2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
A new generation of more selective inhibitors of CDK4 and CDK6 has entered the clinic in combination therapies for patients with breast cancer. These agents also are being investigated in patients with KRAS-mutant malignancies.
Geoffrey I. Shapiro, MD, PhD
Although the roles of cyclin-dependent kinases (CDKs) were established more than 20 years ago, early anticancer inhibitors in the class resulted in high toxicity levels. Now, a new generation of more selective inhibitors of CDK4 and CDK6 has entered the clinic in combination therapies for patients with breast cancer.
These agents also are being investigated in patients with KRAS-mutant malignancies. Early-stage results have been promising, but hopes for an advance in KRAS-mutant non—small cell lung cancer (NSCLC) were dampened recently when the phase III JUNIPER study evaluating abemaciclib (Verzenio) in this population failed to meet its primary endpoint.
There is a pressing need for new therapeutic strategies that target KRAS, the Kirsten rat sarcoma viral oncogene homolog, one of the most frequently mutated genes in human cancers. “Despite some recent advances, mutant KRAS remains a very challenging target,” said Pier Paolo Scaglioni, MD, an associate professor at UT Southwestern Medical Center and a member of the Harold C. Simmons Comprehensive Cancer Center, both in Dallas, Texas. “There is a dearth of treatment options for tumors initiated by this gene.”1
CDKs and KRAS signaling are both players in oncogenic events (Figure).2 CDK4/6 phosphorylates the retinoblastoma tumor suppressor protein through interaction with D-type cyclins, leading to G1 to S phase cell-cycle progression. CDK4 and CDK6 activity is regulated by the INK4 family, and the INK4- CDK4/CDK6-cyclin D axis is frequently genetically or epigenetically disrupted in cancer, leading to increased kinase activity. Furthermore, increased CDK4 or CDK6 activity has led to suppressed senescence, contributing to the initiation and maintenance of transformed cells.
Mutations in KRAS are among the earliest and most important drivers of disease progression. When KRAS is mutated, several downstream pathways, such as the MEK/ ERK and PI3K/AKT signaling pathways, become constitutively activated.
Activating KRAS mutations contribute to approximately 20% of human tumors. Most KRAS mutations are point substitutions in codons 12 and 13 and have been shown to be negative predictors of response to antiEGFR antibodies. KRAS mutations are detected through quantitative polymerase chain reaction (PCR)-based methods. Frequently used methods include allele-specific PCR, real-time PCR with melt-curve analysis, and nucleic acid sequencing (pyrosequencing and dideoxy sequencing). Recent studies have also investigated the detection of circulating tumor cells in peripheral blood, concluding that KRAS mutations were identified in a similar percentage of blood samples compared with tumor tissue. Samples with at least 5% presence of the non—wild-type band have been considered KRAS mutation-positive.
KRAS mutations are found in about 30% of lung adenocarcinomas, corresponding with a worse prognosis compared with patients having KRAS wild-type tumors. Persistent activation of RAS signaling results from these mutations, contributing to uncontrolled growth and eventually, malignant transformation.
In KRAS-induced lung adenocarcinoma models, CDK4 activity is needed for tumor progression. CDK4 and RAS co-expression induces phosphorylation of the retinoblastoma protein, which leads to invasive neoplasms. Importantly, ablation of CDK4 leads to selective senescence of cells expressing KRAS, suggesting that CDK4 may be a worthy drug candidate for patients with KRAS-mutated non—small cell lung cancer (NSCLC).
Activating KRAS mutations have been reported in approximately 50% of colorectal cancers (CRCs), which lead to constitutive activation of the RAF/ MEK/ERK signaling pathway. These mutations are proven predictive biomarkers of resistance to antiEGFR therapy. Patients with KRAS mutations have fewer options for effective therapies, and currently, no therapies successfully exploit KRAS mutations to target malignant cells.
Since February 2015, the FDA has approved 3 CDK4/6 inhibitors for the treatment of hormone receptor (HR)-positive, HER2-negative advanced or metastatic breast cancer: palbociclib (Ibrance), ribociclib (Kisqali), and abemaciclib. The estrogen receptor signaling pathway is closely linked with cyclin D1-CDK4, which may contribute to the sensitivity of breast cancer cells to antiproliferative drugs, particularly when they are combined with endocrine therapy. These agents are also being explored in KRAS-mutant cancers (Table).Abemaciclib is the CDK4/6 furthest along in clinical development for KRAS-mutant cancers; it is being evaluated in NSCLC. The agent is a small-molecule inhibitor of CDK4/6, with greater selectivity for CDK4. Abemaciclib functions to inhibit cell cycle progression via phosphorylation prevention and functional inactivation of the retinoblastoma tumor suppressor protein.
Abemaciclib treatment results in fewer reports of neutropenia, which may be a result of enhanced selection of CDK4 versus CDK6 (CDK6 inhibition is correlated more strongly with neutropenia); therefore, abemaciclib can be given continuously. This administration schedule may more effectively drive cells into permanent G1 arrest and senescence, which may regress over time. Although abemaciclib has been classified as a selective CDK4/6 inhibitor, it has also been shown to inhibit other targets such as CDK9. This promiscuity may contribute to its clinical activity as a single agent.
“Abemaciclib, an oral CDK4/6 inhibitor, is a very different molecule from palbociclib, with distinct attributes that contribute to its discrete therapeutic effects, in particular, its single-agent activity,” noted Geoffrey I. Shapiro, MD, PhD, director of the Early Drug Development Center at the Dana-Farber Cancer Institute. “Abemaciclib also penetrates the central nervous system, whereas palbociclib does not, raising the possibility that it could be used to treat primary or metastatic brain tumors.”
Preclinical evidence has suggested that there is a lethal interaction between KRAS oncogenes and CDK4 inhibition in lung cells. In vitro study results demonstrate effective growth inhibition of cancer cells, and xenograft studies demonstrate efficacy for human cancers such as NSCLC. The rationale for evaluating abemaciclib in patients with KRAS-mutant NSCLC was established through a multicenter study that investigated the safety, pharmacodynamics, pharmacokinetics, and clinical activity of the drug in 225 patients with breast cancer, NSCLC, glioblastoma, CRC, and melanoma.4
There were 68 patients with NSCLC participating in the trial, including 29 who had tumors harboring KRAS mutations. The disease control rate was 55% among patients with mutations (16 of 29) compared with 39% among patients with wild-type tumors (13 of 33). Furthermore, stable disease lasting at least 24 weeks was achieved in 31% of patients with mutated KRAS and in 12% of patients with KRAS wild-type disease. These findings demonstrated that compared with KRAS wild-type tumors, tumors with KRAS mutation are more sensitive to abemaciclib.
An attempt to translate those early findings into later-stage success resulted in the JUNIPER trial, which was launched in 2014. The phase III study tested abemaciclib plus best supportive care versus erlotinib plus best supportive care in patients with previously treated stage IV KRAS-mutant NSCLC. Erlotinib was chosen as a control therapy because it is indicated for second- and third-line treatment of advanced NSCLC.
Patients were randomized to receive 200 mg of abemaciclib orally twice a day on a continuous dosing schedule, every 12 hours or 150 mg of erlotinib administered at its approved dose and schedule until disease progression or unacceptable toxicity. The primary endpoint was overall survival (OS); secondary endpoints included progression-free survival (PFS) and objective response rate (ORR).
However, the study failed to meet its primary endpoint of OS, according to Eli Lilly and Company, which is developing abemaciclib.5 The company, which did not provide details of the findings, noted that the OS rate among patients in the control arm was higher than expected based on historical data.
At the same time, an analysis of PFS and ORR demonstrated that abemaciclib was active as monotherapy, the company said, adding that full results would be presented at a medical meeting in 2018.
“While the outcome is unfortunate for patients with KRAS-mutated, advanced lung cancer, we remain encouraged by the antitumor activity observed with abemaciclib in this form of lung cancer where few clinical advances have been achieved,” Levi Garraway, MD, PhD, senior vice president for global development and medical affairs at Lilly Oncology, said in a statement.5
He noted that other studies with abemaciclib in NSCLC and other malignancies are ongoing. These include a 3-arm phase I trial evaluating the combination of abemaciclib with pembrolizumab in patients with stage IV NSCLC or HR-positive/ HER2-negative breast cancer; 1 arm is restricted to patients with KRAS-mutant, PD-L1—positive NSCLC.On March 31, the FDA granted palbociclib regular approval as a treatment for HR-positive, HER2- negative advanced or metastatic breast cancer in combination with an aromatase inhibitor as initial endocrine-based therapy in postmenopausal women. The decision cemented and expanded accelerated approvals the drug received in 2015 and 2016 in breast cancer settings.
In the realm of KRAS-mutant cancers, palbociclib has shown activity in CRC and NSCLC. In preclinical models of KRAS-mutant CRC, results of in vitro studies demonstrated inhibition of cell growth and patient-derived xenograft models revealed tumor regression when treated with the combination of palbociclib and the MEK inhibitor, trametinib.5
The results of a recent trial presented at the 2017 American Association for Cancer Research Annual Meeting demonstrate the tolerability of palbociclib combined with PD-0325901, a MEK inhibitor, with additional data revealing improved PFS in patients with NSCLC and KRAS-mutant tumors.6 Of 32 patients, 1 had a partial response and 7 exhibited stable disease for more than 6 months.
Separate ongoing phase I/II trials are investigating the combination of palbociclib with PD-00325901 or binimetinib, both MEK inhibitors, in patients with KRAS-mutant NSCLC or other solid tumors. The use of CDK4/6 inhibitors provides antitumor activity with an acceptable safety profile that may be developed further as a monotherapy and in combinations with other agents.
“Like all kinase inhibitors, CDK4/6 inhibitors will not be curative and cells will engineer a way around them,” stated Shapiro. “Defining the mechanisms of resistance will be critical for designing future strategies.”6
Related Content: