Prolongation of survival, disease cure, and palliation of symptoms are all goals of systemic anticancer agents. As current systemic therapies rarely cause complete disease eradication in patients with solid tumors, the primary end point of most phase III randomized chemotherapy trials is prolongation of survival. In contrast, in the majority of early-stage clinical studies, response rates are used as the primary therapeutic end point based on the assumption that agents that induce tumor regression are those most likely to show a survival benefit in future randomized studies. In these trials, response is typically defined as a reduction in tumor size as determined by noninvasive computed tomography and magnetic resonance imaging.
Reliance on quantitative imaging techniques to identify potentially useful anticancer agents has several disadvantages. Changes in tumor size in response to agents that inhibit tumor growth often take several weeks or months to become apparent. As a consequence, patients are often continued on ineffective treatments for months before quantitative imaging can establish that a particular treatment has failed. Second, many agents, particularly those targeting pathways responsible for cell proliferation, do not induce tumor cell death and therefore would not be expected to induce significant tumor regression. These agents, however, may induce cell cycle arrest or senescence and therefore may prolong survival by delaying disease progression. Recent trials with erlotinib in lung cancer and imatinib and sunitinib in gastrointestinal stromal tumors suggest that disease stability and not tumor regression may account for much if not most of the survival benefit associated with the use of these agents (11
). Finally, it is difficult for clinical investigators to distinguish indolent tumor growth that is resistant to a study drug from the stabilized growth of tumors that have responded in a cytostatic manner to the agent. Therefore, in early-stage clinical trials of drugs such as the MEK inhibitor, stable disease is a category that may include both patients who have responded to the drug with disease stabilization and those with resistant tumors that have an indolent natural history.
In the current study, we show that PD0325901, a selective, allosteric inhibitor of MEK1 and MEK2 kinases, induces growth arrest, differentiation, and senescence of cancer cells with V600E BRAF mutation. Mutations in the kinase domain of BRAF have been identified in ~7% of all human cancers, most often in melanoma, papillary thyroid, and colon cancers (16
). The V600E mutation is by far the most commonly observed BRAF mutation in human tumors, accounting for >80% of cases (16
). In tumors with this mutation, cyclin D1 expression and G1
progression are under the control of MEK/ERK and inhibition of MEK causes a rapid down-regulation of cyclin D1 expression, induction of p27, hypophosphorylation of Rb, and G1
cell cycle arrest (4
). Within the class of cell lines harboring V600E BRAF mutations, apoptosis in response to MEK inhibition is variable with some cell lines, including the SKMEL-28 line, showing little if any apoptosis following MEK inhibition (4
). Consistent with this observation, complete tumor responses are rare in mice bearing SKMEL-28 xenografts treated with PD0325901. Rather, PD0325901 treatment of SKMEL-28–bearing mice induces modest tumor regression followed by disease stabilization. In this model, resistance to therapy was not observed even after 8 weeks of treatment. However, tumors remained viable and tumor growth resumed following discontinuation of therapy.
Analysis of tumor tissue from SKMEL-28 xenograft-bearing mice treated with PD0325901 showed little evidence of apoptosis at both early and late time points. However, a significant reduction in tumor cell proliferation was apparent at 1 week in the MEK inhibitor-treated mice. As inhibition of cell proliferation seemed to be the primary response of these tumors to MEK inhibition, we sought to determine whether [18
F]FLT PET could be used as an early predictive marker of response to MEK inhibition. The utility of [18
F]FLT PET as a marker of proliferation is based on the finding that the expression and thus activity of TK1 is regulated in a cell cycle–specific manner (1
). Specifically, TK1 is expressed primarily in S phase and is thus highly expressed in proliferating cells but is expressed at low levels in quiescent cells. TK1 catalyzes the phosphorylation of [18
F]FLT to [18
F]FLT-monophosphate, which, because of its negative charge, is trapped in cells (2
). Therefore, agents such as the MEK inhibitor PD0325901 that selectively arrest tumor cells in G1
would be predicted to cause a decrease in tumor [18
F]FLT uptake and tracer retention.
Consistent with this hypothesis, we observed that treatment of mice with established SKMEL-28 (V600E BRAF) tumors with the MEK inhibitor was associated with an early FLT response, which was maintained throughout the course of drug treatment. In contrast, the change in [18F]FDG uptake in response to MEK inhibition was modest and delayed in this model. Furthermore, no change in [18F]FLT or [18F]FDG uptake was observed in response to PD0325901 treatment in BT-474 xenografts, a PD0325901-resistant model. These data suggest that FLT may have advantages over [18F]FDG as an early predictor of response to MEK inhibition in cancer patients whose tumors are driven by mutant BRAF.
In summary, the data support the use of [18F]FLT PET as a method for imaging the biological consequence of MEK inhibition in vivo. The incorporation of [18F]FLT imaging into early-stage clinical trials of MEK pathway inhibitors should thus accelerate the development of such compounds by aiding in the identification of their optimal biological dose. The use of [18F]FLT PET imaging will also allow for a better assessment of the clinical utility of this class of agents in phase I and II clinical trials by helping to distinguish patients who have responded to MEK inhibition with disease stabilization from those with indolent or partially responsive disease.