The application and prospect of CDK4/6 inhibitors in malignant solid tumors

Cyclin-dependent kinases (CDKs), part of the serine/threonine protein kinase family, are a group of key kinases that regulate the cell cycle; CDKs are activated by cyclins in a time-dependent manner. Twenty kinds of CDKs have been found, and these CDKs bind with their corresponding regulatory subunits (i.e., cyclins) to form active heterodimers [5]. According to their specialized functions, CDKs can be divided into two main categories: CDKs involved in cell cycle regulation, including CDK 1/2/4/6 and CDKs involved in transcriptional regulation, including CDK 7/8/9/11 [6].

CDK4/6 is the main driving factor in cell cycle regulation and plays a key role in the occurrence and progression of various malignant tumors. Among the four cell cycle phases—G1 phase (prophase of DNA synthesis), S phase (DNA synthesis), G2 phase (prophase of mitosis), and M phase (mitosis)—cyclin D-CDK4/6-retinoblastoma (cyclin D-CDK4/6-Rb) signaling pathway is mainly responsible for regulating the G1-S transition [7]. CDK4 and CDK6 share 71% amino acid homology, and both can bind to cyclin D1/2/3. Under the induction of pro-mitosis signal, cyclin D binds to CDK4/6 and promotes retinoblastoma (Rb) phosphorylation, thus separating transcription factor E2F from the Rb-E2F complex, which causes cells to enter S phase and initiates DNA replication [8, 9] (Fig. 1). Changes in the cyclin D-CDK4/6-Rb pathway have been observed in the tumorigenesis processes of many tumors, such as breast cancer, pancreatic cancer, kidney cancer, liver cancer, and hematologic system tumors [10‐17]. Gene amplification, gene mutation, and abnormalities in upstream and downstream regulators of cyclin D, CDK4, and CDK6 can all lead to abnormal activation of the cyclin D-CDK4/6-Rb pathway [8, 18]. The core regulatory effect of CDK4/6 in the cell cycle illustrates its vital role as a target in the treatment of malignant tumors.

The cdk4, cdk6, rb1, and e2f1 genes are expressed in a wide range of cancers according to analysis of the TCGA PANCAN database, which is composed of 12,839 samples. cdk4 has the highest expression in ACC and the lowest expression in KIRC, cdk6 has the highest expression in LAML and the lowest expression in THCA, and e2f1 has the highest expression in DLBC and the lowest expression in PRAD. The expression of rb1 was the highest in KIRC and the lowest in TGCT. Intriguingly, the expression of cdk4, cdk6, e2f1, and rb1 in breast cancer was moderate or low compared to that in all tumors. The expression of the four genes in other tumors is shown in Fig. 2a. UALCAN cancer database analysis indicated that the expression of cdk4 and e2f1 in breast cancer was significantly higher than that in normal tissues (p < 0.01), while the expression of cdk6 and rb1 was not higher than that in normal tissues. We ultimately observed that the expression levels of cdk4, cdk6, e2f1, and rb1 in digestive system tumors such as COAD, ESCA, STAD, LIHC, and HNSC were significantly higher than those in normal tissues (p < 0.01), suggesting that CDK4/CDK6-E2F1/Rb1 signaling may be involved in the occurrence and progression of these tumors and that CDK4/6 inhibitors may have better efficacy in these tumors; the expression of cdk4 and cdk6 in CESC, PAAD, and THYM was not significantly different compared with that in normal tissues (p > 0.05), suggesting that CDK4/6 inhibitors may have poor or no therapeutic effect in these tumors. However, the expression trends of cdk4, cdk6, e2f1, and rb1 in many other cancers were inconsistent, indicating that this signaling pathway may regulate the cell cycle by crossing with other signaling pathways, and CDK4/6 inhibitor combination therapies may lead to considerable antitumor effects in these cancers (Fig. 2b). Moreover, the cdk4, cdk6, and e2f1 genes show extensive gene amplification and gene mutation in various tumors; deep deletion and gene mutation of the rb1 gene were observed in various tumors, while gene fusion and multiple alterations in these four genes are rare (Fig. 2c). These gene changes may explain the difference in the clinical efficacy of and drug resistance to CDK4/6 inhibitors in different tumors.

Based on the analysis from the UALCAN cancer database, moderate expression of the cdk4 gene in KIRC, LGG, KIRP, MESO, KICH, and SKCM was significantly negatively related to overall survival (p < 0.05); high expression of the cdk4 gene in LIHC was closely related to worse overall survival than low expression and may be a sensitive marker for predicting the prognosis of LIHC. The cdk6 gene was expressed at low levels in UCEC and moderately expressed in BLCA, LUAD, PAAD, LGG, SARC, ACC, and MESO, and cdk6 expression was significantly negatively correlated with the overall survival of patients with these tumors. e2f1 was moderately expressed in PAAD, LGG, ACC, and MESO, and highly expressed in KIRC, HNSC, KIRP, ESCA, CHOL, and KICH, which was closely related to the overall survival of those patients. Moreover, in KIRC and LGG, the rb1 gene was moderately expressed, and its upregulation predicted a significant decrease in overall survival (Fig. 3). Thus, cdk4, cdk6, e2f1, and rb1 not only participate in the progression of cancer cells but might also be useful biomarkers for predicting prognosis in some tumors.

Based on the essential regulatory role of CDK4/6 in the cell cycle, CDK4/6 inhibitors have emerged as antitumor drugs. CDK4/6 inhibitors hinder the transition from G1 phase to S phase by inhibiting Rb phosphorylation and E2F release and induce tumor cycle arrest at G1 phase, which can inhibit tumor cell growth and cause tumor regression [18]. Since CDK inhibitors were developed 20 years ago, CDK4/6 inhibitors have achieved great success in breast cancer [19, 20], and cell cycle therapy has gradually matured. Approximately, 75–80% of patients with breast cancer are hormone receptor (HR) positive, and the proliferation of breast cancer cells depends on the activation of estrogen [21]. Endocrine therapy is the main treatment for HR positive breast cancer [22]; however, drug resistance is inevitable in the course of treatment [20, 23, 24]. Endocrine therapy combined with chemotherapy is not as effective as expected due to the limited survival benefits and higher toxicity in HR-positive breast cancer patients. Therefore, emphasis should be placed on improving endocrine therapy efficacy [25]. The cyclin D-CDK4/6 complex is usually highly expressed or abnormally activated in breast cancer [19]. The mutation rate of cell cycle-related genes in breast cancer is as high as 38% [26]. Increased expression of cyclin D causes continuous phosphorylation of Rb and leads to continuous proliferation of breast cancer cells; blocking CDK4/6 exerts a lethal effect on breast cancer cells. Moreover, Finn’s study confirmed that palbociclib combined with tamoxifen sensitizes endocrine-resistant estrogen receptor-positive breast cancer cell lines in vitro [27]. Thus, investigators turned their attention to CDK4/6 inhibitors.

Alvociclib, the first-generation CDK inhibitor, lacks specificity and blocks CDK1/2/4/6/7/9, causing serious adverse effects and limiting its clinical application [20]. As medical science has advanced, three selective CDK4/6 inhibitors (palbociclib, ribociclib, and abemaciclib) have achieved fairly good curative benefits in breast cancer [28]. Palbociclib is the first FDA-approved CDK4/6 inhibitor [29]. The PALOMA-1 study showed that palbociclib combined with letrozole increased the PFS of HR+/HER2− advanced breast cancer patients from 10.2 to 20.2 months compared to letrozole alone, and its side effects were controllable [25]. Based on this study, the FDA accelerated the approval of palbociclib in combination with letrozole for first-line treatment of postmenopausal HR+/HER2− metastatic breast cancer patients in February 2015 [25]. The PALOMA-3 study demonstrated that the mOS of the palbociclib combined fulvestrant group was higher than that of the fulvestrant group (34.9 m vs 28.0 m). Accordingly, the FDA approved palbociclib in combination with fulvestrant for HR+/HER2− postmenopausal women with advanced breast cancer who had failed previous endocrine therapy in February 2016 [30]. The phase III study MONALEESA-2 showed that ribociclib plus letrozole significantly improved PFS in patients with HR+/HER2− advanced breast cancer compared with placebo plus letrozole (NR vs 14.7 m; HR 0.56, 95% CI 0.43–0.72, p < 0.001) [31]. Based on this trial, ribociclib was approved for first-line treatment of postmenopausal women with HR+/HER2− advanced breast cancer in March 2017 [32]. Palbociclib and ribociclib are breakthroughs in the treatment of advanced HR+/HER2− postmenopausal breast cancer, overturning the old pattern of single endocrine therapy for advanced postmenopausal breast cancer in the past decades, laying the foundation of CDK4/6 inhibitors combined with aromatase inhibitors in first-line treatment for postmenopausal women with HR+/HER2− advanced breast cancer. Abemaciclib is the third FDA-approved CDK4/6 inhibitor. In MONARCH-3 trial, the HR+/HER2− postmenopausal advanced breast cancer patients who did not receive systemic therapy as the aimed population, the 18-month interim results indicated that the PFS of abemaciclib group had not been reached, and the median PFS was 14.7 months in the placebo group; the objective response rates were 59% and 44%, respectively (p = 0.004) [33]. According to this study, the FDA approved abemaciclib combined with aromatase inhibitors for the first-line treatment of postmenopausal HR+/HER2− advanced or metastatic breast cancer in February 2018 [34]. Moreover, in the MONARCH-1 trial, the objective response rate of abemaciclib monotherapy reached 19.7% [35]. Compared with palbociclib and ribociclib, abemaciclib can be administered alone, without serious neutropenia toxicity, which provides a new choice for breast cancer patients and highlights its advantages among the highly competitive CDK-targeted drugs. More clinical trial information on the application of CDK4/6 inhibitors in various breast cancers [36‐40] is demonstrated in Table 1.

The major adverse effects of CDK4/6 inhibitors are leukopenia and neutropenia, mainly caused by palbociclib and ribociclib [41]. CDK4/6 inhibitors can also cause gastrointestinal side effects such as diarrhea, nausea, and vomiting. It is worth noting that once neutropenia occurs simultaneously with diarrhea, the risk of infection is greatly increased [42]. Some patients have prolonged QTc intervals, elevated transaminases, thromboembolism, and others [43]. However, these side effects are reversible and can be controlled by dose interruption, dose reduction, and symptomatic supportive treatment [44].

The cdk4/6 genes are generally expressed among various cancers, and their expression is higher or lower than that of normal tissues in most cancer types. High or moderate expression of CDK4/6-related genes often indicates poor survival. The CDK4/6 genes participate in tumorigenesis by synergistically regulating multiple genes in many signaling pathways. CDK4/6 inhibitors may be more effective when combined with other signaling pathway inhibitors, which is consistent with the complexity and intersectionality of tumor signaling pathway interactions. Although the cdk4/6 gene expression level in breast cancer is relatively low, a marked antitumor effect of CDK4/6 inhibitors has been observed in breast cancer, so we speculate that it may also work in other malignancies, especially in those cancers with high cdk4/6-related gene expression. CDK4/6 inhibitors may have the potential to be widely applied in multiple cancers, similar to traditional chemotherapy drugs such as cisplatin or paclitaxel.

Although main battlefield in HR+/HER2− breast cancer, CDK4/6 inhibitors have also been actively explored in other malignancies. Yang’s study showed that CDK4/6 inhibitors increased the sensitivity of acute myeloid leukemia cells to cytotoxic drugs [17]. Patnaik and Taylor’s study showed that CDK4/6 inhibitors achieved disease control rates of 49% and 44%, respectively, in non-small-cell lung cancer patients (n = 68) and melanoma patients (n = 18) [44, 45]. Adkins’ study indicated an objective response rate of 39% for CDK4/6 inhibitors in patients with head and neck squamous cell carcinoma (n = 62) [46]. It is not difficult to conclude that the exploration of CDK4/6 inhibitors in other cancers except breast cancer, such as liposarcoma, lymphoma melanoma, and many other advanced cancers [47‐50], is still at an early stage and is mostly limited to basic experiments and stage I or II clinical trials (See Table 2).

At present, endocrine therapy combined with CDK4/6 inhibitors has achieved significant therapeutic effects and controllable toxicity in many clinical trials, as mentioned above, and combined therapy has become the most promising therapeutic strategy for HR+/HER2− breast cancer patients.

CDK4/6 inhibitors not only arrest the tumor cell cycle, but also trigger antitumor immunity. First, CDK4/6 inhibitors downregulate E2F transcription factor-related gene expression and upregulate major histocompatibility complex class I molecule expression in breast cancer cell lines [52]; CDK4/6 inhibitors activate endogenous retroviral components in tumor cells, stimulating the production of type III interferons to promote tumor antigen presentation [52, 53]. Second, CDK4/6 inhibitors inhibit the proliferation of regulatory T (Treg) cells and DNA methyltransferase 1 expression in Treg cells, resulting in promoter hypomethylation and suppression of E2F release [52]; other studies demonstrated that the expression of CDK6 in Treg cells was higher than that of other T cell subtypes. CDK4/6 inhibitors can downregulate Treg cell proliferation by inhibiting CDK6 [54]. Furthermore, CDK4/6 inhibitors promote tumor cell clearance by enhancing cytotoxic T cells (CTLs). CDK6 is an upstream regulatory element of nuclear factor of activated T cells (NFAT), and CDK4/6 inhibitors suppress NFAT phosphorylation, the activation of CTLs, and its ability to kill tumor cells [55]. Finally, the Cyclin D1-CDK4 complex directly phosphorylates speckle-type POZ protein (SPOP), and CDK4/6 inhibitors can enhance the immune escape of tumors by decreasing the ubiquitination of SPOP and reducing the degradation of PD-L1 [56, 57]. These synergistic mechanisms provide a theoretical basis for the combination therapy of CDK4/6 inhibitors plus immune checkpoint inhibitors [52, 58]. In Zhang’s study, CDK4/6 inhibitor combined with PD-1 antibody significantly reduced the proliferation of tumor cells and improved the survival rate of carcinogenic mic e[57], illustrating the synergistic antitumor effect of the CDK4/6 inhibitor and immune checkpoint-related inhibitor. The synergism, antagonism, and interactions of those related genes deserve further exploration (Fig. 4, red line).

The PI3K-AKT-mTOR pathway proteins and the CDK-Rb-E2F pathway proteins are widely co-expressed in different tumor types, and the two pathways are the relationship between the upstream and downstream, signifying that the combined targeting inhibition of CDK4/6 and PI3K-AKT-mTOR should further improve the efficacy [59]. Cortes’ research indicated that the combination of PI3K inhibitors or mTOR inhibitors might increase the sensitivity of CDK4/6 inhibitors [60]. Teo’s study showed that CDK4/6 inhibitors combined with PI3K inhibitors increased the apoptosis of triple-negative breast cancer cell lines and induced persistent tumor regression in vivo [61]. Moreover, CDK4/6 inhibitors can restore sensitivity to EGFR/HER2 inhibitors by reducing the activity of mTOR [62]. EGFR/HER2 inhibitors combined with CDK4/6 inhibitors may increase the sensitivity to EGFR inhibitor-resistant lung cancer cells [63]. Goel’s study demonstrated that CDK4/6 inhibitors augmented the efficacy of EGFR inhibitors in esophageal squamous cell carcinoma and reversed drug resistance [64]. A phase II clinical trial (NCT02101034) involving patients with HPV-unrelated head and neck squamous-cell carcinomas showed that palbociclib plus cetuximab resulted in an objective response rate of 39% in patients with disease progression on platinum but cetuximab-naive and 19% in patients with disease progression on cetuximab [46]. Teh’s study demonstrated that CDK4/6 inhibitors in combination with BRAF inhibitors or MEK inhibitors are an effective treatment strategy for melanoma, but continued administration increases toxicity. The mechanism of acquired resistance of CDK4/6 inhibitors combined with BRAF inhibitors or MEK inhibitors may be due to mTOR pathway activation, so adding mTOR inhibitors may overcome the resistance of foresaid combined therapy [65]. Chen’s study showed that RAF inhibitors combined with CDK4/6 inhibitors improve the therapeutic effects of RAS or BRAF mutant tumors [66]. Small’s study confirmed that abemaciclib combined with sunitinib significantly decreased tumor size in a preclinical mouse model of renal cell carcinoma without serious adverse effects [67]. Pek’s study demonstrated that in KRAS-dependent and BRAF-mutant metastatic colorectal cancer cell lines, palbociclib combined with MEK inhibitors is an effective treatment strategy [68]. Ribociclib combined with the ALK inhibitor ceritinib showed excellent therapeutic effects in ALK mutant neuroblastoma [69]. However, large-scale clinical trials are needed to confirm the efficacy of the abovementioned combination therapies.

Cyclin D3-CDK6 can phosphorylate two key enzymes (6-phosphofructokinase and pyruvate kinase M2) in the glucose metabolism pathway and restrain its metabolic activity, which consumes the antioxidants NADPH and glutathione, therefore increasing the level of reactive oxygen species and leading to apoptosis of tumor cells. Palbociclib combined with low-dose pentose phosphate pathway inhibitors may exert synergistic antitumor effects [70]. Vijayaraghavan’s research showed that CDK4/6 inhibitors combined with autophagy inhibitors can maintain the integrity of the G1/S checkpoint and may be a new therapeutic pattern for multiple solid tumors [71]. Francis’s study confirmed that CDK4/6 inhibitors resensitize Rb-positive sarcoma cells to the Weel kinase inhibitor AZD1775 [72].