More than 20 years have passed since the cyclin-dependent kinases 4 (CDK4) and CDK6 were discovered to drive oncogenesis in several cancer types.
1D-type cyclins form a complex with CDK4 and CDK6, and promote tumorigenesis by inducing cell cycle transition from G1 phase to S phase. Cyclin D1 is overexpressed in more than 50% of all breast cancers, and CDK4 and cyclin D1 amplification have been reported in up to 25% of breast cancers.2
Because CDK4 and CDK6 regulate the cell cycle, they have been attractive targets for cancer treatment for many years. First-generation CDK inhibitors were broad and nonselective, which likely explained why they did not exhibit activity in clinical trials. Expanding knowledge of the cyclin D:CDK4 and CDK6 complex led to the development of second-generation CDK inhibitors, which selectively target CDK4 and CDK6, leading to greater clinical success.1
History of CDK4&6 Inhibitors
The D-type cyclins were first discovered in 1991. In budding yeast cells and murine macrophages, D-type cyclins were found to induce the transition from G1 phase to S phase of the cell cycle. Subsequently, the corresponding genes and amino acid sequences were identified for cyclins D1, D2, and D3 in humans. The D-type cyclins were determined to have potential oncogenic properties because they responded to mitogen signals and were found to be involved in parathyroid adenoma pathogenesis.1
CDK4 was the first cyclin-dependent kinase that was discovered to bind to and be activated by the 3 D-type cyclins. Shortly thereafter, CDK6 was identified and was found to be similar in structure and function to CDK4. Once activated by D-type cyclins, CDK4 and CDK6 phosphorylated the retinoblastoma protein (pRb), causing it to inactivate and induce the cell to transition from G1 to S phase.1CDK4 in particular has nearly exclusive specificity for pRb.3Other CDKs were subsequently discovered, including transcriptional, mitotic, and regulatory CDKs.2
Because the cyclin D:CDK axis regulates the cell cycle, several attempts have been made to target this pathway in the treatment of cancer since the early 1990s. Flavopiridol, also known as alvocidib, was the first CDK inhibitor to enter clinical trials.4It was also the most studied first-generation CDK inhibitor, with more than 60 clinical trials over the course of 15 years.2,3Flavopiridol is derived from the chromone alkaloid, rohitukine, and is administered intravenously.3,4Although flavopiridol inhibited CDK4 and CDK6, it also suppressed multiple other CDKs, including transcriptional CDK9. As a result, flavopiridol induced not only cell cycle arrest, but also cytotoxicity by inhibiting transcription.3
Despite the fact that flavopiridol demonstrated in vitro activity against cancer, preclinical success did not translate into clinical activity.3Flavopiridol had low-to-minimal efficacy in phase I and II trials in patients with cancer who experienced toxicities similar to those of conventional chemotherapy agents, including infusion site irritation and gastrointestinal toxicity. In addition, patients with breast cancer in a phase I trial experienced overly high rates of neutropenia.2
Despite these disappointing results, flavopiridol demonstrated clinical activity in chronic lymphocytic leukemia, with a 40% response rate in one early-phase trial. However, subsequent phase III studies were not pursued.3
Seliciclib, or R-roscovitine, is a purine-based, nonselective CDK inhibitor. In a phase I trial, seliciclib induced a partial response in only 1 of 56 patients. In APPRAISE, aphase II trial in non-small cell lung cancer (NSCLC), seliciclib did not improve efficacy, and the study was terminated.3
UCN-01, a staurosporine analog and broad CDK inhibitor, showed efficacy in preclinical studies, but had significant toxicities in a phase I trial, including arrhythmia and hyperglycemia. In a phase II trial of patients with breast cancer, UCN-01 did not demonstrate efficacy.2
The poor specificity of early CDK inhibitors was likely the primary reason for their lack of efficacy. Broad coverage combined with off-target effects resulted in a small therapeutic window, with significant toxicities such as diarrhea, neutropenia, and anemia that limited further dose escalation. Finally, the study populations were nonselective in the absence of any biomarkers to identify patients who might be likely to respond to treatment.3
After the failures of multiple CDK inhibitors, efforts were made to develop more selective agents to achieve greater efficacy and reduce off-target effects. From 1993 to 2000, multiple attempts to synthesize a CDK4 inhibitor were unsuccessful.1Then, in the early 2000s, PD0332991 entered development and eventually became the CDK4&6-selective inhibitor, palbociclib. Success in clinical trials led the US Food and Drug Administration (FDA) to approve palbociclib plus fulvestrant for treatment of estrogen receptor (ER)‒positive, human epidermal growth factor receptor (HER2)‒negative, advanced breast cancer that has progressed on prior endocrine therapy.5Other CDK4 and CDK6 inhibitors have emerged as well, including abemaciclib and ribociclib, both of which have demonstrated clinical activity in the clinical setting and have entered phase III trials.2
Abemaciclib plus fulvestant in HR+/HER2- MBC (not recruiting)
Abemaciclib plus an NSAI in HR+/HER2- MBC (not recruiting)
Ribociclib plus ceritinib in ALK+ NSCLC
Abemaciclib versus erlotinib in KRAS-mutant NSCLC (currently recruiting)
Palbociclib with exemestane versus capecitabine in HR+/HER- MBC following NSAI resistance (currently recruiting)
Palbociclib plus endocrine therapy following neoadjuvant chemotherapy in HR+/HER2- breast cancer (currently recruiting)
Palbociclib with adjuvant endocrine therapy in HR+/HER2- breast cancer (currently recruiting)
Ribociclib with fulvestrant in HR+/HER2- MBC (currently recruiting)
Ribociclib plus letrozole for HR+/HER2- MBC (not recruiting)
Ribociclib or placebo with an NSAI/goserelin or tamoxifen/goserelin in HR+/HER2- MBC (currently recruiting)
The Cyclin D:CDK4 and CDK6 Complex
D-type cyclins and CDKs regulate the cell cycle via the Rb pathway and underlie the tumorigenic mechanism for several types of cancer. When cells in G1 phase encounter mitogenic and adhesion signals, they synthesize cyclin D1, which binds to and activates CDK4 and CDK6, known as interphase CDKs. D2 or D3 cyclins may also activate CDK4 and CDK6 by a similar process.2
The cyclin D:CDK4 and CDK6 complex then goes on to phosphorylate and inactivate pRb, a tumor suppressor that sequesters the E2F family of transcription factors. Thus released, E2F facilitates the transcription of S-phase target genes, driving the cell from G1 into S phase. The product of one transcription target, cyclin E, binds to CDK2, forming a complex that further phosphorylates, and deactivates pRb and other molecules that regulate the G1/S checkpoint. A positive feedback loop results, and cell cycle progression and proliferation proceed independent of further mitogen signals. Cyclin D:CDK4 and CDK6 may also phosphorylate components of other proliferation and migration pathways, such as SMAD2 and FOXM1.2
Several negative regulators, known as cyclin kinase inhibitors, block the cyclin D:CDK4 and CDK6 complex. The INK family, which includes p16INK4a, and the CIP/KIP family, which includes p21 and p27, are upregulated in the presence of inhibitory signals, such as transforming growth factor-β and senescence. By binding to CDK4 and CDK6, p16 INK4a prevents D1 from forming a complex with CDK4 and CDK6, thereby inducing cell cycle arrest and senescence. Dysregulation of p16 INK4a activity has been found to drive the pathogenesis of many cancers. In breast cancer, loss of p21 and p27 may mediate resistance to estrogen treatments.2
Overexpression of D-type cyclins appears to be required to initiate tumorigenesis in breast cancer. In fact, mouse models that lack D1 or D3 are resistant to cancers driven by RAS, ErbB-2/HER2/neu, and Notch1 oncogenes. In addition, the cyclin D:CDK4 and CDK6 complex is necessary for continued oncogenesis and tumor proliferation in breast cancer. Inhibiting CDK4 and CDK6 or cyclin D1 in vivo blocks further tumor growth and leads to senescence.2
In a genomic analysis performed by The Cancer Genome Atlas in 510 tumor specimens, CDK4 amplification and cyclin D1 overexpression were most frequently present in luminal A, B, and HER2-enriched tumors. Basal-like tumors were more likely to have pRb loss and less concomitant cyclin D:CDK4 and CDK6 dysregulation, reflecting the dependence of the cyclin D:CDK4 and CDK6 complex on functional pRb.2
In both ER- and HER2-positive breast cancer types, endocrine signaling has downstream effects on the cyclin D:CDK4 and CDK6 axis. In particular, estrogen signaling in ER-positive cancer amplifies cyclin D1 levels and CDK4 and CDK6 activity.2Antiestrogen agents have been shown to induce cell cycle arrest and subsequent decreased expression of cyclin D1.4Among patients with breast cancer who do not respond or develop resistance to hormone-based treatment, other dysregulated mitogenic pathways, including HER2, may enhance cyclin D1:CDK4 and CDK6 activity independent of ER signaling. Thus, the cyclin D1:CDK4 and CDK6 complex may be a reasonable target for treatment of ER-positive breast cancer that is resistant to standard endocrine therapy.2
Current CDK4&6 Inhibitors
Palbociclib is an orally bioavailable, selective, reversible CDK4&6 inhibitor. In preclinical trials, palbociclib achieved tumor stasis by blocking CDK4 and CDK6-mediated pRb phosphorylation in cell lines with functional pRb. Lack of pRb phosphorylation led to downregulation of E2F target genes and loss of proliferation marker Ki67. Among patients with mantle cell lymphoma, posttreatment biopsies showed decreased fluorothymidine-PET uptake compared with pretreatment samples. Xenograft models treated with palbociclib did not exhibit significant toxicity. Models lacking CDKN2A, which encodes tumor suppressor p16, responded to palbociclib, whereas those without functional pRb did not. When palbociclib was withdrawn and then reintroduced, tumors did not exhibit resistance, but they did resume proliferation in the absence of palbociclib, demonstrating the reversible nature of palbociclib-mediated CDK4&6 inhibition.1
In clinical trials, palbociclib significantly prolonged progressionfree survival (PFS) among patients with ER-positive, HER2-negative genetic status. In the phase III trial, PALOMA-3, among women with breast cancer that had progressed on prior endocrine therapy, those who were treated with palbociclib plus fulvestrant had a longer PFS than those treated with fulvestrant only. The results of PALOMA-3 led to FDA approval of palbociclib plus fulvestrant for this indication.5PALOMA-2 is a phase III trial that was reported to have met its primary endpoint of improved PFS in women with previously untreated breast cancer who received palbociclib plus letrozole versus letrozole alone. Final results are pending at this time.6
The high specificity of palbociclib for cyclin D:CDK4 and CDK6 complexes may explain the efficacy and tolerable toxicity profiles seen in clinical trials. The half-maximal inhibitory concentration (IC50) of palbociclib for D1:CDK4, D3:CDK4, and D2:CDK6 are 11 nmol/L, 9 nmol/L, and 15 nmol/L, respectively. In contrast, palbociclib is far less selective for other CDKs, with the IC50 for CDK1 and CDK2 greater than 10 μmol/L for each target. The major dose-limiting toxicity (DLT) for palbociclib is myelosuppression, particularly neutropenia. “Given the reliance of myeloid development in the mouse on cyclin D2- and D3-driven CDK6, these results might have been anticipated,” noted Charles Sherr, MD, PhD, with Howard Hughes Medical Institute in Chevy Chase, MD, and colleagues, in an article reviewing CDK4&6 inhibitors.1As a result, palbociclib must be dosed intermittently at 125 mg daily for 3 weeks followed by 1 week off treatment, or 200 mg daily for 2 weeks with 1 week off treatment. Other adverse effects (AEs) are mild and include anemia, nausea, and fatigue.1
Ribociclib is also an oral selective CDK4&6 inhibitor that primarily targets cyclin D1:CDK4 and cyclin D3:CDK6, with IC50 10 nmol/L and 39 nmol/L, respectively.1,4Like palbociclib, it has poor specificity for CDK1 and CDK2, with a corresponding IC50 of greater than 50 μmol/L each. Preclinical data from paired tumor biopsies demonstrated lower Rb phosphorylation and Ki67 expression posttreatment. As with palbociclib, ribociclib has DLTs of myelosuppression and neutropenia, possibly because ribociclib also targets the cyclin D3:CDK complex. Thus, ribociclib requires intermittent dosing with 600 mg daily for 3 weeks followed by 1 week off per cycle. Other AEs include mucositis and a prolonged QTc interval on electrocardiogram.1
Ongoing phase I/II clinical trials in patients with breast cancer have demonstrated preliminary clinical activity of ribociclib in combination with exemestane and everolimus, as well as with letrozole and PI3K inhibitor, BYL719.4,7Ribociclib development has been expedited based on the success with palbociclib, and is currently in phase III trials for treatment of ER-positive, HER2-negative breast cancer (NCT01958021, NCT02422615, NCT02278120).
Similar to palbociclib and ribociclib, abemaciclib is an oral, selective, small-molecule inhibitor of CDK4&6. Compared with the other agents, abemaciclib is a more potent inhibitor of CDK4 and CDK6 via their complexes with cyclin D1, with IC50 0.6 to 2 nmol/L and 2.4 to 5 nmol/L, respectively. In contrast, the IC50 of palbociclib and ribociclib for their targets are 10 nmol/L or higher. The greater potency of abemaciclib for its targets and its selectivity for cyclin D1 may explain the lack of bone marrow toxicity as a major dose-limiting factor. As a result, abemaciclib may be dosed continuously at 200 mg daily without any off-treatment weeks. Other AEs include diarrhea and neutropenia, which may be alleviated by reducing the dosage.1
In preclinical trials, tumor biopsies after abemaciclib treatment have demonstrated decreased pRb phosphorylation and topoisomerase Iiα expression compared with pretreatment samples.1Phase I studies have shown clinical benefit and disease control rates of more than 70% with abemaciclib monotherapy, and these results led to breakthrough therapy designation by the FDA in 2015.1,2,8A phase II trial is investigating abemaciclib as monotherapy for refractory, ER-positive, HER2-negative breast cancer (NCT02102490). As with ribociclib, abemaciclib development has also been expedited to phase III trials, which are evaluating abemaciclib in combination with fulvestrant or nonsteroidal aromatase inhibitors in patients with breast cancer (NCT02107703, NCT02246621).
While the importance of the cyclin D:CDK pathway in the pathogenesis of many cancers has been recognized for decades, early attempts to develop effective CDK inhibitors have been unsuccessful. In recent years, advances in our understanding of the cyclin D:CDK4 and CDK6 axis have refined our ability to produce agents that selectively target key components while minimizing off-target toxicity.
Continuing to elucidate the intricacies of the cyclin D:CDK4 and CDK6 pathway and its interactions with other molecular drivers of cancer will enable us to identify additional, more-specific targets, and to discover biomarkers that will help guide clinical decision making and personalize treatment.
“Despite the decades required for drug discovery and clinical applications, much remains to be learned,” Sherr concluded in his CDK4&6 inhibitor review article. “Future work will indicate whether the promise of CDK4&6 inhibitors, most advanced for breast cancer, can be validated and extended to other cancers.”1