell cycle regulation, and its amplification was detected in samples with both intrinsic and acquired resistance. AURORA inhibitors are under investigation but toxicities such as myelotoxicity limit the development of these agents [102, 103]. Interestingly, in tissue samples of resistant tumors, two or more alterations were concurrently detected [98]. Upregulation of the proto-oncogene myc was reported as an additional mechanism of resistance [104].

3 Side Effect Profile of CDK4/6i

The varying toxicity profiles of drugs in the class can be explained by the differences in the binding site and or differing binding affinities of the different agents [105].

3.1 Neutropenia

Neutropenia is a common side effect of CDK4/6i: 56.1% of palbociclib patients in the PALOMA-2 trial and 49.7% of ribociclib patients in the MONALEESA-2 trial experienced grade 3 neutropenia [63, 66]. The risk of febrile neutropenia was, however, low, with only 1.8% of cases in the PALOMA-2 study and 1.5% in the MONALEESA-2 trial, respectively [63, 66]. In a pooled analyses of the PALOMA trials, neutropenia was the most common cause of treatment discontinuation, with 1.7% of all patients having discontinued palbocilib due to neutropenia [106]. The high neutropenia rate is linked to CDK6 inhibition, as DDK6 is a known regulator of hematopoietic stem cells [107]. The rate of neutropenia grade 3 was lower with treatment with abemaciclib, at 19.7% [40].

3.2 Cardiac Effects

In a pooled analysis, ribociclib lead to QT interval prolongations in 6.5% if patients (grad 3/4 1.2%), leading to dose reductions in up to 2.5% of patients [108]. Prolongation of the QT interval was documented especially in combination with tamoxifen [68]. Ribociclib is not licensed in combination with tamoxifen.

3.3 Diarrhea

Diarrhea is a common adverse event (AE) with abemaciclib, observed in 81.3–86.4% of patients with a median onset of 8 days and median duration of 8 days in grade 3 cases [37, 40]. Grade 3 diarrhea occurred in 9.5% of patients, leading to treatment discontinuation in 2.3% in the MONARCH 3 trial [40]. Under treatment with ribociclib, the rate of diarrhea was lower, with 30% of patients experiencing all-grade diarrhea including 2% grade 3 events [108]. In the PALOMA-2 trial, diarrhea was seen in 26.1% of patients, with 1.4% cases of grade 3 diarrhea [63].

3.4 Pneumonitis/Interstitial Lung Disease

Interstitial lung disease/pneumonitis was described as a class effect of CDK4/6i and occurred in 3.4% with abemaciclib, in 1.5% with ribociclib, and 1.49% with palbociclib in pooled analyses [106, 108, 109].

3.5 Other Side Effects

AEs leading to discontinuation of ribociclib in 14.6% of patients mostly were due to an increase alanine-aminotransferase (ALT) and/or aspartate-aminotransferase (AST). Ribociclib led to an increase in ALT grade 3 in 7% and grade 4 in 1% of patients [108]. Grade 3 increases in ALT and AST were seen in 5.8% and 3.8% of patients with abemaciclib, respectively [40]. Another common side effect of abemaciclib is a reversible elevation of creatinine levels not reflecting a worsening of kidney function, which was the most common laboratory abnormality, occurring in 19.0% of patients [40]. This is due to abemaciclib-induced inhibition of renal transporters leading to decreased urinary creatinine excretion without changing the glomerular filtration rate [110].

Abemaciclib led to an increased risk of venous thromboembolic events, which were observed in 4.9% of participants receiving abemaciclib in the MONARCH 3 trial [40]. The risk for venous thromboembolic events was seen especially in combination with tamoxifen, with 4.3% compared to 1.8% in the MonarchE trial, respectively [111].

4 CDK4/6i in Early BC

Patients with HR+, HER2− early BC and additional risk factors for recurrence such as node-positive disease, large tumor size, or high tumor grade, have a high probability of recurrence within 5 years or for late relapse [112,113,114,115]. With extended ET the latter risk can be reduced [113, 116,117,118]. A large Dutch trial analyzing retrospectively outcomes of 87,455 patients with early-stage HR+, HER2− BC showed 10-year OS rates of 84.1% and 10-year recurrence-free rates of 98.7% in low-risk patients compared to 10-year OS rates of 63.4% and 10-year recurrence-free rates of 72.3% in high-risk patients [114]. (Neo)adjuvant chemotherapy was administered in 73% of high-risk patients [114]. Therefore, there is still an unmet need to improve outcomes in these high-risk eBC patients.

Beyond clinical markers, molecular biomarkers associated with risk for recurrence have been studied [119, 120]. Because of its particular relevance for HR+ eBC, their predictive ability with respect to late recurrences have also been extensively studied and compared with scores based on clinicopathological prognosticators alone [120,121,122]. An analysis of the BIG 1-98 study showed that amplifications on chromosome 8p11 and BRCA2 mutations were associated with a high risk of late recurrence, whereas the presence of PIK3CA mutations reduced the risk [123]. A copy number gain of FGFR1 was associated with late recurrence risk as well [124].

While senescence is irreversible, quiescence on the other hand represents a state of reversible cell cycle arrest [125]. In vivo experiments showed that CDK4/6i have the ability to induce a senescent state in tumor cells [125,126,127,128,129,130]. It seems to be a cell-, situation- and drug-dependent manner whether CDK4/6 inhibition induces senescence or quiescence in RB1-positive cells [125].

4.1 Neoadjuvant Trials Investigating CDK4/6i

Different scores for response to ET had been established and tested [131]. The preoperative endocrine prognostic index (PEPI) score for example is a risk score to predict risk of relapse and BC-specific death conducted from a multivariable Cox proportional hazard regression model comprising pathological tumor size, pathological node status, clinical response, surgical specimen ES status, histological grade, and Ki67 level [131]. Another marker for response is the complete cell cycle arrest (CCCA), defined as Ki67 ≤ 2.7% [132].

In the WSG-ADAPT HR+/HER2− trial, the genomic risk score in the primary tissue biopsy was combined with an assessment of dynamic response of Ki67 under ET in high-risk luminal patients: Patients received 2–4 weeks of ET before the Ki67 in tissue was reassessed for endocrine response. Patients with pN0-1, recurrence score 0–11 or recurrence score 12–25 and Ki67 responsive disease received further treatment with ET alone. Patients with higher post-endocrine Ki67 levels were randomized between 4x paclitaxel followed by 4x dose-dense epirubicin/cyclophosphamid or 8x nab-paclitaxel followed by 4x dose-dense epirubicin/cyclophosphamid [133]. It was demonstrated that the combination of recurrence score and low post-endocrine Ki67 levels was associated with significantly better iDFS, and could be used even in high-risk patients to de-escalate chemotherapy [133].

Based upon their mechanism of action, CDK4/6i can achieve a substantial reduction of Ki67 and complete cell cycle arrest in a substantial subset of patients [132]. The combination of ET with CDK4/6 inhibition could be a feasible neoadjuvant strategy in luminal BC leading to higher efficacy.

In the NEOPAL phase II trial, palbociclib plus letrozole was compared with third-generation chemotherapy in patients with high-risk HR+, HER2− clinical stage II-III, luminal BC [134]. While pathologic complete remission rate (pCR) was lower in patients receiving targeted therapy, no significant differences in clinical response rates, breast-conserving surgery rates, PFS, and invasive disease-free survival (iDFS) were shown [134, 135]. Palbociclib could not increase the rate of pCR in patients with low genomic risk though [134]. However, pCR might not be a relevant prognostic marker in patients with early luminal BC [136]. In the NeoPalAna trial palbociclib was added after 4 weeks of treatment with anastrozole ± goserelin and administered for four cycles. After exposure to palbociclib, the CCCA rate was significantly increased from 26 to 87% at cycle 1/day 15 (P < 0.001) [132]. However, a rebound of Ki67 in tissue at surgery was seen with significantly higher Ki67 levels compared to day 15 at cycle 1, but not significantly different from the Ki67 at start of treatment [132]. The combination of palbociclib and letrozole was also tested in the phase II trial PALLET [137]. In this study postmenopausal women with luminal eBC with 2 cm or larger size were randomized into four cohorts. Cohort A received letrozole for 14 weeks, cohort B letrozole for 2 weeks, followed by 14 weeks with palbociclib and letrozole, cohort C palbociclib for 14 weeks followed by the combination of palbociclib and letrozole, and cohort D palbociclib and letrozole for 14 weeks [137]. Palbociclib induced a greater change in Ki67, more complete cell-cycle arrests, and suppression of cleaved poly (ADP-ribose) polymerase [137].

The phase II CORALLEEN trial compared ribociclib plus letrozole to chemotherapy in patients with stage I–IIIA HR+, HER2− luminal B disease (assessed by PAM 50) [138]. The rate of patients with PAM 50 low-risk-of recurrence (ROR) score at surgery in both arms was defined as the primary endpoint, which was comparable between both arms [138].

In the (placebo-controlled) FELINE trial, ribociclib with 600 mg versus a continuous dose of 400 mg in combination with letrozole was administered for six cycles in postmenopausal women with luminal > 2 cm or node positive eBC [139]. Ribociclib led to a significant increase in CCCA at day 14 with 92% and 52% (P < 0.0001), respectively. Nonetheless, there was a rebound effect at surgery with no significance difference in CCCA [139]. The PEPI sore 0 did not differ with a rate of 25% in both cohorts [139].

In the single-arm phase II neoMONARCH trial, postmenopausal women with stage II, IIIA, or IIIB BC were randomized to receive a 2-week lead-in of anastrozole, abemaciclib, or abemaciclib plus anastrozole followed by 16 weeks of the combination therapy [140]. A significant decrease of Ki67 and a significant reduction in the expression of the cell-cycle genes FOXM1, RRM2, CCNE, MKI67, and TOPO2A were detected in the tumor biopsy taken after 2 weeks on treatment in patients who received abemaciclib alone or in combination with anastrozole compared to anastrozole alone [140]; 8% of patients reached a pCR at the time of surgery [140].

The phase III NCT03969121 trial was conducted to assess early biological response to ET [141]. 141 women with operable HR+, HER- BC were randomized to 16 weeks of palbociclib versus placebo in combination with ET. Inclusion criteria were tumor size at least 15 mm, T1c-3, N0-1, Ki67 ≥ 14%. Letrozole was administered in postmenopausal patients or pre- or perimenopausal women received tamoxifen plus OFS. No significance between ET response such as PEPI score and EPclin risk score after treatment was detected at surgery. However, patients treated with palbociclib had a lower rate of a high-risk EPclin Risk Score [141].

In the phase III SAFIA trial, 354 patients with early luminal BC were enrolled to receive palbociclib versus placebo in combination with fulvestrant with or without goserelin for 5 months [142]. No statistically significant difference in outcomes was detected between the palbociclib and the placebo cohort. A clinical benefit of 96% was seen in patients with a low oncotype DX recurrence score [142]. Neoadjuvant trials with reported outcome data are shown in Table 2.

Table 2 Neoadjuvant CDK4/6i trials

4.2 Adjuvant Trials Investigating CDK4/6i

Clinical data are available from four randomized clinical trials investigating palbociclib, ribociclib, and abemaciclib, respectively, in the adjuvant setting (Table 3).

Table 3 Adjuvant CDK4/6i trials

4.2.1 PALLAS

Overall 5796 patients with stage II or stage III HR+, HER2− early BC were randomized to receive 2 years of palbociclib in addition to ET or Et alone [144]. Of these, 17.9% of patients had stage I or stage IIA disease, and 82% had stage IIB or III disease. Adjuvant or neoadjuvant chemotherapy had been administered in 82.5% of patients; 13.0% of patients had node-negative disease, 49.3% N1, 24.5% N2, and 13.2% N3 stage; 45.6% of patients were pre- or perimenopausal and 0.6% were male. Adding 2 years of palbociclib to standard ET did not improve iDFS (4 years’ iDFS 84.2% vs. 84.5%; HR 0.96; 95% CI 0.81–1.14; P = 0.65). No benefit of palbociclib addition to ET was observed in the intention-to-treat or different subgroups [144].

4.2.2 PENELOPE-B

PENELOPE-B investigated the role of palbociclib in the post-neoadjuvant setting. Eligible patients had residual disease after neoadjuvant chemotherapy and high risk of relapse defined as clinical pathological staging-estrogen receptor grading (CPS-EG) score ≥ 3 or 2 and ypN+ [145]. 1,250 women were randomized to 13 cycles of palbociclib or placebo plus ET [145]; 49.3% of the patients were pre- or perimenopausal. After a median follow-up of 42.8 months, palbociclib failed to improve iDFS (HR 0.93; 95% CI 0.74–1.17; P = 0.525) [145].

4.2.3 NATALEE

Ribociclib was given at an intermittent dose of 400 mg plus an AI for 3 years, and was compared with single-agent AI in 5101 patients with stage IIB–III or stage IIA with at least one positive lymph node or a Ki67 > 20%, G2, G3, or high genomic risk [47]; 43.9% of the patients were pre-or perimenopausal and 0.4% were male. At a median follow-up of 34 months, addition of ribociclib to ET yielded a significant iDFS improvement with an absolute benefit of 3.3% and an HR of 0.75 (95% CI 0.62–0.91; P = 0.003) [146, 147]. At the 4-year landmark analysis, 62.8% of patients had completed 3 years of ribociclib. iDFS was improved from 83.6% at 88.5% (HR 0.715; 95% CI 0.609–0.840; P < 0.001), again suggestive of a carry-over effect. The absolute benefit in patients with node negative disease was 5.1% and 5.0% in node positive disease and 4.3% in stage II versus 5.9% in stage III disease, respectively. OS was still immature with an HR 0.827 favoring ribociclib (95% CI 0.636–1.074). Ribociclib was discontinued in 37.2% of patients, due to AEs in 20.0% of patients [147].

4.2.4 MonarchE

In the MonarchE trial, 5,637 patients with N2 stage or 1–3 positive lymph nodes and a risk factor such as T3, Ki67≥20%, or G3 were included [46]; 43.5% were pre- or perimenopausal. Abemaciclib improved iDFS significantly with a HR of 0.68 (95% CI 0.60–0.77) and a difference in iDFS of 7.6% at 5 years compared to a difference of 6% at 4 years and 4.8% at 3 years [148].The difference in iDFS between the study cohorts increased over the follow-up time, suggesting a carry-over effect. Ki67% is a well-known risk factor in BC, but was not revealed as predictive marker in the MonarchE trial [149] while retaining its prognostic signal. OS did not reach statistical significance but fewer deaths were reported in the abemaciclib cohort (208 vs. 234), more patients in the control arm were alive with metastatic disease without any difference in overall survival; 30% of patients discontinued adjuvant abemaciclib therapy before the 2-year duration was reached [148]. Based on these results, abemaciclib was the first CDK4/6i to be approved by regulators in the adjuvant treatment setting [150].

An overview of adjuvant trials of CDK4/6i and ET is given in Table 3.

4.3 Safety and Tolerability of CDK4/6i in Early BC (eBC) Trials

The most frequent side effects in the PALLAS trial were neutropenia (83.5%), leukopenia (55.1%), and fatigue (41.0%) [144, 151]. In the PENELOPE-B trial, the most common AEs were leukopenia (99.2%), neutropenia (95.7%), and anemia (73.9%) [145]. Grade 3-4 AEs included neutropenia (70.0%), leukopenia (56.1%), and infections (3.2%) [145]. Regarding non-hematological AEs, no significant difference between the CDK4/6i cohort and the control group waas reported and no potential study medication-related deaths were described [145]. Drug discontinuation overall occurred in 42%, of which 72% were due to toxicity. Side-effect incidence and discontinuation rate may depend on individual patient factors such as obesity [152].

In the MonarchE trial, the most common AEs were diarrhea, fatigue, and abdominal pain [46]. Diarrhea was mostly of grade 1–2 with 75.27% and grade 3 in 7.8% of patients. A venous thromboembolic event occurred in 1.2% of patients compared to 0.3% in the ET-alone arm. The most common side effects in the control group were arthralgia, hot flushes, and fatigue [46]. The most common grade 3–4 AEs were neutropenia (19.8%), leukopenia (11.4%), and diarrhea (7.8%); 6.4% discontinued both abemaciclib and ET, mostly due to diarrhea (2.4%) and fatigue (1.0%). Arthralgia was the most common reason for drug discontinuation in the control group, with 0.2% permanently discontinuing treatment. However, two deaths—one due to diarrhea and one due to pneumonitis—were considered as related to study treatment [46]. AEs such as diarrhea, neutropenia, and fatigue led to dose interruption in 61.7% of patients and to dose reduction in 43.6%. Overall, 30% of patients stopped their treatment before 2 years were reached [46].

In the NATALEE trial, the most frequent AEs were neutropenia, arthralgia, and liver-related events (transaminitis). Grade 3 neutropenia occurred in 41.8%, grade 3 elevation of ALT and AST occurred in 6.1%, and AST in 3.8% of patients, respectively. Liver-related events were the most common AEs that led to drug discontinuation, with 8.9%. Arthralgia led to discontinuation in 1.3%. QT interval prolongation occurred in 5.2% of patients [146]. No case of potential study treatment-related deaths were reported. Overall, 32% of patients stopped their treatment before 3 years were reached [146, 147]. The lower dose with 400 mg ribociclib led to less QT interval prolongation and lower grade 3 neutropenia, compared to the 600 mg dose used in the metastatic setting. Differences and similarities of the MonarchE and NATALEE trial are shown in Fig. 2.

Fig. 2
figure 2

Similarities and differences between the monarchE and NATALEE trials [147, 148]. RTX radiotherapy, cont. continuous, BID twice daily, QD once a day, G grade, iDFS invasive disease-free survival, ET endocrine treatment, HR hazard ratio, CI confidence interval, vs versus

AEs and discontinuation rates reported in adjuvant CDK4/6i trials are listed in Table 4.

Table 4 Adverse events in CDK4/6i adjuvant trials

4.4 Additional Trials and Combinations in eBC

The impact of additional chemotherapy versus ET alone on outcomes of luminal BC has not been fully elucidated to date. Outcomes largely depend on baseline risk factors, and these prognosticators are still relevant even for very late recurrences [153]. Genomic risk scores such as Oncotype and Mammaprint can further improve risk assessment [154,155,156]. In premenopausal patients, these tests apparently overestimate the role of chemotherapy, raising the question to what extent the potential benefit of chemotherapy in premenopausal women is directly linked to the endocrine properties of chemotherapy in terms of induction of menopause [157,158,159]. A recent analysis of the RxPonder trial showed at ASCO 2024 that only premenopausal patients with a high anti-Müllerian hormone levels (i.e., truly premenopausal) before chemotherapy derived benefit from adjuvant chemotherapy, supporting the primarily endocrine (side) effect of adjuvant chemotherapy in premenopausal patients [158].

The phase III ADAPTcylce trial is currently being conducted to evaluate the benefit of (neo)adjuvant treatment with ribociclib plus ET compared to chemotherapy in patients with genomic intermediate risk by Recurrence-Score and luminal B like BC with endocrine sensitivity testing (assessed by a drop in Ki67 after 3–4 weeks of endocrine therapy). Preliminary results were presented at the San Antonio Breast Cancer Conference 2023 (SABCC), demonstrating that premenopausal patients had a significantly higher endocrine response rate if they received additional OFS in combination with tamoxifen or AI, irrespective of the recurrence score [160]. The ongoing ADAPTlate trial compares the benefit of adjuvant abemaciclib for 2 years with standard ET after completed locoregional therapy with or without chemotherapy and 2–6 years of prior adjuvant standard treatment in patients with clinical or genomic high-risk HR+, HER2− BC [161], and the planned NoLEEta trial will investigate whether adjuvant chemotherapy is necessary at all in HR+ eBC patients who receive ET and ribociclib in the upfront adjuvant setting [162].

In the future, ctDNA could be used as a biomarker for prediction of risk of recurrence and adverse prognosis [163]. Loi et al. showed in an analysis of the MonarchE trial that patients who remained ctDNA positive or gained ctDNA positivity during adjuvant treatment had a significantly worse outcome [164]. In the ongoing TRAK-ER phase II trial, patients with luminal eBC with an increased risk of recurrence are undergoing ctDNA-based surveillance during standard ET and detection of ctDNA with exclusion of clinically overt metastatic disease triggers, randomized to continuing ET or switching to fulvestrant and palbociclib [165]. In the upcoming neoadjuvant ERIKA trial (ABCSG-63, EU-CT 2023-505758-17), the combination of the SERD elacestrant with ribociclib will be assessed against the combination of ribociclib with AI in a patient population selected by endocrine responsiveness and absence of ctDNA. A number of ongoing adjuvant SERD trials (e.g., ABCSG-62/CAMBRIA-2 and EMBER-4) also include patients with adjuvant CDK4/6i [166].

In the CheckMate 7A8 trial, palbociclib and anastrozole were combined with the immune-checkpoint inhibitor nivolumab in patients with HR+, HER2− early BC [35]. However, the trial was closed early due to AEs such as hepatotoxicity, febrile neutropenia, ILD, and rash [35]. The ORR was 71.4%, including 14 patients with partial response (PR) and one patient with pCR of 21 patients [35].

A combination of palbociclib and fulvestrant plus trastuzumab/pertuzumab was administered preoperatively in the phase II NA-PHER2 trial. Patients with HR+, HER2+ eBC experienced a significant reduction in Ki67 and an overall response of 97%, with 15 patients having a complete clinical response and 14 a partial clinical response; eight (27%) patients had a pCR at surgery [167].

5 Next-Generation CDK Inhibitors

Next-generation CDK inhibitors targeting different CDKs are under clinical development, for example, CDK2 inhibitors. The selective CDK2 inhibitor INCB123667 led to growth inhibition in CCNE1 amplified cell lines and showed antitumor activity in a BC xenograft model [168]. In a phase I trial comprising patients with metastatic solid tumors, INCB123667 induced responses in CCNE1 amplified or cyclin E expressing tumors [169]. A phase Ib/II trial showed antitumor activity and acceptable toxicity of the combination of a selective CDK2 inhibitor and a selective CDK 4 inhibitor in heavily pretreated patients with HR+, HER2− BC [170]. It was suggested that this combination could overcome endocrine resistance [170]. Another agent, fadraciclib, has high selectivity for CDK2 and CDK 9, and induces apoptosis rapidly [171]. Fadraciclib showed higher selectivity than the CDK2/7/9 inhibitor seliciclib, which had already been tested in different malignancies [172,173,174]. The CDK 9/4/6 inhibitor myrtleciclib is thought to selectively induce apoptosis in cancer cells [175]. Furthermore, CDK 7 might be another promising target in HR+ BC. It was shown that inhibition of CDK7 could inhibit cell proliferation and myc expression in CDK4/6i-resistant cells [176, 177]. However, all these agents require further extensive clinical development.

6 Guideline Recommendations for Adjuvant CDK4/6i Treatment

Adjuvant abemaciclib is an established agent in the treatment of node-positive HR+, HER2− eBC, and recommended in all established guidelines [178, 179]. The ESMO guidelines suggest administering abemaciclib in patients without germline BRCA1 or BRCA2 mutations or who had not been tested [156]. In case of a known germline BRCA1 or BRCA2 mutation and stage III or high-risk non-pCR disease, adjuvant olaparib should be given priority [156], because olaparib not only improved DFS significantly (luminal: HR 0.70; (95% CI 0.38–1.27); triple negative BC: HR 0.56 (95% CI 0.43–0.73), but also OS (4-year OS 89.8% vs. 86.4%) [180, 181], NCCN guidelines recommend adjuvant abemaciclib according to the inclusion criteria of the MonarchE trial; however, no clear indication was given regarding sequential administration in patients who have both an abemaciclib and an olaparib indication [77]. Half of the panelists of the St. Gallen Consensus suggested administering olaparib to all patients suiting the OlympiA inclusion criteria [182]. The recently FDA- and EMA-approved CDK4/6i ribociclib will likely find inclusion in guidelines as well, since recent NATALEE results suggest a benefit of adjuvant CDK4/6 inhibition in an intermediate risk population as well, including patients with node-negative disease. Considering treatment-emergent toxicity, costs, and workload, individualized risk-adapted strategies are required as well as improved patient education about factual and perceived risks and potential treatment burden [183].

7 Summary

CDK4/6i have revolutionized the treatment of HR+, HER2− BC both in metastatic/advanced and early disease. A clinically relevant and highly consistent improvement in terms of PFS was reported in aBC trials with a favorable toxicity profile; in addition, some trials observed a significant benefit in terms of OS as well, leading to ET and CDK4/6i being the first-line treatment standard for the vast majority of patients with HR+/HER2− aBC. The CDK4/6i display different properties with regards to binding affinity to different CDKs, and switching to another agent upon progression might represent a feasible strategy in the absence of other druggable targets. In eBC, abemaciclib and ribociclib showed significant benefits when added to adjuvant ET, whereas palbociclib failed to improve iDFS in eBC. In any case, a treatment decision as to whether the additional toxicity is justified in the adjuvant setting must rely on shared decision-making with the individual patient, based on a thorough assessment of expected relative and absolute benefits as well as individual risk of (distant) recurrence. Patients with lower risk profiles (among those with increased risk) might carefully weigh the potential additional treatment burden versus the absolute benefits to expect. Likewise, the decision between abemaciclib and ribociclib will probably be based on risk profile, toxicity profile, and patients’ comorbidities, but also individual preferences. Finally, from a health economy point of view, healthcare systems will also have to assess the benefit and the burden of additional cost of these new agents added to standard endocrine treatment.