The rationale for targeting CDK2, CDK4 and CDK6 has been solid since the discovery of these kinases almost 40 years ago. They sit at the bottom of the oncogenic signaling funnel, meaning that they are the final recipient and executioners of most of the oncogenic signals driving cancer cell proliferation. Cell surface receptors, like HER2 and EGFR, signal through cytoplasmic pathways, including the RAS/MAPK and AKT/PI3K families, to control cyclin D levels and CDK2/4/6 activities.
Robust EGFR or Her2 inhibitors, like Tarceva and Tratusamab, prevent this signaling cascade and turn CDK2/4/6 off, preventing proliferation. But the field has long recognized that another way to prevent the signaling from overactive receptors or constitutive cytoplasmic proteins is to “drug” CDK2/4/6 directly. These three kinases work together to control the G1 to S phase transition, and after their activation, the cell is committed to at least one round of DNA replication and proliferation, so they define the restriction point in whether a cell is committed to divide or alternatively remain quiescent.
Drawbacks of traditional small molecule CDK inhibitors
While these G1 phase CDKs are also required for proliferation in our rapidly proliferating tissues, such as our hematopoietic lineages, they are not active in most other solid tissues and their aberrant activation is a viable therapeutic target. This concept was validated with the approval of the CDK4/6 inhibitors, such as palbociclib (Ibrance®), abemaciclib (Verzinio®) and ribociclib (Kisqali®), for the treatment of advanced hormone receptor+ breast cancer.
In combination with endocrine modulation therapy, turning off CDK4/6 activity extends progression free survival and has revolutionized the treatment of metastatic breast cancer. Recently, abemaciclib and ribociclib have been approved for use in high-risk Stage 2, 3 breast cancer patients to delay the development of metastatic disease.
However, with time almost all patients become resistant to this form of therapy, because of the activation of the other G1 CDK, CDK2, demonstrating that targeting one half of the CDK see-saw is just not enough. This has led to renewed interest in drugging CDK2, and more than 10 companies are pushing new CDK2 inhibitors into clinical trials. Drugging CDK2 has historically been difficult, due to its homology with other essential CDKs such as CDK1.
Why p27 has therapeutic promise
The toxicity seen in more than 50 clinical trials with earlier forms of CDK2i precluded their approval. Recent results from ASCO and ESMO suggest that similar problems still plague newer CDK2i: where Pfizer’s CDK2 inhibitor PF-07104091 demonstrated neutropenia and fatigue. And one very clear data point that has emerged from clinical and preclinical studies is that targeting all three kinases with coordination will be necessary. Resistance seen in the presence of CDK4/6i is due to CDK2 activation, and CDK2 monotherapies have shown little to no response, due to CDK4/6 activation. In most cases this monotherapy approach has been discontinued in clinical trials. Only combined inhibition will fully shut down this proliferation hub.
Instead of generating more “me too” drugs, Concarlo has taken a different approach. They asked how does nature regulate CDK2/4/6 and shifted their focus away from the conserved CDKs and instead targeted the master regulator of these kinases: p27. p27 binds specifically to CDK2/4/6, avoiding the other members of the CDK kinome. Its endogenous function is to turn these kinases on and off, a function required during the cell’s natural cycling between proliferative and quiescent state.
One way to think about p27 is to compare it to a door controlling access to the active site of these kinases. When the key or the phosphorylation modification is present on p27 itself, the door swings open and CDK4/6 and CDK2 are on, but when the key is absent, the door slams shut, blocking the activity of these kinases. This dynamic opening and closing of the door is the way that CDK4/6 and CDK2 are regulated normally and p27 acts as the conduit between the left and right half of the CDK see-saw, linking the activation or inhibition of CDK4/6 and CDK2 together.
The transition between “ON” and “OFF” is controlled by non-receptor bound kinases in different tumor types: BRK (breast tumor-related kinase) controls p27 phosphorylation in breast epithelial cells, while Abelson tyrosine kinase (ABL1), Janus kinase 2 (JAK2), and SRC family can also phosphorylate p27 in other tumor types and the receptor FLT3 directly phosphorylates p27 in AML. This causes p27’s C-terminal tail to leave CDK active site, turning the kinases on; without phosphorylation, p27’s C-terminal tail remains in the CDK active site, keeping the kinases off.
When p27 is phosphorylated, it also becomes a target for Ubiquitin-mediated degradation reducing the amount of p27 and reinforcing the unconstrained activity of the G1 CDKs. To block p27 phosphorylation, inhibitors aimed at either these kinases or p27 should be effective. But targeting BRK or other kinases with small molecules was unsuccessful because of a lack of specificity and severe toxicity. Additionally, targeting upstream at the receptor or cytoplasmic level, still permits mutation and escape downstream of the blockage. Only targeting the CDKs, via p27 directly, solves this issue.
There are no approved therapeutics against p27 and Concarlo is aiming to change that with its p27 inhibitors. p27 is an intrinsically disordered protein, meaning that it has no structure when it is not bound to its CDK substrates, which has made it a difficult drug target. It is not a kinase and as such does not have a readily druggable pocket. Additionally, p27 is a tumor suppressor: Its loss has been shown to accelerate tumor development in mouse models, and its levels are frequently reduced in human tumors, and it may serve as a prognostic indicator of therapeutic response as well.
However, unlike other tumor suppressors, including the famous, still undrugged ones, RB or p53, p27 is never mutated, and the locus is never silenced or lost. Its levels are controlled instead by degradation, suggesting that a therapeutic which could decrease p27 degradation and stabilize it would be effective against this tumor suppressor. That is exactly what Concarlo’s p27 inhibitors do.
Concarlo has been investigating modalities to inhibit p27 directly, to inhibit CDK2/4/6 indirectly. ALT, a naturally occurring BRK variant lacking the kinase domain, is an endogenous p27 inhibitor that binds to p27 and prevents BRK-mediated p27 and CDK2/4/6 kinase activation. ALT’s specificity for CDK2/4/6 is determined by p27 itself as p27 binds primarily to these three CDKs.
Concarlo generated lpY.20, a smaller synthetic peptide variant of ALT packaged in a liposome. Published and ongoing studies demonstrates the feasibility and tolerability of inhibiting p27 in breast cancer, both in naïve and palbociclib resistant cell lines. lpY.20 significantly reduced tumor growth rates/volumes in BC xenograft models, with evidence of CDK2 inactivation, tumor necrosis and increased survival. Moreover, it was well-tolerated during 30-day repeat dosing in mice.
Thus, Concarlo has validated the p27 target. But recognized the potential issues using a liposomal-peptide drug, Concarlo identified small molecule p27 inhibitors, which act as molecular glues and increase the stability of the trimeric p27-CyclinD-CDK4 complex. They displace ALT binding from the trimeric p27-cyclin D-CDK4 complex and block BRK-dependent phosphorylation of p27. They potently inhibited cell proliferation in treatment-naïve and palbociclib-resistant breast cancer cells, and induced a G1-Phase cell cycle arrest, demonstrating specificity for the G1 Phase CDKs.
As predicted, total p27 levels were increased, demonstrating that when p27 phosphorylation was blocked, p27 degradation was reduced and more p27 was present to inhibit CDK2/4/6. We also demonstrated that p27i enhance palbociclib’s response in breast cancer cells, enabling a non-pharmacologically relevant dose of this FDA approved drug to be effective. These inhibitors demonstrate promising plasma stability and low off-target binding capability, and encouraging oral bioavailability.
Concarlo will continue to optimize and validate their p27 inhibitors, with the goal of bringing this approach to patients. Instead of “more of the same” approaches, Concarlo’s inhibitors and its focus on p27 will bring true innovation to the field. While many early-stage companies are trying to identify new targets in oncology, most of the really important targets have been identified and as a field we need to do the hard work to really drug those targets well and efficiently.
CDK2, CDK4 and CDK6 are three of the oldest and most important targets in oncology: they are well validated and regulate the activity of almost all oncologic signals. They sit downstream of all signaling pathways regulating growth and proliferation and are the effectors of almost all targeted therapies in the oncology space. It’s time for different approaches to turn these kinases off.
Stacy Blain, Ph.D., is a scientist, educator, and entrepreneur with nearly three decades of expertise in cell cycle and cancer biology. She earned her Bachelor’s in Molecular Biology from Princeton University, a Ph.D. from Columbia University, and completed a postdoctoral fellowship at Memorial Sloan-Kettering Cancer Center. Dr. Blain served as an Associate Professor at SUNY Downstate Medical Center for 20 years and in 2017, she co-founded Concarlo Therapeutics, where she is the CEO, leading the development of innovative small molecule therapeutics that target p27 to combat drug-resistant cancers. Committed to professional growth, Blain actively participates in leadership organizations such as Springboard Enterprises Healthtech cohort and Chief, a women’s leadership network.
Filed Under: Oncology
