Aberrant control of the cell cycle, leading to sustained cellular proliferation, is a hallmark of cancer (Watt and Goel, 2022). The progression through the distinct phases of the cell cycle is meticulously regulated by a complex system of cyclin proteins and their partner cyclin-dependent kinases (CDKs), thus making them attractive targets in cancer treatment (Ghafouri-Fard et al., 2022). CDK inhibitors (CDKi) encompass small molecules or antibodies that inhibit the enzymatic activity of CDKs, halting cell cycle progression and inducing cell death (proving effective even in quiescent cancer cells) (Criscitiello et al., 2014). Specifically, cyclin-dependent kinase 4/6 (CDK4/6) and their D-type cyclins control the transition from gap 1 (G1) to synthesis (S) cell cycle phase, playing a crucial role in hormone-receptor positive (HR+) breast cancer (BC) tumorigenesis and endocrine resistance. Hence, blocking this pathway with CDK4/6 inhibitors (CDK4/6i) combined with endocrine therapy (ET), has demonstrated substantial improvements in progression-free survival (PFS) and overall survival (OS) in HR+ metastatic breast cancer (mBC), leading to their clinical approval (Mittal et al., 2023, Cogliati et al., 2022). This pivotal achievement has generated greater interest in targeting other members of the CDK family, including cyclin-dependent kinase 2 (CDK2) (Tadesse et al., 2020).

The traditional cell-cycle progression model posits that mitogens stimulate the mitogen-activated protein kinase (MAPK) pathway, triggering the expression of D-type cyclins and activation of CDK4/6. Subsequently, CDK4/6-cyclin D complexes phosphorylate retinoblastoma tumor suppressor (Rb), releasing the adenoviral early region 2 binding factor (E2F) and promoting transcription of cyclins E1/E2 and A. CDK2 activation, by cyclins E1/2 and A, leads to hyperphosphorylation of Rb, establishing a positive feedback loop that ensures sustained expression of essential proteins for S phase, committing cells to the complete cell cycle (Arora et al., 2023)

With a few exceptions, CDK2 is generally not upregulated or amplified in cancer, but rather, its activity is altered through its binding partners or by alterations to post-translational modifications. For example, certain oncogenic molecular pathways, including the upregulation of cyclin E1/amplification of the G1/S-specific cyclin-E1 encoding gene (CCNE1) and overexpression of the basic helix–loop–helix transcription factor (MYC), converge on CDK2, thereby modifying its activity as a crucial node in cell cycle control (Fig. 1.) (Tadesse et al., 2020, Panagiotou et al., 2022, Gomatou et al., 2021). Notably, in tumours marked by MYC overexpression, as approximately 70% of triple-negative breast cancer (TNBC), CDK2 activity appears indeed to be vital for preventing senescence and permitting the immortalization of cancer cells (Agostinetto et al., 2021, Freeman-Cook et al., 2021). On the other hand, a broad range of aggressive cancers overexpress cyclin E and/or harbour CCNE1 gene amplifications (such as high-grade serous ovarian cancers), with preclinical evidence suggesting that the addiction to CDK2/cyclin E activity results in high sensitivity to CDK2 inhibition (Patel et al., 2023). Zi-Ming Zhao et al. brilliantly described that overexpression of CCNE1 was a significantly frequent event in TNBC patients (48,7% and 42.1%, for TCGA and METABRIC databases, respectively) and may confer resistance to chemotherapy, as it is associated with poor overall survival (Zhao et al., 2019). Despite the extensive heterogeneity in the mechanisms driving TNBC, encompassing various potentially targetable pathways, the lack of alternative targeted therapies beyond poly ADP ribose polymerase (PARP) inhibitors and the newly emerging antibody-drug conjugates (ADCs), is remarkable and challenging (Zhu et al., 2023) Notably, the use of CDK4/6i as monotherapy in metastatic TNBC (mTNBC) has yielded unsatisfactory results, likely due to the frequent Rb loss event, distinguishing it from the luminal subtype. Despite preclinical indications suggesting potential responsiveness in certain TNBC subtypes (such as Rb-proficient, Luminal Androgen Receptor [LAR] or Mesenchymal Stem-like [MSL]), recent findings reported that abemaciclib monotherapy lacked meaningful clinical activity in Rb or androgene receptor (AR)-positive mTNBC. Future trials exploring CDK4/6i monotherapy in TNBC may not be warranted, and the combined approach with chemotherapy appears controversial and not worth pursuing (Agostinetto et al., 2021, Jovanović (2) et al., 2023). Nonetheless, the emergence of CDK2 aberrant activation as a key oncogenic driver in mTNBC represents a potentially appealing avenue for a novel therapeutic approach.

Broadening the scope of analysis beyond patients with mTNBC, consistent preclinical discoveries have also unveiled CDK2 crucial role in driving endocrine and CDK4/6 resistance in HR+ breast cancer (Ma et al., 2023). Generally, mechanisms of resistance to CDK4/6i (independent of potential primary or secondary resistance to the ET, which may occur simultaneously) can be broadly categorized as aberrations affecting cell cycle progression or activation of other signaling pathways (Fig. 1.). The latter include the activation of the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) pathway, neurofibromin 1 (NF1) / mitogen-activated protein kinase (MAPK) cascade, as well as activating mutations or amplifications in other growth factor receptor genes like epidermal growth factor receptor 2 (HER2) and fibroblast growth factor receptor (FGFR) (Ma et al., 2023). Among resistance mechanisms driven by alterations in cell cycle regulators, including Rb loss-of-function mutations, increased expression of cyclin-dependent kinase 6 (CDK6) or 7 (CDK7) and Aurora kinase A, CDK2 aberrant activation hold a pivotal role, primarily steaming from the same CCNE1 and C-MYC alterations mentioned earlier. For instance, the upregulation of cyclin E1/E2, that interact with CDK2 and trigger its activation, has been shown in preclinical CDK4/6i resistant models and thus suggested as a main resistance mechanism. Interestingly, the overexpression of cyclin E may arise from the upregulation of other established signaling pathways associated with resistance to CDK4/6i and ET, including the PI3K/AKT/mTOR pathway (Gomatou et al., 2021). From 302 patients enrolled in the PALOMA-3 trial, palbociclib efficacy was indeed lower in patients with high cyclin E expression (Gomatou et al., 2021). Furthermore, Al-Qasem et al. have observed that high levels of CDK6, p-CDK2, and/or cyclin E1 were associated with adaptation and resistance to endocrine therapy (ET) and CDK4/6i in HR+ mBC. Hence, their combined expression was found to be an independent prognostic factor in these patients (Al-Qasem et al., 2022). Therefore, Freeman-Cook et al. uncovered that in both preclinical and clinical contexts, C-MYC overexpression results in resistance to endocrine therapy and CDK4/6 inhibitors. This resistance occurs by suppressing the cyclin dependent kinase inhibitor 1 A (CDKN1A) gene, which encodes p21, a well-known G1-CDK blockade protein. Consequently, this suppression releases CDK2–cyclin-E complexes, facilitating cell cycle progression (Tadesse et al., 2020).

While CDK2 gained prominence in cancer drug development during the 1990 s, the initial enthusiasm was tempered by the limited specificity and off-target effects of both first and second-generation CDKi. Nonetheless, the emerging preclinical evidence underscoring the pivotal role of abnormal CDK2 activity in breast cancer (as well as in other advanced solid tumors), as just highlighted, has revitalized the interest in developing novel more selective inhibitors and testing their clinical activity (Panagiotou et al., 2022). Therefore, considering emerging preliminary results, this review aims to fill a current gap in the scientific literature by providing an overview of clinical impact and ongoing clinical trials involving novel CDK2 inhibitors (CDK2i) in the context of mBC. On the same trajectory, it is worth mentioning that other selective inhibitors targeting different CDK, such as cyclin-dependent kinase 7 (CDK7) and 9 (CDK9), are currently under evaluation in early-phase clinical trials in the context of breast disease. In both instances, preliminary promising efficacy data along with a favourable toxicity profile, prompt further investigations and research (Patnaik (1) et al., 2023, Clack (1) et al., 2023, Mita (1) et al., 2023).

Moreover, informed by robust preclinical evidence, there has been a pursuit of combination therapeutic strategies to restore cell cycle control effectively, through the simultaneous inhibition of CDKs, particularly those crucially involved in the G1-S phase transition. In preclinical models resistant to ET and/or CDK4/6i, Al-Qasem et al. have indeed demonstrated that co-targeting of CDK2 and CDK4/6 in a triple combination with ET has a synergistic effect. This combination effectively inhibits cellular growth, induces cell cycle arrest, promotes apoptosis, and delays disease progression (Al-Qasem et al., 2022). Moreover, Arora et al. have demonstrated that acute response to selective CDK2 inhibition alone, despite an immediate reduction kinase activity, lead to cells rapid adaptation via a CDK2/4/6-Rb-E2F-dependent mechanism that circumvents CDK2 block and enables cell-cycle completion (Arora et al., 2023). Indeed the maintenance of Rb1 hyperphosphorylation by unblocked CDK4/6, results in active E2F transcription therefore sustaining cyclin A2 production, enabling a final paradoxical CDK2 reactivation that maintain the positive feedback loop with cell-cycle commitment. Interestingly, the novel CDK2i preferentially inhibits CDK2-cyclin E complex over CDK2-cyclin A2 and, despite the rebound on short timescale, long-term CDK2 inhibition is particularly effective in cyclin E-amplified patient-derived mouse cancers xenografts. Hence, at the light of these findings, Arora et al. serendipitously emphasize the usefulness of the current CDK2i for cancers that are heavily reliant on cyclin E. On the other hand, the co-inhibition of both CDK2 and CDK4/6 stops the rebound, undermining cell’s CDK2-increasing proliferative trajectory and breaking the positive feedback loop that reinforces Rb1 phosphorylation. Furthermore, on extended timescales of CDK2 inhibition, there is a shift from CDK4/6 to CDK1 reliance for cell proliferation. Given that CDK1 inhibition is expected to be poorly tolerated in people, co-targeting CDK2 and CDK4/6 represents a potential treatment strategy to ablate the early adaptive rebound and even stave off more problematic CDK1-mediated adaptation to CDK2 inhibition (Arora et al., 2023).

Taking into account these preclinical findings, the review expands its scope beyond CDK2i to encompass the new combined CDK2/4/6 inhibitors (CDK2/4/6i) and the novel selective CDK4 inhibitors (CDK4i). Thanks to the sparing of CDK6, CDK4i hold promise for a more favorable toxicity profile and permit higher dosages, making them indeed ideal partners for a combination strategy with CDK2i.

Therefore, encompassing the older broad-spectrum CDK inhibitors (pan-CDKi) ,the review also describes the impact of non-selective CDK2 targeting along with concurrent inhibition of multiple other CDKs. Specifically, it reviews those that include CDK2 in their spectrum of action, that have available clinical data in mBC setting.

Remarkably, other alternative therapeutic strategies target in a different way CDK2 activity, rather than inhibiting it, such as the novel MK-8776 drug, which inhibits checkpoint kinase 1 (CHK1). Ordinarily, CHK1 facilitates cell cycle arrest to aid DNA damage repair, and its inhibition in sensitive cells, particularly those with CDK2 activation, permits the accumulation of DNA breaks, ultimately leading to cytotoxicity (Ma et al., 2023). A comprehensive exploration of such alternative strategies is beyond the scope of this review.